* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-11-17 13:41 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Hi,
As we discussed a couple of weeks ago, this is the third attempt at
creating a generic behaviour for slave capabilities retrieval so that
generic layers using dmaengine can actually rely on that.
That has been done mostly through two steps: by moving out the
sub-commands of the device_control callback, so that the dmaengine
core can then infer from that wether a sub-command is implemented, and
then by moving the slave properties, such as the supported buswidth,
to the structure dma_device itself.
Comments are as usual appreciated!
Thanks,
Maxime
Changes from v4:
- Rebased on top of current dma/next
- Converted AT91 XDMAC driver to the new API
- Updated the documentation to reflect our changes
Changes from v3:
- Removed the generic_slave_caps flag
- Merged the patch introducing a generic slave caps functions with
the one introducing the dma_device capabilities
Changes from v2:
- Reworked dma_chan_get/dma_chan_put in order to optionally use
device_alloc_chan_resources/device_free_chan_resources
- Modified a few client drivers that were still calling
device_control directly to use the dmaengine API instead
- Totally remove device_control and device_slave_config
- Move device_control BUG_ON removal earlier in the patch set to
avoid breaking the bisectability
- Converted rapidio tsi721 driver too.
- Minor cosmetics changes and fixes suggested by Laurent Pinchart
and Andy Shevchenko
- Fixed a few build warnings
- Collected the various Acked-by
- Rebased on top of 3.18-rc1
Changes from v1:
- Add a flag to trigger the generic slave caps mechanism
- Add a warning whenever this flag is not set, or when a
device_control callback is still defined
- Migrate all existing users to use the new callbacks, and the
generic slave capabilities
Ludovic Desroches (2):
dmaengine: at_xdmac: split device_control
dmaengine: at_xdmac: Declare slave capabilities for the generic code
Maxime Ripard (59):
serial: at91: Use dmaengine_slave_config API
crypto: ux500: Use dmaengine_terminate_all API
dmaengine: Make the destination abbreviation coherent
dmaengine: Rework dma_chan_get
dmaengine: Make channel allocation callbacks optional
dmaengine: Introduce a device_config callback
dmaengine: split out pause/resume operations from device_control
dmaengine: Add device_terminate_all callback
dmaengine: Remove the need to declare device_control
dmaengine: Create a generic dma_slave_caps callback
dmaengine: pl08x: Split device_control
dmaengine: hdmac: Split device_control
dmaengine: bcm2835: Split device_control
dmaengine: coh901318: Split device_control
dmaengine: cppi41: Split device_control
dmaengine: jz4740: Split device_control
dmaengine: dw: Split device_control
dmaengine: edma: Split device_control
dmaengine: ep93xx: Split device_control
dmaengine: fsl-edma: Split device_control
dmaengine: imx: Split device_control
dmaengine: imx-sdma: Split device_control
dmaengine: intel-mid-dma: Split device_control
dmaengine: ipu-idmac: Split device_control
dmaengine: k3: Split device_control
dmaengine: mmp-pdma: Split device_control
dmaengine: mmp-tdma: Split device_control
dmaengine: moxart: Split device_control
dmaengine: fsl-dma: Split device_control
dmaengine: mpc512x: Split device_control
dmaengine: mxs: Split device_control
dmaengine: nbpfaxi: Split device_control
dmaengine: omap: Split device_control
dmaengine: pl330: Split device_control
dmaengine: bam-dma: Split device_control
dmaengine: s3c24xx: Split device_control
dmaengine: sa11x0: Split device_control
dmaengine: sh: Split device_control
dmaengine: sirf: Split device_control
dmaengine: sun6i: Split device_control
dmaengine: d40: Split device_control
dmaengine: tegra20: Split device_control
dmaengine: xilinx: Split device_control
dmaengine: mv_xor: Remove device_control
dmaengine: pch-dma: Rename device_control
dmaengine: td: Rename device_control
dmaengine: txx9: Rename device_control
dmaengine: rapidio: tsi721: Rename device_control
dmaengine: bcm2835: Declare slave capabilities for the generic code
dmaengine: fsl-edma: Declare slave capabilities for the generic code
dmaengine: edma: Declare slave capabilities for the generic code
dmaengine: nbpfaxi: Declare slave capabilities for the generic code
dmaengine: omap: Declare slave capabilities for the generic code
dmaengine: pl330: Declare slave capabilities for the generic code
dmaengine: sirf: Declare slave capabilities for the generic code
dmaengine: sun6i: Declare slave capabilities for the generic code
dmaengine: Add a warning for drivers not using the generic slave caps
retrieval
dmaengine: Remove device_control and device_slave_caps
Documentation: dmaengine: Update the documentation
Documentation/dmaengine/provider.txt | 93 ++++++++--------
drivers/crypto/ux500/cryp/cryp_core.c | 4 +-
drivers/crypto/ux500/hash/hash_core.c | 2 +-
drivers/dma/amba-pl08x.c | 156 +++++++++++++++------------
drivers/dma/at_hdmac.c | 121 ++++++++++++---------
drivers/dma/at_xdmac.c | 127 ++++++++++++----------
drivers/dma/bcm2835-dma.c | 46 ++------
drivers/dma/coh901318.c | 137 +++++++++++------------
drivers/dma/cppi41.c | 30 +-----
drivers/dma/dma-jz4740.c | 20 +---
drivers/dma/dmaengine.c | 51 +++++----
drivers/dma/dw/core.c | 82 +++++++-------
drivers/dma/edma.c | 70 ++++--------
drivers/dma/ep93xx_dma.c | 41 ++-----
drivers/dma/fsl-edma.c | 123 ++++++++++-----------
drivers/dma/fsldma.c | 91 ++++++----------
drivers/dma/imx-dma.c | 103 +++++++++---------
drivers/dma/imx-sdma.c | 66 ++++++------
drivers/dma/intel_mid_dma.c | 25 ++---
drivers/dma/ipu/ipu_idmac.c | 96 +++++++++--------
drivers/dma/k3dma.c | 197 ++++++++++++++++++----------------
drivers/dma/mmp_pdma.c | 109 ++++++++++---------
drivers/dma/mmp_tdma.c | 82 +++++++-------
drivers/dma/moxart-dma.c | 25 +----
drivers/dma/mpc512x_dma.c | 111 +++++++++----------
drivers/dma/mv_xor.c | 9 --
drivers/dma/mxs-dma.c | 59 ++++------
drivers/dma/nbpfaxi.c | 110 +++++++++----------
drivers/dma/omap-dma.c | 69 ++++--------
drivers/dma/pch_dma.c | 8 +-
drivers/dma/pl330.c | 126 ++++++++++------------
drivers/dma/qcom_bam_dma.c | 85 +++++++--------
drivers/dma/s3c24xx-dma.c | 75 +++++++------
drivers/dma/sa11x0-dma.c | 158 ++++++++++++++-------------
drivers/dma/sh/shdma-base.c | 72 ++++++-------
drivers/dma/sirf-dma.c | 59 +++-------
drivers/dma/ste_dma40.c | 60 +++++------
drivers/dma/sun6i-dma.c | 159 ++++++++++++++-------------
drivers/dma/tegra20-apb-dma.c | 22 +---
drivers/dma/timb_dma.c | 8 +-
drivers/dma/txx9dmac.c | 9 +-
drivers/dma/xilinx/xilinx_vdma.c | 29 ++---
drivers/rapidio/devices/tsi721_dma.c | 8 +-
drivers/tty/serial/atmel_serial.c | 10 +-
include/linux/dmaengine.h | 121 ++++++++++++---------
sound/soc/soc-generic-dmaengine-pcm.c | 2 +-
46 files changed, 1520 insertions(+), 1746 deletions(-)
--
2.1.1
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-11-17 13:41 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: linux-arm-kernel
Hi,
As we discussed a couple of weeks ago, this is the third attempt at
creating a generic behaviour for slave capabilities retrieval so that
generic layers using dmaengine can actually rely on that.
That has been done mostly through two steps: by moving out the
sub-commands of the device_control callback, so that the dmaengine
core can then infer from that wether a sub-command is implemented, and
then by moving the slave properties, such as the supported buswidth,
to the structure dma_device itself.
Comments are as usual appreciated!
Thanks,
Maxime
Changes from v4:
- Rebased on top of current dma/next
- Converted AT91 XDMAC driver to the new API
- Updated the documentation to reflect our changes
Changes from v3:
- Removed the generic_slave_caps flag
- Merged the patch introducing a generic slave caps functions with
the one introducing the dma_device capabilities
Changes from v2:
- Reworked dma_chan_get/dma_chan_put in order to optionally use
device_alloc_chan_resources/device_free_chan_resources
- Modified a few client drivers that were still calling
device_control directly to use the dmaengine API instead
- Totally remove device_control and device_slave_config
- Move device_control BUG_ON removal earlier in the patch set to
avoid breaking the bisectability
- Converted rapidio tsi721 driver too.
- Minor cosmetics changes and fixes suggested by Laurent Pinchart
and Andy Shevchenko
- Fixed a few build warnings
- Collected the various Acked-by
- Rebased on top of 3.18-rc1
Changes from v1:
- Add a flag to trigger the generic slave caps mechanism
- Add a warning whenever this flag is not set, or when a
device_control callback is still defined
- Migrate all existing users to use the new callbacks, and the
generic slave capabilities
Ludovic Desroches (2):
dmaengine: at_xdmac: split device_control
dmaengine: at_xdmac: Declare slave capabilities for the generic code
Maxime Ripard (59):
serial: at91: Use dmaengine_slave_config API
crypto: ux500: Use dmaengine_terminate_all API
dmaengine: Make the destination abbreviation coherent
dmaengine: Rework dma_chan_get
dmaengine: Make channel allocation callbacks optional
dmaengine: Introduce a device_config callback
dmaengine: split out pause/resume operations from device_control
dmaengine: Add device_terminate_all callback
dmaengine: Remove the need to declare device_control
dmaengine: Create a generic dma_slave_caps callback
dmaengine: pl08x: Split device_control
dmaengine: hdmac: Split device_control
dmaengine: bcm2835: Split device_control
dmaengine: coh901318: Split device_control
dmaengine: cppi41: Split device_control
dmaengine: jz4740: Split device_control
dmaengine: dw: Split device_control
dmaengine: edma: Split device_control
dmaengine: ep93xx: Split device_control
dmaengine: fsl-edma: Split device_control
dmaengine: imx: Split device_control
dmaengine: imx-sdma: Split device_control
dmaengine: intel-mid-dma: Split device_control
dmaengine: ipu-idmac: Split device_control
dmaengine: k3: Split device_control
dmaengine: mmp-pdma: Split device_control
dmaengine: mmp-tdma: Split device_control
dmaengine: moxart: Split device_control
dmaengine: fsl-dma: Split device_control
dmaengine: mpc512x: Split device_control
dmaengine: mxs: Split device_control
dmaengine: nbpfaxi: Split device_control
dmaengine: omap: Split device_control
dmaengine: pl330: Split device_control
dmaengine: bam-dma: Split device_control
dmaengine: s3c24xx: Split device_control
dmaengine: sa11x0: Split device_control
dmaengine: sh: Split device_control
dmaengine: sirf: Split device_control
dmaengine: sun6i: Split device_control
dmaengine: d40: Split device_control
dmaengine: tegra20: Split device_control
dmaengine: xilinx: Split device_control
dmaengine: mv_xor: Remove device_control
dmaengine: pch-dma: Rename device_control
dmaengine: td: Rename device_control
dmaengine: txx9: Rename device_control
dmaengine: rapidio: tsi721: Rename device_control
dmaengine: bcm2835: Declare slave capabilities for the generic code
dmaengine: fsl-edma: Declare slave capabilities for the generic code
dmaengine: edma: Declare slave capabilities for the generic code
dmaengine: nbpfaxi: Declare slave capabilities for the generic code
dmaengine: omap: Declare slave capabilities for the generic code
dmaengine: pl330: Declare slave capabilities for the generic code
dmaengine: sirf: Declare slave capabilities for the generic code
dmaengine: sun6i: Declare slave capabilities for the generic code
dmaengine: Add a warning for drivers not using the generic slave caps
retrieval
dmaengine: Remove device_control and device_slave_caps
Documentation: dmaengine: Update the documentation
Documentation/dmaengine/provider.txt | 93 ++++++++--------
drivers/crypto/ux500/cryp/cryp_core.c | 4 +-
drivers/crypto/ux500/hash/hash_core.c | 2 +-
drivers/dma/amba-pl08x.c | 156 +++++++++++++++------------
drivers/dma/at_hdmac.c | 121 ++++++++++++---------
drivers/dma/at_xdmac.c | 127 ++++++++++++----------
drivers/dma/bcm2835-dma.c | 46 ++------
drivers/dma/coh901318.c | 137 +++++++++++------------
drivers/dma/cppi41.c | 30 +-----
drivers/dma/dma-jz4740.c | 20 +---
drivers/dma/dmaengine.c | 51 +++++----
drivers/dma/dw/core.c | 82 +++++++-------
drivers/dma/edma.c | 70 ++++--------
drivers/dma/ep93xx_dma.c | 41 ++-----
drivers/dma/fsl-edma.c | 123 ++++++++++-----------
drivers/dma/fsldma.c | 91 ++++++----------
drivers/dma/imx-dma.c | 103 +++++++++---------
drivers/dma/imx-sdma.c | 66 ++++++------
drivers/dma/intel_mid_dma.c | 25 ++---
drivers/dma/ipu/ipu_idmac.c | 96 +++++++++--------
drivers/dma/k3dma.c | 197 ++++++++++++++++++----------------
drivers/dma/mmp_pdma.c | 109 ++++++++++---------
drivers/dma/mmp_tdma.c | 82 +++++++-------
drivers/dma/moxart-dma.c | 25 +----
drivers/dma/mpc512x_dma.c | 111 +++++++++----------
drivers/dma/mv_xor.c | 9 --
drivers/dma/mxs-dma.c | 59 ++++------
drivers/dma/nbpfaxi.c | 110 +++++++++----------
drivers/dma/omap-dma.c | 69 ++++--------
drivers/dma/pch_dma.c | 8 +-
drivers/dma/pl330.c | 126 ++++++++++------------
drivers/dma/qcom_bam_dma.c | 85 +++++++--------
drivers/dma/s3c24xx-dma.c | 75 +++++++------
drivers/dma/sa11x0-dma.c | 158 ++++++++++++++-------------
drivers/dma/sh/shdma-base.c | 72 ++++++-------
drivers/dma/sirf-dma.c | 59 +++-------
drivers/dma/ste_dma40.c | 60 +++++------
drivers/dma/sun6i-dma.c | 159 ++++++++++++++-------------
drivers/dma/tegra20-apb-dma.c | 22 +---
drivers/dma/timb_dma.c | 8 +-
drivers/dma/txx9dmac.c | 9 +-
drivers/dma/xilinx/xilinx_vdma.c | 29 ++---
drivers/rapidio/devices/tsi721_dma.c | 8 +-
drivers/tty/serial/atmel_serial.c | 10 +-
include/linux/dmaengine.h | 121 ++++++++++++---------
sound/soc/soc-generic-dmaengine-pcm.c | 2 +-
46 files changed, 1520 insertions(+), 1746 deletions(-)
--
2.1.1
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 01/61] serial: at91: Use dmaengine_slave_config API
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:41 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard,
Greg Kroah-Hartman
We are removing the dmaengine_device_control API, that shouldn't even have been
exposed in the first place. Change the callers to use the proper API.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/tty/serial/atmel_serial.c | 10 ++++------
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c
index edde3eca055d..10e6b382d68f 100644
--- a/drivers/tty/serial/atmel_serial.c
+++ b/drivers/tty/serial/atmel_serial.c
@@ -862,9 +862,8 @@ static int atmel_prepare_tx_dma(struct uart_port *port)
config.dst_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
config.dst_addr = port->mapbase + ATMEL_US_THR;
- ret = dmaengine_device_control(atmel_port->chan_tx,
- DMA_SLAVE_CONFIG,
- (unsigned long)&config);
+ ret = dmaengine_slave_config(atmel_port->chan_tx,
+ &config);
if (ret) {
dev_err(port->dev, "DMA tx slave configuration failed\n");
goto chan_err;
@@ -1026,9 +1025,8 @@ static int atmel_prepare_rx_dma(struct uart_port *port)
config.src_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
config.src_addr = port->mapbase + ATMEL_US_RHR;
- ret = dmaengine_device_control(atmel_port->chan_rx,
- DMA_SLAVE_CONFIG,
- (unsigned long)&config);
+ ret = dmaengine_slave_config(atmel_port->chan_rx,
+ &config);
if (ret) {
dev_err(port->dev, "DMA rx slave configuration failed\n");
goto chan_err;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 01/61] serial: at91: Use dmaengine_slave_config API
@ 2014-11-17 13:41 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: linux-arm-kernel
We are removing the dmaengine_device_control API, that shouldn't even have been
exposed in the first place. Change the callers to use the proper API.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/tty/serial/atmel_serial.c | 10 ++++------
1 file changed, 4 insertions(+), 6 deletions(-)
diff --git a/drivers/tty/serial/atmel_serial.c b/drivers/tty/serial/atmel_serial.c
index edde3eca055d..10e6b382d68f 100644
--- a/drivers/tty/serial/atmel_serial.c
+++ b/drivers/tty/serial/atmel_serial.c
@@ -862,9 +862,8 @@ static int atmel_prepare_tx_dma(struct uart_port *port)
config.dst_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
config.dst_addr = port->mapbase + ATMEL_US_THR;
- ret = dmaengine_device_control(atmel_port->chan_tx,
- DMA_SLAVE_CONFIG,
- (unsigned long)&config);
+ ret = dmaengine_slave_config(atmel_port->chan_tx,
+ &config);
if (ret) {
dev_err(port->dev, "DMA tx slave configuration failed\n");
goto chan_err;
@@ -1026,9 +1025,8 @@ static int atmel_prepare_rx_dma(struct uart_port *port)
config.src_addr_width = DMA_SLAVE_BUSWIDTH_1_BYTE;
config.src_addr = port->mapbase + ATMEL_US_RHR;
- ret = dmaengine_device_control(atmel_port->chan_rx,
- DMA_SLAVE_CONFIG,
- (unsigned long)&config);
+ ret = dmaengine_slave_config(atmel_port->chan_rx,
+ &config);
if (ret) {
dev_err(port->dev, "DMA rx slave configuration failed\n");
goto chan_err;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 02/61] crypto: ux500: Use dmaengine_terminate_all API
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:41 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
We are removing the dmaengine_device_control API, that shouldn't even have been
exposed in the first place. Change the callers to use the proper API.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/crypto/ux500/cryp/cryp_core.c | 4 ++--
drivers/crypto/ux500/hash/hash_core.c | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/crypto/ux500/cryp/cryp_core.c b/drivers/crypto/ux500/cryp/cryp_core.c
index 92105f3dc8e0..75bd0f1e9a94 100644
--- a/drivers/crypto/ux500/cryp/cryp_core.c
+++ b/drivers/crypto/ux500/cryp/cryp_core.c
@@ -606,12 +606,12 @@ static void cryp_dma_done(struct cryp_ctx *ctx)
dev_dbg(ctx->device->dev, "[%s]: ", __func__);
chan = ctx->device->dma.chan_mem2cryp;
- dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
+ dmaengine_terminate_all(chan);
dma_unmap_sg(chan->device->dev, ctx->device->dma.sg_src,
ctx->device->dma.sg_src_len, DMA_TO_DEVICE);
chan = ctx->device->dma.chan_cryp2mem;
- dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
+ dmaengine_terminate_all(chan);
dma_unmap_sg(chan->device->dev, ctx->device->dma.sg_dst,
ctx->device->dma.sg_dst_len, DMA_FROM_DEVICE);
}
diff --git a/drivers/crypto/ux500/hash/hash_core.c b/drivers/crypto/ux500/hash/hash_core.c
index 1c73f4fbc252..48bdc5df7b3b 100644
--- a/drivers/crypto/ux500/hash/hash_core.c
+++ b/drivers/crypto/ux500/hash/hash_core.c
@@ -202,7 +202,7 @@ static void hash_dma_done(struct hash_ctx *ctx)
struct dma_chan *chan;
chan = ctx->device->dma.chan_mem2hash;
- dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
+ dmaengine_terminate_all(chan);
dma_unmap_sg(chan->device->dev, ctx->device->dma.sg,
ctx->device->dma.sg_len, DMA_TO_DEVICE);
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 02/61] crypto: ux500: Use dmaengine_terminate_all API
@ 2014-11-17 13:41 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: linux-arm-kernel
We are removing the dmaengine_device_control API, that shouldn't even have been
exposed in the first place. Change the callers to use the proper API.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/crypto/ux500/cryp/cryp_core.c | 4 ++--
drivers/crypto/ux500/hash/hash_core.c | 2 +-
2 files changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/crypto/ux500/cryp/cryp_core.c b/drivers/crypto/ux500/cryp/cryp_core.c
index 92105f3dc8e0..75bd0f1e9a94 100644
--- a/drivers/crypto/ux500/cryp/cryp_core.c
+++ b/drivers/crypto/ux500/cryp/cryp_core.c
@@ -606,12 +606,12 @@ static void cryp_dma_done(struct cryp_ctx *ctx)
dev_dbg(ctx->device->dev, "[%s]: ", __func__);
chan = ctx->device->dma.chan_mem2cryp;
- dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
+ dmaengine_terminate_all(chan);
dma_unmap_sg(chan->device->dev, ctx->device->dma.sg_src,
ctx->device->dma.sg_src_len, DMA_TO_DEVICE);
chan = ctx->device->dma.chan_cryp2mem;
- dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
+ dmaengine_terminate_all(chan);
dma_unmap_sg(chan->device->dev, ctx->device->dma.sg_dst,
ctx->device->dma.sg_dst_len, DMA_FROM_DEVICE);
}
diff --git a/drivers/crypto/ux500/hash/hash_core.c b/drivers/crypto/ux500/hash/hash_core.c
index 1c73f4fbc252..48bdc5df7b3b 100644
--- a/drivers/crypto/ux500/hash/hash_core.c
+++ b/drivers/crypto/ux500/hash/hash_core.c
@@ -202,7 +202,7 @@ static void hash_dma_done(struct hash_ctx *ctx)
struct dma_chan *chan;
chan = ctx->device->dma.chan_mem2hash;
- dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
+ dmaengine_terminate_all(chan);
dma_unmap_sg(chan->device->dev, ctx->device->dma.sg,
ctx->device->dma.sg_len, DMA_TO_DEVICE);
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 03/61] dmaengine: Make the destination abbreviation coherent
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:41 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
The dmaengine header abbreviates destination as at least two different strings.
Make a coherent use of a single one.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Mark Brown <broonie@kernel.org>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
---
drivers/dma/at_xdmac.c | 2 +-
drivers/dma/bcm2835-dma.c | 2 +-
drivers/dma/edma.c | 2 +-
drivers/dma/fsl-edma.c | 2 +-
drivers/dma/nbpfaxi.c | 2 +-
drivers/dma/omap-dma.c | 2 +-
drivers/dma/pl330.c | 2 +-
drivers/dma/sirf-dma.c | 2 +-
include/linux/dmaengine.h | 8 ++++----
sound/soc/soc-generic-dmaengine-pcm.c | 2 +-
10 files changed, 13 insertions(+), 13 deletions(-)
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index 89c43be20a38..bdc193cb82b7 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -1219,7 +1219,7 @@ static int at_xdmac_device_slave_caps(struct dma_chan *dchan,
{
caps->src_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/drivers/dma/bcm2835-dma.c b/drivers/dma/bcm2835-dma.c
index 918b7b3f766f..3feba6c25836 100644
--- a/drivers/dma/bcm2835-dma.c
+++ b/drivers/dma/bcm2835-dma.c
@@ -569,7 +569,7 @@ static int bcm2835_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
- caps->dstn_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ caps->dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = false;
caps->cmd_terminate = true;
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 123f578d6dd3..790054eef76c 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -998,7 +998,7 @@ static int edma_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = EDMA_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = EDMA_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = EDMA_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/drivers/dma/fsl-edma.c b/drivers/dma/fsl-edma.c
index 3c5711d5fe97..13df85a16438 100644
--- a/drivers/dma/fsl-edma.c
+++ b/drivers/dma/fsl-edma.c
@@ -781,7 +781,7 @@ static int fsl_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = FSL_EDMA_BUSWIDTHS;
- caps->dstn_addr_widths = FSL_EDMA_BUSWIDTHS;
+ caps->dst_addr_widths = FSL_EDMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/drivers/dma/nbpfaxi.c b/drivers/dma/nbpfaxi.c
index a7a36835a233..9d20de73ebae 100644
--- a/drivers/dma/nbpfaxi.c
+++ b/drivers/dma/nbpfaxi.c
@@ -1076,7 +1076,7 @@ static int nbpf_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = NBPF_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = NBPF_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = NBPF_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = false;
caps->cmd_terminate = true;
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index 6ea1aded7e74..a8aa5b6bb5e2 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c
@@ -1098,7 +1098,7 @@ static int omap_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = OMAP_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = OMAP_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = OMAP_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 38c961705836..3c9f715cb8d7 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2588,7 +2588,7 @@ static int pl330_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = PL330_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = PL330_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = PL330_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = false;
caps->cmd_terminate = true;
diff --git a/drivers/dma/sirf-dma.c b/drivers/dma/sirf-dma.c
index feb1e8ab8d7b..e0aa759139f3 100644
--- a/drivers/dma/sirf-dma.c
+++ b/drivers/dma/sirf-dma.c
@@ -652,7 +652,7 @@ static int sirfsoc_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 653a1fd07ae8..4d912575fec0 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -387,7 +387,7 @@ enum dma_residue_granularity {
/* struct dma_slave_caps - expose capabilities of a slave channel only
*
* @src_addr_widths: bit mask of src addr widths the channel supports
- * @dstn_addr_widths: bit mask of dstn addr widths the channel supports
+ * @dst_addr_widths: bit mask of dstn addr widths the channel supports
* @directions: bit mask of slave direction the channel supported
* since the enum dma_transfer_direction is not defined as bits for each
* type of direction, the dma controller should fill (1 << <TYPE>) and same
@@ -398,7 +398,7 @@ enum dma_residue_granularity {
*/
struct dma_slave_caps {
u32 src_addr_widths;
- u32 dstn_addr_widths;
+ u32 dst_addr_widths;
u32 directions;
bool cmd_pause;
bool cmd_terminate;
@@ -638,10 +638,10 @@ struct dma_device {
void (*device_free_chan_resources)(struct dma_chan *chan);
struct dma_async_tx_descriptor *(*device_prep_dma_memcpy)(
- struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
+ struct dma_chan *chan, dma_addr_t dst, dma_addr_t src,
size_t len, unsigned long flags);
struct dma_async_tx_descriptor *(*device_prep_dma_xor)(
- struct dma_chan *chan, dma_addr_t dest, dma_addr_t *src,
+ struct dma_chan *chan, dma_addr_t dst, dma_addr_t *src,
unsigned int src_cnt, size_t len, unsigned long flags);
struct dma_async_tx_descriptor *(*device_prep_dma_xor_val)(
struct dma_chan *chan, dma_addr_t *src, unsigned int src_cnt,
diff --git a/sound/soc/soc-generic-dmaengine-pcm.c b/sound/soc/soc-generic-dmaengine-pcm.c
index b329b84bc5af..851f7afcd5dc 100644
--- a/sound/soc/soc-generic-dmaengine-pcm.c
+++ b/sound/soc/soc-generic-dmaengine-pcm.c
@@ -151,7 +151,7 @@ static int dmaengine_pcm_set_runtime_hwparams(struct snd_pcm_substream *substrea
hw.info |= SNDRV_PCM_INFO_BATCH;
if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
- addr_widths = dma_caps.dstn_addr_widths;
+ addr_widths = dma_caps.dst_addr_widths;
else
addr_widths = dma_caps.src_addr_widths;
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 03/61] dmaengine: Make the destination abbreviation coherent
@ 2014-11-17 13:41 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: linux-arm-kernel
The dmaengine header abbreviates destination as at least two different strings.
Make a coherent use of a single one.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Mark Brown <broonie@kernel.org>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
---
drivers/dma/at_xdmac.c | 2 +-
drivers/dma/bcm2835-dma.c | 2 +-
drivers/dma/edma.c | 2 +-
drivers/dma/fsl-edma.c | 2 +-
drivers/dma/nbpfaxi.c | 2 +-
drivers/dma/omap-dma.c | 2 +-
drivers/dma/pl330.c | 2 +-
drivers/dma/sirf-dma.c | 2 +-
include/linux/dmaengine.h | 8 ++++----
sound/soc/soc-generic-dmaengine-pcm.c | 2 +-
10 files changed, 13 insertions(+), 13 deletions(-)
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index 89c43be20a38..bdc193cb82b7 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -1219,7 +1219,7 @@ static int at_xdmac_device_slave_caps(struct dma_chan *dchan,
{
caps->src_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/drivers/dma/bcm2835-dma.c b/drivers/dma/bcm2835-dma.c
index 918b7b3f766f..3feba6c25836 100644
--- a/drivers/dma/bcm2835-dma.c
+++ b/drivers/dma/bcm2835-dma.c
@@ -569,7 +569,7 @@ static int bcm2835_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
- caps->dstn_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ caps->dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = false;
caps->cmd_terminate = true;
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 123f578d6dd3..790054eef76c 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -998,7 +998,7 @@ static int edma_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = EDMA_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = EDMA_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = EDMA_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/drivers/dma/fsl-edma.c b/drivers/dma/fsl-edma.c
index 3c5711d5fe97..13df85a16438 100644
--- a/drivers/dma/fsl-edma.c
+++ b/drivers/dma/fsl-edma.c
@@ -781,7 +781,7 @@ static int fsl_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = FSL_EDMA_BUSWIDTHS;
- caps->dstn_addr_widths = FSL_EDMA_BUSWIDTHS;
+ caps->dst_addr_widths = FSL_EDMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/drivers/dma/nbpfaxi.c b/drivers/dma/nbpfaxi.c
index a7a36835a233..9d20de73ebae 100644
--- a/drivers/dma/nbpfaxi.c
+++ b/drivers/dma/nbpfaxi.c
@@ -1076,7 +1076,7 @@ static int nbpf_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = NBPF_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = NBPF_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = NBPF_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = false;
caps->cmd_terminate = true;
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index 6ea1aded7e74..a8aa5b6bb5e2 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c
@@ -1098,7 +1098,7 @@ static int omap_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = OMAP_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = OMAP_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = OMAP_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 38c961705836..3c9f715cb8d7 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2588,7 +2588,7 @@ static int pl330_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = PL330_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = PL330_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = PL330_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = false;
caps->cmd_terminate = true;
diff --git a/drivers/dma/sirf-dma.c b/drivers/dma/sirf-dma.c
index feb1e8ab8d7b..e0aa759139f3 100644
--- a/drivers/dma/sirf-dma.c
+++ b/drivers/dma/sirf-dma.c
@@ -652,7 +652,7 @@ static int sirfsoc_dma_device_slave_caps(struct dma_chan *dchan,
struct dma_slave_caps *caps)
{
caps->src_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
- caps->dstn_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
+ caps->dst_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
caps->cmd_pause = true;
caps->cmd_terminate = true;
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 653a1fd07ae8..4d912575fec0 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -387,7 +387,7 @@ enum dma_residue_granularity {
/* struct dma_slave_caps - expose capabilities of a slave channel only
*
* @src_addr_widths: bit mask of src addr widths the channel supports
- * @dstn_addr_widths: bit mask of dstn addr widths the channel supports
+ * @dst_addr_widths: bit mask of dstn addr widths the channel supports
* @directions: bit mask of slave direction the channel supported
* since the enum dma_transfer_direction is not defined as bits for each
* type of direction, the dma controller should fill (1 << <TYPE>) and same
@@ -398,7 +398,7 @@ enum dma_residue_granularity {
*/
struct dma_slave_caps {
u32 src_addr_widths;
- u32 dstn_addr_widths;
+ u32 dst_addr_widths;
u32 directions;
bool cmd_pause;
bool cmd_terminate;
@@ -638,10 +638,10 @@ struct dma_device {
void (*device_free_chan_resources)(struct dma_chan *chan);
struct dma_async_tx_descriptor *(*device_prep_dma_memcpy)(
- struct dma_chan *chan, dma_addr_t dest, dma_addr_t src,
+ struct dma_chan *chan, dma_addr_t dst, dma_addr_t src,
size_t len, unsigned long flags);
struct dma_async_tx_descriptor *(*device_prep_dma_xor)(
- struct dma_chan *chan, dma_addr_t dest, dma_addr_t *src,
+ struct dma_chan *chan, dma_addr_t dst, dma_addr_t *src,
unsigned int src_cnt, size_t len, unsigned long flags);
struct dma_async_tx_descriptor *(*device_prep_dma_xor_val)(
struct dma_chan *chan, dma_addr_t *src, unsigned int src_cnt,
diff --git a/sound/soc/soc-generic-dmaengine-pcm.c b/sound/soc/soc-generic-dmaengine-pcm.c
index b329b84bc5af..851f7afcd5dc 100644
--- a/sound/soc/soc-generic-dmaengine-pcm.c
+++ b/sound/soc/soc-generic-dmaengine-pcm.c
@@ -151,7 +151,7 @@ static int dmaengine_pcm_set_runtime_hwparams(struct snd_pcm_substream *substrea
hw.info |= SNDRV_PCM_INFO_BATCH;
if (substream->stream == SNDRV_PCM_STREAM_PLAYBACK)
- addr_widths = dma_caps.dstn_addr_widths;
+ addr_widths = dma_caps.dst_addr_widths;
else
addr_widths = dma_caps.src_addr_widths;
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 04/61] dmaengine: Rework dma_chan_get
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:41 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
dma_chan_get uses a rather interesting error handling and code path.
Change it to something more usual in the kernel.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/dmaengine.c | 36 +++++++++++++++++++-----------------
1 file changed, 19 insertions(+), 17 deletions(-)
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 24bfaf0b92ba..72f10e2f5c32 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -222,31 +222,33 @@ static void balance_ref_count(struct dma_chan *chan)
*/
static int dma_chan_get(struct dma_chan *chan)
{
- int err = -ENODEV;
struct module *owner = dma_chan_to_owner(chan);
+ int ret;
+ /* The channel is already in use, update client count */
if (chan->client_count) {
__module_get(owner);
- err = 0;
- } else if (try_module_get(owner))
- err = 0;
+ goto out;
+ }
- if (err == 0)
- chan->client_count++;
+ if (!try_module_get(owner))
+ return -ENODEV;
/* allocate upon first client reference */
- if (chan->client_count == 1 && err == 0) {
- int desc_cnt = chan->device->device_alloc_chan_resources(chan);
-
- if (desc_cnt < 0) {
- err = desc_cnt;
- chan->client_count = 0;
- module_put(owner);
- } else if (!dma_has_cap(DMA_PRIVATE, chan->device->cap_mask))
- balance_ref_count(chan);
- }
+ ret = chan->device->device_alloc_chan_resources(chan);
+ if (ret < 0)
+ goto err_out;
- return err;
+ if (!dma_has_cap(DMA_PRIVATE, chan->device->cap_mask))
+ balance_ref_count(chan);
+
+out:
+ chan->client_count++;
+ return 0;
+
+err_out:
+ module_put(owner);
+ return ret;
}
/**
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 04/61] dmaengine: Rework dma_chan_get
@ 2014-11-17 13:41 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: linux-arm-kernel
dma_chan_get uses a rather interesting error handling and code path.
Change it to something more usual in the kernel.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/dmaengine.c | 36 +++++++++++++++++++-----------------
1 file changed, 19 insertions(+), 17 deletions(-)
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 24bfaf0b92ba..72f10e2f5c32 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -222,31 +222,33 @@ static void balance_ref_count(struct dma_chan *chan)
*/
static int dma_chan_get(struct dma_chan *chan)
{
- int err = -ENODEV;
struct module *owner = dma_chan_to_owner(chan);
+ int ret;
+ /* The channel is already in use, update client count */
if (chan->client_count) {
__module_get(owner);
- err = 0;
- } else if (try_module_get(owner))
- err = 0;
+ goto out;
+ }
- if (err == 0)
- chan->client_count++;
+ if (!try_module_get(owner))
+ return -ENODEV;
/* allocate upon first client reference */
- if (chan->client_count == 1 && err == 0) {
- int desc_cnt = chan->device->device_alloc_chan_resources(chan);
-
- if (desc_cnt < 0) {
- err = desc_cnt;
- chan->client_count = 0;
- module_put(owner);
- } else if (!dma_has_cap(DMA_PRIVATE, chan->device->cap_mask))
- balance_ref_count(chan);
- }
+ ret = chan->device->device_alloc_chan_resources(chan);
+ if (ret < 0)
+ goto err_out;
- return err;
+ if (!dma_has_cap(DMA_PRIVATE, chan->device->cap_mask))
+ balance_ref_count(chan);
+
+out:
+ chan->client_count++;
+ return 0;
+
+err_out:
+ module_put(owner);
+ return ret;
}
/**
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 05/61] dmaengine: Make channel allocation callbacks optional
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:41 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Nowadays, some drivers don't have anything in there channel allocation
callbacks anymore.
Remove the BUG_ON if those callbacks aren't implemented, in order to allow
drivers to not implement them.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/dmaengine.c | 18 +++++++++++-------
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 72f10e2f5c32..9b2cd74b8f2e 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -235,9 +235,11 @@ static int dma_chan_get(struct dma_chan *chan)
return -ENODEV;
/* allocate upon first client reference */
- ret = chan->device->device_alloc_chan_resources(chan);
- if (ret < 0)
- goto err_out;
+ if (chan->device->device_alloc_chan_resources) {
+ ret = chan->device->device_alloc_chan_resources(chan);
+ if (ret < 0)
+ goto err_out;
+ }
if (!dma_has_cap(DMA_PRIVATE, chan->device->cap_mask))
balance_ref_count(chan);
@@ -259,11 +261,15 @@ err_out:
*/
static void dma_chan_put(struct dma_chan *chan)
{
+ /* This channel is not in use, bail out */
if (!chan->client_count)
- return; /* this channel failed alloc_chan_resources */
+ return;
+
chan->client_count--;
module_put(dma_chan_to_owner(chan));
- if (chan->client_count == 0)
+
+ /* This channel is not in use anymore, free it */
+ if (!chan->client_count && chan->device->device_free_chan_resources)
chan->device->device_free_chan_resources(chan);
}
@@ -819,8 +825,6 @@ int dma_async_device_register(struct dma_device *device)
BUG_ON(dma_has_cap(DMA_INTERLEAVE, device->cap_mask) &&
!device->device_prep_interleaved_dma);
- BUG_ON(!device->device_alloc_chan_resources);
- BUG_ON(!device->device_free_chan_resources);
BUG_ON(!device->device_tx_status);
BUG_ON(!device->device_issue_pending);
BUG_ON(!device->dev);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 05/61] dmaengine: Make channel allocation callbacks optional
@ 2014-11-17 13:41 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:41 UTC (permalink / raw)
To: linux-arm-kernel
Nowadays, some drivers don't have anything in there channel allocation
callbacks anymore.
Remove the BUG_ON if those callbacks aren't implemented, in order to allow
drivers to not implement them.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/dmaengine.c | 18 +++++++++++-------
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 72f10e2f5c32..9b2cd74b8f2e 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -235,9 +235,11 @@ static int dma_chan_get(struct dma_chan *chan)
return -ENODEV;
/* allocate upon first client reference */
- ret = chan->device->device_alloc_chan_resources(chan);
- if (ret < 0)
- goto err_out;
+ if (chan->device->device_alloc_chan_resources) {
+ ret = chan->device->device_alloc_chan_resources(chan);
+ if (ret < 0)
+ goto err_out;
+ }
if (!dma_has_cap(DMA_PRIVATE, chan->device->cap_mask))
balance_ref_count(chan);
@@ -259,11 +261,15 @@ err_out:
*/
static void dma_chan_put(struct dma_chan *chan)
{
+ /* This channel is not in use, bail out */
if (!chan->client_count)
- return; /* this channel failed alloc_chan_resources */
+ return;
+
chan->client_count--;
module_put(dma_chan_to_owner(chan));
- if (chan->client_count == 0)
+
+ /* This channel is not in use anymore, free it */
+ if (!chan->client_count && chan->device->device_free_chan_resources)
chan->device->device_free_chan_resources(chan);
}
@@ -819,8 +825,6 @@ int dma_async_device_register(struct dma_device *device)
BUG_ON(dma_has_cap(DMA_INTERLEAVE, device->cap_mask) &&
!device->device_prep_interleaved_dma);
- BUG_ON(!device->device_alloc_chan_resources);
- BUG_ON(!device->device_free_chan_resources);
BUG_ON(!device->device_tx_status);
BUG_ON(!device->device_issue_pending);
BUG_ON(!device->dev);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 06/61] dmaengine: Introduce a device_config callback
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
The fact that the channel configuration is done in device_control is rather
misleading, since it's not really advertised as such, plus, the fact that the
framework exposes a function of its own makes it not really intuitive, while
we're losing the type checking whenever we pass that unsigned long argument.
Add a device_config callback to dma_device, with a fallback on the old
behaviour for now for existing drivers to opt in.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 4d912575fec0..f797cb81355f 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -607,6 +607,8 @@ struct dma_tx_state {
* The function takes a buffer of size buf_len. The callback function will
* be called after period_len bytes have been transferred.
* @device_prep_interleaved_dma: Transfer expression in a generic way.
+ * @device_config: Pushes a new configuration to a channel, return 0 or an error
+ * code
* @device_control: manipulate all pending operations on a channel, returns
* zero or error code
* @device_tx_status: poll for transaction completion, the optional
@@ -673,6 +675,9 @@ struct dma_device {
struct dma_async_tx_descriptor *(*device_prep_interleaved_dma)(
struct dma_chan *chan, struct dma_interleaved_template *xt,
unsigned long flags);
+
+ int (*device_config)(struct dma_chan *chan,
+ struct dma_slave_config *config);
int (*device_control)(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
unsigned long arg);
@@ -696,6 +701,9 @@ static inline int dmaengine_device_control(struct dma_chan *chan,
static inline int dmaengine_slave_config(struct dma_chan *chan,
struct dma_slave_config *config)
{
+ if (chan->device->device_config)
+ return chan->device->device_config(chan, config);
+
return dmaengine_device_control(chan, DMA_SLAVE_CONFIG,
(unsigned long)config);
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 06/61] dmaengine: Introduce a device_config callback
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
The fact that the channel configuration is done in device_control is rather
misleading, since it's not really advertised as such, plus, the fact that the
framework exposes a function of its own makes it not really intuitive, while
we're losing the type checking whenever we pass that unsigned long argument.
Add a device_config callback to dma_device, with a fallback on the old
behaviour for now for existing drivers to opt in.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 8 ++++++++
1 file changed, 8 insertions(+)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 4d912575fec0..f797cb81355f 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -607,6 +607,8 @@ struct dma_tx_state {
* The function takes a buffer of size buf_len. The callback function will
* be called after period_len bytes have been transferred.
* @device_prep_interleaved_dma: Transfer expression in a generic way.
+ * @device_config: Pushes a new configuration to a channel, return 0 or an error
+ * code
* @device_control: manipulate all pending operations on a channel, returns
* zero or error code
* @device_tx_status: poll for transaction completion, the optional
@@ -673,6 +675,9 @@ struct dma_device {
struct dma_async_tx_descriptor *(*device_prep_interleaved_dma)(
struct dma_chan *chan, struct dma_interleaved_template *xt,
unsigned long flags);
+
+ int (*device_config)(struct dma_chan *chan,
+ struct dma_slave_config *config);
int (*device_control)(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
unsigned long arg);
@@ -696,6 +701,9 @@ static inline int dmaengine_device_control(struct dma_chan *chan,
static inline int dmaengine_slave_config(struct dma_chan *chan,
struct dma_slave_config *config)
{
+ if (chan->device->device_config)
+ return chan->device->device_config(chan, config);
+
return dmaengine_device_control(chan, DMA_SLAVE_CONFIG,
(unsigned long)config);
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 07/61] dmaengine: split out pause/resume operations from device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split out the pause and resume operations to callbacks of their own. In order
to preserve some backwark compatibility, the dmaengine_pause/dmaengine_resume
are still falling back on dmaengine_device_control.
Eventually, that will allow to get the device capabilities in a generic way,
removing the need to implement device_slave_caps.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index f797cb81355f..4583c55f3355 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -611,6 +611,10 @@ struct dma_tx_state {
* code
* @device_control: manipulate all pending operations on a channel, returns
* zero or error code
+ * @device_pause: Pauses any transfer happening on a channel. Returns
+ * 0 or an error code
+ * @device_resume: Resumes any transfer on a channel previously
+ * paused. Returns 0 or an error code
* @device_tx_status: poll for transaction completion, the optional
* txstate parameter can be supplied with a pointer to get a
* struct with auxiliary transfer status information, otherwise the call
@@ -680,6 +684,8 @@ struct dma_device {
struct dma_slave_config *config);
int (*device_control)(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
unsigned long arg);
+ int (*device_pause)(struct dma_chan *chan);
+ int (*device_resume)(struct dma_chan *chan);
enum dma_status (*device_tx_status)(struct dma_chan *chan,
dma_cookie_t cookie,
@@ -794,11 +800,17 @@ static inline int dmaengine_terminate_all(struct dma_chan *chan)
static inline int dmaengine_pause(struct dma_chan *chan)
{
+ if (chan->device->device_pause)
+ return chan->device->device_pause(chan);
+
return dmaengine_device_control(chan, DMA_PAUSE, 0);
}
static inline int dmaengine_resume(struct dma_chan *chan)
{
+ if (chan->device->device_resume)
+ return chan->device->device_resume(chan);
+
return dmaengine_device_control(chan, DMA_RESUME, 0);
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 07/61] dmaengine: split out pause/resume operations from device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split out the pause and resume operations to callbacks of their own. In order
to preserve some backwark compatibility, the dmaengine_pause/dmaengine_resume
are still falling back on dmaengine_device_control.
Eventually, that will allow to get the device capabilities in a generic way,
removing the need to implement device_slave_caps.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index f797cb81355f..4583c55f3355 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -611,6 +611,10 @@ struct dma_tx_state {
* code
* @device_control: manipulate all pending operations on a channel, returns
* zero or error code
+ * @device_pause: Pauses any transfer happening on a channel. Returns
+ * 0 or an error code
+ * @device_resume: Resumes any transfer on a channel previously
+ * paused. Returns 0 or an error code
* @device_tx_status: poll for transaction completion, the optional
* txstate parameter can be supplied with a pointer to get a
* struct with auxiliary transfer status information, otherwise the call
@@ -680,6 +684,8 @@ struct dma_device {
struct dma_slave_config *config);
int (*device_control)(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
unsigned long arg);
+ int (*device_pause)(struct dma_chan *chan);
+ int (*device_resume)(struct dma_chan *chan);
enum dma_status (*device_tx_status)(struct dma_chan *chan,
dma_cookie_t cookie,
@@ -794,11 +800,17 @@ static inline int dmaengine_terminate_all(struct dma_chan *chan)
static inline int dmaengine_pause(struct dma_chan *chan)
{
+ if (chan->device->device_pause)
+ return chan->device->device_pause(chan);
+
return dmaengine_device_control(chan, DMA_PAUSE, 0);
}
static inline int dmaengine_resume(struct dma_chan *chan)
{
+ if (chan->device->device_resume)
+ return chan->device->device_resume(chan);
+
return dmaengine_device_control(chan, DMA_RESUME, 0);
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 08/61] dmaengine: Add device_terminate_all callback
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split out the terminate_all command from device_control to a dma_device
callback. In order to preserve backward capability, still rely on
device_control if no such callback has been implemented.
Eventually, this will allow to create a generic dma_slave_caps callback.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 4583c55f3355..b0efc805937a 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -615,6 +615,8 @@ struct dma_tx_state {
* 0 or an error code
* @device_resume: Resumes any transfer on a channel previously
* paused. Returns 0 or an error code
+ * @device_terminate_all: Aborts all transfers on a channel. Returns 0
+ * or an error code
* @device_tx_status: poll for transaction completion, the optional
* txstate parameter can be supplied with a pointer to get a
* struct with auxiliary transfer status information, otherwise the call
@@ -686,6 +688,7 @@ struct dma_device {
unsigned long arg);
int (*device_pause)(struct dma_chan *chan);
int (*device_resume)(struct dma_chan *chan);
+ int (*device_terminate_all)(struct dma_chan *chan);
enum dma_status (*device_tx_status)(struct dma_chan *chan,
dma_cookie_t cookie,
@@ -795,6 +798,9 @@ static inline int dma_get_slave_caps(struct dma_chan *chan, struct dma_slave_cap
static inline int dmaengine_terminate_all(struct dma_chan *chan)
{
+ if (chan->device->device_terminate_all)
+ return chan->device->device_terminate_all(chan);
+
return dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 08/61] dmaengine: Add device_terminate_all callback
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split out the terminate_all command from device_control to a dma_device
callback. In order to preserve backward capability, still rely on
device_control if no such callback has been implemented.
Eventually, this will allow to create a generic dma_slave_caps callback.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 6 ++++++
1 file changed, 6 insertions(+)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index 4583c55f3355..b0efc805937a 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -615,6 +615,8 @@ struct dma_tx_state {
* 0 or an error code
* @device_resume: Resumes any transfer on a channel previously
* paused. Returns 0 or an error code
+ * @device_terminate_all: Aborts all transfers on a channel. Returns 0
+ * or an error code
* @device_tx_status: poll for transaction completion, the optional
* txstate parameter can be supplied with a pointer to get a
* struct with auxiliary transfer status information, otherwise the call
@@ -686,6 +688,7 @@ struct dma_device {
unsigned long arg);
int (*device_pause)(struct dma_chan *chan);
int (*device_resume)(struct dma_chan *chan);
+ int (*device_terminate_all)(struct dma_chan *chan);
enum dma_status (*device_tx_status)(struct dma_chan *chan,
dma_cookie_t cookie,
@@ -795,6 +798,9 @@ static inline int dma_get_slave_caps(struct dma_chan *chan, struct dma_slave_cap
static inline int dmaengine_terminate_all(struct dma_chan *chan)
{
+ if (chan->device->device_terminate_all)
+ return chan->device->device_terminate_all(chan);
+
return dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 09/61] dmaengine: Remove the need to declare device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
In order to migrate the drivers without triggering a BUG_ON for the converted
drivers, which would cause bisectability issues, we need to remove that check
before removing the device_control function entirely.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/dmaengine.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 9b2cd74b8f2e..98e9431f85ec 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -820,8 +820,6 @@ int dma_async_device_register(struct dma_device *device)
!device->device_prep_dma_sg);
BUG_ON(dma_has_cap(DMA_CYCLIC, device->cap_mask) &&
!device->device_prep_dma_cyclic);
- BUG_ON(dma_has_cap(DMA_SLAVE, device->cap_mask) &&
- !device->device_control);
BUG_ON(dma_has_cap(DMA_INTERLEAVE, device->cap_mask) &&
!device->device_prep_interleaved_dma);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 09/61] dmaengine: Remove the need to declare device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
In order to migrate the drivers without triggering a BUG_ON for the converted
drivers, which would cause bisectability issues, we need to remove that check
before removing the device_control function entirely.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/dmaengine.c | 2 --
1 file changed, 2 deletions(-)
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 9b2cd74b8f2e..98e9431f85ec 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -820,8 +820,6 @@ int dma_async_device_register(struct dma_device *device)
!device->device_prep_dma_sg);
BUG_ON(dma_has_cap(DMA_CYCLIC, device->cap_mask) &&
!device->device_prep_dma_cyclic);
- BUG_ON(dma_has_cap(DMA_SLAVE, device->cap_mask) &&
- !device->device_control);
BUG_ON(dma_has_cap(DMA_INTERLEAVE, device->cap_mask) &&
!device->device_prep_interleaved_dma);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 10/61] dmaengine: Create a generic dma_slave_caps callback
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
dma_slave_caps is very important to the generic layers that might interact with
dmaengine, such as ASoC. Unfortunately, it has been added as yet another
dma_device callback, and most of the existing drivers haven't implemented it,
reducing its reliability.
Introduce a generic behaviour to implement this, that rely on both the split of
device_control to derive which functions are supported and on new variables to
be set in the dma_device structure.
These variables holds what used to be the capabilities, that were set
per-channel. However, this proved to be a bit overkill, since every driver
filling these so far were hardcoding it, disregarding which channel was
actually given.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 41 +++++++++++++++++++++++++++++++++++++----
1 file changed, 37 insertions(+), 4 deletions(-)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index b0efc805937a..be9e60aa8f5e 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -593,6 +593,14 @@ struct dma_tx_state {
* @fill_align: alignment shift for memset operations
* @dev_id: unique device ID
* @dev: struct device reference for dma mapping api
+ * @src_addr_widths: bit mask of src addr widths the device supports
+ * @dst_addr_widths: bit mask of dst addr widths the device supports
+ * @directions: bit mask of slave direction the device supports since
+ * the enum dma_transfer_direction is not defined as bits for
+ * each type of direction, the dma controller should fill (1 <<
+ * <TYPE>) and same should be checked by controller as well
+ * @residue_granularity: granularity of the transfer residue reported
+ * by tx_status
* @device_alloc_chan_resources: allocate resources and return the
* number of allocated descriptors
* @device_free_chan_resources: release DMA channel's resources
@@ -642,6 +650,11 @@ struct dma_device {
int dev_id;
struct device *dev;
+ u32 src_addr_widths;
+ u32 dst_addr_widths;
+ u32 directions;
+ enum dma_residue_granularity residue_granularity;
+
int (*device_alloc_chan_resources)(struct dma_chan *chan);
void (*device_free_chan_resources)(struct dma_chan *chan);
@@ -783,17 +796,37 @@ static inline struct dma_async_tx_descriptor *dmaengine_prep_dma_sg(
static inline int dma_get_slave_caps(struct dma_chan *chan, struct dma_slave_caps *caps)
{
+ struct dma_device *device;
+
if (!chan || !caps)
return -EINVAL;
+ device = chan->device;
+
/* check if the channel supports slave transactions */
- if (!test_bit(DMA_SLAVE, chan->device->cap_mask.bits))
+ if (!test_bit(DMA_SLAVE, device->cap_mask.bits))
+ return -ENXIO;
+
+ if (device->device_slave_caps)
+ return device->device_slave_caps(chan, caps);
+
+ /*
+ * Check whether it reports it uses the generic slave
+ * capabilities, if not, that means it doesn't support any
+ * kind of slave capabilities reporting.
+ */
+ if (!device->directions)
return -ENXIO;
- if (chan->device->device_slave_caps)
- return chan->device->device_slave_caps(chan, caps);
+ caps->src_addr_widths = device->src_addr_widths;
+ caps->dst_addr_widths = device->dst_addr_widths;
+ caps->directions = device->directions;
+ caps->residue_granularity = device->residue_granularity;
+
+ caps->cmd_pause = !!device->device_pause;
+ caps->cmd_terminate = !!device->device_terminate_all;
- return -ENXIO;
+ return 0;
}
static inline int dmaengine_terminate_all(struct dma_chan *chan)
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 10/61] dmaengine: Create a generic dma_slave_caps callback
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
dma_slave_caps is very important to the generic layers that might interact with
dmaengine, such as ASoC. Unfortunately, it has been added as yet another
dma_device callback, and most of the existing drivers haven't implemented it,
reducing its reliability.
Introduce a generic behaviour to implement this, that rely on both the split of
device_control to derive which functions are supported and on new variables to
be set in the dma_device structure.
These variables holds what used to be the capabilities, that were set
per-channel. However, this proved to be a bit overkill, since every driver
filling these so far were hardcoding it, disregarding which channel was
actually given.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 41 +++++++++++++++++++++++++++++++++++++----
1 file changed, 37 insertions(+), 4 deletions(-)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index b0efc805937a..be9e60aa8f5e 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -593,6 +593,14 @@ struct dma_tx_state {
* @fill_align: alignment shift for memset operations
* @dev_id: unique device ID
* @dev: struct device reference for dma mapping api
+ * @src_addr_widths: bit mask of src addr widths the device supports
+ * @dst_addr_widths: bit mask of dst addr widths the device supports
+ * @directions: bit mask of slave direction the device supports since
+ * the enum dma_transfer_direction is not defined as bits for
+ * each type of direction, the dma controller should fill (1 <<
+ * <TYPE>) and same should be checked by controller as well
+ * @residue_granularity: granularity of the transfer residue reported
+ * by tx_status
* @device_alloc_chan_resources: allocate resources and return the
* number of allocated descriptors
* @device_free_chan_resources: release DMA channel's resources
@@ -642,6 +650,11 @@ struct dma_device {
int dev_id;
struct device *dev;
+ u32 src_addr_widths;
+ u32 dst_addr_widths;
+ u32 directions;
+ enum dma_residue_granularity residue_granularity;
+
int (*device_alloc_chan_resources)(struct dma_chan *chan);
void (*device_free_chan_resources)(struct dma_chan *chan);
@@ -783,17 +796,37 @@ static inline struct dma_async_tx_descriptor *dmaengine_prep_dma_sg(
static inline int dma_get_slave_caps(struct dma_chan *chan, struct dma_slave_caps *caps)
{
+ struct dma_device *device;
+
if (!chan || !caps)
return -EINVAL;
+ device = chan->device;
+
/* check if the channel supports slave transactions */
- if (!test_bit(DMA_SLAVE, chan->device->cap_mask.bits))
+ if (!test_bit(DMA_SLAVE, device->cap_mask.bits))
+ return -ENXIO;
+
+ if (device->device_slave_caps)
+ return device->device_slave_caps(chan, caps);
+
+ /*
+ * Check whether it reports it uses the generic slave
+ * capabilities, if not, that means it doesn't support any
+ * kind of slave capabilities reporting.
+ */
+ if (!device->directions)
return -ENXIO;
- if (chan->device->device_slave_caps)
- return chan->device->device_slave_caps(chan, caps);
+ caps->src_addr_widths = device->src_addr_widths;
+ caps->dst_addr_widths = device->dst_addr_widths;
+ caps->directions = device->directions;
+ caps->residue_granularity = device->residue_granularity;
+
+ caps->cmd_pause = !!device->device_pause;
+ caps->cmd_terminate = !!device->device_terminate_all;
- return -ENXIO;
+ return 0;
}
static inline int dmaengine_terminate_all(struct dma_chan *chan)
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 11/61] dmaengine: pl08x: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the AMBA PL08x DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/amba-pl08x.c | 156 +++++++++++++++++++++++++++--------------------
1 file changed, 90 insertions(+), 66 deletions(-)
diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c
index 1364d00881dd..4a5fd245014e 100644
--- a/drivers/dma/amba-pl08x.c
+++ b/drivers/dma/amba-pl08x.c
@@ -1386,32 +1386,6 @@ static u32 pl08x_get_cctl(struct pl08x_dma_chan *plchan,
return pl08x_cctl(cctl);
}
-static int dma_set_runtime_config(struct dma_chan *chan,
- struct dma_slave_config *config)
-{
- struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
- struct pl08x_driver_data *pl08x = plchan->host;
-
- if (!plchan->slave)
- return -EINVAL;
-
- /* Reject definitely invalid configurations */
- if (config->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
- config->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
- return -EINVAL;
-
- if (config->device_fc && pl08x->vd->pl080s) {
- dev_err(&pl08x->adev->dev,
- "%s: PL080S does not support peripheral flow control\n",
- __func__);
- return -EINVAL;
- }
-
- plchan->cfg = *config;
-
- return 0;
-}
-
/*
* Slave transactions callback to the slave device to allow
* synchronization of slave DMA signals with the DMAC enable
@@ -1693,20 +1667,71 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_cyclic(
return vchan_tx_prep(&plchan->vc, &txd->vd, flags);
}
-static int pl08x_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int pl08x_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
+{
+ struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
+ struct pl08x_driver_data *pl08x = plchan->host;
+
+ if (!plchan->slave)
+ return -EINVAL;
+
+ /* Reject definitely invalid configurations */
+ if (config->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
+ config->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
+ return -EINVAL;
+
+ if (config->device_fc && pl08x->vd->pl080s) {
+ dev_err(&pl08x->adev->dev,
+ "%s: PL080S does not support peripheral flow control\n",
+ __func__);
+ return -EINVAL;
+ }
+
+ plchan->cfg = *config;
+
+ return 0;
+}
+
+static int pl08x_terminate_all(struct dma_chan *chan)
{
struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
struct pl08x_driver_data *pl08x = plchan->host;
unsigned long flags;
- int ret = 0;
- /* Controls applicable to inactive channels */
- if (cmd == DMA_SLAVE_CONFIG) {
- return dma_set_runtime_config(chan,
- (struct dma_slave_config *)arg);
+ spin_lock_irqsave(&plchan->vc.lock, flags);
+ if (!plchan->phychan && !plchan->at) {
+ spin_unlock_irqrestore(&plchan->vc.lock, flags);
+ return 0;
}
+ plchan->state = PL08X_CHAN_IDLE;
+
+ if (plchan->phychan) {
+ /*
+ * Mark physical channel as free and free any slave
+ * signal
+ */
+ pl08x_phy_free(plchan);
+ }
+ /* Dequeue jobs and free LLIs */
+ if (plchan->at) {
+ pl08x_desc_free(&plchan->at->vd);
+ plchan->at = NULL;
+ }
+ /* Dequeue jobs not yet fired as well */
+ pl08x_free_txd_list(pl08x, plchan);
+
+ spin_unlock_irqrestore(&plchan->vc.lock, flags);
+
+ return 0;
+}
+
+static int pl08x_pause(struct dma_chan *chan)
+{
+ struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
+ unsigned long flags;
+
/*
* Anything succeeds on channels with no physical allocation and
* no queued transfers.
@@ -1717,42 +1742,35 @@ static int pl08x_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
return 0;
}
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- plchan->state = PL08X_CHAN_IDLE;
+ pl08x_pause_phy_chan(plchan->phychan);
+ plchan->state = PL08X_CHAN_PAUSED;
- if (plchan->phychan) {
- /*
- * Mark physical channel as free and free any slave
- * signal
- */
- pl08x_phy_free(plchan);
- }
- /* Dequeue jobs and free LLIs */
- if (plchan->at) {
- pl08x_desc_free(&plchan->at->vd);
- plchan->at = NULL;
- }
- /* Dequeue jobs not yet fired as well */
- pl08x_free_txd_list(pl08x, plchan);
- break;
- case DMA_PAUSE:
- pl08x_pause_phy_chan(plchan->phychan);
- plchan->state = PL08X_CHAN_PAUSED;
- break;
- case DMA_RESUME:
- pl08x_resume_phy_chan(plchan->phychan);
- plchan->state = PL08X_CHAN_RUNNING;
- break;
- default:
- /* Unknown command */
- ret = -ENXIO;
- break;
+ spin_unlock_irqrestore(&plchan->vc.lock, flags);
+
+ return 0;
+}
+
+static int pl08x_resume(struct dma_chan *chan)
+{
+ struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
+ unsigned long flags;
+
+ /*
+ * Anything succeeds on channels with no physical allocation and
+ * no queued transfers.
+ */
+ spin_lock_irqsave(&plchan->vc.lock, flags);
+ if (!plchan->phychan && !plchan->at) {
+ spin_unlock_irqrestore(&plchan->vc.lock, flags);
+ return 0;
}
+ pl08x_resume_phy_chan(plchan->phychan);
+ plchan->state = PL08X_CHAN_RUNNING;
+
spin_unlock_irqrestore(&plchan->vc.lock, flags);
- return ret;
+ return 0;
}
bool pl08x_filter_id(struct dma_chan *chan, void *chan_id)
@@ -2048,7 +2066,10 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
pl08x->memcpy.device_prep_dma_interrupt = pl08x_prep_dma_interrupt;
pl08x->memcpy.device_tx_status = pl08x_dma_tx_status;
pl08x->memcpy.device_issue_pending = pl08x_issue_pending;
- pl08x->memcpy.device_control = pl08x_control;
+ pl08x->memcpy.device_config = pl08x_config;
+ pl08x->memcpy.device_pause = pl08x_pause;
+ pl08x->memcpy.device_resume = pl08x_resume;
+ pl08x->memcpy.device_terminate_all = pl08x_terminate_all;
/* Initialize slave engine */
dma_cap_set(DMA_SLAVE, pl08x->slave.cap_mask);
@@ -2061,7 +2082,10 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
pl08x->slave.device_issue_pending = pl08x_issue_pending;
pl08x->slave.device_prep_slave_sg = pl08x_prep_slave_sg;
pl08x->slave.device_prep_dma_cyclic = pl08x_prep_dma_cyclic;
- pl08x->slave.device_control = pl08x_control;
+ pl08x->slave.device_config = pl08x_config;
+ pl08x->slave.device_pause = pl08x_pause;
+ pl08x->slave.device_resume = pl08x_resume;
+ pl08x->slave.device_terminate_all = pl08x_terminate_all;
/* Get the platform data */
pl08x->pd = dev_get_platdata(&adev->dev);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 11/61] dmaengine: pl08x: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the AMBA PL08x DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/amba-pl08x.c | 156 +++++++++++++++++++++++++++--------------------
1 file changed, 90 insertions(+), 66 deletions(-)
diff --git a/drivers/dma/amba-pl08x.c b/drivers/dma/amba-pl08x.c
index 1364d00881dd..4a5fd245014e 100644
--- a/drivers/dma/amba-pl08x.c
+++ b/drivers/dma/amba-pl08x.c
@@ -1386,32 +1386,6 @@ static u32 pl08x_get_cctl(struct pl08x_dma_chan *plchan,
return pl08x_cctl(cctl);
}
-static int dma_set_runtime_config(struct dma_chan *chan,
- struct dma_slave_config *config)
-{
- struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
- struct pl08x_driver_data *pl08x = plchan->host;
-
- if (!plchan->slave)
- return -EINVAL;
-
- /* Reject definitely invalid configurations */
- if (config->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
- config->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
- return -EINVAL;
-
- if (config->device_fc && pl08x->vd->pl080s) {
- dev_err(&pl08x->adev->dev,
- "%s: PL080S does not support peripheral flow control\n",
- __func__);
- return -EINVAL;
- }
-
- plchan->cfg = *config;
-
- return 0;
-}
-
/*
* Slave transactions callback to the slave device to allow
* synchronization of slave DMA signals with the DMAC enable
@@ -1693,20 +1667,71 @@ static struct dma_async_tx_descriptor *pl08x_prep_dma_cyclic(
return vchan_tx_prep(&plchan->vc, &txd->vd, flags);
}
-static int pl08x_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int pl08x_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
+{
+ struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
+ struct pl08x_driver_data *pl08x = plchan->host;
+
+ if (!plchan->slave)
+ return -EINVAL;
+
+ /* Reject definitely invalid configurations */
+ if (config->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
+ config->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
+ return -EINVAL;
+
+ if (config->device_fc && pl08x->vd->pl080s) {
+ dev_err(&pl08x->adev->dev,
+ "%s: PL080S does not support peripheral flow control\n",
+ __func__);
+ return -EINVAL;
+ }
+
+ plchan->cfg = *config;
+
+ return 0;
+}
+
+static int pl08x_terminate_all(struct dma_chan *chan)
{
struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
struct pl08x_driver_data *pl08x = plchan->host;
unsigned long flags;
- int ret = 0;
- /* Controls applicable to inactive channels */
- if (cmd == DMA_SLAVE_CONFIG) {
- return dma_set_runtime_config(chan,
- (struct dma_slave_config *)arg);
+ spin_lock_irqsave(&plchan->vc.lock, flags);
+ if (!plchan->phychan && !plchan->at) {
+ spin_unlock_irqrestore(&plchan->vc.lock, flags);
+ return 0;
}
+ plchan->state = PL08X_CHAN_IDLE;
+
+ if (plchan->phychan) {
+ /*
+ * Mark physical channel as free and free any slave
+ * signal
+ */
+ pl08x_phy_free(plchan);
+ }
+ /* Dequeue jobs and free LLIs */
+ if (plchan->at) {
+ pl08x_desc_free(&plchan->at->vd);
+ plchan->at = NULL;
+ }
+ /* Dequeue jobs not yet fired as well */
+ pl08x_free_txd_list(pl08x, plchan);
+
+ spin_unlock_irqrestore(&plchan->vc.lock, flags);
+
+ return 0;
+}
+
+static int pl08x_pause(struct dma_chan *chan)
+{
+ struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
+ unsigned long flags;
+
/*
* Anything succeeds on channels with no physical allocation and
* no queued transfers.
@@ -1717,42 +1742,35 @@ static int pl08x_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
return 0;
}
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- plchan->state = PL08X_CHAN_IDLE;
+ pl08x_pause_phy_chan(plchan->phychan);
+ plchan->state = PL08X_CHAN_PAUSED;
- if (plchan->phychan) {
- /*
- * Mark physical channel as free and free any slave
- * signal
- */
- pl08x_phy_free(plchan);
- }
- /* Dequeue jobs and free LLIs */
- if (plchan->at) {
- pl08x_desc_free(&plchan->at->vd);
- plchan->at = NULL;
- }
- /* Dequeue jobs not yet fired as well */
- pl08x_free_txd_list(pl08x, plchan);
- break;
- case DMA_PAUSE:
- pl08x_pause_phy_chan(plchan->phychan);
- plchan->state = PL08X_CHAN_PAUSED;
- break;
- case DMA_RESUME:
- pl08x_resume_phy_chan(plchan->phychan);
- plchan->state = PL08X_CHAN_RUNNING;
- break;
- default:
- /* Unknown command */
- ret = -ENXIO;
- break;
+ spin_unlock_irqrestore(&plchan->vc.lock, flags);
+
+ return 0;
+}
+
+static int pl08x_resume(struct dma_chan *chan)
+{
+ struct pl08x_dma_chan *plchan = to_pl08x_chan(chan);
+ unsigned long flags;
+
+ /*
+ * Anything succeeds on channels with no physical allocation and
+ * no queued transfers.
+ */
+ spin_lock_irqsave(&plchan->vc.lock, flags);
+ if (!plchan->phychan && !plchan->at) {
+ spin_unlock_irqrestore(&plchan->vc.lock, flags);
+ return 0;
}
+ pl08x_resume_phy_chan(plchan->phychan);
+ plchan->state = PL08X_CHAN_RUNNING;
+
spin_unlock_irqrestore(&plchan->vc.lock, flags);
- return ret;
+ return 0;
}
bool pl08x_filter_id(struct dma_chan *chan, void *chan_id)
@@ -2048,7 +2066,10 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
pl08x->memcpy.device_prep_dma_interrupt = pl08x_prep_dma_interrupt;
pl08x->memcpy.device_tx_status = pl08x_dma_tx_status;
pl08x->memcpy.device_issue_pending = pl08x_issue_pending;
- pl08x->memcpy.device_control = pl08x_control;
+ pl08x->memcpy.device_config = pl08x_config;
+ pl08x->memcpy.device_pause = pl08x_pause;
+ pl08x->memcpy.device_resume = pl08x_resume;
+ pl08x->memcpy.device_terminate_all = pl08x_terminate_all;
/* Initialize slave engine */
dma_cap_set(DMA_SLAVE, pl08x->slave.cap_mask);
@@ -2061,7 +2082,10 @@ static int pl08x_probe(struct amba_device *adev, const struct amba_id *id)
pl08x->slave.device_issue_pending = pl08x_issue_pending;
pl08x->slave.device_prep_slave_sg = pl08x_prep_slave_sg;
pl08x->slave.device_prep_dma_cyclic = pl08x_prep_dma_cyclic;
- pl08x->slave.device_control = pl08x_control;
+ pl08x->slave.device_config = pl08x_config;
+ pl08x->slave.device_pause = pl08x_pause;
+ pl08x->slave.device_resume = pl08x_resume;
+ pl08x->slave.device_terminate_all = pl08x_terminate_all;
/* Get the platform data */
pl08x->pd = dev_get_platdata(&adev->dev);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 12/61] dmaengine: hdmac: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Atmel HDMAC driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
---
drivers/dma/at_hdmac.c | 121 +++++++++++++++++++++++++++++--------------------
1 file changed, 73 insertions(+), 48 deletions(-)
diff --git a/drivers/dma/at_hdmac.c b/drivers/dma/at_hdmac.c
index ca9dd2613283..86450b3442f2 100644
--- a/drivers/dma/at_hdmac.c
+++ b/drivers/dma/at_hdmac.c
@@ -972,11 +972,13 @@ err_out:
return NULL;
}
-static int set_runtime_config(struct dma_chan *chan,
- struct dma_slave_config *sconfig)
+static int atc_config(struct dma_chan *chan,
+ struct dma_slave_config *sconfig)
{
struct at_dma_chan *atchan = to_at_dma_chan(chan);
+ dev_vdbg(chan2dev(chan), "%s\n", __func__);
+
/* Check if it is chan is configured for slave transfers */
if (!chan->private)
return -EINVAL;
@@ -989,9 +991,28 @@ static int set_runtime_config(struct dma_chan *chan,
return 0;
}
+static int atc_pause(struct dma_chan *chan)
+{
+ struct at_dma_chan *atchan = to_at_dma_chan(chan);
+ struct at_dma *atdma = to_at_dma(chan->device);
+ int chan_id = atchan->chan_common.chan_id;
+ unsigned long flags;
-static int atc_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+ LIST_HEAD(list);
+
+ dev_vdbg(chan2dev(chan), "%s\n", __func__);
+
+ spin_lock_irqsave(&atchan->lock, flags);
+
+ dma_writel(atdma, CHER, AT_DMA_SUSP(chan_id));
+ set_bit(ATC_IS_PAUSED, &atchan->status);
+
+ spin_unlock_irqrestore(&atchan->lock, flags);
+
+ return 0;
+}
+
+static int atc_resume(struct dma_chan *chan)
{
struct at_dma_chan *atchan = to_at_dma_chan(chan);
struct at_dma *atdma = to_at_dma(chan->device);
@@ -1000,60 +1021,61 @@ static int atc_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
LIST_HEAD(list);
- dev_vdbg(chan2dev(chan), "atc_control (%d)\n", cmd);
+ dev_vdbg(chan2dev(chan), "%s\n", __func__);
- if (cmd == DMA_PAUSE) {
- spin_lock_irqsave(&atchan->lock, flags);
+ if (!atc_chan_is_paused(atchan))
+ return 0;
- dma_writel(atdma, CHER, AT_DMA_SUSP(chan_id));
- set_bit(ATC_IS_PAUSED, &atchan->status);
+ spin_lock_irqsave(&atchan->lock, flags);
- spin_unlock_irqrestore(&atchan->lock, flags);
- } else if (cmd == DMA_RESUME) {
- if (!atc_chan_is_paused(atchan))
- return 0;
+ dma_writel(atdma, CHDR, AT_DMA_RES(chan_id));
+ clear_bit(ATC_IS_PAUSED, &atchan->status);
- spin_lock_irqsave(&atchan->lock, flags);
+ spin_unlock_irqrestore(&atchan->lock, flags);
- dma_writel(atdma, CHDR, AT_DMA_RES(chan_id));
- clear_bit(ATC_IS_PAUSED, &atchan->status);
+ return 0;
+}
- spin_unlock_irqrestore(&atchan->lock, flags);
- } else if (cmd == DMA_TERMINATE_ALL) {
- struct at_desc *desc, *_desc;
- /*
- * This is only called when something went wrong elsewhere, so
- * we don't really care about the data. Just disable the
- * channel. We still have to poll the channel enable bit due
- * to AHB/HSB limitations.
- */
- spin_lock_irqsave(&atchan->lock, flags);
+static int atc_terminate_all(struct dma_chan *chan)
+{
+ struct at_dma_chan *atchan = to_at_dma_chan(chan);
+ struct at_dma *atdma = to_at_dma(chan->device);
+ int chan_id = atchan->chan_common.chan_id;
+ struct at_desc *desc, *_desc;
+ unsigned long flags;
- /* disabling channel: must also remove suspend state */
- dma_writel(atdma, CHDR, AT_DMA_RES(chan_id) | atchan->mask);
+ LIST_HEAD(list);
- /* confirm that this channel is disabled */
- while (dma_readl(atdma, CHSR) & atchan->mask)
- cpu_relax();
+ dev_vdbg(chan2dev(chan), "%s\n", __func__);
- /* active_list entries will end up before queued entries */
- list_splice_init(&atchan->queue, &list);
- list_splice_init(&atchan->active_list, &list);
+ /*
+ * This is only called when something went wrong elsewhere, so
+ * we don't really care about the data. Just disable the
+ * channel. We still have to poll the channel enable bit due
+ * to AHB/HSB limitations.
+ */
+ spin_lock_irqsave(&atchan->lock, flags);
- /* Flush all pending and queued descriptors */
- list_for_each_entry_safe(desc, _desc, &list, desc_node)
- atc_chain_complete(atchan, desc);
+ /* disabling channel: must also remove suspend state */
+ dma_writel(atdma, CHDR, AT_DMA_RES(chan_id) | atchan->mask);
- clear_bit(ATC_IS_PAUSED, &atchan->status);
- /* if channel dedicated to cyclic operations, free it */
- clear_bit(ATC_IS_CYCLIC, &atchan->status);
+ /* confirm that this channel is disabled */
+ while (dma_readl(atdma, CHSR) & atchan->mask)
+ cpu_relax();
- spin_unlock_irqrestore(&atchan->lock, flags);
- } else if (cmd == DMA_SLAVE_CONFIG) {
- return set_runtime_config(chan, (struct dma_slave_config *)arg);
- } else {
- return -ENXIO;
- }
+ /* active_list entries will end up before queued entries */
+ list_splice_init(&atchan->queue, &list);
+ list_splice_init(&atchan->active_list, &list);
+
+ /* Flush all pending and queued descriptors */
+ list_for_each_entry_safe(desc, _desc, &list, desc_node)
+ atc_chain_complete(atchan, desc);
+
+ clear_bit(ATC_IS_PAUSED, &atchan->status);
+ /* if channel dedicated to cyclic operations, free it */
+ clear_bit(ATC_IS_CYCLIC, &atchan->status);
+
+ spin_unlock_irqrestore(&atchan->lock, flags);
return 0;
}
@@ -1505,7 +1527,10 @@ static int __init at_dma_probe(struct platform_device *pdev)
/* controller can do slave DMA: can trigger cyclic transfers */
dma_cap_set(DMA_CYCLIC, atdma->dma_common.cap_mask);
atdma->dma_common.device_prep_dma_cyclic = atc_prep_dma_cyclic;
- atdma->dma_common.device_control = atc_control;
+ atdma->dma_common.device_config = atc_config;
+ atdma->dma_common.device_pause = atc_pause;
+ atdma->dma_common.device_resume = atc_resume;
+ atdma->dma_common.device_terminate_all = atc_terminate_all;
}
dma_writel(atdma, EN, AT_DMA_ENABLE);
@@ -1622,7 +1647,7 @@ static void atc_suspend_cyclic(struct at_dma_chan *atchan)
if (!atc_chan_is_paused(atchan)) {
dev_warn(chan2dev(chan),
"cyclic channel not paused, should be done by channel user\n");
- atc_control(chan, DMA_PAUSE, 0);
+ atc_pause(chan);
}
/* now preserve additional data for cyclic operations */
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 12/61] dmaengine: hdmac: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Atmel HDMAC driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Nicolas Ferre <nicolas.ferre@atmel.com>
---
drivers/dma/at_hdmac.c | 121 +++++++++++++++++++++++++++++--------------------
1 file changed, 73 insertions(+), 48 deletions(-)
diff --git a/drivers/dma/at_hdmac.c b/drivers/dma/at_hdmac.c
index ca9dd2613283..86450b3442f2 100644
--- a/drivers/dma/at_hdmac.c
+++ b/drivers/dma/at_hdmac.c
@@ -972,11 +972,13 @@ err_out:
return NULL;
}
-static int set_runtime_config(struct dma_chan *chan,
- struct dma_slave_config *sconfig)
+static int atc_config(struct dma_chan *chan,
+ struct dma_slave_config *sconfig)
{
struct at_dma_chan *atchan = to_at_dma_chan(chan);
+ dev_vdbg(chan2dev(chan), "%s\n", __func__);
+
/* Check if it is chan is configured for slave transfers */
if (!chan->private)
return -EINVAL;
@@ -989,9 +991,28 @@ static int set_runtime_config(struct dma_chan *chan,
return 0;
}
+static int atc_pause(struct dma_chan *chan)
+{
+ struct at_dma_chan *atchan = to_at_dma_chan(chan);
+ struct at_dma *atdma = to_at_dma(chan->device);
+ int chan_id = atchan->chan_common.chan_id;
+ unsigned long flags;
-static int atc_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+ LIST_HEAD(list);
+
+ dev_vdbg(chan2dev(chan), "%s\n", __func__);
+
+ spin_lock_irqsave(&atchan->lock, flags);
+
+ dma_writel(atdma, CHER, AT_DMA_SUSP(chan_id));
+ set_bit(ATC_IS_PAUSED, &atchan->status);
+
+ spin_unlock_irqrestore(&atchan->lock, flags);
+
+ return 0;
+}
+
+static int atc_resume(struct dma_chan *chan)
{
struct at_dma_chan *atchan = to_at_dma_chan(chan);
struct at_dma *atdma = to_at_dma(chan->device);
@@ -1000,60 +1021,61 @@ static int atc_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
LIST_HEAD(list);
- dev_vdbg(chan2dev(chan), "atc_control (%d)\n", cmd);
+ dev_vdbg(chan2dev(chan), "%s\n", __func__);
- if (cmd == DMA_PAUSE) {
- spin_lock_irqsave(&atchan->lock, flags);
+ if (!atc_chan_is_paused(atchan))
+ return 0;
- dma_writel(atdma, CHER, AT_DMA_SUSP(chan_id));
- set_bit(ATC_IS_PAUSED, &atchan->status);
+ spin_lock_irqsave(&atchan->lock, flags);
- spin_unlock_irqrestore(&atchan->lock, flags);
- } else if (cmd == DMA_RESUME) {
- if (!atc_chan_is_paused(atchan))
- return 0;
+ dma_writel(atdma, CHDR, AT_DMA_RES(chan_id));
+ clear_bit(ATC_IS_PAUSED, &atchan->status);
- spin_lock_irqsave(&atchan->lock, flags);
+ spin_unlock_irqrestore(&atchan->lock, flags);
- dma_writel(atdma, CHDR, AT_DMA_RES(chan_id));
- clear_bit(ATC_IS_PAUSED, &atchan->status);
+ return 0;
+}
- spin_unlock_irqrestore(&atchan->lock, flags);
- } else if (cmd == DMA_TERMINATE_ALL) {
- struct at_desc *desc, *_desc;
- /*
- * This is only called when something went wrong elsewhere, so
- * we don't really care about the data. Just disable the
- * channel. We still have to poll the channel enable bit due
- * to AHB/HSB limitations.
- */
- spin_lock_irqsave(&atchan->lock, flags);
+static int atc_terminate_all(struct dma_chan *chan)
+{
+ struct at_dma_chan *atchan = to_at_dma_chan(chan);
+ struct at_dma *atdma = to_at_dma(chan->device);
+ int chan_id = atchan->chan_common.chan_id;
+ struct at_desc *desc, *_desc;
+ unsigned long flags;
- /* disabling channel: must also remove suspend state */
- dma_writel(atdma, CHDR, AT_DMA_RES(chan_id) | atchan->mask);
+ LIST_HEAD(list);
- /* confirm that this channel is disabled */
- while (dma_readl(atdma, CHSR) & atchan->mask)
- cpu_relax();
+ dev_vdbg(chan2dev(chan), "%s\n", __func__);
- /* active_list entries will end up before queued entries */
- list_splice_init(&atchan->queue, &list);
- list_splice_init(&atchan->active_list, &list);
+ /*
+ * This is only called when something went wrong elsewhere, so
+ * we don't really care about the data. Just disable the
+ * channel. We still have to poll the channel enable bit due
+ * to AHB/HSB limitations.
+ */
+ spin_lock_irqsave(&atchan->lock, flags);
- /* Flush all pending and queued descriptors */
- list_for_each_entry_safe(desc, _desc, &list, desc_node)
- atc_chain_complete(atchan, desc);
+ /* disabling channel: must also remove suspend state */
+ dma_writel(atdma, CHDR, AT_DMA_RES(chan_id) | atchan->mask);
- clear_bit(ATC_IS_PAUSED, &atchan->status);
- /* if channel dedicated to cyclic operations, free it */
- clear_bit(ATC_IS_CYCLIC, &atchan->status);
+ /* confirm that this channel is disabled */
+ while (dma_readl(atdma, CHSR) & atchan->mask)
+ cpu_relax();
- spin_unlock_irqrestore(&atchan->lock, flags);
- } else if (cmd == DMA_SLAVE_CONFIG) {
- return set_runtime_config(chan, (struct dma_slave_config *)arg);
- } else {
- return -ENXIO;
- }
+ /* active_list entries will end up before queued entries */
+ list_splice_init(&atchan->queue, &list);
+ list_splice_init(&atchan->active_list, &list);
+
+ /* Flush all pending and queued descriptors */
+ list_for_each_entry_safe(desc, _desc, &list, desc_node)
+ atc_chain_complete(atchan, desc);
+
+ clear_bit(ATC_IS_PAUSED, &atchan->status);
+ /* if channel dedicated to cyclic operations, free it */
+ clear_bit(ATC_IS_CYCLIC, &atchan->status);
+
+ spin_unlock_irqrestore(&atchan->lock, flags);
return 0;
}
@@ -1505,7 +1527,10 @@ static int __init at_dma_probe(struct platform_device *pdev)
/* controller can do slave DMA: can trigger cyclic transfers */
dma_cap_set(DMA_CYCLIC, atdma->dma_common.cap_mask);
atdma->dma_common.device_prep_dma_cyclic = atc_prep_dma_cyclic;
- atdma->dma_common.device_control = atc_control;
+ atdma->dma_common.device_config = atc_config;
+ atdma->dma_common.device_pause = atc_pause;
+ atdma->dma_common.device_resume = atc_resume;
+ atdma->dma_common.device_terminate_all = atc_terminate_all;
}
dma_writel(atdma, EN, AT_DMA_ENABLE);
@@ -1622,7 +1647,7 @@ static void atc_suspend_cyclic(struct at_dma_chan *atchan)
if (!atc_chan_is_paused(atchan)) {
dev_warn(chan2dev(chan),
"cyclic channel not paused, should be done by channel user\n");
- atc_control(chan, DMA_PAUSE, 0);
+ atc_pause(chan);
}
/* now preserve additional data for cyclic operations */
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 13/61] dmaengine: at_xdmac: split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Ludovic Desroches,
Maxime Ripard
From: Ludovic Desroches <ludovic.desroches@atmel.com>
Use newly introduced callbacks.
Signed-off-by: Ludovic Desroches <ludovic.desroches@atmel.com>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/at_xdmac.c | 94 ++++++++++++++++++++++++++++++--------------------
1 file changed, 57 insertions(+), 37 deletions(-)
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index bdc193cb82b7..9906e7cfbe37 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -1097,58 +1097,75 @@ static void at_xdmac_issue_pending(struct dma_chan *chan)
return;
}
-static int at_xdmac_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int at_xdmac_device_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
+{
+ struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
+ int ret;
+
+ dev_dbg(chan2dev(chan), "%s\n", __func__);
+
+ spin_lock_bh(&atchan->lock);
+ ret = at_xdmac_set_slave_config(chan, config);
+ spin_unlock_bh(&atchan->lock);
+
+ return ret;
+}
+
+static int at_xdmac_device_pause(struct dma_chan *chan)
{
- struct at_xdmac_desc *desc, *_desc;
struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
struct at_xdmac *atxdmac = to_at_xdmac(atchan->chan.device);
- int ret = 0;
- dev_dbg(chan2dev(chan), "%s: cmd=%d\n", __func__, cmd);
+ dev_dbg(chan2dev(chan), "%s\n", __func__);
spin_lock_bh(&atchan->lock);
+ at_xdmac_write(atxdmac, AT_XDMAC_GRWS, atchan->mask);
+ set_bit(AT_XDMAC_CHAN_IS_PAUSED, &atchan->status);
+ spin_unlock_bh(&atchan->lock);
- switch (cmd) {
- case DMA_PAUSE:
- at_xdmac_write(atxdmac, AT_XDMAC_GRWS, atchan->mask);
- set_bit(AT_XDMAC_CHAN_IS_PAUSED, &atchan->status);
- break;
+ return 0;
+}
- case DMA_RESUME:
- if (!at_xdmac_chan_is_paused(atchan))
- break;
+static int at_xdmac_device_resume(struct dma_chan *chan)
+{
+ struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
+ struct at_xdmac *atxdmac = to_at_xdmac(atchan->chan.device);
- at_xdmac_write(atxdmac, AT_XDMAC_GRWR, atchan->mask);
- clear_bit(AT_XDMAC_CHAN_IS_PAUSED, &atchan->status);
- break;
+ dev_dbg(chan2dev(chan), "%s\n", __func__);
- case DMA_TERMINATE_ALL:
- at_xdmac_write(atxdmac, AT_XDMAC_GD, atchan->mask);
- while (at_xdmac_read(atxdmac, AT_XDMAC_GS) & atchan->mask)
- cpu_relax();
+ spin_lock_bh(&atchan->lock);
+ if (!at_xdmac_chan_is_paused(atchan))
+ return 0;
- /* Cancel all pending transfers. */
- list_for_each_entry_safe(desc, _desc, &atchan->xfers_list, xfer_node)
- at_xdmac_remove_xfer(atchan, desc);
+ at_xdmac_write(atxdmac, AT_XDMAC_GRWR, atchan->mask);
+ clear_bit(AT_XDMAC_CHAN_IS_PAUSED, &atchan->status);
+ spin_unlock_bh(&atchan->lock);
+
+ return 0;
+}
+
+static int at_xdmac_device_terminate_all(struct dma_chan *chan)
+{
+ struct at_xdmac_desc *desc, *_desc;
+ struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
+ struct at_xdmac *atxdmac = to_at_xdmac(atchan->chan.device);
- clear_bit(AT_XDMAC_CHAN_IS_CYCLIC, &atchan->status);
- break;
+ dev_dbg(chan2dev(chan), "%s\n", __func__);
- case DMA_SLAVE_CONFIG:
- ret = at_xdmac_set_slave_config(chan,
- (struct dma_slave_config *)arg);
- break;
+ spin_lock_bh(&atchan->lock);
+ at_xdmac_write(atxdmac, AT_XDMAC_GD, atchan->mask);
+ while (at_xdmac_read(atxdmac, AT_XDMAC_GS) & atchan->mask)
+ cpu_relax();
- default:
- dev_err(chan2dev(chan),
- "unmanaged or unknown dma control cmd: %d\n", cmd);
- ret = -ENXIO;
- }
+ /* Cancel all pending transfers. */
+ list_for_each_entry_safe(desc, _desc, &atchan->xfers_list, xfer_node)
+ at_xdmac_remove_xfer(atchan, desc);
+ clear_bit(AT_XDMAC_CHAN_IS_CYCLIC, &atchan->status);
spin_unlock_bh(&atchan->lock);
- return ret;
+ return 0;
}
static int at_xdmac_alloc_chan_resources(struct dma_chan *chan)
@@ -1260,7 +1277,7 @@ static int atmel_xdmac_suspend(struct device *dev)
if (at_xdmac_chan_is_cyclic(atchan)) {
if (!at_xdmac_chan_is_paused(atchan))
- at_xdmac_control(chan, DMA_PAUSE, 0);
+ at_xdmac_device_pause(chan);
atchan->save_cim = at_xdmac_chan_read(atchan, AT_XDMAC_CIM);
atchan->save_cnda = at_xdmac_chan_read(atchan, AT_XDMAC_CNDA);
atchan->save_cndc = at_xdmac_chan_read(atchan, AT_XDMAC_CNDC);
@@ -1392,7 +1409,10 @@ static int at_xdmac_probe(struct platform_device *pdev)
atxdmac->dma.device_prep_dma_cyclic = at_xdmac_prep_dma_cyclic;
atxdmac->dma.device_prep_dma_memcpy = at_xdmac_prep_dma_memcpy;
atxdmac->dma.device_prep_slave_sg = at_xdmac_prep_slave_sg;
- atxdmac->dma.device_control = at_xdmac_control;
+ atxdmac->dma.device_config = at_xdmac_device_config;
+ atxdmac->dma.device_pause = at_xdmac_device_pause;
+ atxdmac->dma.device_resume = at_xdmac_device_resume;
+ atxdmac->dma.device_terminate_all = at_xdmac_device_terminate_all;
atxdmac->dma.chancnt = nr_channels;
atxdmac->dma.device_slave_caps = at_xdmac_device_slave_caps;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 13/61] dmaengine: at_xdmac: split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
From: Ludovic Desroches <ludovic.desroches@atmel.com>
Use newly introduced callbacks.
Signed-off-by: Ludovic Desroches <ludovic.desroches@atmel.com>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/at_xdmac.c | 94 ++++++++++++++++++++++++++++++--------------------
1 file changed, 57 insertions(+), 37 deletions(-)
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index bdc193cb82b7..9906e7cfbe37 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -1097,58 +1097,75 @@ static void at_xdmac_issue_pending(struct dma_chan *chan)
return;
}
-static int at_xdmac_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int at_xdmac_device_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
+{
+ struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
+ int ret;
+
+ dev_dbg(chan2dev(chan), "%s\n", __func__);
+
+ spin_lock_bh(&atchan->lock);
+ ret = at_xdmac_set_slave_config(chan, config);
+ spin_unlock_bh(&atchan->lock);
+
+ return ret;
+}
+
+static int at_xdmac_device_pause(struct dma_chan *chan)
{
- struct at_xdmac_desc *desc, *_desc;
struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
struct at_xdmac *atxdmac = to_at_xdmac(atchan->chan.device);
- int ret = 0;
- dev_dbg(chan2dev(chan), "%s: cmd=%d\n", __func__, cmd);
+ dev_dbg(chan2dev(chan), "%s\n", __func__);
spin_lock_bh(&atchan->lock);
+ at_xdmac_write(atxdmac, AT_XDMAC_GRWS, atchan->mask);
+ set_bit(AT_XDMAC_CHAN_IS_PAUSED, &atchan->status);
+ spin_unlock_bh(&atchan->lock);
- switch (cmd) {
- case DMA_PAUSE:
- at_xdmac_write(atxdmac, AT_XDMAC_GRWS, atchan->mask);
- set_bit(AT_XDMAC_CHAN_IS_PAUSED, &atchan->status);
- break;
+ return 0;
+}
- case DMA_RESUME:
- if (!at_xdmac_chan_is_paused(atchan))
- break;
+static int at_xdmac_device_resume(struct dma_chan *chan)
+{
+ struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
+ struct at_xdmac *atxdmac = to_at_xdmac(atchan->chan.device);
- at_xdmac_write(atxdmac, AT_XDMAC_GRWR, atchan->mask);
- clear_bit(AT_XDMAC_CHAN_IS_PAUSED, &atchan->status);
- break;
+ dev_dbg(chan2dev(chan), "%s\n", __func__);
- case DMA_TERMINATE_ALL:
- at_xdmac_write(atxdmac, AT_XDMAC_GD, atchan->mask);
- while (at_xdmac_read(atxdmac, AT_XDMAC_GS) & atchan->mask)
- cpu_relax();
+ spin_lock_bh(&atchan->lock);
+ if (!at_xdmac_chan_is_paused(atchan))
+ return 0;
- /* Cancel all pending transfers. */
- list_for_each_entry_safe(desc, _desc, &atchan->xfers_list, xfer_node)
- at_xdmac_remove_xfer(atchan, desc);
+ at_xdmac_write(atxdmac, AT_XDMAC_GRWR, atchan->mask);
+ clear_bit(AT_XDMAC_CHAN_IS_PAUSED, &atchan->status);
+ spin_unlock_bh(&atchan->lock);
+
+ return 0;
+}
+
+static int at_xdmac_device_terminate_all(struct dma_chan *chan)
+{
+ struct at_xdmac_desc *desc, *_desc;
+ struct at_xdmac_chan *atchan = to_at_xdmac_chan(chan);
+ struct at_xdmac *atxdmac = to_at_xdmac(atchan->chan.device);
- clear_bit(AT_XDMAC_CHAN_IS_CYCLIC, &atchan->status);
- break;
+ dev_dbg(chan2dev(chan), "%s\n", __func__);
- case DMA_SLAVE_CONFIG:
- ret = at_xdmac_set_slave_config(chan,
- (struct dma_slave_config *)arg);
- break;
+ spin_lock_bh(&atchan->lock);
+ at_xdmac_write(atxdmac, AT_XDMAC_GD, atchan->mask);
+ while (at_xdmac_read(atxdmac, AT_XDMAC_GS) & atchan->mask)
+ cpu_relax();
- default:
- dev_err(chan2dev(chan),
- "unmanaged or unknown dma control cmd: %d\n", cmd);
- ret = -ENXIO;
- }
+ /* Cancel all pending transfers. */
+ list_for_each_entry_safe(desc, _desc, &atchan->xfers_list, xfer_node)
+ at_xdmac_remove_xfer(atchan, desc);
+ clear_bit(AT_XDMAC_CHAN_IS_CYCLIC, &atchan->status);
spin_unlock_bh(&atchan->lock);
- return ret;
+ return 0;
}
static int at_xdmac_alloc_chan_resources(struct dma_chan *chan)
@@ -1260,7 +1277,7 @@ static int atmel_xdmac_suspend(struct device *dev)
if (at_xdmac_chan_is_cyclic(atchan)) {
if (!at_xdmac_chan_is_paused(atchan))
- at_xdmac_control(chan, DMA_PAUSE, 0);
+ at_xdmac_device_pause(chan);
atchan->save_cim = at_xdmac_chan_read(atchan, AT_XDMAC_CIM);
atchan->save_cnda = at_xdmac_chan_read(atchan, AT_XDMAC_CNDA);
atchan->save_cndc = at_xdmac_chan_read(atchan, AT_XDMAC_CNDC);
@@ -1392,7 +1409,10 @@ static int at_xdmac_probe(struct platform_device *pdev)
atxdmac->dma.device_prep_dma_cyclic = at_xdmac_prep_dma_cyclic;
atxdmac->dma.device_prep_dma_memcpy = at_xdmac_prep_dma_memcpy;
atxdmac->dma.device_prep_slave_sg = at_xdmac_prep_slave_sg;
- atxdmac->dma.device_control = at_xdmac_control;
+ atxdmac->dma.device_config = at_xdmac_device_config;
+ atxdmac->dma.device_pause = at_xdmac_device_pause;
+ atxdmac->dma.device_resume = at_xdmac_device_resume;
+ atxdmac->dma.device_terminate_all = at_xdmac_device_terminate_all;
atxdmac->dma.chancnt = nr_channels;
atxdmac->dma.device_slave_caps = at_xdmac_device_slave_caps;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 14/61] dmaengine: bcm2835: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Broadcom BCM2835 DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
---
drivers/dma/bcm2835-dma.c | 31 ++++++++-----------------------
1 file changed, 8 insertions(+), 23 deletions(-)
diff --git a/drivers/dma/bcm2835-dma.c b/drivers/dma/bcm2835-dma.c
index 3feba6c25836..13b05c1673b1 100644
--- a/drivers/dma/bcm2835-dma.c
+++ b/drivers/dma/bcm2835-dma.c
@@ -436,9 +436,11 @@ static struct dma_async_tx_descriptor *bcm2835_dma_prep_dma_cyclic(
return vchan_tx_prep(&c->vc, &d->vd, flags);
}
-static int bcm2835_dma_slave_config(struct bcm2835_chan *c,
- struct dma_slave_config *cfg)
+static int bcm2835_dma_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
{
+ struct bcm2835_chan *c = to_bcm2835_dma_chan(chan);
+
if ((cfg->direction == DMA_DEV_TO_MEM &&
cfg->src_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES) ||
(cfg->direction == DMA_MEM_TO_DEV &&
@@ -452,8 +454,9 @@ static int bcm2835_dma_slave_config(struct bcm2835_chan *c,
return 0;
}
-static int bcm2835_dma_terminate_all(struct bcm2835_chan *c)
+static int bcm2835_dma_terminate_all(struct dma_chan *chan)
{
+ struct bcm2835_chan *c = to_bcm2835_dma_chan(chan);
struct bcm2835_dmadev *d = to_bcm2835_dma_dev(c->vc.chan.device);
unsigned long flags;
int timeout = 10000;
@@ -495,24 +498,6 @@ static int bcm2835_dma_terminate_all(struct bcm2835_chan *c)
return 0;
}
-static int bcm2835_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct bcm2835_chan *c = to_bcm2835_dma_chan(chan);
-
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- return bcm2835_dma_slave_config(c,
- (struct dma_slave_config *)arg);
-
- case DMA_TERMINATE_ALL:
- return bcm2835_dma_terminate_all(c);
-
- default:
- return -ENXIO;
- }
-}
-
static int bcm2835_dma_chan_init(struct bcm2835_dmadev *d, int chan_id, int irq)
{
struct bcm2835_chan *c;
@@ -615,9 +600,9 @@ static int bcm2835_dma_probe(struct platform_device *pdev)
od->ddev.device_free_chan_resources = bcm2835_dma_free_chan_resources;
od->ddev.device_tx_status = bcm2835_dma_tx_status;
od->ddev.device_issue_pending = bcm2835_dma_issue_pending;
- od->ddev.device_slave_caps = bcm2835_dma_device_slave_caps;
od->ddev.device_prep_dma_cyclic = bcm2835_dma_prep_dma_cyclic;
- od->ddev.device_control = bcm2835_dma_control;
+ od->ddev.device_config = bcm2835_dma_slave_config;
+ od->ddev.device_terminate_all = bcm2835_dma_terminate_all;
od->ddev.dev = &pdev->dev;
INIT_LIST_HEAD(&od->ddev.channels);
spin_lock_init(&od->lock);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 14/61] dmaengine: bcm2835: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Broadcom BCM2835 DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
---
drivers/dma/bcm2835-dma.c | 31 ++++++++-----------------------
1 file changed, 8 insertions(+), 23 deletions(-)
diff --git a/drivers/dma/bcm2835-dma.c b/drivers/dma/bcm2835-dma.c
index 3feba6c25836..13b05c1673b1 100644
--- a/drivers/dma/bcm2835-dma.c
+++ b/drivers/dma/bcm2835-dma.c
@@ -436,9 +436,11 @@ static struct dma_async_tx_descriptor *bcm2835_dma_prep_dma_cyclic(
return vchan_tx_prep(&c->vc, &d->vd, flags);
}
-static int bcm2835_dma_slave_config(struct bcm2835_chan *c,
- struct dma_slave_config *cfg)
+static int bcm2835_dma_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
{
+ struct bcm2835_chan *c = to_bcm2835_dma_chan(chan);
+
if ((cfg->direction == DMA_DEV_TO_MEM &&
cfg->src_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES) ||
(cfg->direction == DMA_MEM_TO_DEV &&
@@ -452,8 +454,9 @@ static int bcm2835_dma_slave_config(struct bcm2835_chan *c,
return 0;
}
-static int bcm2835_dma_terminate_all(struct bcm2835_chan *c)
+static int bcm2835_dma_terminate_all(struct dma_chan *chan)
{
+ struct bcm2835_chan *c = to_bcm2835_dma_chan(chan);
struct bcm2835_dmadev *d = to_bcm2835_dma_dev(c->vc.chan.device);
unsigned long flags;
int timeout = 10000;
@@ -495,24 +498,6 @@ static int bcm2835_dma_terminate_all(struct bcm2835_chan *c)
return 0;
}
-static int bcm2835_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct bcm2835_chan *c = to_bcm2835_dma_chan(chan);
-
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- return bcm2835_dma_slave_config(c,
- (struct dma_slave_config *)arg);
-
- case DMA_TERMINATE_ALL:
- return bcm2835_dma_terminate_all(c);
-
- default:
- return -ENXIO;
- }
-}
-
static int bcm2835_dma_chan_init(struct bcm2835_dmadev *d, int chan_id, int irq)
{
struct bcm2835_chan *c;
@@ -615,9 +600,9 @@ static int bcm2835_dma_probe(struct platform_device *pdev)
od->ddev.device_free_chan_resources = bcm2835_dma_free_chan_resources;
od->ddev.device_tx_status = bcm2835_dma_tx_status;
od->ddev.device_issue_pending = bcm2835_dma_issue_pending;
- od->ddev.device_slave_caps = bcm2835_dma_device_slave_caps;
od->ddev.device_prep_dma_cyclic = bcm2835_dma_prep_dma_cyclic;
- od->ddev.device_control = bcm2835_dma_control;
+ od->ddev.device_config = bcm2835_dma_slave_config;
+ od->ddev.device_terminate_all = bcm2835_dma_terminate_all;
od->ddev.dev = &pdev->dev;
INIT_LIST_HEAD(&od->ddev.channels);
spin_lock_init(&od->lock);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 15/61] dmaengine: coh901318: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the ST-Ericsson COH901318 DMA driver to
make use of the newly introduced callbacks, that will eventually be used to
retrieve slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
---
drivers/dma/coh901318.c | 137 +++++++++++++++++++++---------------------------
1 file changed, 60 insertions(+), 77 deletions(-)
diff --git a/drivers/dma/coh901318.c b/drivers/dma/coh901318.c
index e88588d8ecd3..418e4e4fb7ba 100644
--- a/drivers/dma/coh901318.c
+++ b/drivers/dma/coh901318.c
@@ -2114,6 +2114,57 @@ static irqreturn_t dma_irq_handler(int irq, void *dev_id)
return IRQ_HANDLED;
}
+static int coh901318_terminate_all(struct dma_chan *chan)
+{
+ unsigned long flags;
+ struct coh901318_chan *cohc = to_coh901318_chan(chan);
+ struct coh901318_desc *cohd;
+ void __iomem *virtbase = cohc->base->virtbase;
+
+ /* The remainder of this function terminates the transfer */
+ coh901318_pause(chan);
+ spin_lock_irqsave(&cohc->lock, flags);
+
+ /* Clear any pending BE or TC interrupt */
+ if (cohc->id < 32) {
+ writel(1 << cohc->id, virtbase + COH901318_BE_INT_CLEAR1);
+ writel(1 << cohc->id, virtbase + COH901318_TC_INT_CLEAR1);
+ } else {
+ writel(1 << (cohc->id - 32), virtbase +
+ COH901318_BE_INT_CLEAR2);
+ writel(1 << (cohc->id - 32), virtbase +
+ COH901318_TC_INT_CLEAR2);
+ }
+
+ enable_powersave(cohc);
+
+ while ((cohd = coh901318_first_active_get(cohc))) {
+ /* release the lli allocation*/
+ coh901318_lli_free(&cohc->base->pool, &cohd->lli);
+
+ /* return desc to free-list */
+ coh901318_desc_remove(cohd);
+ coh901318_desc_free(cohc, cohd);
+ }
+
+ while ((cohd = coh901318_first_queued(cohc))) {
+ /* release the lli allocation*/
+ coh901318_lli_free(&cohc->base->pool, &cohd->lli);
+
+ /* return desc to free-list */
+ coh901318_desc_remove(cohd);
+ coh901318_desc_free(cohc, cohd);
+ }
+
+
+ cohc->nbr_active_done = 0;
+ cohc->busy = 0;
+
+ spin_unlock_irqrestore(&cohc->lock, flags);
+
+ return 0;
+}
+
static int coh901318_alloc_chan_resources(struct dma_chan *chan)
{
struct coh901318_chan *cohc = to_coh901318_chan(chan);
@@ -2156,7 +2207,7 @@ coh901318_free_chan_resources(struct dma_chan *chan)
spin_unlock_irqrestore(&cohc->lock, flags);
- dmaengine_terminate_all(chan);
+ coh901318_terminate_all(chan);
}
@@ -2540,80 +2591,6 @@ static void coh901318_dma_set_runtimeconfig(struct dma_chan *chan,
cohc->ctrl = ctrl;
}
-static int
-coh901318_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- unsigned long flags;
- struct coh901318_chan *cohc = to_coh901318_chan(chan);
- struct coh901318_desc *cohd;
- void __iomem *virtbase = cohc->base->virtbase;
-
- if (cmd == DMA_SLAVE_CONFIG) {
- struct dma_slave_config *config =
- (struct dma_slave_config *) arg;
-
- coh901318_dma_set_runtimeconfig(chan, config);
- return 0;
- }
-
- if (cmd == DMA_PAUSE) {
- coh901318_pause(chan);
- return 0;
- }
-
- if (cmd == DMA_RESUME) {
- coh901318_resume(chan);
- return 0;
- }
-
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
- /* The remainder of this function terminates the transfer */
- coh901318_pause(chan);
- spin_lock_irqsave(&cohc->lock, flags);
-
- /* Clear any pending BE or TC interrupt */
- if (cohc->id < 32) {
- writel(1 << cohc->id, virtbase + COH901318_BE_INT_CLEAR1);
- writel(1 << cohc->id, virtbase + COH901318_TC_INT_CLEAR1);
- } else {
- writel(1 << (cohc->id - 32), virtbase +
- COH901318_BE_INT_CLEAR2);
- writel(1 << (cohc->id - 32), virtbase +
- COH901318_TC_INT_CLEAR2);
- }
-
- enable_powersave(cohc);
-
- while ((cohd = coh901318_first_active_get(cohc))) {
- /* release the lli allocation*/
- coh901318_lli_free(&cohc->base->pool, &cohd->lli);
-
- /* return desc to free-list */
- coh901318_desc_remove(cohd);
- coh901318_desc_free(cohc, cohd);
- }
-
- while ((cohd = coh901318_first_queued(cohc))) {
- /* release the lli allocation*/
- coh901318_lli_free(&cohc->base->pool, &cohd->lli);
-
- /* return desc to free-list */
- coh901318_desc_remove(cohd);
- coh901318_desc_free(cohc, cohd);
- }
-
-
- cohc->nbr_active_done = 0;
- cohc->busy = 0;
-
- spin_unlock_irqrestore(&cohc->lock, flags);
-
- return 0;
-}
-
void coh901318_base_init(struct dma_device *dma, const int *pick_chans,
struct coh901318_base *base)
{
@@ -2717,7 +2694,10 @@ static int __init coh901318_probe(struct platform_device *pdev)
base->dma_slave.device_prep_slave_sg = coh901318_prep_slave_sg;
base->dma_slave.device_tx_status = coh901318_tx_status;
base->dma_slave.device_issue_pending = coh901318_issue_pending;
- base->dma_slave.device_control = coh901318_control;
+ base->dma_slave.device_config = coh901318_dma_set_runtimeconfig;
+ base->dma_slave.device_pause = coh901318_pause;
+ base->dma_slave.device_resume = coh901318_resume;
+ base->dma_slave.device_terminate_all = coh901318_terminate_all;
base->dma_slave.dev = &pdev->dev;
err = dma_async_device_register(&base->dma_slave);
@@ -2737,7 +2717,10 @@ static int __init coh901318_probe(struct platform_device *pdev)
base->dma_memcpy.device_prep_dma_memcpy = coh901318_prep_memcpy;
base->dma_memcpy.device_tx_status = coh901318_tx_status;
base->dma_memcpy.device_issue_pending = coh901318_issue_pending;
- base->dma_memcpy.device_control = coh901318_control;
+ base->dma_memcpy.device_config = coh901318_dma_set_runtimeconfig;
+ base->dma_memcpy.device_pause = coh901318_pause;
+ base->dma_memcpy.device_resume = coh901318_resume;
+ base->dma_memcpy.device_terminate_all = coh901318_terminate_all;
base->dma_memcpy.dev = &pdev->dev;
/*
* This controller can only access address at even 32bit boundaries,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 15/61] dmaengine: coh901318: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the ST-Ericsson COH901318 DMA driver to
make use of the newly introduced callbacks, that will eventually be used to
retrieve slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
---
drivers/dma/coh901318.c | 137 +++++++++++++++++++++---------------------------
1 file changed, 60 insertions(+), 77 deletions(-)
diff --git a/drivers/dma/coh901318.c b/drivers/dma/coh901318.c
index e88588d8ecd3..418e4e4fb7ba 100644
--- a/drivers/dma/coh901318.c
+++ b/drivers/dma/coh901318.c
@@ -2114,6 +2114,57 @@ static irqreturn_t dma_irq_handler(int irq, void *dev_id)
return IRQ_HANDLED;
}
+static int coh901318_terminate_all(struct dma_chan *chan)
+{
+ unsigned long flags;
+ struct coh901318_chan *cohc = to_coh901318_chan(chan);
+ struct coh901318_desc *cohd;
+ void __iomem *virtbase = cohc->base->virtbase;
+
+ /* The remainder of this function terminates the transfer */
+ coh901318_pause(chan);
+ spin_lock_irqsave(&cohc->lock, flags);
+
+ /* Clear any pending BE or TC interrupt */
+ if (cohc->id < 32) {
+ writel(1 << cohc->id, virtbase + COH901318_BE_INT_CLEAR1);
+ writel(1 << cohc->id, virtbase + COH901318_TC_INT_CLEAR1);
+ } else {
+ writel(1 << (cohc->id - 32), virtbase +
+ COH901318_BE_INT_CLEAR2);
+ writel(1 << (cohc->id - 32), virtbase +
+ COH901318_TC_INT_CLEAR2);
+ }
+
+ enable_powersave(cohc);
+
+ while ((cohd = coh901318_first_active_get(cohc))) {
+ /* release the lli allocation*/
+ coh901318_lli_free(&cohc->base->pool, &cohd->lli);
+
+ /* return desc to free-list */
+ coh901318_desc_remove(cohd);
+ coh901318_desc_free(cohc, cohd);
+ }
+
+ while ((cohd = coh901318_first_queued(cohc))) {
+ /* release the lli allocation*/
+ coh901318_lli_free(&cohc->base->pool, &cohd->lli);
+
+ /* return desc to free-list */
+ coh901318_desc_remove(cohd);
+ coh901318_desc_free(cohc, cohd);
+ }
+
+
+ cohc->nbr_active_done = 0;
+ cohc->busy = 0;
+
+ spin_unlock_irqrestore(&cohc->lock, flags);
+
+ return 0;
+}
+
static int coh901318_alloc_chan_resources(struct dma_chan *chan)
{
struct coh901318_chan *cohc = to_coh901318_chan(chan);
@@ -2156,7 +2207,7 @@ coh901318_free_chan_resources(struct dma_chan *chan)
spin_unlock_irqrestore(&cohc->lock, flags);
- dmaengine_terminate_all(chan);
+ coh901318_terminate_all(chan);
}
@@ -2540,80 +2591,6 @@ static void coh901318_dma_set_runtimeconfig(struct dma_chan *chan,
cohc->ctrl = ctrl;
}
-static int
-coh901318_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- unsigned long flags;
- struct coh901318_chan *cohc = to_coh901318_chan(chan);
- struct coh901318_desc *cohd;
- void __iomem *virtbase = cohc->base->virtbase;
-
- if (cmd == DMA_SLAVE_CONFIG) {
- struct dma_slave_config *config =
- (struct dma_slave_config *) arg;
-
- coh901318_dma_set_runtimeconfig(chan, config);
- return 0;
- }
-
- if (cmd == DMA_PAUSE) {
- coh901318_pause(chan);
- return 0;
- }
-
- if (cmd == DMA_RESUME) {
- coh901318_resume(chan);
- return 0;
- }
-
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
- /* The remainder of this function terminates the transfer */
- coh901318_pause(chan);
- spin_lock_irqsave(&cohc->lock, flags);
-
- /* Clear any pending BE or TC interrupt */
- if (cohc->id < 32) {
- writel(1 << cohc->id, virtbase + COH901318_BE_INT_CLEAR1);
- writel(1 << cohc->id, virtbase + COH901318_TC_INT_CLEAR1);
- } else {
- writel(1 << (cohc->id - 32), virtbase +
- COH901318_BE_INT_CLEAR2);
- writel(1 << (cohc->id - 32), virtbase +
- COH901318_TC_INT_CLEAR2);
- }
-
- enable_powersave(cohc);
-
- while ((cohd = coh901318_first_active_get(cohc))) {
- /* release the lli allocation*/
- coh901318_lli_free(&cohc->base->pool, &cohd->lli);
-
- /* return desc to free-list */
- coh901318_desc_remove(cohd);
- coh901318_desc_free(cohc, cohd);
- }
-
- while ((cohd = coh901318_first_queued(cohc))) {
- /* release the lli allocation*/
- coh901318_lli_free(&cohc->base->pool, &cohd->lli);
-
- /* return desc to free-list */
- coh901318_desc_remove(cohd);
- coh901318_desc_free(cohc, cohd);
- }
-
-
- cohc->nbr_active_done = 0;
- cohc->busy = 0;
-
- spin_unlock_irqrestore(&cohc->lock, flags);
-
- return 0;
-}
-
void coh901318_base_init(struct dma_device *dma, const int *pick_chans,
struct coh901318_base *base)
{
@@ -2717,7 +2694,10 @@ static int __init coh901318_probe(struct platform_device *pdev)
base->dma_slave.device_prep_slave_sg = coh901318_prep_slave_sg;
base->dma_slave.device_tx_status = coh901318_tx_status;
base->dma_slave.device_issue_pending = coh901318_issue_pending;
- base->dma_slave.device_control = coh901318_control;
+ base->dma_slave.device_config = coh901318_dma_set_runtimeconfig;
+ base->dma_slave.device_pause = coh901318_pause;
+ base->dma_slave.device_resume = coh901318_resume;
+ base->dma_slave.device_terminate_all = coh901318_terminate_all;
base->dma_slave.dev = &pdev->dev;
err = dma_async_device_register(&base->dma_slave);
@@ -2737,7 +2717,10 @@ static int __init coh901318_probe(struct platform_device *pdev)
base->dma_memcpy.device_prep_dma_memcpy = coh901318_prep_memcpy;
base->dma_memcpy.device_tx_status = coh901318_tx_status;
base->dma_memcpy.device_issue_pending = coh901318_issue_pending;
- base->dma_memcpy.device_control = coh901318_control;
+ base->dma_memcpy.device_config = coh901318_dma_set_runtimeconfig;
+ base->dma_memcpy.device_pause = coh901318_pause;
+ base->dma_memcpy.device_resume = coh901318_resume;
+ base->dma_memcpy.device_terminate_all = coh901318_terminate_all;
base->dma_memcpy.dev = &pdev->dev;
/*
* This controller can only access address at even 32bit boundaries,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 16/61] dmaengine: cppi41: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the TI CPPI41 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/cppi41.c | 30 +-----------------------------
1 file changed, 1 insertion(+), 29 deletions(-)
diff --git a/drivers/dma/cppi41.c b/drivers/dma/cppi41.c
index cf7633f6d474..98b7dc8948a9 100644
--- a/drivers/dma/cppi41.c
+++ b/drivers/dma/cppi41.c
@@ -524,12 +524,6 @@ static struct dma_async_tx_descriptor *cppi41_dma_prep_slave_sg(
return &c->txd;
}
-static int cpp41_cfg_chan(struct cppi41_channel *c,
- struct dma_slave_config *cfg)
-{
- return 0;
-}
-
static void cppi41_compute_td_desc(struct cppi41_desc *d)
{
d->pd0 = DESC_TYPE_TEARD << DESC_TYPE;
@@ -642,28 +636,6 @@ static int cppi41_stop_chan(struct dma_chan *chan)
return 0;
}
-static int cppi41_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct cppi41_channel *c = to_cpp41_chan(chan);
- int ret;
-
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- ret = cpp41_cfg_chan(c, (struct dma_slave_config *) arg);
- break;
-
- case DMA_TERMINATE_ALL:
- ret = cppi41_stop_chan(chan);
- break;
-
- default:
- ret = -ENXIO;
- break;
- }
- return ret;
-}
-
static void cleanup_chans(struct cppi41_dd *cdd)
{
while (!list_empty(&cdd->ddev.channels)) {
@@ -948,7 +920,7 @@ static int cppi41_dma_probe(struct platform_device *pdev)
cdd->ddev.device_tx_status = cppi41_dma_tx_status;
cdd->ddev.device_issue_pending = cppi41_dma_issue_pending;
cdd->ddev.device_prep_slave_sg = cppi41_dma_prep_slave_sg;
- cdd->ddev.device_control = cppi41_dma_control;
+ cdd->ddev.device_terminate_all = cppi41_stop_chan;
cdd->ddev.dev = dev;
INIT_LIST_HEAD(&cdd->ddev.channels);
cpp41_dma_info.dma_cap = cdd->ddev.cap_mask;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 16/61] dmaengine: cppi41: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the TI CPPI41 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/cppi41.c | 30 +-----------------------------
1 file changed, 1 insertion(+), 29 deletions(-)
diff --git a/drivers/dma/cppi41.c b/drivers/dma/cppi41.c
index cf7633f6d474..98b7dc8948a9 100644
--- a/drivers/dma/cppi41.c
+++ b/drivers/dma/cppi41.c
@@ -524,12 +524,6 @@ static struct dma_async_tx_descriptor *cppi41_dma_prep_slave_sg(
return &c->txd;
}
-static int cpp41_cfg_chan(struct cppi41_channel *c,
- struct dma_slave_config *cfg)
-{
- return 0;
-}
-
static void cppi41_compute_td_desc(struct cppi41_desc *d)
{
d->pd0 = DESC_TYPE_TEARD << DESC_TYPE;
@@ -642,28 +636,6 @@ static int cppi41_stop_chan(struct dma_chan *chan)
return 0;
}
-static int cppi41_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct cppi41_channel *c = to_cpp41_chan(chan);
- int ret;
-
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- ret = cpp41_cfg_chan(c, (struct dma_slave_config *) arg);
- break;
-
- case DMA_TERMINATE_ALL:
- ret = cppi41_stop_chan(chan);
- break;
-
- default:
- ret = -ENXIO;
- break;
- }
- return ret;
-}
-
static void cleanup_chans(struct cppi41_dd *cdd)
{
while (!list_empty(&cdd->ddev.channels)) {
@@ -948,7 +920,7 @@ static int cppi41_dma_probe(struct platform_device *pdev)
cdd->ddev.device_tx_status = cppi41_dma_tx_status;
cdd->ddev.device_issue_pending = cppi41_dma_issue_pending;
cdd->ddev.device_prep_slave_sg = cppi41_dma_prep_slave_sg;
- cdd->ddev.device_control = cppi41_dma_control;
+ cdd->ddev.device_terminate_all = cppi41_stop_chan;
cdd->ddev.dev = dev;
INIT_LIST_HEAD(&cdd->ddev.channels);
cpp41_dma_info.dma_cap = cdd->ddev.cap_mask;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 17/61] dmaengine: jz4740: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the JZ4740 DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/dma-jz4740.c | 20 +++-----------------
1 file changed, 3 insertions(+), 17 deletions(-)
diff --git a/drivers/dma/dma-jz4740.c b/drivers/dma/dma-jz4740.c
index bdeafeefa5f6..4527a3ebeac4 100644
--- a/drivers/dma/dma-jz4740.c
+++ b/drivers/dma/dma-jz4740.c
@@ -210,7 +210,7 @@ static enum jz4740_dma_transfer_size jz4740_dma_maxburst(u32 maxburst)
}
static int jz4740_dma_slave_config(struct dma_chan *c,
- const struct dma_slave_config *config)
+ struct dma_slave_config *config)
{
struct jz4740_dmaengine_chan *chan = to_jz4740_dma_chan(c);
struct jz4740_dma_dev *dmadev = jz4740_dma_chan_get_dev(chan);
@@ -290,21 +290,6 @@ static int jz4740_dma_terminate_all(struct dma_chan *c)
return 0;
}
-static int jz4740_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct dma_slave_config *config = (struct dma_slave_config *)arg;
-
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- return jz4740_dma_slave_config(chan, config);
- case DMA_TERMINATE_ALL:
- return jz4740_dma_terminate_all(chan);
- default:
- return -ENOSYS;
- }
-}
-
static int jz4740_dma_start_transfer(struct jz4740_dmaengine_chan *chan)
{
struct jz4740_dma_dev *dmadev = jz4740_dma_chan_get_dev(chan);
@@ -561,7 +546,8 @@ static int jz4740_dma_probe(struct platform_device *pdev)
dd->device_issue_pending = jz4740_dma_issue_pending;
dd->device_prep_slave_sg = jz4740_dma_prep_slave_sg;
dd->device_prep_dma_cyclic = jz4740_dma_prep_dma_cyclic;
- dd->device_control = jz4740_dma_control;
+ dd->device_config = jz4740_dma_slave_config;
+ dd->device_terminate_all = jz4740_dma_terminate_all;
dd->dev = &pdev->dev;
INIT_LIST_HEAD(&dd->channels);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 17/61] dmaengine: jz4740: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the JZ4740 DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/dma-jz4740.c | 20 +++-----------------
1 file changed, 3 insertions(+), 17 deletions(-)
diff --git a/drivers/dma/dma-jz4740.c b/drivers/dma/dma-jz4740.c
index bdeafeefa5f6..4527a3ebeac4 100644
--- a/drivers/dma/dma-jz4740.c
+++ b/drivers/dma/dma-jz4740.c
@@ -210,7 +210,7 @@ static enum jz4740_dma_transfer_size jz4740_dma_maxburst(u32 maxburst)
}
static int jz4740_dma_slave_config(struct dma_chan *c,
- const struct dma_slave_config *config)
+ struct dma_slave_config *config)
{
struct jz4740_dmaengine_chan *chan = to_jz4740_dma_chan(c);
struct jz4740_dma_dev *dmadev = jz4740_dma_chan_get_dev(chan);
@@ -290,21 +290,6 @@ static int jz4740_dma_terminate_all(struct dma_chan *c)
return 0;
}
-static int jz4740_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct dma_slave_config *config = (struct dma_slave_config *)arg;
-
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- return jz4740_dma_slave_config(chan, config);
- case DMA_TERMINATE_ALL:
- return jz4740_dma_terminate_all(chan);
- default:
- return -ENOSYS;
- }
-}
-
static int jz4740_dma_start_transfer(struct jz4740_dmaengine_chan *chan)
{
struct jz4740_dma_dev *dmadev = jz4740_dma_chan_get_dev(chan);
@@ -561,7 +546,8 @@ static int jz4740_dma_probe(struct platform_device *pdev)
dd->device_issue_pending = jz4740_dma_issue_pending;
dd->device_prep_slave_sg = jz4740_dma_prep_slave_sg;
dd->device_prep_dma_cyclic = jz4740_dma_prep_dma_cyclic;
- dd->device_control = jz4740_dma_control;
+ dd->device_config = jz4740_dma_slave_config;
+ dd->device_terminate_all = jz4740_dma_terminate_all;
dd->dev = &pdev->dev;
INIT_LIST_HEAD(&dd->channels);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 18/61] dmaengine: dw: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the DesignWare DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/dw/core.c | 82 +++++++++++++++++++++++++++------------------------
1 file changed, 44 insertions(+), 38 deletions(-)
diff --git a/drivers/dma/dw/core.c b/drivers/dma/dw/core.c
index 244722170410..0bcf3f37fd5d 100644
--- a/drivers/dma/dw/core.c
+++ b/drivers/dma/dw/core.c
@@ -954,8 +954,7 @@ static inline void convert_burst(u32 *maxburst)
*maxburst = 0;
}
-static int
-set_runtime_config(struct dma_chan *chan, struct dma_slave_config *sconfig)
+static int dwc_config(struct dma_chan *chan, struct dma_slave_config *sconfig)
{
struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
@@ -972,16 +971,25 @@ set_runtime_config(struct dma_chan *chan, struct dma_slave_config *sconfig)
return 0;
}
-static inline void dwc_chan_pause(struct dw_dma_chan *dwc)
+static int dwc_pause(struct dma_chan *chan)
{
- u32 cfglo = channel_readl(dwc, CFG_LO);
- unsigned int count = 20; /* timeout iterations */
+ struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
+ unsigned long flags;
+ unsigned int count = 20; /* timeout iterations */
+ u32 cfglo;
+
+ spin_lock_irqsave(&dwc->lock, flags);
+ cfglo = channel_readl(dwc, CFG_LO);
channel_writel(dwc, CFG_LO, cfglo | DWC_CFGL_CH_SUSP);
while (!(channel_readl(dwc, CFG_LO) & DWC_CFGL_FIFO_EMPTY) && count--)
udelay(2);
dwc->paused = true;
+
+ spin_unlock_irqrestore(&dwc->lock, flags);
+
+ return 0;
}
static inline void dwc_chan_resume(struct dw_dma_chan *dwc)
@@ -993,53 +1001,48 @@ static inline void dwc_chan_resume(struct dw_dma_chan *dwc)
dwc->paused = false;
}
-static int dwc_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int dwc_resume(struct dma_chan *chan)
{
struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
- struct dw_dma *dw = to_dw_dma(chan->device);
- struct dw_desc *desc, *_desc;
unsigned long flags;
- LIST_HEAD(list);
- if (cmd == DMA_PAUSE) {
- spin_lock_irqsave(&dwc->lock, flags);
+ if (!dwc->paused)
+ return 0;
- dwc_chan_pause(dwc);
+ spin_lock_irqsave(&dwc->lock, flags);
- spin_unlock_irqrestore(&dwc->lock, flags);
- } else if (cmd == DMA_RESUME) {
- if (!dwc->paused)
- return 0;
+ dwc_chan_resume(dwc);
- spin_lock_irqsave(&dwc->lock, flags);
+ spin_unlock_irqrestore(&dwc->lock, flags);
- dwc_chan_resume(dwc);
+ return 0;
+}
- spin_unlock_irqrestore(&dwc->lock, flags);
- } else if (cmd == DMA_TERMINATE_ALL) {
- spin_lock_irqsave(&dwc->lock, flags);
+static int dwc_terminate_all(struct dma_chan *chan)
+{
+ struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
+ struct dw_dma *dw = to_dw_dma(chan->device);
+ struct dw_desc *desc, *_desc;
+ unsigned long flags;
+ LIST_HEAD(list);
+
+ spin_lock_irqsave(&dwc->lock, flags);
- clear_bit(DW_DMA_IS_SOFT_LLP, &dwc->flags);
+ clear_bit(DW_DMA_IS_SOFT_LLP, &dwc->flags);
- dwc_chan_disable(dw, dwc);
+ dwc_chan_disable(dw, dwc);
- dwc_chan_resume(dwc);
+ dwc_chan_resume(dwc);
- /* active_list entries will end up before queued entries */
- list_splice_init(&dwc->queue, &list);
- list_splice_init(&dwc->active_list, &list);
+ /* active_list entries will end up before queued entries */
+ list_splice_init(&dwc->queue, &list);
+ list_splice_init(&dwc->active_list, &list);
- spin_unlock_irqrestore(&dwc->lock, flags);
+ spin_unlock_irqrestore(&dwc->lock, flags);
- /* Flush all pending and queued descriptors */
- list_for_each_entry_safe(desc, _desc, &list, desc_node)
- dwc_descriptor_complete(dwc, desc, false);
- } else if (cmd == DMA_SLAVE_CONFIG) {
- return set_runtime_config(chan, (struct dma_slave_config *)arg);
- } else {
- return -ENXIO;
- }
+ /* Flush all pending and queued descriptors */
+ list_for_each_entry_safe(desc, _desc, &list, desc_node)
+ dwc_descriptor_complete(dwc, desc, false);
return 0;
}
@@ -1655,7 +1658,10 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
dw->dma.device_prep_dma_memcpy = dwc_prep_dma_memcpy;
dw->dma.device_prep_slave_sg = dwc_prep_slave_sg;
- dw->dma.device_control = dwc_control;
+ dw->dma.device_config = dwc_config;
+ dw->dma.device_pause = dwc_pause;
+ dw->dma.device_resume = dwc_resume;
+ dw->dma.device_terminate_all = dwc_terminate_all;
dw->dma.device_tx_status = dwc_tx_status;
dw->dma.device_issue_pending = dwc_issue_pending;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 18/61] dmaengine: dw: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the DesignWare DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/dw/core.c | 82 +++++++++++++++++++++++++++------------------------
1 file changed, 44 insertions(+), 38 deletions(-)
diff --git a/drivers/dma/dw/core.c b/drivers/dma/dw/core.c
index 244722170410..0bcf3f37fd5d 100644
--- a/drivers/dma/dw/core.c
+++ b/drivers/dma/dw/core.c
@@ -954,8 +954,7 @@ static inline void convert_burst(u32 *maxburst)
*maxburst = 0;
}
-static int
-set_runtime_config(struct dma_chan *chan, struct dma_slave_config *sconfig)
+static int dwc_config(struct dma_chan *chan, struct dma_slave_config *sconfig)
{
struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
@@ -972,16 +971,25 @@ set_runtime_config(struct dma_chan *chan, struct dma_slave_config *sconfig)
return 0;
}
-static inline void dwc_chan_pause(struct dw_dma_chan *dwc)
+static int dwc_pause(struct dma_chan *chan)
{
- u32 cfglo = channel_readl(dwc, CFG_LO);
- unsigned int count = 20; /* timeout iterations */
+ struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
+ unsigned long flags;
+ unsigned int count = 20; /* timeout iterations */
+ u32 cfglo;
+
+ spin_lock_irqsave(&dwc->lock, flags);
+ cfglo = channel_readl(dwc, CFG_LO);
channel_writel(dwc, CFG_LO, cfglo | DWC_CFGL_CH_SUSP);
while (!(channel_readl(dwc, CFG_LO) & DWC_CFGL_FIFO_EMPTY) && count--)
udelay(2);
dwc->paused = true;
+
+ spin_unlock_irqrestore(&dwc->lock, flags);
+
+ return 0;
}
static inline void dwc_chan_resume(struct dw_dma_chan *dwc)
@@ -993,53 +1001,48 @@ static inline void dwc_chan_resume(struct dw_dma_chan *dwc)
dwc->paused = false;
}
-static int dwc_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int dwc_resume(struct dma_chan *chan)
{
struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
- struct dw_dma *dw = to_dw_dma(chan->device);
- struct dw_desc *desc, *_desc;
unsigned long flags;
- LIST_HEAD(list);
- if (cmd == DMA_PAUSE) {
- spin_lock_irqsave(&dwc->lock, flags);
+ if (!dwc->paused)
+ return 0;
- dwc_chan_pause(dwc);
+ spin_lock_irqsave(&dwc->lock, flags);
- spin_unlock_irqrestore(&dwc->lock, flags);
- } else if (cmd == DMA_RESUME) {
- if (!dwc->paused)
- return 0;
+ dwc_chan_resume(dwc);
- spin_lock_irqsave(&dwc->lock, flags);
+ spin_unlock_irqrestore(&dwc->lock, flags);
- dwc_chan_resume(dwc);
+ return 0;
+}
- spin_unlock_irqrestore(&dwc->lock, flags);
- } else if (cmd == DMA_TERMINATE_ALL) {
- spin_lock_irqsave(&dwc->lock, flags);
+static int dwc_terminate_all(struct dma_chan *chan)
+{
+ struct dw_dma_chan *dwc = to_dw_dma_chan(chan);
+ struct dw_dma *dw = to_dw_dma(chan->device);
+ struct dw_desc *desc, *_desc;
+ unsigned long flags;
+ LIST_HEAD(list);
+
+ spin_lock_irqsave(&dwc->lock, flags);
- clear_bit(DW_DMA_IS_SOFT_LLP, &dwc->flags);
+ clear_bit(DW_DMA_IS_SOFT_LLP, &dwc->flags);
- dwc_chan_disable(dw, dwc);
+ dwc_chan_disable(dw, dwc);
- dwc_chan_resume(dwc);
+ dwc_chan_resume(dwc);
- /* active_list entries will end up before queued entries */
- list_splice_init(&dwc->queue, &list);
- list_splice_init(&dwc->active_list, &list);
+ /* active_list entries will end up before queued entries */
+ list_splice_init(&dwc->queue, &list);
+ list_splice_init(&dwc->active_list, &list);
- spin_unlock_irqrestore(&dwc->lock, flags);
+ spin_unlock_irqrestore(&dwc->lock, flags);
- /* Flush all pending and queued descriptors */
- list_for_each_entry_safe(desc, _desc, &list, desc_node)
- dwc_descriptor_complete(dwc, desc, false);
- } else if (cmd == DMA_SLAVE_CONFIG) {
- return set_runtime_config(chan, (struct dma_slave_config *)arg);
- } else {
- return -ENXIO;
- }
+ /* Flush all pending and queued descriptors */
+ list_for_each_entry_safe(desc, _desc, &list, desc_node)
+ dwc_descriptor_complete(dwc, desc, false);
return 0;
}
@@ -1655,7 +1658,10 @@ int dw_dma_probe(struct dw_dma_chip *chip, struct dw_dma_platform_data *pdata)
dw->dma.device_prep_dma_memcpy = dwc_prep_dma_memcpy;
dw->dma.device_prep_slave_sg = dwc_prep_slave_sg;
- dw->dma.device_control = dwc_control;
+ dw->dma.device_config = dwc_config;
+ dw->dma.device_pause = dwc_pause;
+ dw->dma.device_resume = dwc_resume;
+ dw->dma.device_terminate_all = dwc_terminate_all;
dw->dma.device_tx_status = dwc_tx_status;
dw->dma.device_issue_pending = dwc_issue_pending;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 19/61] dmaengine: edma: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the TI EDMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/edma.c | 50 +++++++++++++++-----------------------------------
1 file changed, 15 insertions(+), 35 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 790054eef76c..56735a0aa3a8 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -244,8 +244,9 @@ static void edma_execute(struct edma_chan *echan)
}
}
-static int edma_terminate_all(struct edma_chan *echan)
+static int edma_terminate_all(struct dma_chan *chan)
{
+ struct edma_chan *echan = to_edma_chan(chan);
unsigned long flags;
LIST_HEAD(head);
@@ -273,9 +274,11 @@ static int edma_terminate_all(struct edma_chan *echan)
return 0;
}
-static int edma_slave_config(struct edma_chan *echan,
+static int edma_slave_config(struct dma_chan *chan,
struct dma_slave_config *cfg)
{
+ struct edma_chan *echan = to_edma_chan(chan);
+
if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
return -EINVAL;
@@ -285,8 +288,10 @@ static int edma_slave_config(struct edma_chan *echan,
return 0;
}
-static int edma_dma_pause(struct edma_chan *echan)
+static int edma_dma_pause(struct dma_chan *chan)
{
+ struct edma_chan *echan = to_edma_chan(chan);
+
/* Pause/Resume only allowed with cyclic mode */
if (!echan->edesc || !echan->edesc->cyclic)
return -EINVAL;
@@ -295,8 +300,10 @@ static int edma_dma_pause(struct edma_chan *echan)
return 0;
}
-static int edma_dma_resume(struct edma_chan *echan)
+static int edma_dma_resume(struct dma_chan *chan)
{
+ struct edma_chan *echan = to_edma_chan(chan);
+
/* Pause/Resume only allowed with cyclic mode */
if (!echan->edesc->cyclic)
return -EINVAL;
@@ -305,36 +312,6 @@ static int edma_dma_resume(struct edma_chan *echan)
return 0;
}
-static int edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- int ret = 0;
- struct dma_slave_config *config;
- struct edma_chan *echan = to_edma_chan(chan);
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- edma_terminate_all(echan);
- break;
- case DMA_SLAVE_CONFIG:
- config = (struct dma_slave_config *)arg;
- ret = edma_slave_config(echan, config);
- break;
- case DMA_PAUSE:
- ret = edma_dma_pause(echan);
- break;
-
- case DMA_RESUME:
- ret = edma_dma_resume(echan);
- break;
-
- default:
- ret = -ENOSYS;
- }
-
- return ret;
-}
-
/*
* A PaRAM set configuration abstraction used by other modes
* @chan: Channel who's PaRAM set we're configuring
@@ -1017,7 +994,10 @@ static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
dma->device_free_chan_resources = edma_free_chan_resources;
dma->device_issue_pending = edma_issue_pending;
dma->device_tx_status = edma_tx_status;
- dma->device_control = edma_control;
+ dma->device_config = edma_slave_config;
+ dma->device_pause = edma_dma_pause;
+ dma->device_resume = edma_dma_resume;
+ dma->device_terminate_all = edma_terminate_all;
dma->device_slave_caps = edma_dma_device_slave_caps;
dma->dev = dev;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 19/61] dmaengine: edma: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the TI EDMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/edma.c | 50 +++++++++++++++-----------------------------------
1 file changed, 15 insertions(+), 35 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 790054eef76c..56735a0aa3a8 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -244,8 +244,9 @@ static void edma_execute(struct edma_chan *echan)
}
}
-static int edma_terminate_all(struct edma_chan *echan)
+static int edma_terminate_all(struct dma_chan *chan)
{
+ struct edma_chan *echan = to_edma_chan(chan);
unsigned long flags;
LIST_HEAD(head);
@@ -273,9 +274,11 @@ static int edma_terminate_all(struct edma_chan *echan)
return 0;
}
-static int edma_slave_config(struct edma_chan *echan,
+static int edma_slave_config(struct dma_chan *chan,
struct dma_slave_config *cfg)
{
+ struct edma_chan *echan = to_edma_chan(chan);
+
if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
return -EINVAL;
@@ -285,8 +288,10 @@ static int edma_slave_config(struct edma_chan *echan,
return 0;
}
-static int edma_dma_pause(struct edma_chan *echan)
+static int edma_dma_pause(struct dma_chan *chan)
{
+ struct edma_chan *echan = to_edma_chan(chan);
+
/* Pause/Resume only allowed with cyclic mode */
if (!echan->edesc || !echan->edesc->cyclic)
return -EINVAL;
@@ -295,8 +300,10 @@ static int edma_dma_pause(struct edma_chan *echan)
return 0;
}
-static int edma_dma_resume(struct edma_chan *echan)
+static int edma_dma_resume(struct dma_chan *chan)
{
+ struct edma_chan *echan = to_edma_chan(chan);
+
/* Pause/Resume only allowed with cyclic mode */
if (!echan->edesc->cyclic)
return -EINVAL;
@@ -305,36 +312,6 @@ static int edma_dma_resume(struct edma_chan *echan)
return 0;
}
-static int edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- int ret = 0;
- struct dma_slave_config *config;
- struct edma_chan *echan = to_edma_chan(chan);
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- edma_terminate_all(echan);
- break;
- case DMA_SLAVE_CONFIG:
- config = (struct dma_slave_config *)arg;
- ret = edma_slave_config(echan, config);
- break;
- case DMA_PAUSE:
- ret = edma_dma_pause(echan);
- break;
-
- case DMA_RESUME:
- ret = edma_dma_resume(echan);
- break;
-
- default:
- ret = -ENOSYS;
- }
-
- return ret;
-}
-
/*
* A PaRAM set configuration abstraction used by other modes
* @chan: Channel who's PaRAM set we're configuring
@@ -1017,7 +994,10 @@ static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
dma->device_free_chan_resources = edma_free_chan_resources;
dma->device_issue_pending = edma_issue_pending;
dma->device_tx_status = edma_tx_status;
- dma->device_control = edma_control;
+ dma->device_config = edma_slave_config;
+ dma->device_pause = edma_dma_pause;
+ dma->device_resume = edma_dma_resume;
+ dma->device_terminate_all = edma_terminate_all;
dma->device_slave_caps = edma_dma_device_slave_caps;
dma->dev = dev;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 20/61] dmaengine: ep93xx: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Cirrus Logic EP93xx driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/ep93xx_dma.c | 41 +++++++----------------------------------
1 file changed, 7 insertions(+), 34 deletions(-)
diff --git a/drivers/dma/ep93xx_dma.c b/drivers/dma/ep93xx_dma.c
index 7650470196c4..a8bcbb5bc0ed 100644
--- a/drivers/dma/ep93xx_dma.c
+++ b/drivers/dma/ep93xx_dma.c
@@ -1164,13 +1164,14 @@ fail:
/**
* ep93xx_dma_terminate_all - terminate all transactions
- * @edmac: channel
+ * @chan: channel
*
* Stops all DMA transactions. All descriptors are put back to the
* @edmac->free_list and callbacks are _not_ called.
*/
-static int ep93xx_dma_terminate_all(struct ep93xx_dma_chan *edmac)
+static int ep93xx_dma_terminate_all(struct dma_chan *chan)
{
+ struct ep93xx_dma_chan *edmac = to_ep93xx_dma_chan(chan);
struct ep93xx_dma_desc *desc, *_d;
unsigned long flags;
LIST_HEAD(list);
@@ -1194,9 +1195,10 @@ static int ep93xx_dma_terminate_all(struct ep93xx_dma_chan *edmac)
return 0;
}
-static int ep93xx_dma_slave_config(struct ep93xx_dma_chan *edmac,
+static int ep93xx_dma_slave_config(struct dma_chan *chan,
struct dma_slave_config *config)
{
+ struct ep93xx_dma_chan *edmac = to_ep93xx_dma_chan(chan);
enum dma_slave_buswidth width;
unsigned long flags;
u32 addr, ctrl;
@@ -1242,36 +1244,6 @@ static int ep93xx_dma_slave_config(struct ep93xx_dma_chan *edmac,
}
/**
- * ep93xx_dma_control - manipulate all pending operations on a channel
- * @chan: channel
- * @cmd: control command to perform
- * @arg: optional argument
- *
- * Controls the channel. Function returns %0 in case of success or negative
- * error in case of failure.
- */
-static int ep93xx_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct ep93xx_dma_chan *edmac = to_ep93xx_dma_chan(chan);
- struct dma_slave_config *config;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- return ep93xx_dma_terminate_all(edmac);
-
- case DMA_SLAVE_CONFIG:
- config = (struct dma_slave_config *)arg;
- return ep93xx_dma_slave_config(edmac, config);
-
- default:
- break;
- }
-
- return -ENOSYS;
-}
-
-/**
* ep93xx_dma_tx_status - check if a transaction is completed
* @chan: channel
* @cookie: transaction specific cookie
@@ -1352,7 +1324,8 @@ static int __init ep93xx_dma_probe(struct platform_device *pdev)
dma_dev->device_free_chan_resources = ep93xx_dma_free_chan_resources;
dma_dev->device_prep_slave_sg = ep93xx_dma_prep_slave_sg;
dma_dev->device_prep_dma_cyclic = ep93xx_dma_prep_dma_cyclic;
- dma_dev->device_control = ep93xx_dma_control;
+ dma_dev->device_config = ep93xx_dma_slave_config;
+ dma_dev->device_terminate_all = ep93xx_dma_terminate_all;
dma_dev->device_issue_pending = ep93xx_dma_issue_pending;
dma_dev->device_tx_status = ep93xx_dma_tx_status;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 20/61] dmaengine: ep93xx: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Cirrus Logic EP93xx driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/ep93xx_dma.c | 41 +++++++----------------------------------
1 file changed, 7 insertions(+), 34 deletions(-)
diff --git a/drivers/dma/ep93xx_dma.c b/drivers/dma/ep93xx_dma.c
index 7650470196c4..a8bcbb5bc0ed 100644
--- a/drivers/dma/ep93xx_dma.c
+++ b/drivers/dma/ep93xx_dma.c
@@ -1164,13 +1164,14 @@ fail:
/**
* ep93xx_dma_terminate_all - terminate all transactions
- * @edmac: channel
+ * @chan: channel
*
* Stops all DMA transactions. All descriptors are put back to the
* @edmac->free_list and callbacks are _not_ called.
*/
-static int ep93xx_dma_terminate_all(struct ep93xx_dma_chan *edmac)
+static int ep93xx_dma_terminate_all(struct dma_chan *chan)
{
+ struct ep93xx_dma_chan *edmac = to_ep93xx_dma_chan(chan);
struct ep93xx_dma_desc *desc, *_d;
unsigned long flags;
LIST_HEAD(list);
@@ -1194,9 +1195,10 @@ static int ep93xx_dma_terminate_all(struct ep93xx_dma_chan *edmac)
return 0;
}
-static int ep93xx_dma_slave_config(struct ep93xx_dma_chan *edmac,
+static int ep93xx_dma_slave_config(struct dma_chan *chan,
struct dma_slave_config *config)
{
+ struct ep93xx_dma_chan *edmac = to_ep93xx_dma_chan(chan);
enum dma_slave_buswidth width;
unsigned long flags;
u32 addr, ctrl;
@@ -1242,36 +1244,6 @@ static int ep93xx_dma_slave_config(struct ep93xx_dma_chan *edmac,
}
/**
- * ep93xx_dma_control - manipulate all pending operations on a channel
- * @chan: channel
- * @cmd: control command to perform
- * @arg: optional argument
- *
- * Controls the channel. Function returns %0 in case of success or negative
- * error in case of failure.
- */
-static int ep93xx_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct ep93xx_dma_chan *edmac = to_ep93xx_dma_chan(chan);
- struct dma_slave_config *config;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- return ep93xx_dma_terminate_all(edmac);
-
- case DMA_SLAVE_CONFIG:
- config = (struct dma_slave_config *)arg;
- return ep93xx_dma_slave_config(edmac, config);
-
- default:
- break;
- }
-
- return -ENOSYS;
-}
-
-/**
* ep93xx_dma_tx_status - check if a transaction is completed
* @chan: channel
* @cookie: transaction specific cookie
@@ -1352,7 +1324,8 @@ static int __init ep93xx_dma_probe(struct platform_device *pdev)
dma_dev->device_free_chan_resources = ep93xx_dma_free_chan_resources;
dma_dev->device_prep_slave_sg = ep93xx_dma_prep_slave_sg;
dma_dev->device_prep_dma_cyclic = ep93xx_dma_prep_dma_cyclic;
- dma_dev->device_control = ep93xx_dma_control;
+ dma_dev->device_config = ep93xx_dma_slave_config;
+ dma_dev->device_terminate_all = ep93xx_dma_terminate_all;
dma_dev->device_issue_pending = ep93xx_dma_issue_pending;
dma_dev->device_tx_status = ep93xx_dma_tx_status;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 21/61] dmaengine: fsl-edma: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Freescale EDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/fsl-edma.c | 106 +++++++++++++++++++++++++++----------------------
1 file changed, 58 insertions(+), 48 deletions(-)
diff --git a/drivers/dma/fsl-edma.c b/drivers/dma/fsl-edma.c
index 13df85a16438..b5b7474f8859 100644
--- a/drivers/dma/fsl-edma.c
+++ b/drivers/dma/fsl-edma.c
@@ -292,62 +292,69 @@ static void fsl_edma_free_desc(struct virt_dma_desc *vdesc)
kfree(fsl_desc);
}
-static int fsl_edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int fsl_edma_terminate_all(struct dma_chan *chan)
{
struct fsl_edma_chan *fsl_chan = to_fsl_edma_chan(chan);
- struct dma_slave_config *cfg = (void *)arg;
unsigned long flags;
LIST_HEAD(head);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
+ spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
+ fsl_edma_disable_request(fsl_chan);
+ fsl_chan->edesc = NULL;
+ vchan_get_all_descriptors(&fsl_chan->vchan, &head);
+ spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
+ vchan_dma_desc_free_list(&fsl_chan->vchan, &head);
+ return 0;
+}
+
+static int fsl_edma_pause(struct dma_chan *chan)
+{
+ struct fsl_edma_chan *fsl_chan = to_fsl_edma_chan(chan);
+ unsigned long flags;
+
+ spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
+ if (fsl_chan->edesc) {
fsl_edma_disable_request(fsl_chan);
- fsl_chan->edesc = NULL;
- vchan_get_all_descriptors(&fsl_chan->vchan, &head);
- spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
- vchan_dma_desc_free_list(&fsl_chan->vchan, &head);
- return 0;
-
- case DMA_SLAVE_CONFIG:
- fsl_chan->fsc.dir = cfg->direction;
- if (cfg->direction == DMA_DEV_TO_MEM) {
- fsl_chan->fsc.dev_addr = cfg->src_addr;
- fsl_chan->fsc.addr_width = cfg->src_addr_width;
- fsl_chan->fsc.burst = cfg->src_maxburst;
- fsl_chan->fsc.attr = fsl_edma_get_tcd_attr(cfg->src_addr_width);
- } else if (cfg->direction == DMA_MEM_TO_DEV) {
- fsl_chan->fsc.dev_addr = cfg->dst_addr;
- fsl_chan->fsc.addr_width = cfg->dst_addr_width;
- fsl_chan->fsc.burst = cfg->dst_maxburst;
- fsl_chan->fsc.attr = fsl_edma_get_tcd_attr(cfg->dst_addr_width);
- } else {
- return -EINVAL;
- }
- return 0;
+ fsl_chan->status = DMA_PAUSED;
+ }
+ spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
+ return 0;
+}
- case DMA_PAUSE:
- spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
- if (fsl_chan->edesc) {
- fsl_edma_disable_request(fsl_chan);
- fsl_chan->status = DMA_PAUSED;
- }
- spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
- return 0;
-
- case DMA_RESUME:
- spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
- if (fsl_chan->edesc) {
- fsl_edma_enable_request(fsl_chan);
- fsl_chan->status = DMA_IN_PROGRESS;
- }
- spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
- return 0;
+static int fsl_edma_resume(struct dma_chan *chan)
+{
+ struct fsl_edma_chan *fsl_chan = to_fsl_edma_chan(chan);
+ unsigned long flags;
- default:
- return -ENXIO;
+ spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
+ if (fsl_chan->edesc) {
+ fsl_edma_enable_request(fsl_chan);
+ fsl_chan->status = DMA_IN_PROGRESS;
}
+ spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
+ return 0;
+}
+
+static int fsl_edma_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
+{
+ struct fsl_edma_chan *fsl_chan = to_fsl_edma_chan(chan);
+
+ fsl_chan->fsc.dir = cfg->direction;
+ if (cfg->direction == DMA_DEV_TO_MEM) {
+ fsl_chan->fsc.dev_addr = cfg->src_addr;
+ fsl_chan->fsc.addr_width = cfg->src_addr_width;
+ fsl_chan->fsc.burst = cfg->src_maxburst;
+ fsl_chan->fsc.attr = fsl_edma_get_tcd_attr(cfg->src_addr_width);
+ } else if (cfg->direction == DMA_MEM_TO_DEV) {
+ fsl_chan->fsc.dev_addr = cfg->dst_addr;
+ fsl_chan->fsc.addr_width = cfg->dst_addr_width;
+ fsl_chan->fsc.burst = cfg->dst_maxburst;
+ fsl_chan->fsc.attr = fsl_edma_get_tcd_attr(cfg->dst_addr_width);
+ } else {
+ return -EINVAL;
+ }
+ return 0;
}
static size_t fsl_edma_desc_residue(struct fsl_edma_chan *fsl_chan,
@@ -914,7 +921,10 @@ static int fsl_edma_probe(struct platform_device *pdev)
fsl_edma->dma_dev.device_tx_status = fsl_edma_tx_status;
fsl_edma->dma_dev.device_prep_slave_sg = fsl_edma_prep_slave_sg;
fsl_edma->dma_dev.device_prep_dma_cyclic = fsl_edma_prep_dma_cyclic;
- fsl_edma->dma_dev.device_control = fsl_edma_control;
+ fsl_edma->dma_dev.device_config = fsl_edma_slave_config;
+ fsl_edma->dma_dev.device_pause = fsl_edma_pause;
+ fsl_edma->dma_dev.device_resume = fsl_edma_resume;
+ fsl_edma->dma_dev.device_terminate_all = fsl_edma_terminate_all;
fsl_edma->dma_dev.device_issue_pending = fsl_edma_issue_pending;
fsl_edma->dma_dev.device_slave_caps = fsl_dma_device_slave_caps;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 21/61] dmaengine: fsl-edma: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Freescale EDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/fsl-edma.c | 106 +++++++++++++++++++++++++++----------------------
1 file changed, 58 insertions(+), 48 deletions(-)
diff --git a/drivers/dma/fsl-edma.c b/drivers/dma/fsl-edma.c
index 13df85a16438..b5b7474f8859 100644
--- a/drivers/dma/fsl-edma.c
+++ b/drivers/dma/fsl-edma.c
@@ -292,62 +292,69 @@ static void fsl_edma_free_desc(struct virt_dma_desc *vdesc)
kfree(fsl_desc);
}
-static int fsl_edma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int fsl_edma_terminate_all(struct dma_chan *chan)
{
struct fsl_edma_chan *fsl_chan = to_fsl_edma_chan(chan);
- struct dma_slave_config *cfg = (void *)arg;
unsigned long flags;
LIST_HEAD(head);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
+ spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
+ fsl_edma_disable_request(fsl_chan);
+ fsl_chan->edesc = NULL;
+ vchan_get_all_descriptors(&fsl_chan->vchan, &head);
+ spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
+ vchan_dma_desc_free_list(&fsl_chan->vchan, &head);
+ return 0;
+}
+
+static int fsl_edma_pause(struct dma_chan *chan)
+{
+ struct fsl_edma_chan *fsl_chan = to_fsl_edma_chan(chan);
+ unsigned long flags;
+
+ spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
+ if (fsl_chan->edesc) {
fsl_edma_disable_request(fsl_chan);
- fsl_chan->edesc = NULL;
- vchan_get_all_descriptors(&fsl_chan->vchan, &head);
- spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
- vchan_dma_desc_free_list(&fsl_chan->vchan, &head);
- return 0;
-
- case DMA_SLAVE_CONFIG:
- fsl_chan->fsc.dir = cfg->direction;
- if (cfg->direction == DMA_DEV_TO_MEM) {
- fsl_chan->fsc.dev_addr = cfg->src_addr;
- fsl_chan->fsc.addr_width = cfg->src_addr_width;
- fsl_chan->fsc.burst = cfg->src_maxburst;
- fsl_chan->fsc.attr = fsl_edma_get_tcd_attr(cfg->src_addr_width);
- } else if (cfg->direction == DMA_MEM_TO_DEV) {
- fsl_chan->fsc.dev_addr = cfg->dst_addr;
- fsl_chan->fsc.addr_width = cfg->dst_addr_width;
- fsl_chan->fsc.burst = cfg->dst_maxburst;
- fsl_chan->fsc.attr = fsl_edma_get_tcd_attr(cfg->dst_addr_width);
- } else {
- return -EINVAL;
- }
- return 0;
+ fsl_chan->status = DMA_PAUSED;
+ }
+ spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
+ return 0;
+}
- case DMA_PAUSE:
- spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
- if (fsl_chan->edesc) {
- fsl_edma_disable_request(fsl_chan);
- fsl_chan->status = DMA_PAUSED;
- }
- spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
- return 0;
-
- case DMA_RESUME:
- spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
- if (fsl_chan->edesc) {
- fsl_edma_enable_request(fsl_chan);
- fsl_chan->status = DMA_IN_PROGRESS;
- }
- spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
- return 0;
+static int fsl_edma_resume(struct dma_chan *chan)
+{
+ struct fsl_edma_chan *fsl_chan = to_fsl_edma_chan(chan);
+ unsigned long flags;
- default:
- return -ENXIO;
+ spin_lock_irqsave(&fsl_chan->vchan.lock, flags);
+ if (fsl_chan->edesc) {
+ fsl_edma_enable_request(fsl_chan);
+ fsl_chan->status = DMA_IN_PROGRESS;
}
+ spin_unlock_irqrestore(&fsl_chan->vchan.lock, flags);
+ return 0;
+}
+
+static int fsl_edma_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
+{
+ struct fsl_edma_chan *fsl_chan = to_fsl_edma_chan(chan);
+
+ fsl_chan->fsc.dir = cfg->direction;
+ if (cfg->direction == DMA_DEV_TO_MEM) {
+ fsl_chan->fsc.dev_addr = cfg->src_addr;
+ fsl_chan->fsc.addr_width = cfg->src_addr_width;
+ fsl_chan->fsc.burst = cfg->src_maxburst;
+ fsl_chan->fsc.attr = fsl_edma_get_tcd_attr(cfg->src_addr_width);
+ } else if (cfg->direction == DMA_MEM_TO_DEV) {
+ fsl_chan->fsc.dev_addr = cfg->dst_addr;
+ fsl_chan->fsc.addr_width = cfg->dst_addr_width;
+ fsl_chan->fsc.burst = cfg->dst_maxburst;
+ fsl_chan->fsc.attr = fsl_edma_get_tcd_attr(cfg->dst_addr_width);
+ } else {
+ return -EINVAL;
+ }
+ return 0;
}
static size_t fsl_edma_desc_residue(struct fsl_edma_chan *fsl_chan,
@@ -914,7 +921,10 @@ static int fsl_edma_probe(struct platform_device *pdev)
fsl_edma->dma_dev.device_tx_status = fsl_edma_tx_status;
fsl_edma->dma_dev.device_prep_slave_sg = fsl_edma_prep_slave_sg;
fsl_edma->dma_dev.device_prep_dma_cyclic = fsl_edma_prep_dma_cyclic;
- fsl_edma->dma_dev.device_control = fsl_edma_control;
+ fsl_edma->dma_dev.device_config = fsl_edma_slave_config;
+ fsl_edma->dma_dev.device_pause = fsl_edma_pause;
+ fsl_edma->dma_dev.device_resume = fsl_edma_resume;
+ fsl_edma->dma_dev.device_terminate_all = fsl_edma_terminate_all;
fsl_edma->dma_dev.device_issue_pending = fsl_edma_issue_pending;
fsl_edma->dma_dev.device_slave_caps = fsl_dma_device_slave_caps;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 22/61] dmaengine: imx: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Freescale IMX DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/imx-dma.c | 103 +++++++++++++++++++++++++-------------------------
1 file changed, 51 insertions(+), 52 deletions(-)
diff --git a/drivers/dma/imx-dma.c b/drivers/dma/imx-dma.c
index 9d2c9e7374dc..dfd1557384ca 100644
--- a/drivers/dma/imx-dma.c
+++ b/drivers/dma/imx-dma.c
@@ -669,69 +669,67 @@ out:
}
-static int imxdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int imxdma_terminate_all(struct dma_chan *chan)
{
struct imxdma_channel *imxdmac = to_imxdma_chan(chan);
- struct dma_slave_config *dmaengine_cfg = (void *)arg;
struct imxdma_engine *imxdma = imxdmac->imxdma;
unsigned long flags;
- unsigned int mode = 0;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- imxdma_disable_hw(imxdmac);
- spin_lock_irqsave(&imxdma->lock, flags);
- list_splice_tail_init(&imxdmac->ld_active, &imxdmac->ld_free);
- list_splice_tail_init(&imxdmac->ld_queue, &imxdmac->ld_free);
- spin_unlock_irqrestore(&imxdma->lock, flags);
- return 0;
- case DMA_SLAVE_CONFIG:
- if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
- imxdmac->per_address = dmaengine_cfg->src_addr;
- imxdmac->watermark_level = dmaengine_cfg->src_maxburst;
- imxdmac->word_size = dmaengine_cfg->src_addr_width;
- } else {
- imxdmac->per_address = dmaengine_cfg->dst_addr;
- imxdmac->watermark_level = dmaengine_cfg->dst_maxburst;
- imxdmac->word_size = dmaengine_cfg->dst_addr_width;
- }
-
- switch (imxdmac->word_size) {
- case DMA_SLAVE_BUSWIDTH_1_BYTE:
- mode = IMX_DMA_MEMSIZE_8;
- break;
- case DMA_SLAVE_BUSWIDTH_2_BYTES:
- mode = IMX_DMA_MEMSIZE_16;
- break;
- default:
- case DMA_SLAVE_BUSWIDTH_4_BYTES:
- mode = IMX_DMA_MEMSIZE_32;
- break;
- }
+ imxdma_disable_hw(imxdmac);
- imxdmac->hw_chaining = 0;
+ spin_lock_irqsave(&imxdma->lock, flags);
+ list_splice_tail_init(&imxdmac->ld_active, &imxdmac->ld_free);
+ list_splice_tail_init(&imxdmac->ld_queue, &imxdmac->ld_free);
+ spin_unlock_irqrestore(&imxdma->lock, flags);
+ return 0;
+}
- imxdmac->ccr_from_device = (mode | IMX_DMA_TYPE_FIFO) |
- ((IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR) << 2) |
- CCR_REN;
- imxdmac->ccr_to_device =
- (IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR) |
- ((mode | IMX_DMA_TYPE_FIFO) << 2) | CCR_REN;
- imx_dmav1_writel(imxdma, imxdmac->dma_request,
- DMA_RSSR(imxdmac->channel));
+static int imxdma_config(struct dma_chan *chan,
+ struct dma_slave_config *dmaengine_cfg)
+{
+ struct imxdma_channel *imxdmac = to_imxdma_chan(chan);
+ struct imxdma_engine *imxdma = imxdmac->imxdma;
+ unsigned int mode = 0;
- /* Set burst length */
- imx_dmav1_writel(imxdma, imxdmac->watermark_level *
- imxdmac->word_size, DMA_BLR(imxdmac->channel));
+ if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
+ imxdmac->per_address = dmaengine_cfg->src_addr;
+ imxdmac->watermark_level = dmaengine_cfg->src_maxburst;
+ imxdmac->word_size = dmaengine_cfg->src_addr_width;
+ } else {
+ imxdmac->per_address = dmaengine_cfg->dst_addr;
+ imxdmac->watermark_level = dmaengine_cfg->dst_maxburst;
+ imxdmac->word_size = dmaengine_cfg->dst_addr_width;
+ }
- return 0;
+ switch (imxdmac->word_size) {
+ case DMA_SLAVE_BUSWIDTH_1_BYTE:
+ mode = IMX_DMA_MEMSIZE_8;
+ break;
+ case DMA_SLAVE_BUSWIDTH_2_BYTES:
+ mode = IMX_DMA_MEMSIZE_16;
+ break;
default:
- return -ENOSYS;
+ case DMA_SLAVE_BUSWIDTH_4_BYTES:
+ mode = IMX_DMA_MEMSIZE_32;
+ break;
}
- return -EINVAL;
+ imxdmac->hw_chaining = 0;
+
+ imxdmac->ccr_from_device = (mode | IMX_DMA_TYPE_FIFO) |
+ ((IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR) << 2) |
+ CCR_REN;
+ imxdmac->ccr_to_device =
+ (IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR) |
+ ((mode | IMX_DMA_TYPE_FIFO) << 2) | CCR_REN;
+ imx_dmav1_writel(imxdma, imxdmac->dma_request,
+ DMA_RSSR(imxdmac->channel));
+
+ /* Set burst length */
+ imx_dmav1_writel(imxdma, imxdmac->watermark_level *
+ imxdmac->word_size, DMA_BLR(imxdmac->channel));
+
+ return 0;
}
static enum dma_status imxdma_tx_status(struct dma_chan *chan,
@@ -1184,7 +1182,8 @@ static int __init imxdma_probe(struct platform_device *pdev)
imxdma->dma_device.device_prep_dma_cyclic = imxdma_prep_dma_cyclic;
imxdma->dma_device.device_prep_dma_memcpy = imxdma_prep_dma_memcpy;
imxdma->dma_device.device_prep_interleaved_dma = imxdma_prep_dma_interleaved;
- imxdma->dma_device.device_control = imxdma_control;
+ imxdma->dma_device.device_config = imxdma_config;
+ imxdma->dma_device.device_terminate_all = imxdma_terminate_all;
imxdma->dma_device.device_issue_pending = imxdma_issue_pending;
platform_set_drvdata(pdev, imxdma);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 22/61] dmaengine: imx: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Freescale IMX DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/imx-dma.c | 103 +++++++++++++++++++++++++-------------------------
1 file changed, 51 insertions(+), 52 deletions(-)
diff --git a/drivers/dma/imx-dma.c b/drivers/dma/imx-dma.c
index 9d2c9e7374dc..dfd1557384ca 100644
--- a/drivers/dma/imx-dma.c
+++ b/drivers/dma/imx-dma.c
@@ -669,69 +669,67 @@ out:
}
-static int imxdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int imxdma_terminate_all(struct dma_chan *chan)
{
struct imxdma_channel *imxdmac = to_imxdma_chan(chan);
- struct dma_slave_config *dmaengine_cfg = (void *)arg;
struct imxdma_engine *imxdma = imxdmac->imxdma;
unsigned long flags;
- unsigned int mode = 0;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- imxdma_disable_hw(imxdmac);
- spin_lock_irqsave(&imxdma->lock, flags);
- list_splice_tail_init(&imxdmac->ld_active, &imxdmac->ld_free);
- list_splice_tail_init(&imxdmac->ld_queue, &imxdmac->ld_free);
- spin_unlock_irqrestore(&imxdma->lock, flags);
- return 0;
- case DMA_SLAVE_CONFIG:
- if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
- imxdmac->per_address = dmaengine_cfg->src_addr;
- imxdmac->watermark_level = dmaengine_cfg->src_maxburst;
- imxdmac->word_size = dmaengine_cfg->src_addr_width;
- } else {
- imxdmac->per_address = dmaengine_cfg->dst_addr;
- imxdmac->watermark_level = dmaengine_cfg->dst_maxburst;
- imxdmac->word_size = dmaengine_cfg->dst_addr_width;
- }
-
- switch (imxdmac->word_size) {
- case DMA_SLAVE_BUSWIDTH_1_BYTE:
- mode = IMX_DMA_MEMSIZE_8;
- break;
- case DMA_SLAVE_BUSWIDTH_2_BYTES:
- mode = IMX_DMA_MEMSIZE_16;
- break;
- default:
- case DMA_SLAVE_BUSWIDTH_4_BYTES:
- mode = IMX_DMA_MEMSIZE_32;
- break;
- }
+ imxdma_disable_hw(imxdmac);
- imxdmac->hw_chaining = 0;
+ spin_lock_irqsave(&imxdma->lock, flags);
+ list_splice_tail_init(&imxdmac->ld_active, &imxdmac->ld_free);
+ list_splice_tail_init(&imxdmac->ld_queue, &imxdmac->ld_free);
+ spin_unlock_irqrestore(&imxdma->lock, flags);
+ return 0;
+}
- imxdmac->ccr_from_device = (mode | IMX_DMA_TYPE_FIFO) |
- ((IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR) << 2) |
- CCR_REN;
- imxdmac->ccr_to_device =
- (IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR) |
- ((mode | IMX_DMA_TYPE_FIFO) << 2) | CCR_REN;
- imx_dmav1_writel(imxdma, imxdmac->dma_request,
- DMA_RSSR(imxdmac->channel));
+static int imxdma_config(struct dma_chan *chan,
+ struct dma_slave_config *dmaengine_cfg)
+{
+ struct imxdma_channel *imxdmac = to_imxdma_chan(chan);
+ struct imxdma_engine *imxdma = imxdmac->imxdma;
+ unsigned int mode = 0;
- /* Set burst length */
- imx_dmav1_writel(imxdma, imxdmac->watermark_level *
- imxdmac->word_size, DMA_BLR(imxdmac->channel));
+ if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
+ imxdmac->per_address = dmaengine_cfg->src_addr;
+ imxdmac->watermark_level = dmaengine_cfg->src_maxburst;
+ imxdmac->word_size = dmaengine_cfg->src_addr_width;
+ } else {
+ imxdmac->per_address = dmaengine_cfg->dst_addr;
+ imxdmac->watermark_level = dmaengine_cfg->dst_maxburst;
+ imxdmac->word_size = dmaengine_cfg->dst_addr_width;
+ }
- return 0;
+ switch (imxdmac->word_size) {
+ case DMA_SLAVE_BUSWIDTH_1_BYTE:
+ mode = IMX_DMA_MEMSIZE_8;
+ break;
+ case DMA_SLAVE_BUSWIDTH_2_BYTES:
+ mode = IMX_DMA_MEMSIZE_16;
+ break;
default:
- return -ENOSYS;
+ case DMA_SLAVE_BUSWIDTH_4_BYTES:
+ mode = IMX_DMA_MEMSIZE_32;
+ break;
}
- return -EINVAL;
+ imxdmac->hw_chaining = 0;
+
+ imxdmac->ccr_from_device = (mode | IMX_DMA_TYPE_FIFO) |
+ ((IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR) << 2) |
+ CCR_REN;
+ imxdmac->ccr_to_device =
+ (IMX_DMA_MEMSIZE_32 | IMX_DMA_TYPE_LINEAR) |
+ ((mode | IMX_DMA_TYPE_FIFO) << 2) | CCR_REN;
+ imx_dmav1_writel(imxdma, imxdmac->dma_request,
+ DMA_RSSR(imxdmac->channel));
+
+ /* Set burst length */
+ imx_dmav1_writel(imxdma, imxdmac->watermark_level *
+ imxdmac->word_size, DMA_BLR(imxdmac->channel));
+
+ return 0;
}
static enum dma_status imxdma_tx_status(struct dma_chan *chan,
@@ -1184,7 +1182,8 @@ static int __init imxdma_probe(struct platform_device *pdev)
imxdma->dma_device.device_prep_dma_cyclic = imxdma_prep_dma_cyclic;
imxdma->dma_device.device_prep_dma_memcpy = imxdma_prep_dma_memcpy;
imxdma->dma_device.device_prep_interleaved_dma = imxdma_prep_dma_interleaved;
- imxdma->dma_device.device_control = imxdma_control;
+ imxdma->dma_device.device_config = imxdma_config;
+ imxdma->dma_device.device_terminate_all = imxdma_terminate_all;
imxdma->dma_device.device_issue_pending = imxdma_issue_pending;
platform_set_drvdata(pdev, imxdma);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 23/61] dmaengine: imx-sdma: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Freescale IMX SDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/imx-sdma.c | 66 +++++++++++++++++++++++---------------------------
1 file changed, 30 insertions(+), 36 deletions(-)
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
index 5b38f2bbd42a..d0f9364ab06c 100644
--- a/drivers/dma/imx-sdma.c
+++ b/drivers/dma/imx-sdma.c
@@ -830,20 +830,29 @@ static int sdma_load_context(struct sdma_channel *sdmac)
return ret;
}
-static void sdma_disable_channel(struct sdma_channel *sdmac)
+static struct sdma_channel *to_sdma_chan(struct dma_chan *chan)
+{
+ return container_of(chan, struct sdma_channel, chan);
+}
+
+static int sdma_disable_channel(struct dma_chan *chan)
{
+ struct sdma_channel *sdmac = to_sdma_chan(chan);
struct sdma_engine *sdma = sdmac->sdma;
int channel = sdmac->channel;
writel_relaxed(BIT(channel), sdma->regs + SDMA_H_STATSTOP);
sdmac->status = DMA_ERROR;
+
+ return 0;
}
-static int sdma_config_channel(struct sdma_channel *sdmac)
+static int sdma_config_channel(struct dma_chan *chan)
{
+ struct sdma_channel *sdmac = to_sdma_chan(chan);
int ret;
- sdma_disable_channel(sdmac);
+ sdma_disable_channel(chan);
sdmac->event_mask[0] = 0;
sdmac->event_mask[1] = 0;
@@ -935,11 +944,6 @@ out:
return ret;
}
-static struct sdma_channel *to_sdma_chan(struct dma_chan *chan)
-{
- return container_of(chan, struct sdma_channel, chan);
-}
-
static dma_cookie_t sdma_tx_submit(struct dma_async_tx_descriptor *tx)
{
unsigned long flags;
@@ -1004,7 +1008,7 @@ static void sdma_free_chan_resources(struct dma_chan *chan)
struct sdma_channel *sdmac = to_sdma_chan(chan);
struct sdma_engine *sdma = sdmac->sdma;
- sdma_disable_channel(sdmac);
+ sdma_disable_channel(chan);
if (sdmac->event_id0)
sdma_event_disable(sdmac, sdmac->event_id0);
@@ -1203,35 +1207,24 @@ err_out:
return NULL;
}
-static int sdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int sdma_config(struct dma_chan *chan,
+ struct dma_slave_config *dmaengine_cfg)
{
struct sdma_channel *sdmac = to_sdma_chan(chan);
- struct dma_slave_config *dmaengine_cfg = (void *)arg;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- sdma_disable_channel(sdmac);
- return 0;
- case DMA_SLAVE_CONFIG:
- if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
- sdmac->per_address = dmaengine_cfg->src_addr;
- sdmac->watermark_level = dmaengine_cfg->src_maxburst *
- dmaengine_cfg->src_addr_width;
- sdmac->word_size = dmaengine_cfg->src_addr_width;
- } else {
- sdmac->per_address = dmaengine_cfg->dst_addr;
- sdmac->watermark_level = dmaengine_cfg->dst_maxburst *
- dmaengine_cfg->dst_addr_width;
- sdmac->word_size = dmaengine_cfg->dst_addr_width;
- }
- sdmac->direction = dmaengine_cfg->direction;
- return sdma_config_channel(sdmac);
- default:
- return -ENOSYS;
- }
- return -EINVAL;
+ if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
+ sdmac->per_address = dmaengine_cfg->src_addr;
+ sdmac->watermark_level = dmaengine_cfg->src_maxburst *
+ dmaengine_cfg->src_addr_width;
+ sdmac->word_size = dmaengine_cfg->src_addr_width;
+ } else {
+ sdmac->per_address = dmaengine_cfg->dst_addr;
+ sdmac->watermark_level = dmaengine_cfg->dst_maxburst *
+ dmaengine_cfg->dst_addr_width;
+ sdmac->word_size = dmaengine_cfg->dst_addr_width;
+ }
+ sdmac->direction = dmaengine_cfg->direction;
+ return sdma_config_channel(chan);
}
static enum dma_status sdma_tx_status(struct dma_chan *chan,
@@ -1599,7 +1592,8 @@ static int sdma_probe(struct platform_device *pdev)
sdma->dma_device.device_tx_status = sdma_tx_status;
sdma->dma_device.device_prep_slave_sg = sdma_prep_slave_sg;
sdma->dma_device.device_prep_dma_cyclic = sdma_prep_dma_cyclic;
- sdma->dma_device.device_control = sdma_control;
+ sdma->dma_device.device_config = sdma_config;
+ sdma->dma_device.device_terminate_all = sdma_disable_channel;
sdma->dma_device.device_issue_pending = sdma_issue_pending;
sdma->dma_device.dev->dma_parms = &sdma->dma_parms;
dma_set_max_seg_size(sdma->dma_device.dev, 65535);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 23/61] dmaengine: imx-sdma: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Freescale IMX SDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/imx-sdma.c | 66 +++++++++++++++++++++++---------------------------
1 file changed, 30 insertions(+), 36 deletions(-)
diff --git a/drivers/dma/imx-sdma.c b/drivers/dma/imx-sdma.c
index 5b38f2bbd42a..d0f9364ab06c 100644
--- a/drivers/dma/imx-sdma.c
+++ b/drivers/dma/imx-sdma.c
@@ -830,20 +830,29 @@ static int sdma_load_context(struct sdma_channel *sdmac)
return ret;
}
-static void sdma_disable_channel(struct sdma_channel *sdmac)
+static struct sdma_channel *to_sdma_chan(struct dma_chan *chan)
+{
+ return container_of(chan, struct sdma_channel, chan);
+}
+
+static int sdma_disable_channel(struct dma_chan *chan)
{
+ struct sdma_channel *sdmac = to_sdma_chan(chan);
struct sdma_engine *sdma = sdmac->sdma;
int channel = sdmac->channel;
writel_relaxed(BIT(channel), sdma->regs + SDMA_H_STATSTOP);
sdmac->status = DMA_ERROR;
+
+ return 0;
}
-static int sdma_config_channel(struct sdma_channel *sdmac)
+static int sdma_config_channel(struct dma_chan *chan)
{
+ struct sdma_channel *sdmac = to_sdma_chan(chan);
int ret;
- sdma_disable_channel(sdmac);
+ sdma_disable_channel(chan);
sdmac->event_mask[0] = 0;
sdmac->event_mask[1] = 0;
@@ -935,11 +944,6 @@ out:
return ret;
}
-static struct sdma_channel *to_sdma_chan(struct dma_chan *chan)
-{
- return container_of(chan, struct sdma_channel, chan);
-}
-
static dma_cookie_t sdma_tx_submit(struct dma_async_tx_descriptor *tx)
{
unsigned long flags;
@@ -1004,7 +1008,7 @@ static void sdma_free_chan_resources(struct dma_chan *chan)
struct sdma_channel *sdmac = to_sdma_chan(chan);
struct sdma_engine *sdma = sdmac->sdma;
- sdma_disable_channel(sdmac);
+ sdma_disable_channel(chan);
if (sdmac->event_id0)
sdma_event_disable(sdmac, sdmac->event_id0);
@@ -1203,35 +1207,24 @@ err_out:
return NULL;
}
-static int sdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int sdma_config(struct dma_chan *chan,
+ struct dma_slave_config *dmaengine_cfg)
{
struct sdma_channel *sdmac = to_sdma_chan(chan);
- struct dma_slave_config *dmaengine_cfg = (void *)arg;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- sdma_disable_channel(sdmac);
- return 0;
- case DMA_SLAVE_CONFIG:
- if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
- sdmac->per_address = dmaengine_cfg->src_addr;
- sdmac->watermark_level = dmaengine_cfg->src_maxburst *
- dmaengine_cfg->src_addr_width;
- sdmac->word_size = dmaengine_cfg->src_addr_width;
- } else {
- sdmac->per_address = dmaengine_cfg->dst_addr;
- sdmac->watermark_level = dmaengine_cfg->dst_maxburst *
- dmaengine_cfg->dst_addr_width;
- sdmac->word_size = dmaengine_cfg->dst_addr_width;
- }
- sdmac->direction = dmaengine_cfg->direction;
- return sdma_config_channel(sdmac);
- default:
- return -ENOSYS;
- }
- return -EINVAL;
+ if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
+ sdmac->per_address = dmaengine_cfg->src_addr;
+ sdmac->watermark_level = dmaengine_cfg->src_maxburst *
+ dmaengine_cfg->src_addr_width;
+ sdmac->word_size = dmaengine_cfg->src_addr_width;
+ } else {
+ sdmac->per_address = dmaengine_cfg->dst_addr;
+ sdmac->watermark_level = dmaengine_cfg->dst_maxburst *
+ dmaengine_cfg->dst_addr_width;
+ sdmac->word_size = dmaengine_cfg->dst_addr_width;
+ }
+ sdmac->direction = dmaengine_cfg->direction;
+ return sdma_config_channel(chan);
}
static enum dma_status sdma_tx_status(struct dma_chan *chan,
@@ -1599,7 +1592,8 @@ static int sdma_probe(struct platform_device *pdev)
sdma->dma_device.device_tx_status = sdma_tx_status;
sdma->dma_device.device_prep_slave_sg = sdma_prep_slave_sg;
sdma->dma_device.device_prep_dma_cyclic = sdma_prep_dma_cyclic;
- sdma->dma_device.device_control = sdma_control;
+ sdma->dma_device.device_config = sdma_config;
+ sdma->dma_device.device_terminate_all = sdma_disable_channel;
sdma->dma_device.device_issue_pending = sdma_issue_pending;
sdma->dma_device.dev->dma_parms = &sdma->dma_parms;
dma_set_max_seg_size(sdma->dma_device.dev, 65535);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 24/61] dmaengine: intel-mid-dma: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Intel MID DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/intel_mid_dma.c | 25 ++++++-------------------
1 file changed, 6 insertions(+), 19 deletions(-)
diff --git a/drivers/dma/intel_mid_dma.c b/drivers/dma/intel_mid_dma.c
index 1aab8130efa1..5aaead9b56f7 100644
--- a/drivers/dma/intel_mid_dma.c
+++ b/drivers/dma/intel_mid_dma.c
@@ -492,10 +492,10 @@ static enum dma_status intel_mid_dma_tx_status(struct dma_chan *chan,
return ret;
}
-static int dma_slave_control(struct dma_chan *chan, unsigned long arg)
+static int intel_mid_dma_config(struct dma_chan *chan,
+ struct dma_slave_config *slave)
{
struct intel_mid_dma_chan *midc = to_intel_mid_dma_chan(chan);
- struct dma_slave_config *slave = (struct dma_slave_config *)arg;
struct intel_mid_dma_slave *mid_slave;
BUG_ON(!midc);
@@ -509,28 +509,14 @@ static int dma_slave_control(struct dma_chan *chan, unsigned long arg)
midc->mid_slave = mid_slave;
return 0;
}
-/**
- * intel_mid_dma_device_control - DMA device control
- * @chan: chan for DMA control
- * @cmd: control cmd
- * @arg: cmd arg value
- *
- * Perform DMA control command
- */
-static int intel_mid_dma_device_control(struct dma_chan *chan,
- enum dma_ctrl_cmd cmd, unsigned long arg)
+
+static int intel_mid_dma_terminate_all(struct dma_chan *chan)
{
struct intel_mid_dma_chan *midc = to_intel_mid_dma_chan(chan);
struct middma_device *mid = to_middma_device(chan->device);
struct intel_mid_dma_desc *desc, *_desc;
union intel_mid_dma_cfg_lo cfg_lo;
- if (cmd == DMA_SLAVE_CONFIG)
- return dma_slave_control(chan, arg);
-
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
spin_lock_bh(&midc->lock);
if (midc->busy == false) {
spin_unlock_bh(&midc->lock);
@@ -1148,7 +1134,8 @@ static int mid_setup_dma(struct pci_dev *pdev)
dma->common.device_prep_dma_memcpy = intel_mid_dma_prep_memcpy;
dma->common.device_issue_pending = intel_mid_dma_issue_pending;
dma->common.device_prep_slave_sg = intel_mid_dma_prep_slave_sg;
- dma->common.device_control = intel_mid_dma_device_control;
+ dma->common.device_config = intel_mid_dma_config;
+ dma->common.device_terminate_all = intel_mid_dma_terminate_all;
/*enable dma cntrl*/
iowrite32(REG_BIT0, dma->dma_base + DMA_CFG);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 24/61] dmaengine: intel-mid-dma: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Intel MID DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/intel_mid_dma.c | 25 ++++++-------------------
1 file changed, 6 insertions(+), 19 deletions(-)
diff --git a/drivers/dma/intel_mid_dma.c b/drivers/dma/intel_mid_dma.c
index 1aab8130efa1..5aaead9b56f7 100644
--- a/drivers/dma/intel_mid_dma.c
+++ b/drivers/dma/intel_mid_dma.c
@@ -492,10 +492,10 @@ static enum dma_status intel_mid_dma_tx_status(struct dma_chan *chan,
return ret;
}
-static int dma_slave_control(struct dma_chan *chan, unsigned long arg)
+static int intel_mid_dma_config(struct dma_chan *chan,
+ struct dma_slave_config *slave)
{
struct intel_mid_dma_chan *midc = to_intel_mid_dma_chan(chan);
- struct dma_slave_config *slave = (struct dma_slave_config *)arg;
struct intel_mid_dma_slave *mid_slave;
BUG_ON(!midc);
@@ -509,28 +509,14 @@ static int dma_slave_control(struct dma_chan *chan, unsigned long arg)
midc->mid_slave = mid_slave;
return 0;
}
-/**
- * intel_mid_dma_device_control - DMA device control
- * @chan: chan for DMA control
- * @cmd: control cmd
- * @arg: cmd arg value
- *
- * Perform DMA control command
- */
-static int intel_mid_dma_device_control(struct dma_chan *chan,
- enum dma_ctrl_cmd cmd, unsigned long arg)
+
+static int intel_mid_dma_terminate_all(struct dma_chan *chan)
{
struct intel_mid_dma_chan *midc = to_intel_mid_dma_chan(chan);
struct middma_device *mid = to_middma_device(chan->device);
struct intel_mid_dma_desc *desc, *_desc;
union intel_mid_dma_cfg_lo cfg_lo;
- if (cmd == DMA_SLAVE_CONFIG)
- return dma_slave_control(chan, arg);
-
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
spin_lock_bh(&midc->lock);
if (midc->busy == false) {
spin_unlock_bh(&midc->lock);
@@ -1148,7 +1134,8 @@ static int mid_setup_dma(struct pci_dev *pdev)
dma->common.device_prep_dma_memcpy = intel_mid_dma_prep_memcpy;
dma->common.device_issue_pending = intel_mid_dma_issue_pending;
dma->common.device_prep_slave_sg = intel_mid_dma_prep_slave_sg;
- dma->common.device_control = intel_mid_dma_device_control;
+ dma->common.device_config = intel_mid_dma_config;
+ dma->common.device_terminate_all = intel_mid_dma_terminate_all;
/*enable dma cntrl*/
iowrite32(REG_BIT0, dma->dma_base + DMA_CFG);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 25/61] dmaengine: ipu-idmac: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the IPU IDMAC driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/ipu/ipu_idmac.c | 96 ++++++++++++++++++++++++---------------------
1 file changed, 51 insertions(+), 45 deletions(-)
diff --git a/drivers/dma/ipu/ipu_idmac.c b/drivers/dma/ipu/ipu_idmac.c
index bbf62927bd72..b36f5a47578a 100644
--- a/drivers/dma/ipu/ipu_idmac.c
+++ b/drivers/dma/ipu/ipu_idmac.c
@@ -1398,76 +1398,81 @@ static void idmac_issue_pending(struct dma_chan *chan)
*/
}
-static int __idmac_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int idmac_pause(struct dma_chan *chan)
{
struct idmac_channel *ichan = to_idmac_chan(chan);
struct idmac *idmac = to_idmac(chan->device);
struct ipu *ipu = to_ipu(idmac);
struct list_head *list, *tmp;
unsigned long flags;
- int i;
- switch (cmd) {
- case DMA_PAUSE:
- spin_lock_irqsave(&ipu->lock, flags);
- ipu_ic_disable_task(ipu, chan->chan_id);
+ mutex_lock(&ichan->chan_mutex);
- /* Return all descriptors into "prepared" state */
- list_for_each_safe(list, tmp, &ichan->queue)
- list_del_init(list);
+ spin_lock_irqsave(&ipu->lock, flags);
+ ipu_ic_disable_task(ipu, chan->chan_id);
- ichan->sg[0] = NULL;
- ichan->sg[1] = NULL;
+ /* Return all descriptors into "prepared" state */
+ list_for_each_safe(list, tmp, &ichan->queue)
+ list_del_init(list);
- spin_unlock_irqrestore(&ipu->lock, flags);
+ ichan->sg[0] = NULL;
+ ichan->sg[1] = NULL;
- ichan->status = IPU_CHANNEL_INITIALIZED;
- break;
- case DMA_TERMINATE_ALL:
- ipu_disable_channel(idmac, ichan,
- ichan->status >= IPU_CHANNEL_ENABLED);
+ spin_unlock_irqrestore(&ipu->lock, flags);
- tasklet_disable(&ipu->tasklet);
+ ichan->status = IPU_CHANNEL_INITIALIZED;
- /* ichan->queue is modified in ISR, have to spinlock */
- spin_lock_irqsave(&ichan->lock, flags);
- list_splice_init(&ichan->queue, &ichan->free_list);
+ mutex_unlock(&ichan->chan_mutex);
- if (ichan->desc)
- for (i = 0; i < ichan->n_tx_desc; i++) {
- struct idmac_tx_desc *desc = ichan->desc + i;
- if (list_empty(&desc->list))
- /* Descriptor was prepared, but not submitted */
- list_add(&desc->list, &ichan->free_list);
+ return 0;
+}
- async_tx_clear_ack(&desc->txd);
- }
+static int __idmac_terminate_all(struct dma_chan *chan)
+{
+ struct idmac_channel *ichan = to_idmac_chan(chan);
+ struct idmac *idmac = to_idmac(chan->device);
+ struct ipu *ipu = to_ipu(idmac);
+ unsigned long flags;
+ int i;
- ichan->sg[0] = NULL;
- ichan->sg[1] = NULL;
- spin_unlock_irqrestore(&ichan->lock, flags);
+ ipu_disable_channel(idmac, ichan,
+ ichan->status >= IPU_CHANNEL_ENABLED);
- tasklet_enable(&ipu->tasklet);
+ tasklet_disable(&ipu->tasklet);
- ichan->status = IPU_CHANNEL_INITIALIZED;
- break;
- default:
- return -ENOSYS;
- }
+ /* ichan->queue is modified in ISR, have to spinlock */
+ spin_lock_irqsave(&ichan->lock, flags);
+ list_splice_init(&ichan->queue, &ichan->free_list);
+
+ if (ichan->desc)
+ for (i = 0; i < ichan->n_tx_desc; i++) {
+ struct idmac_tx_desc *desc = ichan->desc + i;
+ if (list_empty(&desc->list))
+ /* Descriptor was prepared, but not submitted */
+ list_add(&desc->list, &ichan->free_list);
+
+ async_tx_clear_ack(&desc->txd);
+ }
+
+ ichan->sg[0] = NULL;
+ ichan->sg[1] = NULL;
+ spin_unlock_irqrestore(&ichan->lock, flags);
+
+ tasklet_enable(&ipu->tasklet);
+
+ ichan->status = IPU_CHANNEL_INITIALIZED;
return 0;
}
-static int idmac_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int idmac_terminate_all(struct dma_chan *chan)
{
struct idmac_channel *ichan = to_idmac_chan(chan);
int ret;
mutex_lock(&ichan->chan_mutex);
- ret = __idmac_control(chan, cmd, arg);
+ ret = __idmac_terminate_all(chan);
mutex_unlock(&ichan->chan_mutex);
@@ -1568,7 +1573,7 @@ static void idmac_free_chan_resources(struct dma_chan *chan)
mutex_lock(&ichan->chan_mutex);
- __idmac_control(chan, DMA_TERMINATE_ALL, 0);
+ __idmac_terminate_all(chan);
if (ichan->status > IPU_CHANNEL_FREE) {
#ifdef DEBUG
@@ -1622,7 +1627,8 @@ static int __init ipu_idmac_init(struct ipu *ipu)
/* Compulsory for DMA_SLAVE fields */
dma->device_prep_slave_sg = idmac_prep_slave_sg;
- dma->device_control = idmac_control;
+ dma->device_pause = idmac_pause;
+ dma->device_terminate_all = idmac_terminate_all;
INIT_LIST_HEAD(&dma->channels);
for (i = 0; i < IPU_CHANNELS_NUM; i++) {
@@ -1655,7 +1661,7 @@ static void ipu_idmac_exit(struct ipu *ipu)
for (i = 0; i < IPU_CHANNELS_NUM; i++) {
struct idmac_channel *ichan = ipu->channel + i;
- idmac_control(&ichan->dma_chan, DMA_TERMINATE_ALL, 0);
+ idmac_terminate_all(&ichan->dma_chan);
}
dma_async_device_unregister(&idmac->dma);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 25/61] dmaengine: ipu-idmac: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the IPU IDMAC driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/ipu/ipu_idmac.c | 96 ++++++++++++++++++++++++---------------------
1 file changed, 51 insertions(+), 45 deletions(-)
diff --git a/drivers/dma/ipu/ipu_idmac.c b/drivers/dma/ipu/ipu_idmac.c
index bbf62927bd72..b36f5a47578a 100644
--- a/drivers/dma/ipu/ipu_idmac.c
+++ b/drivers/dma/ipu/ipu_idmac.c
@@ -1398,76 +1398,81 @@ static void idmac_issue_pending(struct dma_chan *chan)
*/
}
-static int __idmac_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int idmac_pause(struct dma_chan *chan)
{
struct idmac_channel *ichan = to_idmac_chan(chan);
struct idmac *idmac = to_idmac(chan->device);
struct ipu *ipu = to_ipu(idmac);
struct list_head *list, *tmp;
unsigned long flags;
- int i;
- switch (cmd) {
- case DMA_PAUSE:
- spin_lock_irqsave(&ipu->lock, flags);
- ipu_ic_disable_task(ipu, chan->chan_id);
+ mutex_lock(&ichan->chan_mutex);
- /* Return all descriptors into "prepared" state */
- list_for_each_safe(list, tmp, &ichan->queue)
- list_del_init(list);
+ spin_lock_irqsave(&ipu->lock, flags);
+ ipu_ic_disable_task(ipu, chan->chan_id);
- ichan->sg[0] = NULL;
- ichan->sg[1] = NULL;
+ /* Return all descriptors into "prepared" state */
+ list_for_each_safe(list, tmp, &ichan->queue)
+ list_del_init(list);
- spin_unlock_irqrestore(&ipu->lock, flags);
+ ichan->sg[0] = NULL;
+ ichan->sg[1] = NULL;
- ichan->status = IPU_CHANNEL_INITIALIZED;
- break;
- case DMA_TERMINATE_ALL:
- ipu_disable_channel(idmac, ichan,
- ichan->status >= IPU_CHANNEL_ENABLED);
+ spin_unlock_irqrestore(&ipu->lock, flags);
- tasklet_disable(&ipu->tasklet);
+ ichan->status = IPU_CHANNEL_INITIALIZED;
- /* ichan->queue is modified in ISR, have to spinlock */
- spin_lock_irqsave(&ichan->lock, flags);
- list_splice_init(&ichan->queue, &ichan->free_list);
+ mutex_unlock(&ichan->chan_mutex);
- if (ichan->desc)
- for (i = 0; i < ichan->n_tx_desc; i++) {
- struct idmac_tx_desc *desc = ichan->desc + i;
- if (list_empty(&desc->list))
- /* Descriptor was prepared, but not submitted */
- list_add(&desc->list, &ichan->free_list);
+ return 0;
+}
- async_tx_clear_ack(&desc->txd);
- }
+static int __idmac_terminate_all(struct dma_chan *chan)
+{
+ struct idmac_channel *ichan = to_idmac_chan(chan);
+ struct idmac *idmac = to_idmac(chan->device);
+ struct ipu *ipu = to_ipu(idmac);
+ unsigned long flags;
+ int i;
- ichan->sg[0] = NULL;
- ichan->sg[1] = NULL;
- spin_unlock_irqrestore(&ichan->lock, flags);
+ ipu_disable_channel(idmac, ichan,
+ ichan->status >= IPU_CHANNEL_ENABLED);
- tasklet_enable(&ipu->tasklet);
+ tasklet_disable(&ipu->tasklet);
- ichan->status = IPU_CHANNEL_INITIALIZED;
- break;
- default:
- return -ENOSYS;
- }
+ /* ichan->queue is modified in ISR, have to spinlock */
+ spin_lock_irqsave(&ichan->lock, flags);
+ list_splice_init(&ichan->queue, &ichan->free_list);
+
+ if (ichan->desc)
+ for (i = 0; i < ichan->n_tx_desc; i++) {
+ struct idmac_tx_desc *desc = ichan->desc + i;
+ if (list_empty(&desc->list))
+ /* Descriptor was prepared, but not submitted */
+ list_add(&desc->list, &ichan->free_list);
+
+ async_tx_clear_ack(&desc->txd);
+ }
+
+ ichan->sg[0] = NULL;
+ ichan->sg[1] = NULL;
+ spin_unlock_irqrestore(&ichan->lock, flags);
+
+ tasklet_enable(&ipu->tasklet);
+
+ ichan->status = IPU_CHANNEL_INITIALIZED;
return 0;
}
-static int idmac_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int idmac_terminate_all(struct dma_chan *chan)
{
struct idmac_channel *ichan = to_idmac_chan(chan);
int ret;
mutex_lock(&ichan->chan_mutex);
- ret = __idmac_control(chan, cmd, arg);
+ ret = __idmac_terminate_all(chan);
mutex_unlock(&ichan->chan_mutex);
@@ -1568,7 +1573,7 @@ static void idmac_free_chan_resources(struct dma_chan *chan)
mutex_lock(&ichan->chan_mutex);
- __idmac_control(chan, DMA_TERMINATE_ALL, 0);
+ __idmac_terminate_all(chan);
if (ichan->status > IPU_CHANNEL_FREE) {
#ifdef DEBUG
@@ -1622,7 +1627,8 @@ static int __init ipu_idmac_init(struct ipu *ipu)
/* Compulsory for DMA_SLAVE fields */
dma->device_prep_slave_sg = idmac_prep_slave_sg;
- dma->device_control = idmac_control;
+ dma->device_pause = idmac_pause;
+ dma->device_terminate_all = idmac_terminate_all;
INIT_LIST_HEAD(&dma->channels);
for (i = 0; i < IPU_CHANNELS_NUM; i++) {
@@ -1655,7 +1661,7 @@ static void ipu_idmac_exit(struct ipu *ipu)
for (i = 0; i < IPU_CHANNELS_NUM; i++) {
struct idmac_channel *ichan = ipu->channel + i;
- idmac_control(&ichan->dma_chan, DMA_TERMINATE_ALL, 0);
+ idmac_terminate_all(&ichan->dma_chan);
}
dma_async_device_unregister(&idmac->dma);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 26/61] dmaengine: k3: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Hisilicon K3 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/k3dma.c | 197 ++++++++++++++++++++++++++++------------------------
1 file changed, 107 insertions(+), 90 deletions(-)
diff --git a/drivers/dma/k3dma.c b/drivers/dma/k3dma.c
index a1de14ab2c51..49be7f687c4c 100644
--- a/drivers/dma/k3dma.c
+++ b/drivers/dma/k3dma.c
@@ -441,7 +441,7 @@ static struct dma_async_tx_descriptor *k3_dma_prep_memcpy(
num = 0;
if (!c->ccfg) {
- /* default is memtomem, without calling device_control */
+ /* default is memtomem, without calling device_config */
c->ccfg = CX_CFG_SRCINCR | CX_CFG_DSTINCR | CX_CFG_EN;
c->ccfg |= (0xf << 20) | (0xf << 24); /* burst = 16 */
c->ccfg |= (0x3 << 12) | (0x3 << 16); /* width = 64 bit */
@@ -523,112 +523,126 @@ static struct dma_async_tx_descriptor *k3_dma_prep_slave_sg(
return vchan_tx_prep(&c->vc, &ds->vd, flags);
}
-static int k3_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int k3_dma_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
+{
+ struct k3_dma_chan *c = to_k3_chan(chan);
+ u32 maxburst = 0, val = 0;
+ enum dma_slave_buswidth width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
+
+ if (cfg == NULL)
+ return -EINVAL;
+ c->dir = cfg->direction;
+ if (c->dir == DMA_DEV_TO_MEM) {
+ c->ccfg = CX_CFG_DSTINCR;
+ c->dev_addr = cfg->src_addr;
+ maxburst = cfg->src_maxburst;
+ width = cfg->src_addr_width;
+ } else if (c->dir == DMA_MEM_TO_DEV) {
+ c->ccfg = CX_CFG_SRCINCR;
+ c->dev_addr = cfg->dst_addr;
+ maxburst = cfg->dst_maxburst;
+ width = cfg->dst_addr_width;
+ }
+ switch (width) {
+ case DMA_SLAVE_BUSWIDTH_1_BYTE:
+ case DMA_SLAVE_BUSWIDTH_2_BYTES:
+ case DMA_SLAVE_BUSWIDTH_4_BYTES:
+ case DMA_SLAVE_BUSWIDTH_8_BYTES:
+ val = __ffs(width);
+ break;
+ default:
+ val = 3;
+ break;
+ }
+ c->ccfg |= (val << 12) | (val << 16);
+
+ if ((maxburst == 0) || (maxburst > 16))
+ val = 16;
+ else
+ val = maxburst - 1;
+ c->ccfg |= (val << 20) | (val << 24);
+ c->ccfg |= CX_CFG_MEM2PER | CX_CFG_EN;
+
+ /* specific request line */
+ c->ccfg |= c->vc.chan.chan_id << 4;
+
+ return 0;
+}
+
+static int k3_dma_terminate_all(struct dma_chan *chan)
{
struct k3_dma_chan *c = to_k3_chan(chan);
struct k3_dma_dev *d = to_k3_dma(chan->device);
- struct dma_slave_config *cfg = (void *)arg;
struct k3_dma_phy *p = c->phy;
unsigned long flags;
- u32 maxburst = 0, val = 0;
- enum dma_slave_buswidth width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
LIST_HEAD(head);
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- if (cfg == NULL)
- return -EINVAL;
- c->dir = cfg->direction;
- if (c->dir == DMA_DEV_TO_MEM) {
- c->ccfg = CX_CFG_DSTINCR;
- c->dev_addr = cfg->src_addr;
- maxburst = cfg->src_maxburst;
- width = cfg->src_addr_width;
- } else if (c->dir == DMA_MEM_TO_DEV) {
- c->ccfg = CX_CFG_SRCINCR;
- c->dev_addr = cfg->dst_addr;
- maxburst = cfg->dst_maxburst;
- width = cfg->dst_addr_width;
- }
- switch (width) {
- case DMA_SLAVE_BUSWIDTH_1_BYTE:
- case DMA_SLAVE_BUSWIDTH_2_BYTES:
- case DMA_SLAVE_BUSWIDTH_4_BYTES:
- case DMA_SLAVE_BUSWIDTH_8_BYTES:
- val = __ffs(width);
- break;
- default:
- val = 3;
- break;
- }
- c->ccfg |= (val << 12) | (val << 16);
+ dev_dbg(d->slave.dev, "vchan %p: terminate all\n", &c->vc);
- if ((maxburst == 0) || (maxburst > 16))
- val = 16;
- else
- val = maxburst - 1;
- c->ccfg |= (val << 20) | (val << 24);
- c->ccfg |= CX_CFG_MEM2PER | CX_CFG_EN;
+ /* Prevent this channel being scheduled */
+ spin_lock(&d->lock);
+ list_del_init(&c->node);
+ spin_unlock(&d->lock);
- /* specific request line */
- c->ccfg |= c->vc.chan.chan_id << 4;
- break;
+ /* Clear the tx descriptor lists */
+ spin_lock_irqsave(&c->vc.lock, flags);
+ vchan_get_all_descriptors(&c->vc, &head);
+ if (p) {
+ /* vchan is assigned to a pchan - stop the channel */
+ k3_dma_terminate_chan(p, d);
+ c->phy = NULL;
+ p->vchan = NULL;
+ p->ds_run = p->ds_done = NULL;
+ }
+ spin_unlock_irqrestore(&c->vc.lock, flags);
+ vchan_dma_desc_free_list(&c->vc, &head);
- case DMA_TERMINATE_ALL:
- dev_dbg(d->slave.dev, "vchan %p: terminate all\n", &c->vc);
+ return 0;
+}
- /* Prevent this channel being scheduled */
- spin_lock(&d->lock);
- list_del_init(&c->node);
- spin_unlock(&d->lock);
+static int k3_dma_pause(struct dma_chan *chan)
+{
+ struct k3_dma_chan *c = to_k3_chan(chan);
+ struct k3_dma_dev *d = to_k3_dma(chan->device);
+ struct k3_dma_phy *p = c->phy;
- /* Clear the tx descriptor lists */
- spin_lock_irqsave(&c->vc.lock, flags);
- vchan_get_all_descriptors(&c->vc, &head);
+ dev_dbg(d->slave.dev, "vchan %p: pause\n", &c->vc);
+ if (c->status == DMA_IN_PROGRESS) {
+ c->status = DMA_PAUSED;
if (p) {
- /* vchan is assigned to a pchan - stop the channel */
- k3_dma_terminate_chan(p, d);
- c->phy = NULL;
- p->vchan = NULL;
- p->ds_run = p->ds_done = NULL;
+ k3_dma_pause_dma(p, false);
+ } else {
+ spin_lock(&d->lock);
+ list_del_init(&c->node);
+ spin_unlock(&d->lock);
}
- spin_unlock_irqrestore(&c->vc.lock, flags);
- vchan_dma_desc_free_list(&c->vc, &head);
- break;
+ }
- case DMA_PAUSE:
- dev_dbg(d->slave.dev, "vchan %p: pause\n", &c->vc);
- if (c->status == DMA_IN_PROGRESS) {
- c->status = DMA_PAUSED;
- if (p) {
- k3_dma_pause_dma(p, false);
- } else {
- spin_lock(&d->lock);
- list_del_init(&c->node);
- spin_unlock(&d->lock);
- }
- }
- break;
+ return 0;
+}
- case DMA_RESUME:
- dev_dbg(d->slave.dev, "vchan %p: resume\n", &c->vc);
- spin_lock_irqsave(&c->vc.lock, flags);
- if (c->status == DMA_PAUSED) {
- c->status = DMA_IN_PROGRESS;
- if (p) {
- k3_dma_pause_dma(p, true);
- } else if (!list_empty(&c->vc.desc_issued)) {
- spin_lock(&d->lock);
- list_add_tail(&c->node, &d->chan_pending);
- spin_unlock(&d->lock);
- }
+static int k3_dma_resume(struct dma_chan *chan)
+{
+ struct k3_dma_chan *c = to_k3_chan(chan);
+ struct k3_dma_dev *d = to_k3_dma(chan->device);
+ struct k3_dma_phy *p = c->phy;
+ unsigned long flags;
+
+ dev_dbg(d->slave.dev, "vchan %p: resume\n", &c->vc);
+ spin_lock_irqsave(&c->vc.lock, flags);
+ if (c->status == DMA_PAUSED) {
+ c->status = DMA_IN_PROGRESS;
+ if (p) {
+ k3_dma_pause_dma(p, true);
+ } else if (!list_empty(&c->vc.desc_issued)) {
+ spin_lock(&d->lock);
+ list_add_tail(&c->node, &d->chan_pending);
+ spin_unlock(&d->lock);
}
- spin_unlock_irqrestore(&c->vc.lock, flags);
- break;
- default:
- return -ENXIO;
}
+ spin_unlock_irqrestore(&c->vc.lock, flags);
+
return 0;
}
@@ -720,7 +734,10 @@ static int k3_dma_probe(struct platform_device *op)
d->slave.device_prep_dma_memcpy = k3_dma_prep_memcpy;
d->slave.device_prep_slave_sg = k3_dma_prep_slave_sg;
d->slave.device_issue_pending = k3_dma_issue_pending;
- d->slave.device_control = k3_dma_control;
+ d->slave.device_config = k3_dma_config;
+ d->slave.device_pause = k3_dma_pause;
+ d->slave.device_resume = k3_dma_resume;
+ d->slave.device_terminate_all = k3_dma_terminate_all;
d->slave.copy_align = DMA_ALIGN;
/* init virtual channel */
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 26/61] dmaengine: k3: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Hisilicon K3 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/k3dma.c | 197 ++++++++++++++++++++++++++++------------------------
1 file changed, 107 insertions(+), 90 deletions(-)
diff --git a/drivers/dma/k3dma.c b/drivers/dma/k3dma.c
index a1de14ab2c51..49be7f687c4c 100644
--- a/drivers/dma/k3dma.c
+++ b/drivers/dma/k3dma.c
@@ -441,7 +441,7 @@ static struct dma_async_tx_descriptor *k3_dma_prep_memcpy(
num = 0;
if (!c->ccfg) {
- /* default is memtomem, without calling device_control */
+ /* default is memtomem, without calling device_config */
c->ccfg = CX_CFG_SRCINCR | CX_CFG_DSTINCR | CX_CFG_EN;
c->ccfg |= (0xf << 20) | (0xf << 24); /* burst = 16 */
c->ccfg |= (0x3 << 12) | (0x3 << 16); /* width = 64 bit */
@@ -523,112 +523,126 @@ static struct dma_async_tx_descriptor *k3_dma_prep_slave_sg(
return vchan_tx_prep(&c->vc, &ds->vd, flags);
}
-static int k3_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int k3_dma_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
+{
+ struct k3_dma_chan *c = to_k3_chan(chan);
+ u32 maxburst = 0, val = 0;
+ enum dma_slave_buswidth width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
+
+ if (cfg == NULL)
+ return -EINVAL;
+ c->dir = cfg->direction;
+ if (c->dir == DMA_DEV_TO_MEM) {
+ c->ccfg = CX_CFG_DSTINCR;
+ c->dev_addr = cfg->src_addr;
+ maxburst = cfg->src_maxburst;
+ width = cfg->src_addr_width;
+ } else if (c->dir == DMA_MEM_TO_DEV) {
+ c->ccfg = CX_CFG_SRCINCR;
+ c->dev_addr = cfg->dst_addr;
+ maxburst = cfg->dst_maxburst;
+ width = cfg->dst_addr_width;
+ }
+ switch (width) {
+ case DMA_SLAVE_BUSWIDTH_1_BYTE:
+ case DMA_SLAVE_BUSWIDTH_2_BYTES:
+ case DMA_SLAVE_BUSWIDTH_4_BYTES:
+ case DMA_SLAVE_BUSWIDTH_8_BYTES:
+ val = __ffs(width);
+ break;
+ default:
+ val = 3;
+ break;
+ }
+ c->ccfg |= (val << 12) | (val << 16);
+
+ if ((maxburst == 0) || (maxburst > 16))
+ val = 16;
+ else
+ val = maxburst - 1;
+ c->ccfg |= (val << 20) | (val << 24);
+ c->ccfg |= CX_CFG_MEM2PER | CX_CFG_EN;
+
+ /* specific request line */
+ c->ccfg |= c->vc.chan.chan_id << 4;
+
+ return 0;
+}
+
+static int k3_dma_terminate_all(struct dma_chan *chan)
{
struct k3_dma_chan *c = to_k3_chan(chan);
struct k3_dma_dev *d = to_k3_dma(chan->device);
- struct dma_slave_config *cfg = (void *)arg;
struct k3_dma_phy *p = c->phy;
unsigned long flags;
- u32 maxburst = 0, val = 0;
- enum dma_slave_buswidth width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
LIST_HEAD(head);
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- if (cfg == NULL)
- return -EINVAL;
- c->dir = cfg->direction;
- if (c->dir == DMA_DEV_TO_MEM) {
- c->ccfg = CX_CFG_DSTINCR;
- c->dev_addr = cfg->src_addr;
- maxburst = cfg->src_maxburst;
- width = cfg->src_addr_width;
- } else if (c->dir == DMA_MEM_TO_DEV) {
- c->ccfg = CX_CFG_SRCINCR;
- c->dev_addr = cfg->dst_addr;
- maxburst = cfg->dst_maxburst;
- width = cfg->dst_addr_width;
- }
- switch (width) {
- case DMA_SLAVE_BUSWIDTH_1_BYTE:
- case DMA_SLAVE_BUSWIDTH_2_BYTES:
- case DMA_SLAVE_BUSWIDTH_4_BYTES:
- case DMA_SLAVE_BUSWIDTH_8_BYTES:
- val = __ffs(width);
- break;
- default:
- val = 3;
- break;
- }
- c->ccfg |= (val << 12) | (val << 16);
+ dev_dbg(d->slave.dev, "vchan %p: terminate all\n", &c->vc);
- if ((maxburst == 0) || (maxburst > 16))
- val = 16;
- else
- val = maxburst - 1;
- c->ccfg |= (val << 20) | (val << 24);
- c->ccfg |= CX_CFG_MEM2PER | CX_CFG_EN;
+ /* Prevent this channel being scheduled */
+ spin_lock(&d->lock);
+ list_del_init(&c->node);
+ spin_unlock(&d->lock);
- /* specific request line */
- c->ccfg |= c->vc.chan.chan_id << 4;
- break;
+ /* Clear the tx descriptor lists */
+ spin_lock_irqsave(&c->vc.lock, flags);
+ vchan_get_all_descriptors(&c->vc, &head);
+ if (p) {
+ /* vchan is assigned to a pchan - stop the channel */
+ k3_dma_terminate_chan(p, d);
+ c->phy = NULL;
+ p->vchan = NULL;
+ p->ds_run = p->ds_done = NULL;
+ }
+ spin_unlock_irqrestore(&c->vc.lock, flags);
+ vchan_dma_desc_free_list(&c->vc, &head);
- case DMA_TERMINATE_ALL:
- dev_dbg(d->slave.dev, "vchan %p: terminate all\n", &c->vc);
+ return 0;
+}
- /* Prevent this channel being scheduled */
- spin_lock(&d->lock);
- list_del_init(&c->node);
- spin_unlock(&d->lock);
+static int k3_dma_pause(struct dma_chan *chan)
+{
+ struct k3_dma_chan *c = to_k3_chan(chan);
+ struct k3_dma_dev *d = to_k3_dma(chan->device);
+ struct k3_dma_phy *p = c->phy;
- /* Clear the tx descriptor lists */
- spin_lock_irqsave(&c->vc.lock, flags);
- vchan_get_all_descriptors(&c->vc, &head);
+ dev_dbg(d->slave.dev, "vchan %p: pause\n", &c->vc);
+ if (c->status == DMA_IN_PROGRESS) {
+ c->status = DMA_PAUSED;
if (p) {
- /* vchan is assigned to a pchan - stop the channel */
- k3_dma_terminate_chan(p, d);
- c->phy = NULL;
- p->vchan = NULL;
- p->ds_run = p->ds_done = NULL;
+ k3_dma_pause_dma(p, false);
+ } else {
+ spin_lock(&d->lock);
+ list_del_init(&c->node);
+ spin_unlock(&d->lock);
}
- spin_unlock_irqrestore(&c->vc.lock, flags);
- vchan_dma_desc_free_list(&c->vc, &head);
- break;
+ }
- case DMA_PAUSE:
- dev_dbg(d->slave.dev, "vchan %p: pause\n", &c->vc);
- if (c->status == DMA_IN_PROGRESS) {
- c->status = DMA_PAUSED;
- if (p) {
- k3_dma_pause_dma(p, false);
- } else {
- spin_lock(&d->lock);
- list_del_init(&c->node);
- spin_unlock(&d->lock);
- }
- }
- break;
+ return 0;
+}
- case DMA_RESUME:
- dev_dbg(d->slave.dev, "vchan %p: resume\n", &c->vc);
- spin_lock_irqsave(&c->vc.lock, flags);
- if (c->status == DMA_PAUSED) {
- c->status = DMA_IN_PROGRESS;
- if (p) {
- k3_dma_pause_dma(p, true);
- } else if (!list_empty(&c->vc.desc_issued)) {
- spin_lock(&d->lock);
- list_add_tail(&c->node, &d->chan_pending);
- spin_unlock(&d->lock);
- }
+static int k3_dma_resume(struct dma_chan *chan)
+{
+ struct k3_dma_chan *c = to_k3_chan(chan);
+ struct k3_dma_dev *d = to_k3_dma(chan->device);
+ struct k3_dma_phy *p = c->phy;
+ unsigned long flags;
+
+ dev_dbg(d->slave.dev, "vchan %p: resume\n", &c->vc);
+ spin_lock_irqsave(&c->vc.lock, flags);
+ if (c->status == DMA_PAUSED) {
+ c->status = DMA_IN_PROGRESS;
+ if (p) {
+ k3_dma_pause_dma(p, true);
+ } else if (!list_empty(&c->vc.desc_issued)) {
+ spin_lock(&d->lock);
+ list_add_tail(&c->node, &d->chan_pending);
+ spin_unlock(&d->lock);
}
- spin_unlock_irqrestore(&c->vc.lock, flags);
- break;
- default:
- return -ENXIO;
}
+ spin_unlock_irqrestore(&c->vc.lock, flags);
+
return 0;
}
@@ -720,7 +734,10 @@ static int k3_dma_probe(struct platform_device *op)
d->slave.device_prep_dma_memcpy = k3_dma_prep_memcpy;
d->slave.device_prep_slave_sg = k3_dma_prep_slave_sg;
d->slave.device_issue_pending = k3_dma_issue_pending;
- d->slave.device_control = k3_dma_control;
+ d->slave.device_config = k3_dma_config;
+ d->slave.device_pause = k3_dma_pause;
+ d->slave.device_resume = k3_dma_resume;
+ d->slave.device_terminate_all = k3_dma_terminate_all;
d->slave.copy_align = DMA_ALIGN;
/* init virtual channel */
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 27/61] dmaengine: mmp-pdma: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Marvell MMP PDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mmp_pdma.c | 109 +++++++++++++++++++++++++------------------------
1 file changed, 56 insertions(+), 53 deletions(-)
diff --git a/drivers/dma/mmp_pdma.c b/drivers/dma/mmp_pdma.c
index 8b8952f35e6c..8926f271904e 100644
--- a/drivers/dma/mmp_pdma.c
+++ b/drivers/dma/mmp_pdma.c
@@ -683,68 +683,70 @@ fail:
return NULL;
}
-static int mmp_pdma_control(struct dma_chan *dchan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int mmp_pdma_config(struct dma_chan *dchan,
+ struct dma_slave_config *cfg)
{
struct mmp_pdma_chan *chan = to_mmp_pdma_chan(dchan);
- struct dma_slave_config *cfg = (void *)arg;
- unsigned long flags;
u32 maxburst = 0, addr = 0;
enum dma_slave_buswidth width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
if (!dchan)
return -EINVAL;
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- disable_chan(chan->phy);
- mmp_pdma_free_phy(chan);
- spin_lock_irqsave(&chan->desc_lock, flags);
- mmp_pdma_free_desc_list(chan, &chan->chain_pending);
- mmp_pdma_free_desc_list(chan, &chan->chain_running);
- spin_unlock_irqrestore(&chan->desc_lock, flags);
- chan->idle = true;
- break;
- case DMA_SLAVE_CONFIG:
- if (cfg->direction == DMA_DEV_TO_MEM) {
- chan->dcmd = DCMD_INCTRGADDR | DCMD_FLOWSRC;
- maxburst = cfg->src_maxburst;
- width = cfg->src_addr_width;
- addr = cfg->src_addr;
- } else if (cfg->direction == DMA_MEM_TO_DEV) {
- chan->dcmd = DCMD_INCSRCADDR | DCMD_FLOWTRG;
- maxburst = cfg->dst_maxburst;
- width = cfg->dst_addr_width;
- addr = cfg->dst_addr;
- }
-
- if (width == DMA_SLAVE_BUSWIDTH_1_BYTE)
- chan->dcmd |= DCMD_WIDTH1;
- else if (width == DMA_SLAVE_BUSWIDTH_2_BYTES)
- chan->dcmd |= DCMD_WIDTH2;
- else if (width == DMA_SLAVE_BUSWIDTH_4_BYTES)
- chan->dcmd |= DCMD_WIDTH4;
-
- if (maxburst == 8)
- chan->dcmd |= DCMD_BURST8;
- else if (maxburst == 16)
- chan->dcmd |= DCMD_BURST16;
- else if (maxburst == 32)
- chan->dcmd |= DCMD_BURST32;
-
- chan->dir = cfg->direction;
- chan->dev_addr = addr;
- /* FIXME: drivers should be ported over to use the filter
- * function. Once that's done, the following two lines can
- * be removed.
- */
- if (cfg->slave_id)
- chan->drcmr = cfg->slave_id;
- break;
- default:
- return -ENOSYS;
+ if (cfg->direction == DMA_DEV_TO_MEM) {
+ chan->dcmd = DCMD_INCTRGADDR | DCMD_FLOWSRC;
+ maxburst = cfg->src_maxburst;
+ width = cfg->src_addr_width;
+ addr = cfg->src_addr;
+ } else if (cfg->direction == DMA_MEM_TO_DEV) {
+ chan->dcmd = DCMD_INCSRCADDR | DCMD_FLOWTRG;
+ maxburst = cfg->dst_maxburst;
+ width = cfg->dst_addr_width;
+ addr = cfg->dst_addr;
}
+ if (width == DMA_SLAVE_BUSWIDTH_1_BYTE)
+ chan->dcmd |= DCMD_WIDTH1;
+ else if (width == DMA_SLAVE_BUSWIDTH_2_BYTES)
+ chan->dcmd |= DCMD_WIDTH2;
+ else if (width == DMA_SLAVE_BUSWIDTH_4_BYTES)
+ chan->dcmd |= DCMD_WIDTH4;
+
+ if (maxburst == 8)
+ chan->dcmd |= DCMD_BURST8;
+ else if (maxburst == 16)
+ chan->dcmd |= DCMD_BURST16;
+ else if (maxburst == 32)
+ chan->dcmd |= DCMD_BURST32;
+
+ chan->dir = cfg->direction;
+ chan->dev_addr = addr;
+ /* FIXME: drivers should be ported over to use the filter
+ * function. Once that's done, the following two lines can
+ * be removed.
+ */
+ if (cfg->slave_id)
+ chan->drcmr = cfg->slave_id;
+
+ return 0;
+}
+
+static int mmp_pdma_terminate_all(struct dma_chan *dchan)
+{
+ struct mmp_pdma_chan *chan = to_mmp_pdma_chan(dchan);
+ unsigned long flags;
+
+ if (!dchan)
+ return -EINVAL;
+
+ disable_chan(chan->phy);
+ mmp_pdma_free_phy(chan);
+ spin_lock_irqsave(&chan->desc_lock, flags);
+ mmp_pdma_free_desc_list(chan, &chan->chain_pending);
+ mmp_pdma_free_desc_list(chan, &chan->chain_running);
+ spin_unlock_irqrestore(&chan->desc_lock, flags);
+ chan->idle = true;
+
return 0;
}
@@ -1061,7 +1063,8 @@ static int mmp_pdma_probe(struct platform_device *op)
pdev->device.device_prep_slave_sg = mmp_pdma_prep_slave_sg;
pdev->device.device_prep_dma_cyclic = mmp_pdma_prep_dma_cyclic;
pdev->device.device_issue_pending = mmp_pdma_issue_pending;
- pdev->device.device_control = mmp_pdma_control;
+ pdev->device.device_config = mmp_pdma_config;
+ pdev->device.device_terminate_all = mmp_pdma_terminate_all;
pdev->device.copy_align = PDMA_ALIGNMENT;
if (pdev->dev->coherent_dma_mask)
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 27/61] dmaengine: mmp-pdma: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Marvell MMP PDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mmp_pdma.c | 109 +++++++++++++++++++++++++------------------------
1 file changed, 56 insertions(+), 53 deletions(-)
diff --git a/drivers/dma/mmp_pdma.c b/drivers/dma/mmp_pdma.c
index 8b8952f35e6c..8926f271904e 100644
--- a/drivers/dma/mmp_pdma.c
+++ b/drivers/dma/mmp_pdma.c
@@ -683,68 +683,70 @@ fail:
return NULL;
}
-static int mmp_pdma_control(struct dma_chan *dchan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int mmp_pdma_config(struct dma_chan *dchan,
+ struct dma_slave_config *cfg)
{
struct mmp_pdma_chan *chan = to_mmp_pdma_chan(dchan);
- struct dma_slave_config *cfg = (void *)arg;
- unsigned long flags;
u32 maxburst = 0, addr = 0;
enum dma_slave_buswidth width = DMA_SLAVE_BUSWIDTH_UNDEFINED;
if (!dchan)
return -EINVAL;
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- disable_chan(chan->phy);
- mmp_pdma_free_phy(chan);
- spin_lock_irqsave(&chan->desc_lock, flags);
- mmp_pdma_free_desc_list(chan, &chan->chain_pending);
- mmp_pdma_free_desc_list(chan, &chan->chain_running);
- spin_unlock_irqrestore(&chan->desc_lock, flags);
- chan->idle = true;
- break;
- case DMA_SLAVE_CONFIG:
- if (cfg->direction == DMA_DEV_TO_MEM) {
- chan->dcmd = DCMD_INCTRGADDR | DCMD_FLOWSRC;
- maxburst = cfg->src_maxburst;
- width = cfg->src_addr_width;
- addr = cfg->src_addr;
- } else if (cfg->direction == DMA_MEM_TO_DEV) {
- chan->dcmd = DCMD_INCSRCADDR | DCMD_FLOWTRG;
- maxburst = cfg->dst_maxburst;
- width = cfg->dst_addr_width;
- addr = cfg->dst_addr;
- }
-
- if (width == DMA_SLAVE_BUSWIDTH_1_BYTE)
- chan->dcmd |= DCMD_WIDTH1;
- else if (width == DMA_SLAVE_BUSWIDTH_2_BYTES)
- chan->dcmd |= DCMD_WIDTH2;
- else if (width == DMA_SLAVE_BUSWIDTH_4_BYTES)
- chan->dcmd |= DCMD_WIDTH4;
-
- if (maxburst == 8)
- chan->dcmd |= DCMD_BURST8;
- else if (maxburst == 16)
- chan->dcmd |= DCMD_BURST16;
- else if (maxburst == 32)
- chan->dcmd |= DCMD_BURST32;
-
- chan->dir = cfg->direction;
- chan->dev_addr = addr;
- /* FIXME: drivers should be ported over to use the filter
- * function. Once that's done, the following two lines can
- * be removed.
- */
- if (cfg->slave_id)
- chan->drcmr = cfg->slave_id;
- break;
- default:
- return -ENOSYS;
+ if (cfg->direction == DMA_DEV_TO_MEM) {
+ chan->dcmd = DCMD_INCTRGADDR | DCMD_FLOWSRC;
+ maxburst = cfg->src_maxburst;
+ width = cfg->src_addr_width;
+ addr = cfg->src_addr;
+ } else if (cfg->direction == DMA_MEM_TO_DEV) {
+ chan->dcmd = DCMD_INCSRCADDR | DCMD_FLOWTRG;
+ maxburst = cfg->dst_maxburst;
+ width = cfg->dst_addr_width;
+ addr = cfg->dst_addr;
}
+ if (width == DMA_SLAVE_BUSWIDTH_1_BYTE)
+ chan->dcmd |= DCMD_WIDTH1;
+ else if (width == DMA_SLAVE_BUSWIDTH_2_BYTES)
+ chan->dcmd |= DCMD_WIDTH2;
+ else if (width == DMA_SLAVE_BUSWIDTH_4_BYTES)
+ chan->dcmd |= DCMD_WIDTH4;
+
+ if (maxburst == 8)
+ chan->dcmd |= DCMD_BURST8;
+ else if (maxburst == 16)
+ chan->dcmd |= DCMD_BURST16;
+ else if (maxburst == 32)
+ chan->dcmd |= DCMD_BURST32;
+
+ chan->dir = cfg->direction;
+ chan->dev_addr = addr;
+ /* FIXME: drivers should be ported over to use the filter
+ * function. Once that's done, the following two lines can
+ * be removed.
+ */
+ if (cfg->slave_id)
+ chan->drcmr = cfg->slave_id;
+
+ return 0;
+}
+
+static int mmp_pdma_terminate_all(struct dma_chan *dchan)
+{
+ struct mmp_pdma_chan *chan = to_mmp_pdma_chan(dchan);
+ unsigned long flags;
+
+ if (!dchan)
+ return -EINVAL;
+
+ disable_chan(chan->phy);
+ mmp_pdma_free_phy(chan);
+ spin_lock_irqsave(&chan->desc_lock, flags);
+ mmp_pdma_free_desc_list(chan, &chan->chain_pending);
+ mmp_pdma_free_desc_list(chan, &chan->chain_running);
+ spin_unlock_irqrestore(&chan->desc_lock, flags);
+ chan->idle = true;
+
return 0;
}
@@ -1061,7 +1063,8 @@ static int mmp_pdma_probe(struct platform_device *op)
pdev->device.device_prep_slave_sg = mmp_pdma_prep_slave_sg;
pdev->device.device_prep_dma_cyclic = mmp_pdma_prep_dma_cyclic;
pdev->device.device_issue_pending = mmp_pdma_issue_pending;
- pdev->device.device_control = mmp_pdma_control;
+ pdev->device.device_config = mmp_pdma_config;
+ pdev->device.device_terminate_all = mmp_pdma_terminate_all;
pdev->device.copy_align = PDMA_ALIGNMENT;
if (pdev->dev->coherent_dma_mask)
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 28/61] dmaengine: mmp-tdma: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Marvell MMP TDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mmp_tdma.c | 82 +++++++++++++++++++++++++++-----------------------
1 file changed, 44 insertions(+), 38 deletions(-)
diff --git a/drivers/dma/mmp_tdma.c b/drivers/dma/mmp_tdma.c
index bfb46957c3dc..a8a79b1763d5 100644
--- a/drivers/dma/mmp_tdma.c
+++ b/drivers/dma/mmp_tdma.c
@@ -164,33 +164,46 @@ static void mmp_tdma_enable_chan(struct mmp_tdma_chan *tdmac)
tdmac->status = DMA_IN_PROGRESS;
}
-static void mmp_tdma_disable_chan(struct mmp_tdma_chan *tdmac)
+static int mmp_tdma_disable_chan(struct dma_chan *chan)
{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
+
writel(readl(tdmac->reg_base + TDCR) & ~TDCR_CHANEN,
tdmac->reg_base + TDCR);
tdmac->status = DMA_COMPLETE;
+
+ return 0;
}
-static void mmp_tdma_resume_chan(struct mmp_tdma_chan *tdmac)
+static int mmp_tdma_resume_chan(struct dma_chan *chan)
{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
+
writel(readl(tdmac->reg_base + TDCR) | TDCR_CHANEN,
tdmac->reg_base + TDCR);
tdmac->status = DMA_IN_PROGRESS;
+
+ return 0;
}
-static void mmp_tdma_pause_chan(struct mmp_tdma_chan *tdmac)
+static int mmp_tdma_pause_chan(struct dma_chan *chan)
{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
+
writel(readl(tdmac->reg_base + TDCR) & ~TDCR_CHANEN,
tdmac->reg_base + TDCR);
tdmac->status = DMA_PAUSED;
+
+ return 0;
}
-static int mmp_tdma_config_chan(struct mmp_tdma_chan *tdmac)
+static int mmp_tdma_config_chan(struct dma_chan *chan)
{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
unsigned int tdcr = 0;
- mmp_tdma_disable_chan(tdmac);
+ mmp_tdma_disable_chan(chan);
if (tdmac->dir == DMA_MEM_TO_DEV)
tdcr = TDCR_DSTDIR_ADDR_HOLD | TDCR_SRCDIR_ADDR_INC;
@@ -452,42 +465,32 @@ err_out:
return NULL;
}
-static int mmp_tdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int mmp_tdma_terminate_all(struct dma_chan *chan)
{
struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
- struct dma_slave_config *dmaengine_cfg = (void *)arg;
- int ret = 0;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- mmp_tdma_disable_chan(tdmac);
- /* disable interrupt */
- mmp_tdma_enable_irq(tdmac, false);
- break;
- case DMA_PAUSE:
- mmp_tdma_pause_chan(tdmac);
- break;
- case DMA_RESUME:
- mmp_tdma_resume_chan(tdmac);
- break;
- case DMA_SLAVE_CONFIG:
- if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
- tdmac->dev_addr = dmaengine_cfg->src_addr;
- tdmac->burst_sz = dmaengine_cfg->src_maxburst;
- tdmac->buswidth = dmaengine_cfg->src_addr_width;
- } else {
- tdmac->dev_addr = dmaengine_cfg->dst_addr;
- tdmac->burst_sz = dmaengine_cfg->dst_maxburst;
- tdmac->buswidth = dmaengine_cfg->dst_addr_width;
- }
- tdmac->dir = dmaengine_cfg->direction;
- return mmp_tdma_config_chan(tdmac);
- default:
- ret = -ENOSYS;
+
+ mmp_tdma_disable_chan(chan);
+ /* disable interrupt */
+ mmp_tdma_enable_irq(tdmac, false);
+}
+
+static int mmp_tdma_config(struct dma_chan *chan,
+ struct dma_slave_config *dmaengine_cfg)
+{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
+
+ if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
+ tdmac->dev_addr = dmaengine_cfg->src_addr;
+ tdmac->burst_sz = dmaengine_cfg->src_maxburst;
+ tdmac->buswidth = dmaengine_cfg->src_addr_width;
+ } else {
+ tdmac->dev_addr = dmaengine_cfg->dst_addr;
+ tdmac->burst_sz = dmaengine_cfg->dst_maxburst;
+ tdmac->buswidth = dmaengine_cfg->dst_addr_width;
}
+ tdmac->dir = dmaengine_cfg->direction;
- return ret;
+ return mmp_tdma_config_chan(chan);
}
static enum dma_status mmp_tdma_tx_status(struct dma_chan *chan,
@@ -668,7 +671,10 @@ static int mmp_tdma_probe(struct platform_device *pdev)
tdev->device.device_prep_dma_cyclic = mmp_tdma_prep_dma_cyclic;
tdev->device.device_tx_status = mmp_tdma_tx_status;
tdev->device.device_issue_pending = mmp_tdma_issue_pending;
- tdev->device.device_control = mmp_tdma_control;
+ tdev->device.device_config = mmp_tdma_config;
+ tdev->device.device_pause = mmp_tdma_pause_chan;
+ tdev->device.device_resume = mmp_tdma_resume_chan;
+ tdev->device.device_terminate_all = mmp_tdma_terminate_all;
tdev->device.copy_align = TDMA_ALIGNMENT;
dma_set_mask(&pdev->dev, DMA_BIT_MASK(64));
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 28/61] dmaengine: mmp-tdma: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Marvell MMP TDMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mmp_tdma.c | 82 +++++++++++++++++++++++++++-----------------------
1 file changed, 44 insertions(+), 38 deletions(-)
diff --git a/drivers/dma/mmp_tdma.c b/drivers/dma/mmp_tdma.c
index bfb46957c3dc..a8a79b1763d5 100644
--- a/drivers/dma/mmp_tdma.c
+++ b/drivers/dma/mmp_tdma.c
@@ -164,33 +164,46 @@ static void mmp_tdma_enable_chan(struct mmp_tdma_chan *tdmac)
tdmac->status = DMA_IN_PROGRESS;
}
-static void mmp_tdma_disable_chan(struct mmp_tdma_chan *tdmac)
+static int mmp_tdma_disable_chan(struct dma_chan *chan)
{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
+
writel(readl(tdmac->reg_base + TDCR) & ~TDCR_CHANEN,
tdmac->reg_base + TDCR);
tdmac->status = DMA_COMPLETE;
+
+ return 0;
}
-static void mmp_tdma_resume_chan(struct mmp_tdma_chan *tdmac)
+static int mmp_tdma_resume_chan(struct dma_chan *chan)
{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
+
writel(readl(tdmac->reg_base + TDCR) | TDCR_CHANEN,
tdmac->reg_base + TDCR);
tdmac->status = DMA_IN_PROGRESS;
+
+ return 0;
}
-static void mmp_tdma_pause_chan(struct mmp_tdma_chan *tdmac)
+static int mmp_tdma_pause_chan(struct dma_chan *chan)
{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
+
writel(readl(tdmac->reg_base + TDCR) & ~TDCR_CHANEN,
tdmac->reg_base + TDCR);
tdmac->status = DMA_PAUSED;
+
+ return 0;
}
-static int mmp_tdma_config_chan(struct mmp_tdma_chan *tdmac)
+static int mmp_tdma_config_chan(struct dma_chan *chan)
{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
unsigned int tdcr = 0;
- mmp_tdma_disable_chan(tdmac);
+ mmp_tdma_disable_chan(chan);
if (tdmac->dir == DMA_MEM_TO_DEV)
tdcr = TDCR_DSTDIR_ADDR_HOLD | TDCR_SRCDIR_ADDR_INC;
@@ -452,42 +465,32 @@ err_out:
return NULL;
}
-static int mmp_tdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int mmp_tdma_terminate_all(struct dma_chan *chan)
{
struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
- struct dma_slave_config *dmaengine_cfg = (void *)arg;
- int ret = 0;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- mmp_tdma_disable_chan(tdmac);
- /* disable interrupt */
- mmp_tdma_enable_irq(tdmac, false);
- break;
- case DMA_PAUSE:
- mmp_tdma_pause_chan(tdmac);
- break;
- case DMA_RESUME:
- mmp_tdma_resume_chan(tdmac);
- break;
- case DMA_SLAVE_CONFIG:
- if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
- tdmac->dev_addr = dmaengine_cfg->src_addr;
- tdmac->burst_sz = dmaengine_cfg->src_maxburst;
- tdmac->buswidth = dmaengine_cfg->src_addr_width;
- } else {
- tdmac->dev_addr = dmaengine_cfg->dst_addr;
- tdmac->burst_sz = dmaengine_cfg->dst_maxburst;
- tdmac->buswidth = dmaengine_cfg->dst_addr_width;
- }
- tdmac->dir = dmaengine_cfg->direction;
- return mmp_tdma_config_chan(tdmac);
- default:
- ret = -ENOSYS;
+
+ mmp_tdma_disable_chan(chan);
+ /* disable interrupt */
+ mmp_tdma_enable_irq(tdmac, false);
+}
+
+static int mmp_tdma_config(struct dma_chan *chan,
+ struct dma_slave_config *dmaengine_cfg)
+{
+ struct mmp_tdma_chan *tdmac = to_mmp_tdma_chan(chan);
+
+ if (dmaengine_cfg->direction == DMA_DEV_TO_MEM) {
+ tdmac->dev_addr = dmaengine_cfg->src_addr;
+ tdmac->burst_sz = dmaengine_cfg->src_maxburst;
+ tdmac->buswidth = dmaengine_cfg->src_addr_width;
+ } else {
+ tdmac->dev_addr = dmaengine_cfg->dst_addr;
+ tdmac->burst_sz = dmaengine_cfg->dst_maxburst;
+ tdmac->buswidth = dmaengine_cfg->dst_addr_width;
}
+ tdmac->dir = dmaengine_cfg->direction;
- return ret;
+ return mmp_tdma_config_chan(chan);
}
static enum dma_status mmp_tdma_tx_status(struct dma_chan *chan,
@@ -668,7 +671,10 @@ static int mmp_tdma_probe(struct platform_device *pdev)
tdev->device.device_prep_dma_cyclic = mmp_tdma_prep_dma_cyclic;
tdev->device.device_tx_status = mmp_tdma_tx_status;
tdev->device.device_issue_pending = mmp_tdma_issue_pending;
- tdev->device.device_control = mmp_tdma_control;
+ tdev->device.device_config = mmp_tdma_config;
+ tdev->device.device_pause = mmp_tdma_pause_chan;
+ tdev->device.device_resume = mmp_tdma_resume_chan;
+ tdev->device.device_terminate_all = mmp_tdma_terminate_all;
tdev->device.copy_align = TDMA_ALIGNMENT;
dma_set_mask(&pdev->dev, DMA_BIT_MASK(64));
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 29/61] dmaengine: moxart: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Moxart DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/moxart-dma.c | 25 ++-----------------------
1 file changed, 2 insertions(+), 23 deletions(-)
diff --git a/drivers/dma/moxart-dma.c b/drivers/dma/moxart-dma.c
index 3258e484e4f6..35e1a3ab5fb8 100644
--- a/drivers/dma/moxart-dma.c
+++ b/drivers/dma/moxart-dma.c
@@ -263,28 +263,6 @@ static int moxart_slave_config(struct dma_chan *chan,
return 0;
}
-static int moxart_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- int ret = 0;
-
- switch (cmd) {
- case DMA_PAUSE:
- case DMA_RESUME:
- return -EINVAL;
- case DMA_TERMINATE_ALL:
- moxart_terminate_all(chan);
- break;
- case DMA_SLAVE_CONFIG:
- ret = moxart_slave_config(chan, (struct dma_slave_config *)arg);
- break;
- default:
- ret = -ENOSYS;
- }
-
- return ret;
-}
-
static struct dma_async_tx_descriptor *moxart_prep_slave_sg(
struct dma_chan *chan, struct scatterlist *sgl,
unsigned int sg_len, enum dma_transfer_direction dir,
@@ -531,7 +509,8 @@ static void moxart_dma_init(struct dma_device *dma, struct device *dev)
dma->device_free_chan_resources = moxart_free_chan_resources;
dma->device_issue_pending = moxart_issue_pending;
dma->device_tx_status = moxart_tx_status;
- dma->device_control = moxart_control;
+ dma->device_config = moxart_slave_config;
+ dma->device_terminate_all = moxart_terminate_all;
dma->dev = dev;
INIT_LIST_HEAD(&dma->channels);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 29/61] dmaengine: moxart: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Moxart DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/moxart-dma.c | 25 ++-----------------------
1 file changed, 2 insertions(+), 23 deletions(-)
diff --git a/drivers/dma/moxart-dma.c b/drivers/dma/moxart-dma.c
index 3258e484e4f6..35e1a3ab5fb8 100644
--- a/drivers/dma/moxart-dma.c
+++ b/drivers/dma/moxart-dma.c
@@ -263,28 +263,6 @@ static int moxart_slave_config(struct dma_chan *chan,
return 0;
}
-static int moxart_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- int ret = 0;
-
- switch (cmd) {
- case DMA_PAUSE:
- case DMA_RESUME:
- return -EINVAL;
- case DMA_TERMINATE_ALL:
- moxart_terminate_all(chan);
- break;
- case DMA_SLAVE_CONFIG:
- ret = moxart_slave_config(chan, (struct dma_slave_config *)arg);
- break;
- default:
- ret = -ENOSYS;
- }
-
- return ret;
-}
-
static struct dma_async_tx_descriptor *moxart_prep_slave_sg(
struct dma_chan *chan, struct scatterlist *sgl,
unsigned int sg_len, enum dma_transfer_direction dir,
@@ -531,7 +509,8 @@ static void moxart_dma_init(struct dma_device *dma, struct device *dev)
dma->device_free_chan_resources = moxart_free_chan_resources;
dma->device_issue_pending = moxart_issue_pending;
dma->device_tx_status = moxart_tx_status;
- dma->device_control = moxart_control;
+ dma->device_config = moxart_slave_config;
+ dma->device_terminate_all = moxart_terminate_all;
dma->dev = dev;
INIT_LIST_HEAD(&dma->channels);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 30/61] dmaengine: fsl-dma: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Freescale Elo DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
While we're at it, remove the useless prep_sg callback.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/fsldma.c | 91 ++++++++++++++++++----------------------------------
1 file changed, 32 insertions(+), 59 deletions(-)
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 3d8feb5e4c2f..676555776a46 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -941,37 +941,8 @@ fail:
return NULL;
}
-/**
- * fsl_dma_prep_slave_sg - prepare descriptors for a DMA_SLAVE transaction
- * @chan: DMA channel
- * @sgl: scatterlist to transfer to/from
- * @sg_len: number of entries in @scatterlist
- * @direction: DMA direction
- * @flags: DMAEngine flags
- * @context: transaction context (ignored)
- *
- * Prepare a set of descriptors for a DMA_SLAVE transaction. Following the
- * DMA_SLAVE API, this gets the device-specific information from the
- * chan->private variable.
- */
-static struct dma_async_tx_descriptor *fsl_dma_prep_slave_sg(
- struct dma_chan *dchan, struct scatterlist *sgl, unsigned int sg_len,
- enum dma_transfer_direction direction, unsigned long flags,
- void *context)
-{
- /*
- * This operation is not supported on the Freescale DMA controller
- *
- * However, we need to provide the function pointer to allow the
- * device_control() method to work.
- */
- return NULL;
-}
-
-static int fsl_dma_device_control(struct dma_chan *dchan,
- enum dma_ctrl_cmd cmd, unsigned long arg)
+static int fsl_dma_device_terminate_all(struct dma_chan *dchan)
{
- struct dma_slave_config *config;
struct fsldma_chan *chan;
int size;
@@ -980,45 +951,47 @@ static int fsl_dma_device_control(struct dma_chan *dchan,
chan = to_fsl_chan(dchan);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- spin_lock_bh(&chan->desc_lock);
-
- /* Halt the DMA engine */
- dma_halt(chan);
+ spin_lock_bh(&chan->desc_lock);
- /* Remove and free all of the descriptors in the LD queue */
- fsldma_free_desc_list(chan, &chan->ld_pending);
- fsldma_free_desc_list(chan, &chan->ld_running);
- fsldma_free_desc_list(chan, &chan->ld_completed);
- chan->idle = true;
+ /* Halt the DMA engine */
+ dma_halt(chan);
- spin_unlock_bh(&chan->desc_lock);
- return 0;
+ /* Remove and free all of the descriptors in the LD queue */
+ fsldma_free_desc_list(chan, &chan->ld_pending);
+ fsldma_free_desc_list(chan, &chan->ld_running);
+ fsldma_free_desc_list(chan, &chan->ld_completed);
+ chan->idle = true;
- case DMA_SLAVE_CONFIG:
- config = (struct dma_slave_config *)arg;
+ spin_unlock_bh(&chan->desc_lock);
+ return 0;
+}
- /* make sure the channel supports setting burst size */
- if (!chan->set_request_count)
- return -ENXIO;
+static int fsl_dma_device_config(struct dma_chan *dchan,
+ struct dma_slave_config *config)
+{
+ struct fsldma_chan *chan;
+ int size;
- /* we set the controller burst size depending on direction */
- if (config->direction == DMA_MEM_TO_DEV)
- size = config->dst_addr_width * config->dst_maxburst;
- else
- size = config->src_addr_width * config->src_maxburst;
+ if (!dchan)
+ return -EINVAL;
- chan->set_request_count(chan, size);
- return 0;
+ chan = to_fsl_chan(dchan);
- default:
+ /* make sure the channel supports setting burst size */
+ if (!chan->set_request_count)
return -ENXIO;
- }
+ /* we set the controller burst size depending on direction */
+ if (config->direction == DMA_MEM_TO_DEV)
+ size = config->dst_addr_width * config->dst_maxburst;
+ else
+ size = config->src_addr_width * config->src_maxburst;
+
+ chan->set_request_count(chan, size);
return 0;
}
+
/**
* fsl_dma_memcpy_issue_pending - Issue the DMA start command
* @chan : Freescale DMA channel
@@ -1395,8 +1368,8 @@ static int fsldma_of_probe(struct platform_device *op)
fdev->common.device_prep_dma_sg = fsl_dma_prep_sg;
fdev->common.device_tx_status = fsl_tx_status;
fdev->common.device_issue_pending = fsl_dma_memcpy_issue_pending;
- fdev->common.device_prep_slave_sg = fsl_dma_prep_slave_sg;
- fdev->common.device_control = fsl_dma_device_control;
+ fdev->common.device_config = fsl_dma_device_config;
+ fdev->common.device_terminate_all = fsl_dma_device_terminate_all;
fdev->common.dev = &op->dev;
dma_set_mask(&(op->dev), DMA_BIT_MASK(36));
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 30/61] dmaengine: fsl-dma: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Freescale Elo DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
While we're at it, remove the useless prep_sg callback.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/fsldma.c | 91 ++++++++++++++++++----------------------------------
1 file changed, 32 insertions(+), 59 deletions(-)
diff --git a/drivers/dma/fsldma.c b/drivers/dma/fsldma.c
index 3d8feb5e4c2f..676555776a46 100644
--- a/drivers/dma/fsldma.c
+++ b/drivers/dma/fsldma.c
@@ -941,37 +941,8 @@ fail:
return NULL;
}
-/**
- * fsl_dma_prep_slave_sg - prepare descriptors for a DMA_SLAVE transaction
- * @chan: DMA channel
- * @sgl: scatterlist to transfer to/from
- * @sg_len: number of entries in @scatterlist
- * @direction: DMA direction
- * @flags: DMAEngine flags
- * @context: transaction context (ignored)
- *
- * Prepare a set of descriptors for a DMA_SLAVE transaction. Following the
- * DMA_SLAVE API, this gets the device-specific information from the
- * chan->private variable.
- */
-static struct dma_async_tx_descriptor *fsl_dma_prep_slave_sg(
- struct dma_chan *dchan, struct scatterlist *sgl, unsigned int sg_len,
- enum dma_transfer_direction direction, unsigned long flags,
- void *context)
-{
- /*
- * This operation is not supported on the Freescale DMA controller
- *
- * However, we need to provide the function pointer to allow the
- * device_control() method to work.
- */
- return NULL;
-}
-
-static int fsl_dma_device_control(struct dma_chan *dchan,
- enum dma_ctrl_cmd cmd, unsigned long arg)
+static int fsl_dma_device_terminate_all(struct dma_chan *dchan)
{
- struct dma_slave_config *config;
struct fsldma_chan *chan;
int size;
@@ -980,45 +951,47 @@ static int fsl_dma_device_control(struct dma_chan *dchan,
chan = to_fsl_chan(dchan);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- spin_lock_bh(&chan->desc_lock);
-
- /* Halt the DMA engine */
- dma_halt(chan);
+ spin_lock_bh(&chan->desc_lock);
- /* Remove and free all of the descriptors in the LD queue */
- fsldma_free_desc_list(chan, &chan->ld_pending);
- fsldma_free_desc_list(chan, &chan->ld_running);
- fsldma_free_desc_list(chan, &chan->ld_completed);
- chan->idle = true;
+ /* Halt the DMA engine */
+ dma_halt(chan);
- spin_unlock_bh(&chan->desc_lock);
- return 0;
+ /* Remove and free all of the descriptors in the LD queue */
+ fsldma_free_desc_list(chan, &chan->ld_pending);
+ fsldma_free_desc_list(chan, &chan->ld_running);
+ fsldma_free_desc_list(chan, &chan->ld_completed);
+ chan->idle = true;
- case DMA_SLAVE_CONFIG:
- config = (struct dma_slave_config *)arg;
+ spin_unlock_bh(&chan->desc_lock);
+ return 0;
+}
- /* make sure the channel supports setting burst size */
- if (!chan->set_request_count)
- return -ENXIO;
+static int fsl_dma_device_config(struct dma_chan *dchan,
+ struct dma_slave_config *config)
+{
+ struct fsldma_chan *chan;
+ int size;
- /* we set the controller burst size depending on direction */
- if (config->direction == DMA_MEM_TO_DEV)
- size = config->dst_addr_width * config->dst_maxburst;
- else
- size = config->src_addr_width * config->src_maxburst;
+ if (!dchan)
+ return -EINVAL;
- chan->set_request_count(chan, size);
- return 0;
+ chan = to_fsl_chan(dchan);
- default:
+ /* make sure the channel supports setting burst size */
+ if (!chan->set_request_count)
return -ENXIO;
- }
+ /* we set the controller burst size depending on direction */
+ if (config->direction == DMA_MEM_TO_DEV)
+ size = config->dst_addr_width * config->dst_maxburst;
+ else
+ size = config->src_addr_width * config->src_maxburst;
+
+ chan->set_request_count(chan, size);
return 0;
}
+
/**
* fsl_dma_memcpy_issue_pending - Issue the DMA start command
* @chan : Freescale DMA channel
@@ -1395,8 +1368,8 @@ static int fsldma_of_probe(struct platform_device *op)
fdev->common.device_prep_dma_sg = fsl_dma_prep_sg;
fdev->common.device_tx_status = fsl_tx_status;
fdev->common.device_issue_pending = fsl_dma_memcpy_issue_pending;
- fdev->common.device_prep_slave_sg = fsl_dma_prep_slave_sg;
- fdev->common.device_control = fsl_dma_device_control;
+ fdev->common.device_config = fsl_dma_device_config;
+ fdev->common.device_terminate_all = fsl_dma_device_terminate_all;
fdev->common.dev = &op->dev;
dma_set_mask(&(op->dev), DMA_BIT_MASK(36));
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 31/61] dmaengine: mpc512x: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Freescale MPC512x DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mpc512x_dma.c | 111 +++++++++++++++++++++-------------------------
1 file changed, 51 insertions(+), 60 deletions(-)
diff --git a/drivers/dma/mpc512x_dma.c b/drivers/dma/mpc512x_dma.c
index 01bec4023de2..57d2457545f3 100644
--- a/drivers/dma/mpc512x_dma.c
+++ b/drivers/dma/mpc512x_dma.c
@@ -800,79 +800,69 @@ err_prep:
return NULL;
}
-static int mpc_dma_device_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int mpc_dma_device_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
{
- struct mpc_dma_chan *mchan;
- struct mpc_dma *mdma;
- struct dma_slave_config *cfg;
+ struct mpc_dma_chan *mchan = dma_chan_to_mpc_dma_chan(chan);
unsigned long flags;
- mchan = dma_chan_to_mpc_dma_chan(chan);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- /* Disable channel requests */
- mdma = dma_chan_to_mpc_dma(chan);
-
- spin_lock_irqsave(&mchan->lock, flags);
-
- out_8(&mdma->regs->dmacerq, chan->chan_id);
- list_splice_tail_init(&mchan->prepared, &mchan->free);
- list_splice_tail_init(&mchan->queued, &mchan->free);
- list_splice_tail_init(&mchan->active, &mchan->free);
-
- spin_unlock_irqrestore(&mchan->lock, flags);
+ /*
+ * Software constraints:
+ * - only transfers between a peripheral device and
+ * memory are supported;
+ * - only peripheral devices with 4-byte FIFO access register
+ * are supported;
+ * - minimal transfer chunk is 4 bytes and consequently
+ * source and destination addresses must be 4-byte aligned
+ * and transfer size must be aligned on (4 * maxburst)
+ * boundary;
+ * - during the transfer RAM address is being incremented by
+ * the size of minimal transfer chunk;
+ * - peripheral port's address is constant during the transfer.
+ */
- return 0;
+ if (cfg->src_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
+ cfg->dst_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
+ !IS_ALIGNED(cfg->src_addr, 4) ||
+ !IS_ALIGNED(cfg->dst_addr, 4)) {
+ return -EINVAL;
+ }
- case DMA_SLAVE_CONFIG:
- /*
- * Software constraints:
- * - only transfers between a peripheral device and
- * memory are supported;
- * - only peripheral devices with 4-byte FIFO access register
- * are supported;
- * - minimal transfer chunk is 4 bytes and consequently
- * source and destination addresses must be 4-byte aligned
- * and transfer size must be aligned on (4 * maxburst)
- * boundary;
- * - during the transfer RAM address is being incremented by
- * the size of minimal transfer chunk;
- * - peripheral port's address is constant during the transfer.
- */
+ spin_lock_irqsave(&mchan->lock, flags);
- cfg = (void *)arg;
+ mchan->src_per_paddr = cfg->src_addr;
+ mchan->src_tcd_nunits = cfg->src_maxburst;
+ mchan->dst_per_paddr = cfg->dst_addr;
+ mchan->dst_tcd_nunits = cfg->dst_maxburst;
- if (cfg->src_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
- cfg->dst_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
- !IS_ALIGNED(cfg->src_addr, 4) ||
- !IS_ALIGNED(cfg->dst_addr, 4)) {
- return -EINVAL;
- }
+ /* Apply defaults */
+ if (mchan->src_tcd_nunits == 0)
+ mchan->src_tcd_nunits = 1;
+ if (mchan->dst_tcd_nunits == 0)
+ mchan->dst_tcd_nunits = 1;
- spin_lock_irqsave(&mchan->lock, flags);
+ spin_unlock_irqrestore(&mchan->lock, flags);
- mchan->src_per_paddr = cfg->src_addr;
- mchan->src_tcd_nunits = cfg->src_maxburst;
- mchan->dst_per_paddr = cfg->dst_addr;
- mchan->dst_tcd_nunits = cfg->dst_maxburst;
+ return 0;
+}
- /* Apply defaults */
- if (mchan->src_tcd_nunits == 0)
- mchan->src_tcd_nunits = 1;
- if (mchan->dst_tcd_nunits == 0)
- mchan->dst_tcd_nunits = 1;
+static int mpc_dma_device_terminate_all(struct dma_chan *chan)
+{
+ struct mpc_dma_chan *mchan = dma_chan_to_mpc_dma_chan(chan);
+ struct mpc_dma *mdma = dma_chan_to_mpc_dma(chan);
+ unsigned long flags;
- spin_unlock_irqrestore(&mchan->lock, flags);
+ /* Disable channel requests */
+ spin_lock_irqsave(&mchan->lock, flags);
- return 0;
+ out_8(&mdma->regs->dmacerq, chan->chan_id);
+ list_splice_tail_init(&mchan->prepared, &mchan->free);
+ list_splice_tail_init(&mchan->queued, &mchan->free);
+ list_splice_tail_init(&mchan->active, &mchan->free);
- default:
- /* Unknown command */
- break;
- }
+ spin_unlock_irqrestore(&mchan->lock, flags);
- return -ENXIO;
+ return 0;
}
static int mpc_dma_probe(struct platform_device *op)
@@ -963,7 +953,8 @@ static int mpc_dma_probe(struct platform_device *op)
dma->device_tx_status = mpc_dma_tx_status;
dma->device_prep_dma_memcpy = mpc_dma_prep_memcpy;
dma->device_prep_slave_sg = mpc_dma_prep_slave_sg;
- dma->device_control = mpc_dma_device_control;
+ dma->device_config = mpc_dma_device_config;
+ dma->device_terminate_all = mpc_dma_device_terminate_all;
INIT_LIST_HEAD(&dma->channels);
dma_cap_set(DMA_MEMCPY, dma->cap_mask);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 31/61] dmaengine: mpc512x: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Freescale MPC512x DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mpc512x_dma.c | 111 +++++++++++++++++++++-------------------------
1 file changed, 51 insertions(+), 60 deletions(-)
diff --git a/drivers/dma/mpc512x_dma.c b/drivers/dma/mpc512x_dma.c
index 01bec4023de2..57d2457545f3 100644
--- a/drivers/dma/mpc512x_dma.c
+++ b/drivers/dma/mpc512x_dma.c
@@ -800,79 +800,69 @@ err_prep:
return NULL;
}
-static int mpc_dma_device_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int mpc_dma_device_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
{
- struct mpc_dma_chan *mchan;
- struct mpc_dma *mdma;
- struct dma_slave_config *cfg;
+ struct mpc_dma_chan *mchan = dma_chan_to_mpc_dma_chan(chan);
unsigned long flags;
- mchan = dma_chan_to_mpc_dma_chan(chan);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- /* Disable channel requests */
- mdma = dma_chan_to_mpc_dma(chan);
-
- spin_lock_irqsave(&mchan->lock, flags);
-
- out_8(&mdma->regs->dmacerq, chan->chan_id);
- list_splice_tail_init(&mchan->prepared, &mchan->free);
- list_splice_tail_init(&mchan->queued, &mchan->free);
- list_splice_tail_init(&mchan->active, &mchan->free);
-
- spin_unlock_irqrestore(&mchan->lock, flags);
+ /*
+ * Software constraints:
+ * - only transfers between a peripheral device and
+ * memory are supported;
+ * - only peripheral devices with 4-byte FIFO access register
+ * are supported;
+ * - minimal transfer chunk is 4 bytes and consequently
+ * source and destination addresses must be 4-byte aligned
+ * and transfer size must be aligned on (4 * maxburst)
+ * boundary;
+ * - during the transfer RAM address is being incremented by
+ * the size of minimal transfer chunk;
+ * - peripheral port's address is constant during the transfer.
+ */
- return 0;
+ if (cfg->src_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
+ cfg->dst_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
+ !IS_ALIGNED(cfg->src_addr, 4) ||
+ !IS_ALIGNED(cfg->dst_addr, 4)) {
+ return -EINVAL;
+ }
- case DMA_SLAVE_CONFIG:
- /*
- * Software constraints:
- * - only transfers between a peripheral device and
- * memory are supported;
- * - only peripheral devices with 4-byte FIFO access register
- * are supported;
- * - minimal transfer chunk is 4 bytes and consequently
- * source and destination addresses must be 4-byte aligned
- * and transfer size must be aligned on (4 * maxburst)
- * boundary;
- * - during the transfer RAM address is being incremented by
- * the size of minimal transfer chunk;
- * - peripheral port's address is constant during the transfer.
- */
+ spin_lock_irqsave(&mchan->lock, flags);
- cfg = (void *)arg;
+ mchan->src_per_paddr = cfg->src_addr;
+ mchan->src_tcd_nunits = cfg->src_maxburst;
+ mchan->dst_per_paddr = cfg->dst_addr;
+ mchan->dst_tcd_nunits = cfg->dst_maxburst;
- if (cfg->src_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
- cfg->dst_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES ||
- !IS_ALIGNED(cfg->src_addr, 4) ||
- !IS_ALIGNED(cfg->dst_addr, 4)) {
- return -EINVAL;
- }
+ /* Apply defaults */
+ if (mchan->src_tcd_nunits == 0)
+ mchan->src_tcd_nunits = 1;
+ if (mchan->dst_tcd_nunits == 0)
+ mchan->dst_tcd_nunits = 1;
- spin_lock_irqsave(&mchan->lock, flags);
+ spin_unlock_irqrestore(&mchan->lock, flags);
- mchan->src_per_paddr = cfg->src_addr;
- mchan->src_tcd_nunits = cfg->src_maxburst;
- mchan->dst_per_paddr = cfg->dst_addr;
- mchan->dst_tcd_nunits = cfg->dst_maxburst;
+ return 0;
+}
- /* Apply defaults */
- if (mchan->src_tcd_nunits == 0)
- mchan->src_tcd_nunits = 1;
- if (mchan->dst_tcd_nunits == 0)
- mchan->dst_tcd_nunits = 1;
+static int mpc_dma_device_terminate_all(struct dma_chan *chan)
+{
+ struct mpc_dma_chan *mchan = dma_chan_to_mpc_dma_chan(chan);
+ struct mpc_dma *mdma = dma_chan_to_mpc_dma(chan);
+ unsigned long flags;
- spin_unlock_irqrestore(&mchan->lock, flags);
+ /* Disable channel requests */
+ spin_lock_irqsave(&mchan->lock, flags);
- return 0;
+ out_8(&mdma->regs->dmacerq, chan->chan_id);
+ list_splice_tail_init(&mchan->prepared, &mchan->free);
+ list_splice_tail_init(&mchan->queued, &mchan->free);
+ list_splice_tail_init(&mchan->active, &mchan->free);
- default:
- /* Unknown command */
- break;
- }
+ spin_unlock_irqrestore(&mchan->lock, flags);
- return -ENXIO;
+ return 0;
}
static int mpc_dma_probe(struct platform_device *op)
@@ -963,7 +953,8 @@ static int mpc_dma_probe(struct platform_device *op)
dma->device_tx_status = mpc_dma_tx_status;
dma->device_prep_dma_memcpy = mpc_dma_prep_memcpy;
dma->device_prep_slave_sg = mpc_dma_prep_slave_sg;
- dma->device_control = mpc_dma_device_control;
+ dma->device_config = mpc_dma_device_config;
+ dma->device_terminate_all = mpc_dma_device_terminate_all;
INIT_LIST_HEAD(&dma->channels);
dma_cap_set(DMA_MEMCPY, dma->cap_mask);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 32/61] dmaengine: mxs: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Freescale MXS DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mxs-dma.c | 59 ++++++++++++++++++++-------------------------------
1 file changed, 23 insertions(+), 36 deletions(-)
diff --git a/drivers/dma/mxs-dma.c b/drivers/dma/mxs-dma.c
index 5ea61201dbf0..834041e5a769 100644
--- a/drivers/dma/mxs-dma.c
+++ b/drivers/dma/mxs-dma.c
@@ -202,8 +202,9 @@ static struct mxs_dma_chan *to_mxs_dma_chan(struct dma_chan *chan)
return container_of(chan, struct mxs_dma_chan, chan);
}
-static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_reset_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
int chan_id = mxs_chan->chan.chan_id;
@@ -250,8 +251,9 @@ static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
mxs_chan->status = DMA_COMPLETE;
}
-static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_enable_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
int chan_id = mxs_chan->chan.chan_id;
@@ -272,13 +274,16 @@ static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
mxs_chan->reset = false;
}
-static void mxs_dma_disable_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_disable_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
+
mxs_chan->status = DMA_COMPLETE;
}
-static void mxs_dma_pause_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_pause_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
int chan_id = mxs_chan->chan.chan_id;
@@ -293,8 +298,9 @@ static void mxs_dma_pause_chan(struct mxs_dma_chan *mxs_chan)
mxs_chan->status = DMA_PAUSED;
}
-static void mxs_dma_resume_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_resume_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
int chan_id = mxs_chan->chan.chan_id;
@@ -383,7 +389,7 @@ static irqreturn_t mxs_dma_int_handler(int irq, void *dev_id)
"%s: error in channel %d\n", __func__,
chan);
mxs_chan->status = DMA_ERROR;
- mxs_dma_reset_chan(mxs_chan);
+ mxs_dma_reset_chan(mxs_chan->chan);
} else if (mxs_chan->status != DMA_COMPLETE) {
if (mxs_chan->flags & MXS_DMA_SG_LOOP) {
mxs_chan->status = DMA_IN_PROGRESS;
@@ -432,7 +438,7 @@ static int mxs_dma_alloc_chan_resources(struct dma_chan *chan)
if (ret)
goto err_clk;
- mxs_dma_reset_chan(mxs_chan);
+ mxs_dma_reset_chan(chan);
dma_async_tx_descriptor_init(&mxs_chan->desc, chan);
mxs_chan->desc.tx_submit = mxs_dma_tx_submit;
@@ -456,7 +462,7 @@ static void mxs_dma_free_chan_resources(struct dma_chan *chan)
struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
- mxs_dma_disable_chan(mxs_chan);
+ mxs_dma_disable_chan(chan);
free_irq(mxs_chan->chan_irq, mxs_dma);
@@ -651,28 +657,14 @@ err_out:
return NULL;
}
-static int mxs_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int mxs_dma_terminate_all(struct dma_chan *chan)
{
struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
- int ret = 0;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- mxs_dma_reset_chan(mxs_chan);
- mxs_dma_disable_chan(mxs_chan);
- break;
- case DMA_PAUSE:
- mxs_dma_pause_chan(mxs_chan);
- break;
- case DMA_RESUME:
- mxs_dma_resume_chan(mxs_chan);
- break;
- default:
- ret = -ENOSYS;
- }
- return ret;
+ mxs_dma_reset_chan(chan);
+ mxs_dma_disable_chan(chan);
+
+ return 0;
}
static enum dma_status mxs_dma_tx_status(struct dma_chan *chan,
@@ -701,13 +693,6 @@ static enum dma_status mxs_dma_tx_status(struct dma_chan *chan,
return mxs_chan->status;
}
-static void mxs_dma_issue_pending(struct dma_chan *chan)
-{
- struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
-
- mxs_dma_enable_chan(mxs_chan);
-}
-
static int __init mxs_dma_init(struct mxs_dma_engine *mxs_dma)
{
int ret;
@@ -860,8 +845,10 @@ static int __init mxs_dma_probe(struct platform_device *pdev)
mxs_dma->dma_device.device_tx_status = mxs_dma_tx_status;
mxs_dma->dma_device.device_prep_slave_sg = mxs_dma_prep_slave_sg;
mxs_dma->dma_device.device_prep_dma_cyclic = mxs_dma_prep_dma_cyclic;
- mxs_dma->dma_device.device_control = mxs_dma_control;
- mxs_dma->dma_device.device_issue_pending = mxs_dma_issue_pending;
+ mxs_dma->dma_device.device_pause = mxs_dma_pause_chan;
+ mxs_dma->dma_device.device_resume = mxs_dma_resume_chan;
+ mxs_dma->dma_device.device_terminate_all = mxs_dma_terminate_all;
+ mxs_dma->dma_device.device_issue_pending = mxs_dma_enable_chan;
ret = dma_async_device_register(&mxs_dma->dma_device);
if (ret) {
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 32/61] dmaengine: mxs: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Freescale MXS DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mxs-dma.c | 59 ++++++++++++++++++++-------------------------------
1 file changed, 23 insertions(+), 36 deletions(-)
diff --git a/drivers/dma/mxs-dma.c b/drivers/dma/mxs-dma.c
index 5ea61201dbf0..834041e5a769 100644
--- a/drivers/dma/mxs-dma.c
+++ b/drivers/dma/mxs-dma.c
@@ -202,8 +202,9 @@ static struct mxs_dma_chan *to_mxs_dma_chan(struct dma_chan *chan)
return container_of(chan, struct mxs_dma_chan, chan);
}
-static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_reset_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
int chan_id = mxs_chan->chan.chan_id;
@@ -250,8 +251,9 @@ static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
mxs_chan->status = DMA_COMPLETE;
}
-static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_enable_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
int chan_id = mxs_chan->chan.chan_id;
@@ -272,13 +274,16 @@ static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
mxs_chan->reset = false;
}
-static void mxs_dma_disable_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_disable_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
+
mxs_chan->status = DMA_COMPLETE;
}
-static void mxs_dma_pause_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_pause_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
int chan_id = mxs_chan->chan.chan_id;
@@ -293,8 +298,9 @@ static void mxs_dma_pause_chan(struct mxs_dma_chan *mxs_chan)
mxs_chan->status = DMA_PAUSED;
}
-static void mxs_dma_resume_chan(struct mxs_dma_chan *mxs_chan)
+static void mxs_dma_resume_chan(struct dma_chan *chan)
{
+ struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
int chan_id = mxs_chan->chan.chan_id;
@@ -383,7 +389,7 @@ static irqreturn_t mxs_dma_int_handler(int irq, void *dev_id)
"%s: error in channel %d\n", __func__,
chan);
mxs_chan->status = DMA_ERROR;
- mxs_dma_reset_chan(mxs_chan);
+ mxs_dma_reset_chan(mxs_chan->chan);
} else if (mxs_chan->status != DMA_COMPLETE) {
if (mxs_chan->flags & MXS_DMA_SG_LOOP) {
mxs_chan->status = DMA_IN_PROGRESS;
@@ -432,7 +438,7 @@ static int mxs_dma_alloc_chan_resources(struct dma_chan *chan)
if (ret)
goto err_clk;
- mxs_dma_reset_chan(mxs_chan);
+ mxs_dma_reset_chan(chan);
dma_async_tx_descriptor_init(&mxs_chan->desc, chan);
mxs_chan->desc.tx_submit = mxs_dma_tx_submit;
@@ -456,7 +462,7 @@ static void mxs_dma_free_chan_resources(struct dma_chan *chan)
struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
- mxs_dma_disable_chan(mxs_chan);
+ mxs_dma_disable_chan(chan);
free_irq(mxs_chan->chan_irq, mxs_dma);
@@ -651,28 +657,14 @@ err_out:
return NULL;
}
-static int mxs_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int mxs_dma_terminate_all(struct dma_chan *chan)
{
struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
- int ret = 0;
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- mxs_dma_reset_chan(mxs_chan);
- mxs_dma_disable_chan(mxs_chan);
- break;
- case DMA_PAUSE:
- mxs_dma_pause_chan(mxs_chan);
- break;
- case DMA_RESUME:
- mxs_dma_resume_chan(mxs_chan);
- break;
- default:
- ret = -ENOSYS;
- }
- return ret;
+ mxs_dma_reset_chan(chan);
+ mxs_dma_disable_chan(chan);
+
+ return 0;
}
static enum dma_status mxs_dma_tx_status(struct dma_chan *chan,
@@ -701,13 +693,6 @@ static enum dma_status mxs_dma_tx_status(struct dma_chan *chan,
return mxs_chan->status;
}
-static void mxs_dma_issue_pending(struct dma_chan *chan)
-{
- struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
-
- mxs_dma_enable_chan(mxs_chan);
-}
-
static int __init mxs_dma_init(struct mxs_dma_engine *mxs_dma)
{
int ret;
@@ -860,8 +845,10 @@ static int __init mxs_dma_probe(struct platform_device *pdev)
mxs_dma->dma_device.device_tx_status = mxs_dma_tx_status;
mxs_dma->dma_device.device_prep_slave_sg = mxs_dma_prep_slave_sg;
mxs_dma->dma_device.device_prep_dma_cyclic = mxs_dma_prep_dma_cyclic;
- mxs_dma->dma_device.device_control = mxs_dma_control;
- mxs_dma->dma_device.device_issue_pending = mxs_dma_issue_pending;
+ mxs_dma->dma_device.device_pause = mxs_dma_pause_chan;
+ mxs_dma->dma_device.device_resume = mxs_dma_resume_chan;
+ mxs_dma->dma_device.device_terminate_all = mxs_dma_terminate_all;
+ mxs_dma->dma_device.device_issue_pending = mxs_dma_enable_chan;
ret = dma_async_device_register(&mxs_dma->dma_device);
if (ret) {
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 33/61] dmaengine: nbpfaxi: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the NBPF AXI DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/nbpfaxi.c | 93 +++++++++++++++++++++++++--------------------------
1 file changed, 46 insertions(+), 47 deletions(-)
diff --git a/drivers/dma/nbpfaxi.c b/drivers/dma/nbpfaxi.c
index 9d20de73ebae..9f151dcd4f73 100644
--- a/drivers/dma/nbpfaxi.c
+++ b/drivers/dma/nbpfaxi.c
@@ -565,13 +565,6 @@ static void nbpf_configure(struct nbpf_device *nbpf)
nbpf_write(nbpf, NBPF_CTRL, NBPF_CTRL_LVINT);
}
-static void nbpf_pause(struct nbpf_channel *chan)
-{
- nbpf_chan_write(chan, NBPF_CHAN_CTRL, NBPF_CHAN_CTRL_SETSUS);
- /* See comment in nbpf_prep_one() */
- nbpf_chan_write(chan, NBPF_CHAN_CTRL, NBPF_CHAN_CTRL_CLREN);
-}
-
/* Generic part */
/* DMA ENGINE functions */
@@ -837,54 +830,58 @@ static void nbpf_chan_idle(struct nbpf_channel *chan)
}
}
-static int nbpf_control(struct dma_chan *dchan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int nbpf_pause(struct dma_chan *dchan)
{
struct nbpf_channel *chan = nbpf_to_chan(dchan);
- struct dma_slave_config *config;
- dev_dbg(dchan->device->dev, "Entry %s(%d)\n", __func__, cmd);
+ dev_dbg(dchan->device->dev, "Entry %s\n", __func__);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- dev_dbg(dchan->device->dev, "Terminating\n");
- nbpf_chan_halt(chan);
- nbpf_chan_idle(chan);
- break;
+ chan->paused = true;
+ nbpf_chan_write(chan, NBPF_CHAN_CTRL, NBPF_CHAN_CTRL_SETSUS);
+ /* See comment in nbpf_prep_one() */
+ nbpf_chan_write(chan, NBPF_CHAN_CTRL, NBPF_CHAN_CTRL_CLREN);
- case DMA_SLAVE_CONFIG:
- if (!arg)
- return -EINVAL;
- config = (struct dma_slave_config *)arg;
+ return 0;
+}
- /*
- * We could check config->slave_id to match chan->terminal here,
- * but with DT they would be coming from the same source, so
- * such a check would be superflous
- */
+static int nbpf_terminate_all(struct dma_chan *dchan)
+{
+ struct nbpf_channel *chan = nbpf_to_chan(dchan);
- chan->slave_dst_addr = config->dst_addr;
- chan->slave_dst_width = nbpf_xfer_size(chan->nbpf,
- config->dst_addr_width, 1);
- chan->slave_dst_burst = nbpf_xfer_size(chan->nbpf,
- config->dst_addr_width,
- config->dst_maxburst);
- chan->slave_src_addr = config->src_addr;
- chan->slave_src_width = nbpf_xfer_size(chan->nbpf,
- config->src_addr_width, 1);
- chan->slave_src_burst = nbpf_xfer_size(chan->nbpf,
- config->src_addr_width,
- config->src_maxburst);
- break;
+ dev_dbg(dchan->device->dev, "Entry %s\n", __func__);
+ dev_dbg(dchan->device->dev, "Terminating\n");
- case DMA_PAUSE:
- chan->paused = true;
- nbpf_pause(chan);
- break;
+ nbpf_chan_halt(chan);
+ nbpf_chan_idle(chan);
- default:
- return -ENXIO;
- }
+ return 0;
+}
+
+static int nbpf_config(struct dma_chan *dchan,
+ struct dma_slave_config *config)
+{
+ struct nbpf_channel *chan = nbpf_to_chan(dchan);
+
+ dev_dbg(dchan->device->dev, "Entry %s\n", __func__);
+
+ /*
+ * We could check config->slave_id to match chan->terminal here,
+ * but with DT they would be coming from the same source, so
+ * such a check would be superflous
+ */
+
+ chan->slave_dst_addr = config->dst_addr;
+ chan->slave_dst_width = nbpf_xfer_size(chan->nbpf,
+ config->dst_addr_width, 1);
+ chan->slave_dst_burst = nbpf_xfer_size(chan->nbpf,
+ config->dst_addr_width,
+ config->dst_maxburst);
+ chan->slave_src_addr = config->src_addr;
+ chan->slave_src_width = nbpf_xfer_size(chan->nbpf,
+ config->src_addr_width, 1);
+ chan->slave_src_burst = nbpf_xfer_size(chan->nbpf,
+ config->src_addr_width,
+ config->src_maxburst);
return 0;
}
@@ -1426,7 +1423,9 @@ static int nbpf_probe(struct platform_device *pdev)
/* Compulsory for DMA_SLAVE fields */
dma_dev->device_prep_slave_sg = nbpf_prep_slave_sg;
- dma_dev->device_control = nbpf_control;
+ dma_dev->device_config = nbpf_config;
+ dma_dev->device_pause = nbpf_pause;
+ dma_dev->device_terminate_all = nbpf_terminate_all;
platform_set_drvdata(pdev, nbpf);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 33/61] dmaengine: nbpfaxi: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the NBPF AXI DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/nbpfaxi.c | 93 +++++++++++++++++++++++++--------------------------
1 file changed, 46 insertions(+), 47 deletions(-)
diff --git a/drivers/dma/nbpfaxi.c b/drivers/dma/nbpfaxi.c
index 9d20de73ebae..9f151dcd4f73 100644
--- a/drivers/dma/nbpfaxi.c
+++ b/drivers/dma/nbpfaxi.c
@@ -565,13 +565,6 @@ static void nbpf_configure(struct nbpf_device *nbpf)
nbpf_write(nbpf, NBPF_CTRL, NBPF_CTRL_LVINT);
}
-static void nbpf_pause(struct nbpf_channel *chan)
-{
- nbpf_chan_write(chan, NBPF_CHAN_CTRL, NBPF_CHAN_CTRL_SETSUS);
- /* See comment in nbpf_prep_one() */
- nbpf_chan_write(chan, NBPF_CHAN_CTRL, NBPF_CHAN_CTRL_CLREN);
-}
-
/* Generic part */
/* DMA ENGINE functions */
@@ -837,54 +830,58 @@ static void nbpf_chan_idle(struct nbpf_channel *chan)
}
}
-static int nbpf_control(struct dma_chan *dchan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int nbpf_pause(struct dma_chan *dchan)
{
struct nbpf_channel *chan = nbpf_to_chan(dchan);
- struct dma_slave_config *config;
- dev_dbg(dchan->device->dev, "Entry %s(%d)\n", __func__, cmd);
+ dev_dbg(dchan->device->dev, "Entry %s\n", __func__);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- dev_dbg(dchan->device->dev, "Terminating\n");
- nbpf_chan_halt(chan);
- nbpf_chan_idle(chan);
- break;
+ chan->paused = true;
+ nbpf_chan_write(chan, NBPF_CHAN_CTRL, NBPF_CHAN_CTRL_SETSUS);
+ /* See comment in nbpf_prep_one() */
+ nbpf_chan_write(chan, NBPF_CHAN_CTRL, NBPF_CHAN_CTRL_CLREN);
- case DMA_SLAVE_CONFIG:
- if (!arg)
- return -EINVAL;
- config = (struct dma_slave_config *)arg;
+ return 0;
+}
- /*
- * We could check config->slave_id to match chan->terminal here,
- * but with DT they would be coming from the same source, so
- * such a check would be superflous
- */
+static int nbpf_terminate_all(struct dma_chan *dchan)
+{
+ struct nbpf_channel *chan = nbpf_to_chan(dchan);
- chan->slave_dst_addr = config->dst_addr;
- chan->slave_dst_width = nbpf_xfer_size(chan->nbpf,
- config->dst_addr_width, 1);
- chan->slave_dst_burst = nbpf_xfer_size(chan->nbpf,
- config->dst_addr_width,
- config->dst_maxburst);
- chan->slave_src_addr = config->src_addr;
- chan->slave_src_width = nbpf_xfer_size(chan->nbpf,
- config->src_addr_width, 1);
- chan->slave_src_burst = nbpf_xfer_size(chan->nbpf,
- config->src_addr_width,
- config->src_maxburst);
- break;
+ dev_dbg(dchan->device->dev, "Entry %s\n", __func__);
+ dev_dbg(dchan->device->dev, "Terminating\n");
- case DMA_PAUSE:
- chan->paused = true;
- nbpf_pause(chan);
- break;
+ nbpf_chan_halt(chan);
+ nbpf_chan_idle(chan);
- default:
- return -ENXIO;
- }
+ return 0;
+}
+
+static int nbpf_config(struct dma_chan *dchan,
+ struct dma_slave_config *config)
+{
+ struct nbpf_channel *chan = nbpf_to_chan(dchan);
+
+ dev_dbg(dchan->device->dev, "Entry %s\n", __func__);
+
+ /*
+ * We could check config->slave_id to match chan->terminal here,
+ * but with DT they would be coming from the same source, so
+ * such a check would be superflous
+ */
+
+ chan->slave_dst_addr = config->dst_addr;
+ chan->slave_dst_width = nbpf_xfer_size(chan->nbpf,
+ config->dst_addr_width, 1);
+ chan->slave_dst_burst = nbpf_xfer_size(chan->nbpf,
+ config->dst_addr_width,
+ config->dst_maxburst);
+ chan->slave_src_addr = config->src_addr;
+ chan->slave_src_width = nbpf_xfer_size(chan->nbpf,
+ config->src_addr_width, 1);
+ chan->slave_src_burst = nbpf_xfer_size(chan->nbpf,
+ config->src_addr_width,
+ config->src_maxburst);
return 0;
}
@@ -1426,7 +1423,9 @@ static int nbpf_probe(struct platform_device *pdev)
/* Compulsory for DMA_SLAVE fields */
dma_dev->device_prep_slave_sg = nbpf_prep_slave_sg;
- dma_dev->device_control = nbpf_control;
+ dma_dev->device_config = nbpf_config;
+ dma_dev->device_pause = nbpf_pause;
+ dma_dev->device_terminate_all = nbpf_terminate_all;
platform_set_drvdata(pdev, nbpf);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 34/61] dmaengine: omap: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the TI OMAP DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
1 file changed, 15 insertions(+), 36 deletions(-)
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index a8aa5b6bb5e2..1cadf812f3b3 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c
@@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
return vchan_tx_prep(&c->vc, &d->vd, flags);
}
-static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
+static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
{
+ struct omap_chan *c = to_omap_dma_chan(chan);
+
if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
return -EINVAL;
@@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
return 0;
}
-static int omap_dma_terminate_all(struct omap_chan *c)
+static int omap_dma_terminate_all(struct dma_chan *chan)
{
+ struct omap_chan *c = to_omap_dma_chan(chan);
struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
unsigned long flags;
LIST_HEAD(head);
@@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
return 0;
}
-static int omap_dma_pause(struct omap_chan *c)
+static int omap_dma_pause(struct dma_chan *chan)
{
+ struct omap_chan *c = to_omap_dma_chan(chan);
+
/* Pause/Resume only allowed with cyclic mode */
if (!c->cyclic)
return -EINVAL;
@@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
return 0;
}
-static int omap_dma_resume(struct omap_chan *c)
+static int omap_dma_resume(struct dma_chan *chan)
{
+ struct omap_chan *c = to_omap_dma_chan(chan);
+
/* Pause/Resume only allowed with cyclic mode */
if (!c->cyclic)
return -EINVAL;
@@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
return 0;
}
-static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct omap_chan *c = to_omap_dma_chan(chan);
- int ret;
-
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
- break;
-
- case DMA_TERMINATE_ALL:
- ret = omap_dma_terminate_all(c);
- break;
-
- case DMA_PAUSE:
- ret = omap_dma_pause(c);
- break;
-
- case DMA_RESUME:
- ret = omap_dma_resume(c);
- break;
-
- default:
- ret = -ENXIO;
- break;
- }
-
- return ret;
-}
-
static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
{
struct omap_chan *c;
@@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
od->ddev.device_issue_pending = omap_dma_issue_pending;
od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
- od->ddev.device_control = omap_dma_control;
+ od->ddev.device_config = omap_dma_config;
+ od->ddev.device_pause = omap_dma_pause;
+ od->ddev.device_resume = omap_dma_resume;
+ od->ddev.device_terminate_all = omap_dma_terminate_all;
od->ddev.device_slave_caps = omap_dma_device_slave_caps;
od->ddev.dev = &pdev->dev;
INIT_LIST_HEAD(&od->ddev.channels);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 34/61] dmaengine: omap: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the TI OMAP DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
1 file changed, 15 insertions(+), 36 deletions(-)
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index a8aa5b6bb5e2..1cadf812f3b3 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c
@@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
return vchan_tx_prep(&c->vc, &d->vd, flags);
}
-static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
+static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
{
+ struct omap_chan *c = to_omap_dma_chan(chan);
+
if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
return -EINVAL;
@@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
return 0;
}
-static int omap_dma_terminate_all(struct omap_chan *c)
+static int omap_dma_terminate_all(struct dma_chan *chan)
{
+ struct omap_chan *c = to_omap_dma_chan(chan);
struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
unsigned long flags;
LIST_HEAD(head);
@@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
return 0;
}
-static int omap_dma_pause(struct omap_chan *c)
+static int omap_dma_pause(struct dma_chan *chan)
{
+ struct omap_chan *c = to_omap_dma_chan(chan);
+
/* Pause/Resume only allowed with cyclic mode */
if (!c->cyclic)
return -EINVAL;
@@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
return 0;
}
-static int omap_dma_resume(struct omap_chan *c)
+static int omap_dma_resume(struct dma_chan *chan)
{
+ struct omap_chan *c = to_omap_dma_chan(chan);
+
/* Pause/Resume only allowed with cyclic mode */
if (!c->cyclic)
return -EINVAL;
@@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
return 0;
}
-static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct omap_chan *c = to_omap_dma_chan(chan);
- int ret;
-
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
- break;
-
- case DMA_TERMINATE_ALL:
- ret = omap_dma_terminate_all(c);
- break;
-
- case DMA_PAUSE:
- ret = omap_dma_pause(c);
- break;
-
- case DMA_RESUME:
- ret = omap_dma_resume(c);
- break;
-
- default:
- ret = -ENXIO;
- break;
- }
-
- return ret;
-}
-
static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
{
struct omap_chan *c;
@@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
od->ddev.device_issue_pending = omap_dma_issue_pending;
od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
- od->ddev.device_control = omap_dma_control;
+ od->ddev.device_config = omap_dma_config;
+ od->ddev.device_pause = omap_dma_pause;
+ od->ddev.device_resume = omap_dma_resume;
+ od->ddev.device_terminate_all = omap_dma_terminate_all;
od->ddev.device_slave_caps = omap_dma_device_slave_caps;
od->ddev.dev = &pdev->dev;
INIT_LIST_HEAD(&od->ddev.channels);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 35/61] dmaengine: pl330: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the AMBA PL330 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/pl330.c | 108 ++++++++++++++++++++++++++--------------------------
1 file changed, 53 insertions(+), 55 deletions(-)
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 3c9f715cb8d7..c2db237e455b 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2062,72 +2062,69 @@ static int pl330_alloc_chan_resources(struct dma_chan *chan)
return 1;
}
-static int pl330_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd, unsigned long arg)
+static int pl330_config(struct dma_chan *chan,
+ struct dma_slave_config *slave_config)
+{
+ struct dma_pl330_chan *pch = to_pchan(chan);
+
+ if (slave_config->direction == DMA_MEM_TO_DEV) {
+ if (slave_config->dst_addr)
+ pch->fifo_addr = slave_config->dst_addr;
+ if (slave_config->dst_addr_width)
+ pch->burst_sz = __ffs(slave_config->dst_addr_width);
+ if (slave_config->dst_maxburst)
+ pch->burst_len = slave_config->dst_maxburst;
+ } else if (slave_config->direction == DMA_DEV_TO_MEM) {
+ if (slave_config->src_addr)
+ pch->fifo_addr = slave_config->src_addr;
+ if (slave_config->src_addr_width)
+ pch->burst_sz = __ffs(slave_config->src_addr_width);
+ if (slave_config->src_maxburst)
+ pch->burst_len = slave_config->src_maxburst;
+ }
+
+ return 0;
+}
+
+static int pl330_terminate_all(struct dma_chan *chan)
{
struct dma_pl330_chan *pch = to_pchan(chan);
struct dma_pl330_desc *desc;
unsigned long flags;
struct pl330_dmac *pl330 = pch->dmac;
- struct dma_slave_config *slave_config;
LIST_HEAD(list);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- spin_lock_irqsave(&pch->lock, flags);
-
- spin_lock(&pl330->lock);
- _stop(pch->thread);
- spin_unlock(&pl330->lock);
+ spin_lock_irqsave(&pch->lock, flags);
- pch->thread->req[0].desc = NULL;
- pch->thread->req[1].desc = NULL;
- pch->thread->req_running = -1;
+ spin_lock(&pl330->lock);
+ _stop(pch->thread);
+ spin_unlock(&pl330->lock);
- /* Mark all desc done */
- list_for_each_entry(desc, &pch->submitted_list, node) {
- desc->status = FREE;
- dma_cookie_complete(&desc->txd);
- }
+ pch->thread->req[0].desc = NULL;
+ pch->thread->req[1].desc = NULL;
+ pch->thread->req_running = -1;
- list_for_each_entry(desc, &pch->work_list , node) {
- desc->status = FREE;
- dma_cookie_complete(&desc->txd);
- }
+ /* Mark all desc done */
+ list_for_each_entry(desc, &pch->submitted_list, node) {
+ desc->status = FREE;
+ dma_cookie_complete(&desc->txd);
+ }
- list_for_each_entry(desc, &pch->completed_list , node) {
- desc->status = FREE;
- dma_cookie_complete(&desc->txd);
- }
+ list_for_each_entry(desc, &pch->work_list , node) {
+ desc->status = FREE;
+ dma_cookie_complete(&desc->txd);
+ }
- list_splice_tail_init(&pch->submitted_list, &pl330->desc_pool);
- list_splice_tail_init(&pch->work_list, &pl330->desc_pool);
- list_splice_tail_init(&pch->completed_list, &pl330->desc_pool);
- spin_unlock_irqrestore(&pch->lock, flags);
- break;
- case DMA_SLAVE_CONFIG:
- slave_config = (struct dma_slave_config *)arg;
-
- if (slave_config->direction == DMA_MEM_TO_DEV) {
- if (slave_config->dst_addr)
- pch->fifo_addr = slave_config->dst_addr;
- if (slave_config->dst_addr_width)
- pch->burst_sz = __ffs(slave_config->dst_addr_width);
- if (slave_config->dst_maxburst)
- pch->burst_len = slave_config->dst_maxburst;
- } else if (slave_config->direction == DMA_DEV_TO_MEM) {
- if (slave_config->src_addr)
- pch->fifo_addr = slave_config->src_addr;
- if (slave_config->src_addr_width)
- pch->burst_sz = __ffs(slave_config->src_addr_width);
- if (slave_config->src_maxburst)
- pch->burst_len = slave_config->src_maxburst;
- }
- break;
- default:
- dev_err(pch->dmac->ddma.dev, "Not supported command.\n");
- return -ENXIO;
+ list_for_each_entry(desc, &pch->completed_list , node) {
+ desc->status = FREE;
+ dma_cookie_complete(&desc->txd);
}
+ list_splice_tail_init(&pch->submitted_list, &pl330->desc_pool);
+ list_splice_tail_init(&pch->work_list, &pl330->desc_pool);
+ list_splice_tail_init(&pch->completed_list, &pl330->desc_pool);
+ spin_unlock_irqrestore(&pch->lock, flags);
+
return 0;
}
@@ -2714,7 +2711,8 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
pd->device_prep_dma_cyclic = pl330_prep_dma_cyclic;
pd->device_tx_status = pl330_tx_status;
pd->device_prep_slave_sg = pl330_prep_slave_sg;
- pd->device_control = pl330_control;
+ pd->device_config = pl330_config;
+ pd->device_terminate_all = pl330_terminate_all;
pd->device_issue_pending = pl330_issue_pending;
pd->device_slave_caps = pl330_dma_device_slave_caps;
@@ -2762,7 +2760,7 @@ probe_err3:
/* Flush the channel */
if (pch->thread) {
- pl330_control(&pch->chan, DMA_TERMINATE_ALL, 0);
+ pl330_terminate_all(&pch->chan);
pl330_free_chan_resources(&pch->chan);
}
}
@@ -2791,7 +2789,7 @@ static int pl330_remove(struct amba_device *adev)
/* Flush the channel */
if (pch->thread) {
- pl330_control(&pch->chan, DMA_TERMINATE_ALL, 0);
+ pl330_terminate_all(&pch->chan);
pl330_free_chan_resources(&pch->chan);
}
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 35/61] dmaengine: pl330: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the AMBA PL330 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/pl330.c | 108 ++++++++++++++++++++++++++--------------------------
1 file changed, 53 insertions(+), 55 deletions(-)
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index 3c9f715cb8d7..c2db237e455b 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2062,72 +2062,69 @@ static int pl330_alloc_chan_resources(struct dma_chan *chan)
return 1;
}
-static int pl330_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd, unsigned long arg)
+static int pl330_config(struct dma_chan *chan,
+ struct dma_slave_config *slave_config)
+{
+ struct dma_pl330_chan *pch = to_pchan(chan);
+
+ if (slave_config->direction == DMA_MEM_TO_DEV) {
+ if (slave_config->dst_addr)
+ pch->fifo_addr = slave_config->dst_addr;
+ if (slave_config->dst_addr_width)
+ pch->burst_sz = __ffs(slave_config->dst_addr_width);
+ if (slave_config->dst_maxburst)
+ pch->burst_len = slave_config->dst_maxburst;
+ } else if (slave_config->direction == DMA_DEV_TO_MEM) {
+ if (slave_config->src_addr)
+ pch->fifo_addr = slave_config->src_addr;
+ if (slave_config->src_addr_width)
+ pch->burst_sz = __ffs(slave_config->src_addr_width);
+ if (slave_config->src_maxburst)
+ pch->burst_len = slave_config->src_maxburst;
+ }
+
+ return 0;
+}
+
+static int pl330_terminate_all(struct dma_chan *chan)
{
struct dma_pl330_chan *pch = to_pchan(chan);
struct dma_pl330_desc *desc;
unsigned long flags;
struct pl330_dmac *pl330 = pch->dmac;
- struct dma_slave_config *slave_config;
LIST_HEAD(list);
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- spin_lock_irqsave(&pch->lock, flags);
-
- spin_lock(&pl330->lock);
- _stop(pch->thread);
- spin_unlock(&pl330->lock);
+ spin_lock_irqsave(&pch->lock, flags);
- pch->thread->req[0].desc = NULL;
- pch->thread->req[1].desc = NULL;
- pch->thread->req_running = -1;
+ spin_lock(&pl330->lock);
+ _stop(pch->thread);
+ spin_unlock(&pl330->lock);
- /* Mark all desc done */
- list_for_each_entry(desc, &pch->submitted_list, node) {
- desc->status = FREE;
- dma_cookie_complete(&desc->txd);
- }
+ pch->thread->req[0].desc = NULL;
+ pch->thread->req[1].desc = NULL;
+ pch->thread->req_running = -1;
- list_for_each_entry(desc, &pch->work_list , node) {
- desc->status = FREE;
- dma_cookie_complete(&desc->txd);
- }
+ /* Mark all desc done */
+ list_for_each_entry(desc, &pch->submitted_list, node) {
+ desc->status = FREE;
+ dma_cookie_complete(&desc->txd);
+ }
- list_for_each_entry(desc, &pch->completed_list , node) {
- desc->status = FREE;
- dma_cookie_complete(&desc->txd);
- }
+ list_for_each_entry(desc, &pch->work_list , node) {
+ desc->status = FREE;
+ dma_cookie_complete(&desc->txd);
+ }
- list_splice_tail_init(&pch->submitted_list, &pl330->desc_pool);
- list_splice_tail_init(&pch->work_list, &pl330->desc_pool);
- list_splice_tail_init(&pch->completed_list, &pl330->desc_pool);
- spin_unlock_irqrestore(&pch->lock, flags);
- break;
- case DMA_SLAVE_CONFIG:
- slave_config = (struct dma_slave_config *)arg;
-
- if (slave_config->direction == DMA_MEM_TO_DEV) {
- if (slave_config->dst_addr)
- pch->fifo_addr = slave_config->dst_addr;
- if (slave_config->dst_addr_width)
- pch->burst_sz = __ffs(slave_config->dst_addr_width);
- if (slave_config->dst_maxburst)
- pch->burst_len = slave_config->dst_maxburst;
- } else if (slave_config->direction == DMA_DEV_TO_MEM) {
- if (slave_config->src_addr)
- pch->fifo_addr = slave_config->src_addr;
- if (slave_config->src_addr_width)
- pch->burst_sz = __ffs(slave_config->src_addr_width);
- if (slave_config->src_maxburst)
- pch->burst_len = slave_config->src_maxburst;
- }
- break;
- default:
- dev_err(pch->dmac->ddma.dev, "Not supported command.\n");
- return -ENXIO;
+ list_for_each_entry(desc, &pch->completed_list , node) {
+ desc->status = FREE;
+ dma_cookie_complete(&desc->txd);
}
+ list_splice_tail_init(&pch->submitted_list, &pl330->desc_pool);
+ list_splice_tail_init(&pch->work_list, &pl330->desc_pool);
+ list_splice_tail_init(&pch->completed_list, &pl330->desc_pool);
+ spin_unlock_irqrestore(&pch->lock, flags);
+
return 0;
}
@@ -2714,7 +2711,8 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
pd->device_prep_dma_cyclic = pl330_prep_dma_cyclic;
pd->device_tx_status = pl330_tx_status;
pd->device_prep_slave_sg = pl330_prep_slave_sg;
- pd->device_control = pl330_control;
+ pd->device_config = pl330_config;
+ pd->device_terminate_all = pl330_terminate_all;
pd->device_issue_pending = pl330_issue_pending;
pd->device_slave_caps = pl330_dma_device_slave_caps;
@@ -2762,7 +2760,7 @@ probe_err3:
/* Flush the channel */
if (pch->thread) {
- pl330_control(&pch->chan, DMA_TERMINATE_ALL, 0);
+ pl330_terminate_all(&pch->chan);
pl330_free_chan_resources(&pch->chan);
}
}
@@ -2791,7 +2789,7 @@ static int pl330_remove(struct amba_device *adev)
/* Flush the channel */
if (pch->thread) {
- pl330_control(&pch->chan, DMA_TERMINATE_ALL, 0);
+ pl330_terminate_all(&pch->chan);
pl330_free_chan_resources(&pch->chan);
}
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 36/61] dmaengine: bam-dma: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Qualcomm BAM DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/qcom_bam_dma.c | 85 +++++++++++++++++++++++-----------------------
1 file changed, 43 insertions(+), 42 deletions(-)
diff --git a/drivers/dma/qcom_bam_dma.c b/drivers/dma/qcom_bam_dma.c
index 3122a99ec06b..d7a33b3ac466 100644
--- a/drivers/dma/qcom_bam_dma.c
+++ b/drivers/dma/qcom_bam_dma.c
@@ -530,11 +530,18 @@ static void bam_free_chan(struct dma_chan *chan)
* Sets slave configuration for channel
*
*/
-static void bam_slave_config(struct bam_chan *bchan,
- struct dma_slave_config *cfg)
+static int bam_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
{
+ struct bam_chan *bchan = to_bam_chan(chan);
+ unsigned long flag;
+
+ spin_lock_irqsave(&bchan->vc.lock, flag);
memcpy(&bchan->slave, cfg, sizeof(*cfg));
bchan->reconfigure = 1;
+ spin_unlock_irqrestore(&bchan->vc.lock, flag);
+
+ return 0;
}
/**
@@ -627,8 +634,9 @@ err_out:
* No callbacks are done
*
*/
-static void bam_dma_terminate_all(struct bam_chan *bchan)
+static int bam_dma_terminate_all(struct dma_chan *chan)
{
+ struct bam_chan *bchan = to_bam_chan(chan);
unsigned long flag;
LIST_HEAD(head);
@@ -643,56 +651,46 @@ static void bam_dma_terminate_all(struct bam_chan *bchan)
spin_unlock_irqrestore(&bchan->vc.lock, flag);
vchan_dma_desc_free_list(&bchan->vc, &head);
+
+ return 0;
}
/**
- * bam_control - DMA device control
+ * bam_pause - Pause DMA channel
* @chan: dma channel
- * @cmd: control cmd
- * @arg: cmd argument
*
- * Perform DMA control command
+ */
+static int bam_pause(struct dma_chan *chan)
+{
+ struct bam_chan *bchan = to_bam_chan(chan);
+ struct bam_device *bdev = bchan->bdev;
+ unsigned long flag;
+
+ spin_lock_irqsave(&bchan->vc.lock, flag);
+ writel_relaxed(1, bam_addr(bdev, bchan->id, BAM_P_HALT));
+ bchan->paused = 1;
+ spin_unlock_irqrestore(&bchan->vc.lock, flag);
+
+ return 0;
+}
+
+/**
+ * bam_resume - Resume DMA channel operations
+ * @chan: dma channel
*
*/
-static int bam_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int bam_resume(struct dma_chan *chan)
{
struct bam_chan *bchan = to_bam_chan(chan);
struct bam_device *bdev = bchan->bdev;
- int ret = 0;
unsigned long flag;
- switch (cmd) {
- case DMA_PAUSE:
- spin_lock_irqsave(&bchan->vc.lock, flag);
- writel_relaxed(1, bam_addr(bdev, bchan->id, BAM_P_HALT));
- bchan->paused = 1;
- spin_unlock_irqrestore(&bchan->vc.lock, flag);
- break;
-
- case DMA_RESUME:
- spin_lock_irqsave(&bchan->vc.lock, flag);
- writel_relaxed(0, bam_addr(bdev, bchan->id, BAM_P_HALT));
- bchan->paused = 0;
- spin_unlock_irqrestore(&bchan->vc.lock, flag);
- break;
-
- case DMA_TERMINATE_ALL:
- bam_dma_terminate_all(bchan);
- break;
-
- case DMA_SLAVE_CONFIG:
- spin_lock_irqsave(&bchan->vc.lock, flag);
- bam_slave_config(bchan, (struct dma_slave_config *)arg);
- spin_unlock_irqrestore(&bchan->vc.lock, flag);
- break;
-
- default:
- ret = -ENXIO;
- break;
- }
+ spin_lock_irqsave(&bchan->vc.lock, flag);
+ writel_relaxed(0, bam_addr(bdev, bchan->id, BAM_P_HALT));
+ bchan->paused = 0;
+ spin_unlock_irqrestore(&bchan->vc.lock, flag);
- return ret;
+ return 0;
}
/**
@@ -1148,7 +1146,10 @@ static int bam_dma_probe(struct platform_device *pdev)
bdev->common.device_alloc_chan_resources = bam_alloc_chan;
bdev->common.device_free_chan_resources = bam_free_chan;
bdev->common.device_prep_slave_sg = bam_prep_slave_sg;
- bdev->common.device_control = bam_control;
+ bdev->common.device_config = bam_slave_config;
+ bdev->common.device_pause = bam_pause;
+ bdev->common.device_resume = bam_resume;
+ bdev->common.device_terminate_all = bam_dma_terminate_all;
bdev->common.device_issue_pending = bam_issue_pending;
bdev->common.device_tx_status = bam_tx_status;
bdev->common.dev = bdev->dev;
@@ -1187,7 +1188,7 @@ static int bam_dma_remove(struct platform_device *pdev)
devm_free_irq(bdev->dev, bdev->irq, bdev);
for (i = 0; i < bdev->num_channels; i++) {
- bam_dma_terminate_all(&bdev->channels[i]);
+ bam_dma_terminate_all(&bdev->channels[i].vc.chan);
tasklet_kill(&bdev->channels[i].vc.task);
dma_free_writecombine(bdev->dev, BAM_DESC_FIFO_SIZE,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 36/61] dmaengine: bam-dma: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Qualcomm BAM DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/qcom_bam_dma.c | 85 +++++++++++++++++++++++-----------------------
1 file changed, 43 insertions(+), 42 deletions(-)
diff --git a/drivers/dma/qcom_bam_dma.c b/drivers/dma/qcom_bam_dma.c
index 3122a99ec06b..d7a33b3ac466 100644
--- a/drivers/dma/qcom_bam_dma.c
+++ b/drivers/dma/qcom_bam_dma.c
@@ -530,11 +530,18 @@ static void bam_free_chan(struct dma_chan *chan)
* Sets slave configuration for channel
*
*/
-static void bam_slave_config(struct bam_chan *bchan,
- struct dma_slave_config *cfg)
+static int bam_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
{
+ struct bam_chan *bchan = to_bam_chan(chan);
+ unsigned long flag;
+
+ spin_lock_irqsave(&bchan->vc.lock, flag);
memcpy(&bchan->slave, cfg, sizeof(*cfg));
bchan->reconfigure = 1;
+ spin_unlock_irqrestore(&bchan->vc.lock, flag);
+
+ return 0;
}
/**
@@ -627,8 +634,9 @@ err_out:
* No callbacks are done
*
*/
-static void bam_dma_terminate_all(struct bam_chan *bchan)
+static int bam_dma_terminate_all(struct dma_chan *chan)
{
+ struct bam_chan *bchan = to_bam_chan(chan);
unsigned long flag;
LIST_HEAD(head);
@@ -643,56 +651,46 @@ static void bam_dma_terminate_all(struct bam_chan *bchan)
spin_unlock_irqrestore(&bchan->vc.lock, flag);
vchan_dma_desc_free_list(&bchan->vc, &head);
+
+ return 0;
}
/**
- * bam_control - DMA device control
+ * bam_pause - Pause DMA channel
* @chan: dma channel
- * @cmd: control cmd
- * @arg: cmd argument
*
- * Perform DMA control command
+ */
+static int bam_pause(struct dma_chan *chan)
+{
+ struct bam_chan *bchan = to_bam_chan(chan);
+ struct bam_device *bdev = bchan->bdev;
+ unsigned long flag;
+
+ spin_lock_irqsave(&bchan->vc.lock, flag);
+ writel_relaxed(1, bam_addr(bdev, bchan->id, BAM_P_HALT));
+ bchan->paused = 1;
+ spin_unlock_irqrestore(&bchan->vc.lock, flag);
+
+ return 0;
+}
+
+/**
+ * bam_resume - Resume DMA channel operations
+ * @chan: dma channel
*
*/
-static int bam_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int bam_resume(struct dma_chan *chan)
{
struct bam_chan *bchan = to_bam_chan(chan);
struct bam_device *bdev = bchan->bdev;
- int ret = 0;
unsigned long flag;
- switch (cmd) {
- case DMA_PAUSE:
- spin_lock_irqsave(&bchan->vc.lock, flag);
- writel_relaxed(1, bam_addr(bdev, bchan->id, BAM_P_HALT));
- bchan->paused = 1;
- spin_unlock_irqrestore(&bchan->vc.lock, flag);
- break;
-
- case DMA_RESUME:
- spin_lock_irqsave(&bchan->vc.lock, flag);
- writel_relaxed(0, bam_addr(bdev, bchan->id, BAM_P_HALT));
- bchan->paused = 0;
- spin_unlock_irqrestore(&bchan->vc.lock, flag);
- break;
-
- case DMA_TERMINATE_ALL:
- bam_dma_terminate_all(bchan);
- break;
-
- case DMA_SLAVE_CONFIG:
- spin_lock_irqsave(&bchan->vc.lock, flag);
- bam_slave_config(bchan, (struct dma_slave_config *)arg);
- spin_unlock_irqrestore(&bchan->vc.lock, flag);
- break;
-
- default:
- ret = -ENXIO;
- break;
- }
+ spin_lock_irqsave(&bchan->vc.lock, flag);
+ writel_relaxed(0, bam_addr(bdev, bchan->id, BAM_P_HALT));
+ bchan->paused = 0;
+ spin_unlock_irqrestore(&bchan->vc.lock, flag);
- return ret;
+ return 0;
}
/**
@@ -1148,7 +1146,10 @@ static int bam_dma_probe(struct platform_device *pdev)
bdev->common.device_alloc_chan_resources = bam_alloc_chan;
bdev->common.device_free_chan_resources = bam_free_chan;
bdev->common.device_prep_slave_sg = bam_prep_slave_sg;
- bdev->common.device_control = bam_control;
+ bdev->common.device_config = bam_slave_config;
+ bdev->common.device_pause = bam_pause;
+ bdev->common.device_resume = bam_resume;
+ bdev->common.device_terminate_all = bam_dma_terminate_all;
bdev->common.device_issue_pending = bam_issue_pending;
bdev->common.device_tx_status = bam_tx_status;
bdev->common.dev = bdev->dev;
@@ -1187,7 +1188,7 @@ static int bam_dma_remove(struct platform_device *pdev)
devm_free_irq(bdev->dev, bdev->irq, bdev);
for (i = 0; i < bdev->num_channels; i++) {
- bam_dma_terminate_all(&bdev->channels[i]);
+ bam_dma_terminate_all(&bdev->channels[i].vc.chan);
tasklet_kill(&bdev->channels[i].vc.task);
dma_free_writecombine(bdev->dev, BAM_DESC_FIFO_SIZE,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 37/61] dmaengine: s3c24xx: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Samsung S3C24xxx DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/s3c24xx-dma.c | 75 +++++++++++++++++++++++------------------------
1 file changed, 36 insertions(+), 39 deletions(-)
diff --git a/drivers/dma/s3c24xx-dma.c b/drivers/dma/s3c24xx-dma.c
index 6941a77521c3..231f76acebde 100644
--- a/drivers/dma/s3c24xx-dma.c
+++ b/drivers/dma/s3c24xx-dma.c
@@ -384,20 +384,30 @@ static u32 s3c24xx_dma_getbytes_chan(struct s3c24xx_dma_chan *s3cchan)
return tc * txd->width;
}
-static int s3c24xx_dma_set_runtime_config(struct s3c24xx_dma_chan *s3cchan,
+static int s3c24xx_dma_set_runtime_config(struct dma_chan *chan,
struct dma_slave_config *config)
{
- if (!s3cchan->slave)
- return -EINVAL;
+ struct s3c24xx_dma_chan *s3cchan = to_s3c24xx_dma_chan(chan);
+ unsigned long flags;
+ int ret = 0;
/* Reject definitely invalid configurations */
if (config->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
config->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
return -EINVAL;
+ spin_lock_irqsave(&s3cchan->vc.lock, flags);
+
+ if (!s3cchan->slave) {
+ ret = -EINVAL;
+ goto out;
+ }
+
s3cchan->cfg = *config;
- return 0;
+out:
+ spin_lock_irqrestore(&s3cchan->vc.lock, flags);
+ return ret;
}
/*
@@ -703,53 +713,38 @@ static irqreturn_t s3c24xx_dma_irq(int irq, void *data)
* The DMA ENGINE API
*/
-static int s3c24xx_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int s3c24xx_dma_terminate_all(struct dma_chan *chan)
{
struct s3c24xx_dma_chan *s3cchan = to_s3c24xx_dma_chan(chan);
struct s3c24xx_dma_engine *s3cdma = s3cchan->host;
unsigned long flags;
- int ret = 0;
spin_lock_irqsave(&s3cchan->vc.lock, flags);
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- ret = s3c24xx_dma_set_runtime_config(s3cchan,
- (struct dma_slave_config *)arg);
- break;
- case DMA_TERMINATE_ALL:
- if (!s3cchan->phy && !s3cchan->at) {
- dev_err(&s3cdma->pdev->dev, "trying to terminate already stopped channel %d\n",
- s3cchan->id);
- ret = -EINVAL;
- break;
- }
-
- s3cchan->state = S3C24XX_DMA_CHAN_IDLE;
+ if (!s3cchan->phy && !s3cchan->at) {
+ dev_err(&s3cdma->pdev->dev, "trying to terminate already stopped channel %d\n",
+ s3cchan->id);
+ return -EINVAL;
+ }
- /* Mark physical channel as free */
- if (s3cchan->phy)
- s3c24xx_dma_phy_free(s3cchan);
+ s3cchan->state = S3C24XX_DMA_CHAN_IDLE;
- /* Dequeue current job */
- if (s3cchan->at) {
- s3c24xx_dma_desc_free(&s3cchan->at->vd);
- s3cchan->at = NULL;
- }
+ /* Mark physical channel as free */
+ if (s3cchan->phy)
+ s3c24xx_dma_phy_free(s3cchan);
- /* Dequeue jobs not yet fired as well */
- s3c24xx_dma_free_txd_list(s3cdma, s3cchan);
- break;
- default:
- /* Unknown command */
- ret = -ENXIO;
- break;
+ /* Dequeue current job */
+ if (s3cchan->at) {
+ s3c24xx_dma_desc_free(&s3cchan->at->vd);
+ s3cchan->at = NULL;
}
+ /* Dequeue jobs not yet fired as well */
+ s3c24xx_dma_free_txd_list(s3cdma, s3cchan);
+
spin_unlock_irqrestore(&s3cchan->vc.lock, flags);
- return ret;
+ return 0;
}
static int s3c24xx_dma_alloc_chan_resources(struct dma_chan *chan)
@@ -1300,7 +1295,8 @@ static int s3c24xx_dma_probe(struct platform_device *pdev)
s3cdma->memcpy.device_prep_dma_memcpy = s3c24xx_dma_prep_memcpy;
s3cdma->memcpy.device_tx_status = s3c24xx_dma_tx_status;
s3cdma->memcpy.device_issue_pending = s3c24xx_dma_issue_pending;
- s3cdma->memcpy.device_control = s3c24xx_dma_control;
+ s3cdma->memcpy.device_config = s3c24xx_dma_set_runtime_config;
+ s3cdma->memcpy.device_terminate_all = s3c24xx_dma_terminate_all;
/* Initialize slave engine for SoC internal dedicated peripherals */
dma_cap_set(DMA_SLAVE, s3cdma->slave.cap_mask);
@@ -1315,7 +1311,8 @@ static int s3c24xx_dma_probe(struct platform_device *pdev)
s3cdma->slave.device_issue_pending = s3c24xx_dma_issue_pending;
s3cdma->slave.device_prep_slave_sg = s3c24xx_dma_prep_slave_sg;
s3cdma->slave.device_prep_dma_cyclic = s3c24xx_dma_prep_dma_cyclic;
- s3cdma->slave.device_control = s3c24xx_dma_control;
+ s3cdma->slave.device_config = s3c24xx_dma_set_runtime_config;
+ s3cdma->slave.device_terminate_all = s3c24xx_dma_terminate_all;
/* Register as many memcpy channels as there are physical channels */
ret = s3c24xx_dma_init_virtual_channels(s3cdma, &s3cdma->memcpy,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 37/61] dmaengine: s3c24xx: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Samsung S3C24xxx DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/s3c24xx-dma.c | 75 +++++++++++++++++++++++------------------------
1 file changed, 36 insertions(+), 39 deletions(-)
diff --git a/drivers/dma/s3c24xx-dma.c b/drivers/dma/s3c24xx-dma.c
index 6941a77521c3..231f76acebde 100644
--- a/drivers/dma/s3c24xx-dma.c
+++ b/drivers/dma/s3c24xx-dma.c
@@ -384,20 +384,30 @@ static u32 s3c24xx_dma_getbytes_chan(struct s3c24xx_dma_chan *s3cchan)
return tc * txd->width;
}
-static int s3c24xx_dma_set_runtime_config(struct s3c24xx_dma_chan *s3cchan,
+static int s3c24xx_dma_set_runtime_config(struct dma_chan *chan,
struct dma_slave_config *config)
{
- if (!s3cchan->slave)
- return -EINVAL;
+ struct s3c24xx_dma_chan *s3cchan = to_s3c24xx_dma_chan(chan);
+ unsigned long flags;
+ int ret = 0;
/* Reject definitely invalid configurations */
if (config->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
config->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
return -EINVAL;
+ spin_lock_irqsave(&s3cchan->vc.lock, flags);
+
+ if (!s3cchan->slave) {
+ ret = -EINVAL;
+ goto out;
+ }
+
s3cchan->cfg = *config;
- return 0;
+out:
+ spin_lock_irqrestore(&s3cchan->vc.lock, flags);
+ return ret;
}
/*
@@ -703,53 +713,38 @@ static irqreturn_t s3c24xx_dma_irq(int irq, void *data)
* The DMA ENGINE API
*/
-static int s3c24xx_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int s3c24xx_dma_terminate_all(struct dma_chan *chan)
{
struct s3c24xx_dma_chan *s3cchan = to_s3c24xx_dma_chan(chan);
struct s3c24xx_dma_engine *s3cdma = s3cchan->host;
unsigned long flags;
- int ret = 0;
spin_lock_irqsave(&s3cchan->vc.lock, flags);
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- ret = s3c24xx_dma_set_runtime_config(s3cchan,
- (struct dma_slave_config *)arg);
- break;
- case DMA_TERMINATE_ALL:
- if (!s3cchan->phy && !s3cchan->at) {
- dev_err(&s3cdma->pdev->dev, "trying to terminate already stopped channel %d\n",
- s3cchan->id);
- ret = -EINVAL;
- break;
- }
-
- s3cchan->state = S3C24XX_DMA_CHAN_IDLE;
+ if (!s3cchan->phy && !s3cchan->at) {
+ dev_err(&s3cdma->pdev->dev, "trying to terminate already stopped channel %d\n",
+ s3cchan->id);
+ return -EINVAL;
+ }
- /* Mark physical channel as free */
- if (s3cchan->phy)
- s3c24xx_dma_phy_free(s3cchan);
+ s3cchan->state = S3C24XX_DMA_CHAN_IDLE;
- /* Dequeue current job */
- if (s3cchan->at) {
- s3c24xx_dma_desc_free(&s3cchan->at->vd);
- s3cchan->at = NULL;
- }
+ /* Mark physical channel as free */
+ if (s3cchan->phy)
+ s3c24xx_dma_phy_free(s3cchan);
- /* Dequeue jobs not yet fired as well */
- s3c24xx_dma_free_txd_list(s3cdma, s3cchan);
- break;
- default:
- /* Unknown command */
- ret = -ENXIO;
- break;
+ /* Dequeue current job */
+ if (s3cchan->at) {
+ s3c24xx_dma_desc_free(&s3cchan->at->vd);
+ s3cchan->at = NULL;
}
+ /* Dequeue jobs not yet fired as well */
+ s3c24xx_dma_free_txd_list(s3cdma, s3cchan);
+
spin_unlock_irqrestore(&s3cchan->vc.lock, flags);
- return ret;
+ return 0;
}
static int s3c24xx_dma_alloc_chan_resources(struct dma_chan *chan)
@@ -1300,7 +1295,8 @@ static int s3c24xx_dma_probe(struct platform_device *pdev)
s3cdma->memcpy.device_prep_dma_memcpy = s3c24xx_dma_prep_memcpy;
s3cdma->memcpy.device_tx_status = s3c24xx_dma_tx_status;
s3cdma->memcpy.device_issue_pending = s3c24xx_dma_issue_pending;
- s3cdma->memcpy.device_control = s3c24xx_dma_control;
+ s3cdma->memcpy.device_config = s3c24xx_dma_set_runtime_config;
+ s3cdma->memcpy.device_terminate_all = s3c24xx_dma_terminate_all;
/* Initialize slave engine for SoC internal dedicated peripherals */
dma_cap_set(DMA_SLAVE, s3cdma->slave.cap_mask);
@@ -1315,7 +1311,8 @@ static int s3c24xx_dma_probe(struct platform_device *pdev)
s3cdma->slave.device_issue_pending = s3c24xx_dma_issue_pending;
s3cdma->slave.device_prep_slave_sg = s3c24xx_dma_prep_slave_sg;
s3cdma->slave.device_prep_dma_cyclic = s3c24xx_dma_prep_dma_cyclic;
- s3cdma->slave.device_control = s3c24xx_dma_control;
+ s3cdma->slave.device_config = s3c24xx_dma_set_runtime_config;
+ s3cdma->slave.device_terminate_all = s3c24xx_dma_terminate_all;
/* Register as many memcpy channels as there are physical channels */
ret = s3c24xx_dma_init_virtual_channels(s3cdma, &s3cdma->memcpy,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 38/61] dmaengine: sa11x0: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the SA-11x0 DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sa11x0-dma.c | 158 +++++++++++++++++++++++++----------------------
1 file changed, 84 insertions(+), 74 deletions(-)
diff --git a/drivers/dma/sa11x0-dma.c b/drivers/dma/sa11x0-dma.c
index 2329d295efb5..a2ee869865ec 100644
--- a/drivers/dma/sa11x0-dma.c
+++ b/drivers/dma/sa11x0-dma.c
@@ -669,8 +669,10 @@ static struct dma_async_tx_descriptor *sa11x0_dma_prep_dma_cyclic(
return vchan_tx_prep(&c->vc, &txd->vd, DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
}
-static int sa11x0_dma_slave_config(struct sa11x0_dma_chan *c, struct dma_slave_config *cfg)
+static int sa11x0_dma_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
{
+ struct sa11x0_dma_chan *c = to_sa11x0_dma_chan(chan);
u32 ddar = c->ddar & ((0xf << 4) | DDAR_RW);
dma_addr_t addr;
enum dma_slave_buswidth width;
@@ -704,8 +706,7 @@ static int sa11x0_dma_slave_config(struct sa11x0_dma_chan *c, struct dma_slave_c
return 0;
}
-static int sa11x0_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int sa11x0_dma_pause(struct dma_chan *chan)
{
struct sa11x0_dma_chan *c = to_sa11x0_dma_chan(chan);
struct sa11x0_dma_dev *d = to_sa11x0_dma(chan->device);
@@ -714,89 +715,95 @@ static int sa11x0_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
unsigned long flags;
int ret;
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- return sa11x0_dma_slave_config(c, (struct dma_slave_config *)arg);
-
- case DMA_TERMINATE_ALL:
- dev_dbg(d->slave.dev, "vchan %p: terminate all\n", &c->vc);
- /* Clear the tx descriptor lists */
- spin_lock_irqsave(&c->vc.lock, flags);
- vchan_get_all_descriptors(&c->vc, &head);
+ dev_dbg(d->slave.dev, "vchan %p: pause\n", &c->vc);
+ spin_lock_irqsave(&c->vc.lock, flags);
+ if (c->status == DMA_IN_PROGRESS) {
+ c->status = DMA_PAUSED;
p = c->phy;
if (p) {
- dev_dbg(d->slave.dev, "pchan %u: terminating\n", p->num);
- /* vchan is assigned to a pchan - stop the channel */
- writel(DCSR_RUN | DCSR_IE |
- DCSR_STRTA | DCSR_DONEA |
- DCSR_STRTB | DCSR_DONEB,
- p->base + DMA_DCSR_C);
-
- if (p->txd_load) {
- if (p->txd_load != p->txd_done)
- list_add_tail(&p->txd_load->vd.node, &head);
- p->txd_load = NULL;
- }
- if (p->txd_done) {
- list_add_tail(&p->txd_done->vd.node, &head);
- p->txd_done = NULL;
- }
- c->phy = NULL;
+ writel(DCSR_RUN | DCSR_IE, p->base + DMA_DCSR_C);
+ } else {
spin_lock(&d->lock);
- p->vchan = NULL;
+ list_del_init(&c->node);
spin_unlock(&d->lock);
- tasklet_schedule(&d->task);
}
- spin_unlock_irqrestore(&c->vc.lock, flags);
- vchan_dma_desc_free_list(&c->vc, &head);
- ret = 0;
- break;
+ }
+ spin_unlock_irqrestore(&c->vc.lock, flags);
- case DMA_PAUSE:
- dev_dbg(d->slave.dev, "vchan %p: pause\n", &c->vc);
- spin_lock_irqsave(&c->vc.lock, flags);
- if (c->status == DMA_IN_PROGRESS) {
- c->status = DMA_PAUSED;
+ return 0;
+}
- p = c->phy;
- if (p) {
- writel(DCSR_RUN | DCSR_IE, p->base + DMA_DCSR_C);
- } else {
- spin_lock(&d->lock);
- list_del_init(&c->node);
- spin_unlock(&d->lock);
- }
- }
- spin_unlock_irqrestore(&c->vc.lock, flags);
- ret = 0;
- break;
+static int sa11x0_dma_resume(struct dma_chan *chan)
+{
+ struct sa11x0_dma_chan *c = to_sa11x0_dma_chan(chan);
+ struct sa11x0_dma_dev *d = to_sa11x0_dma(chan->device);
+ struct sa11x0_dma_phy *p;
+ LIST_HEAD(head);
+ unsigned long flags;
+ int ret;
- case DMA_RESUME:
- dev_dbg(d->slave.dev, "vchan %p: resume\n", &c->vc);
- spin_lock_irqsave(&c->vc.lock, flags);
- if (c->status == DMA_PAUSED) {
- c->status = DMA_IN_PROGRESS;
-
- p = c->phy;
- if (p) {
- writel(DCSR_RUN | DCSR_IE, p->base + DMA_DCSR_S);
- } else if (!list_empty(&c->vc.desc_issued)) {
- spin_lock(&d->lock);
- list_add_tail(&c->node, &d->chan_pending);
- spin_unlock(&d->lock);
- }
+ dev_dbg(d->slave.dev, "vchan %p: resume\n", &c->vc);
+ spin_lock_irqsave(&c->vc.lock, flags);
+ if (c->status == DMA_PAUSED) {
+ c->status = DMA_IN_PROGRESS;
+
+ p = c->phy;
+ if (p) {
+ writel(DCSR_RUN | DCSR_IE, p->base + DMA_DCSR_S);
+ } else if (!list_empty(&c->vc.desc_issued)) {
+ spin_lock(&d->lock);
+ list_add_tail(&c->node, &d->chan_pending);
+ spin_unlock(&d->lock);
}
- spin_unlock_irqrestore(&c->vc.lock, flags);
- ret = 0;
- break;
+ }
+ spin_unlock_irqrestore(&c->vc.lock, flags);
- default:
- ret = -ENXIO;
- break;
+ return 0;
+}
+
+static int sa11x0_dma_terminate_all(struct dma_chan *chan)
+{
+ struct sa11x0_dma_chan *c = to_sa11x0_dma_chan(chan);
+ struct sa11x0_dma_dev *d = to_sa11x0_dma(chan->device);
+ struct sa11x0_dma_phy *p;
+ LIST_HEAD(head);
+ unsigned long flags;
+ int ret;
+
+ dev_dbg(d->slave.dev, "vchan %p: terminate all\n", &c->vc);
+ /* Clear the tx descriptor lists */
+ spin_lock_irqsave(&c->vc.lock, flags);
+ vchan_get_all_descriptors(&c->vc, &head);
+
+ p = c->phy;
+ if (p) {
+ dev_dbg(d->slave.dev, "pchan %u: terminating\n", p->num);
+ /* vchan is assigned to a pchan - stop the channel */
+ writel(DCSR_RUN | DCSR_IE |
+ DCSR_STRTA | DCSR_DONEA |
+ DCSR_STRTB | DCSR_DONEB,
+ p->base + DMA_DCSR_C);
+
+ if (p->txd_load) {
+ if (p->txd_load != p->txd_done)
+ list_add_tail(&p->txd_load->vd.node, &head);
+ p->txd_load = NULL;
+ }
+ if (p->txd_done) {
+ list_add_tail(&p->txd_done->vd.node, &head);
+ p->txd_done = NULL;
+ }
+ c->phy = NULL;
+ spin_lock(&d->lock);
+ p->vchan = NULL;
+ spin_unlock(&d->lock);
+ tasklet_schedule(&d->task);
}
+ spin_unlock_irqrestore(&c->vc.lock, flags);
+ vchan_dma_desc_free_list(&c->vc, &head);
- return ret;
+ return 0;
}
struct sa11x0_dma_channel_desc {
@@ -833,7 +840,10 @@ static int sa11x0_dma_init_dmadev(struct dma_device *dmadev,
dmadev->dev = dev;
dmadev->device_alloc_chan_resources = sa11x0_dma_alloc_chan_resources;
dmadev->device_free_chan_resources = sa11x0_dma_free_chan_resources;
- dmadev->device_control = sa11x0_dma_control;
+ dmadev->device_config = sa11x0_dma_slave_config;
+ dmadev->device_pause = sa11x0_dma_pause;
+ dmadev->device_resume = sa11x0_dma_resume;
+ dmadev->device_terminate_all = sa11x0_dma_terminate_all;
dmadev->device_tx_status = sa11x0_dma_tx_status;
dmadev->device_issue_pending = sa11x0_dma_issue_pending;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 38/61] dmaengine: sa11x0: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the SA-11x0 DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sa11x0-dma.c | 158 +++++++++++++++++++++++++----------------------
1 file changed, 84 insertions(+), 74 deletions(-)
diff --git a/drivers/dma/sa11x0-dma.c b/drivers/dma/sa11x0-dma.c
index 2329d295efb5..a2ee869865ec 100644
--- a/drivers/dma/sa11x0-dma.c
+++ b/drivers/dma/sa11x0-dma.c
@@ -669,8 +669,10 @@ static struct dma_async_tx_descriptor *sa11x0_dma_prep_dma_cyclic(
return vchan_tx_prep(&c->vc, &txd->vd, DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
}
-static int sa11x0_dma_slave_config(struct sa11x0_dma_chan *c, struct dma_slave_config *cfg)
+static int sa11x0_dma_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *cfg)
{
+ struct sa11x0_dma_chan *c = to_sa11x0_dma_chan(chan);
u32 ddar = c->ddar & ((0xf << 4) | DDAR_RW);
dma_addr_t addr;
enum dma_slave_buswidth width;
@@ -704,8 +706,7 @@ static int sa11x0_dma_slave_config(struct sa11x0_dma_chan *c, struct dma_slave_c
return 0;
}
-static int sa11x0_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int sa11x0_dma_pause(struct dma_chan *chan)
{
struct sa11x0_dma_chan *c = to_sa11x0_dma_chan(chan);
struct sa11x0_dma_dev *d = to_sa11x0_dma(chan->device);
@@ -714,89 +715,95 @@ static int sa11x0_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
unsigned long flags;
int ret;
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- return sa11x0_dma_slave_config(c, (struct dma_slave_config *)arg);
-
- case DMA_TERMINATE_ALL:
- dev_dbg(d->slave.dev, "vchan %p: terminate all\n", &c->vc);
- /* Clear the tx descriptor lists */
- spin_lock_irqsave(&c->vc.lock, flags);
- vchan_get_all_descriptors(&c->vc, &head);
+ dev_dbg(d->slave.dev, "vchan %p: pause\n", &c->vc);
+ spin_lock_irqsave(&c->vc.lock, flags);
+ if (c->status == DMA_IN_PROGRESS) {
+ c->status = DMA_PAUSED;
p = c->phy;
if (p) {
- dev_dbg(d->slave.dev, "pchan %u: terminating\n", p->num);
- /* vchan is assigned to a pchan - stop the channel */
- writel(DCSR_RUN | DCSR_IE |
- DCSR_STRTA | DCSR_DONEA |
- DCSR_STRTB | DCSR_DONEB,
- p->base + DMA_DCSR_C);
-
- if (p->txd_load) {
- if (p->txd_load != p->txd_done)
- list_add_tail(&p->txd_load->vd.node, &head);
- p->txd_load = NULL;
- }
- if (p->txd_done) {
- list_add_tail(&p->txd_done->vd.node, &head);
- p->txd_done = NULL;
- }
- c->phy = NULL;
+ writel(DCSR_RUN | DCSR_IE, p->base + DMA_DCSR_C);
+ } else {
spin_lock(&d->lock);
- p->vchan = NULL;
+ list_del_init(&c->node);
spin_unlock(&d->lock);
- tasklet_schedule(&d->task);
}
- spin_unlock_irqrestore(&c->vc.lock, flags);
- vchan_dma_desc_free_list(&c->vc, &head);
- ret = 0;
- break;
+ }
+ spin_unlock_irqrestore(&c->vc.lock, flags);
- case DMA_PAUSE:
- dev_dbg(d->slave.dev, "vchan %p: pause\n", &c->vc);
- spin_lock_irqsave(&c->vc.lock, flags);
- if (c->status == DMA_IN_PROGRESS) {
- c->status = DMA_PAUSED;
+ return 0;
+}
- p = c->phy;
- if (p) {
- writel(DCSR_RUN | DCSR_IE, p->base + DMA_DCSR_C);
- } else {
- spin_lock(&d->lock);
- list_del_init(&c->node);
- spin_unlock(&d->lock);
- }
- }
- spin_unlock_irqrestore(&c->vc.lock, flags);
- ret = 0;
- break;
+static int sa11x0_dma_resume(struct dma_chan *chan)
+{
+ struct sa11x0_dma_chan *c = to_sa11x0_dma_chan(chan);
+ struct sa11x0_dma_dev *d = to_sa11x0_dma(chan->device);
+ struct sa11x0_dma_phy *p;
+ LIST_HEAD(head);
+ unsigned long flags;
+ int ret;
- case DMA_RESUME:
- dev_dbg(d->slave.dev, "vchan %p: resume\n", &c->vc);
- spin_lock_irqsave(&c->vc.lock, flags);
- if (c->status == DMA_PAUSED) {
- c->status = DMA_IN_PROGRESS;
-
- p = c->phy;
- if (p) {
- writel(DCSR_RUN | DCSR_IE, p->base + DMA_DCSR_S);
- } else if (!list_empty(&c->vc.desc_issued)) {
- spin_lock(&d->lock);
- list_add_tail(&c->node, &d->chan_pending);
- spin_unlock(&d->lock);
- }
+ dev_dbg(d->slave.dev, "vchan %p: resume\n", &c->vc);
+ spin_lock_irqsave(&c->vc.lock, flags);
+ if (c->status == DMA_PAUSED) {
+ c->status = DMA_IN_PROGRESS;
+
+ p = c->phy;
+ if (p) {
+ writel(DCSR_RUN | DCSR_IE, p->base + DMA_DCSR_S);
+ } else if (!list_empty(&c->vc.desc_issued)) {
+ spin_lock(&d->lock);
+ list_add_tail(&c->node, &d->chan_pending);
+ spin_unlock(&d->lock);
}
- spin_unlock_irqrestore(&c->vc.lock, flags);
- ret = 0;
- break;
+ }
+ spin_unlock_irqrestore(&c->vc.lock, flags);
- default:
- ret = -ENXIO;
- break;
+ return 0;
+}
+
+static int sa11x0_dma_terminate_all(struct dma_chan *chan)
+{
+ struct sa11x0_dma_chan *c = to_sa11x0_dma_chan(chan);
+ struct sa11x0_dma_dev *d = to_sa11x0_dma(chan->device);
+ struct sa11x0_dma_phy *p;
+ LIST_HEAD(head);
+ unsigned long flags;
+ int ret;
+
+ dev_dbg(d->slave.dev, "vchan %p: terminate all\n", &c->vc);
+ /* Clear the tx descriptor lists */
+ spin_lock_irqsave(&c->vc.lock, flags);
+ vchan_get_all_descriptors(&c->vc, &head);
+
+ p = c->phy;
+ if (p) {
+ dev_dbg(d->slave.dev, "pchan %u: terminating\n", p->num);
+ /* vchan is assigned to a pchan - stop the channel */
+ writel(DCSR_RUN | DCSR_IE |
+ DCSR_STRTA | DCSR_DONEA |
+ DCSR_STRTB | DCSR_DONEB,
+ p->base + DMA_DCSR_C);
+
+ if (p->txd_load) {
+ if (p->txd_load != p->txd_done)
+ list_add_tail(&p->txd_load->vd.node, &head);
+ p->txd_load = NULL;
+ }
+ if (p->txd_done) {
+ list_add_tail(&p->txd_done->vd.node, &head);
+ p->txd_done = NULL;
+ }
+ c->phy = NULL;
+ spin_lock(&d->lock);
+ p->vchan = NULL;
+ spin_unlock(&d->lock);
+ tasklet_schedule(&d->task);
}
+ spin_unlock_irqrestore(&c->vc.lock, flags);
+ vchan_dma_desc_free_list(&c->vc, &head);
- return ret;
+ return 0;
}
struct sa11x0_dma_channel_desc {
@@ -833,7 +840,10 @@ static int sa11x0_dma_init_dmadev(struct dma_device *dmadev,
dmadev->dev = dev;
dmadev->device_alloc_chan_resources = sa11x0_dma_alloc_chan_resources;
dmadev->device_free_chan_resources = sa11x0_dma_free_chan_resources;
- dmadev->device_control = sa11x0_dma_control;
+ dmadev->device_config = sa11x0_dma_slave_config;
+ dmadev->device_pause = sa11x0_dma_pause;
+ dmadev->device_resume = sa11x0_dma_resume;
+ dmadev->device_terminate_all = sa11x0_dma_terminate_all;
dmadev->device_tx_status = sa11x0_dma_tx_status;
dmadev->device_issue_pending = sa11x0_dma_issue_pending;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 39/61] dmaengine: sh: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Super-H DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/sh/shdma-base.c | 72 +++++++++++++++++++++------------------------
1 file changed, 33 insertions(+), 39 deletions(-)
diff --git a/drivers/dma/sh/shdma-base.c b/drivers/dma/sh/shdma-base.c
index 3a2adb131d46..8ee383d339a5 100644
--- a/drivers/dma/sh/shdma-base.c
+++ b/drivers/dma/sh/shdma-base.c
@@ -729,57 +729,50 @@ static struct dma_async_tx_descriptor *shdma_prep_dma_cyclic(
return desc;
}
-static int shdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int shdma_terminate_all(struct dma_chan *chan)
{
struct shdma_chan *schan = to_shdma_chan(chan);
struct shdma_dev *sdev = to_shdma_dev(chan->device);
const struct shdma_ops *ops = sdev->ops;
- struct dma_slave_config *config;
unsigned long flags;
- int ret;
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- spin_lock_irqsave(&schan->chan_lock, flags);
- ops->halt_channel(schan);
+ spin_lock_irqsave(&schan->chan_lock, flags);
+ ops->halt_channel(schan);
- if (ops->get_partial && !list_empty(&schan->ld_queue)) {
- /* Record partial transfer */
- struct shdma_desc *desc = list_first_entry(&schan->ld_queue,
- struct shdma_desc, node);
- desc->partial = ops->get_partial(schan, desc);
- }
+ if (ops->get_partial && !list_empty(&schan->ld_queue)) {
+ /* Record partial transfer */
+ struct shdma_desc *desc = list_first_entry(&schan->ld_queue,
+ struct shdma_desc, node);
+ desc->partial = ops->get_partial(schan, desc);
+ }
- spin_unlock_irqrestore(&schan->chan_lock, flags);
+ spin_unlock_irqrestore(&schan->chan_lock, flags);
- shdma_chan_ld_cleanup(schan, true);
- break;
- case DMA_SLAVE_CONFIG:
- /*
- * So far only .slave_id is used, but the slave drivers are
- * encouraged to also set a transfer direction and an address.
- */
- if (!arg)
- return -EINVAL;
- /*
- * We could lock this, but you shouldn't be configuring the
- * channel, while using it...
- */
- config = (struct dma_slave_config *)arg;
- ret = shdma_setup_slave(schan, config->slave_id,
- config->direction == DMA_DEV_TO_MEM ?
- config->src_addr : config->dst_addr);
- if (ret < 0)
- return ret;
- break;
- default:
- return -ENXIO;
- }
+ shdma_chan_ld_cleanup(schan, true);
return 0;
}
+static int shdma_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
+{
+ struct shdma_chan *schan = to_shdma_chan(chan);
+
+ /*
+ * So far only .slave_id is used, but the slave drivers are
+ * encouraged to also set a transfer direction and an address.
+ */
+ if (!config)
+ return -EINVAL;
+ /*
+ * We could lock this, but you shouldn't be configuring the
+ * channel, while using it...
+ */
+ return shdma_setup_slave(schan, config->slave_id,
+ config->direction == DMA_DEV_TO_MEM ?
+ config->src_addr : config->dst_addr);
+}
+
static void shdma_issue_pending(struct dma_chan *chan)
{
struct shdma_chan *schan = to_shdma_chan(chan);
@@ -1002,7 +995,8 @@ int shdma_init(struct device *dev, struct shdma_dev *sdev,
/* Compulsory for DMA_SLAVE fields */
dma_dev->device_prep_slave_sg = shdma_prep_slave_sg;
dma_dev->device_prep_dma_cyclic = shdma_prep_dma_cyclic;
- dma_dev->device_control = shdma_control;
+ dma_dev->device_config = shdma_config;
+ dma_dev->device_terminate_all = shdma_terminate_all;
dma_dev->dev = dev;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 39/61] dmaengine: sh: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Super-H DMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/sh/shdma-base.c | 72 +++++++++++++++++++++------------------------
1 file changed, 33 insertions(+), 39 deletions(-)
diff --git a/drivers/dma/sh/shdma-base.c b/drivers/dma/sh/shdma-base.c
index 3a2adb131d46..8ee383d339a5 100644
--- a/drivers/dma/sh/shdma-base.c
+++ b/drivers/dma/sh/shdma-base.c
@@ -729,57 +729,50 @@ static struct dma_async_tx_descriptor *shdma_prep_dma_cyclic(
return desc;
}
-static int shdma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int shdma_terminate_all(struct dma_chan *chan)
{
struct shdma_chan *schan = to_shdma_chan(chan);
struct shdma_dev *sdev = to_shdma_dev(chan->device);
const struct shdma_ops *ops = sdev->ops;
- struct dma_slave_config *config;
unsigned long flags;
- int ret;
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- spin_lock_irqsave(&schan->chan_lock, flags);
- ops->halt_channel(schan);
+ spin_lock_irqsave(&schan->chan_lock, flags);
+ ops->halt_channel(schan);
- if (ops->get_partial && !list_empty(&schan->ld_queue)) {
- /* Record partial transfer */
- struct shdma_desc *desc = list_first_entry(&schan->ld_queue,
- struct shdma_desc, node);
- desc->partial = ops->get_partial(schan, desc);
- }
+ if (ops->get_partial && !list_empty(&schan->ld_queue)) {
+ /* Record partial transfer */
+ struct shdma_desc *desc = list_first_entry(&schan->ld_queue,
+ struct shdma_desc, node);
+ desc->partial = ops->get_partial(schan, desc);
+ }
- spin_unlock_irqrestore(&schan->chan_lock, flags);
+ spin_unlock_irqrestore(&schan->chan_lock, flags);
- shdma_chan_ld_cleanup(schan, true);
- break;
- case DMA_SLAVE_CONFIG:
- /*
- * So far only .slave_id is used, but the slave drivers are
- * encouraged to also set a transfer direction and an address.
- */
- if (!arg)
- return -EINVAL;
- /*
- * We could lock this, but you shouldn't be configuring the
- * channel, while using it...
- */
- config = (struct dma_slave_config *)arg;
- ret = shdma_setup_slave(schan, config->slave_id,
- config->direction == DMA_DEV_TO_MEM ?
- config->src_addr : config->dst_addr);
- if (ret < 0)
- return ret;
- break;
- default:
- return -ENXIO;
- }
+ shdma_chan_ld_cleanup(schan, true);
return 0;
}
+static int shdma_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
+{
+ struct shdma_chan *schan = to_shdma_chan(chan);
+
+ /*
+ * So far only .slave_id is used, but the slave drivers are
+ * encouraged to also set a transfer direction and an address.
+ */
+ if (!config)
+ return -EINVAL;
+ /*
+ * We could lock this, but you shouldn't be configuring the
+ * channel, while using it...
+ */
+ return shdma_setup_slave(schan, config->slave_id,
+ config->direction == DMA_DEV_TO_MEM ?
+ config->src_addr : config->dst_addr);
+}
+
static void shdma_issue_pending(struct dma_chan *chan)
{
struct shdma_chan *schan = to_shdma_chan(chan);
@@ -1002,7 +995,8 @@ int shdma_init(struct device *dev, struct shdma_dev *sdev,
/* Compulsory for DMA_SLAVE fields */
dma_dev->device_prep_slave_sg = shdma_prep_slave_sg;
dma_dev->device_prep_dma_cyclic = shdma_prep_dma_cyclic;
- dma_dev->device_control = shdma_control;
+ dma_dev->device_config = shdma_config;
+ dma_dev->device_terminate_all = shdma_terminate_all;
dma_dev->dev = dev;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 40/61] dmaengine: sirf: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the SiRF Prima 2 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sirf-dma.c | 43 +++++++++++++------------------------------
1 file changed, 13 insertions(+), 30 deletions(-)
diff --git a/drivers/dma/sirf-dma.c b/drivers/dma/sirf-dma.c
index e0aa759139f3..eeb2af440b9f 100644
--- a/drivers/dma/sirf-dma.c
+++ b/drivers/dma/sirf-dma.c
@@ -281,9 +281,10 @@ static dma_cookie_t sirfsoc_dma_tx_submit(struct dma_async_tx_descriptor *txd)
return cookie;
}
-static int sirfsoc_dma_slave_config(struct sirfsoc_dma_chan *schan,
- struct dma_slave_config *config)
+static int sirfsoc_dma_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
{
+ struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
unsigned long flags;
if ((config->src_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES) ||
@@ -297,8 +298,9 @@ static int sirfsoc_dma_slave_config(struct sirfsoc_dma_chan *schan,
return 0;
}
-static int sirfsoc_dma_terminate_all(struct sirfsoc_dma_chan *schan)
+static int sirfsoc_dma_terminate_all(struct dma_chan *chan)
{
+ struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
struct sirfsoc_dma *sdma = dma_chan_to_sirfsoc_dma(&schan->chan);
int cid = schan->chan.chan_id;
unsigned long flags;
@@ -327,8 +329,9 @@ static int sirfsoc_dma_terminate_all(struct sirfsoc_dma_chan *schan)
return 0;
}
-static int sirfsoc_dma_pause_chan(struct sirfsoc_dma_chan *schan)
+static int sirfsoc_dma_pause_chan(struct dma_chan *chan)
{
+ struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
struct sirfsoc_dma *sdma = dma_chan_to_sirfsoc_dma(&schan->chan);
int cid = schan->chan.chan_id;
unsigned long flags;
@@ -348,8 +351,9 @@ static int sirfsoc_dma_pause_chan(struct sirfsoc_dma_chan *schan)
return 0;
}
-static int sirfsoc_dma_resume_chan(struct sirfsoc_dma_chan *schan)
+static int sirfsoc_dma_resume_chan(struct dma_chan *chan)
{
+ struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
struct sirfsoc_dma *sdma = dma_chan_to_sirfsoc_dma(&schan->chan);
int cid = schan->chan.chan_id;
unsigned long flags;
@@ -369,30 +373,6 @@ static int sirfsoc_dma_resume_chan(struct sirfsoc_dma_chan *schan)
return 0;
}
-static int sirfsoc_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct dma_slave_config *config;
- struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
-
- switch (cmd) {
- case DMA_PAUSE:
- return sirfsoc_dma_pause_chan(schan);
- case DMA_RESUME:
- return sirfsoc_dma_resume_chan(schan);
- case DMA_TERMINATE_ALL:
- return sirfsoc_dma_terminate_all(schan);
- case DMA_SLAVE_CONFIG:
- config = (struct dma_slave_config *)arg;
- return sirfsoc_dma_slave_config(schan, config);
-
- default:
- break;
- }
-
- return -ENOSYS;
-}
-
/* Alloc channel resources */
static int sirfsoc_dma_alloc_chan_resources(struct dma_chan *chan)
{
@@ -739,7 +719,10 @@ static int sirfsoc_dma_probe(struct platform_device *op)
dma->device_alloc_chan_resources = sirfsoc_dma_alloc_chan_resources;
dma->device_free_chan_resources = sirfsoc_dma_free_chan_resources;
dma->device_issue_pending = sirfsoc_dma_issue_pending;
- dma->device_control = sirfsoc_dma_control;
+ dma->device_config = sirfsoc_dma_slave_config;
+ dma->device_pause = sirfsoc_dma_pause_chan;
+ dma->device_resume = sirfsoc_dma_resume_chan;
+ dma->device_terminate_all = sirfsoc_dma_terminate_all;
dma->device_tx_status = sirfsoc_dma_tx_status;
dma->device_prep_interleaved_dma = sirfsoc_dma_prep_interleaved;
dma->device_prep_dma_cyclic = sirfsoc_dma_prep_cyclic;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 40/61] dmaengine: sirf: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the SiRF Prima 2 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sirf-dma.c | 43 +++++++++++++------------------------------
1 file changed, 13 insertions(+), 30 deletions(-)
diff --git a/drivers/dma/sirf-dma.c b/drivers/dma/sirf-dma.c
index e0aa759139f3..eeb2af440b9f 100644
--- a/drivers/dma/sirf-dma.c
+++ b/drivers/dma/sirf-dma.c
@@ -281,9 +281,10 @@ static dma_cookie_t sirfsoc_dma_tx_submit(struct dma_async_tx_descriptor *txd)
return cookie;
}
-static int sirfsoc_dma_slave_config(struct sirfsoc_dma_chan *schan,
- struct dma_slave_config *config)
+static int sirfsoc_dma_slave_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
{
+ struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
unsigned long flags;
if ((config->src_addr_width != DMA_SLAVE_BUSWIDTH_4_BYTES) ||
@@ -297,8 +298,9 @@ static int sirfsoc_dma_slave_config(struct sirfsoc_dma_chan *schan,
return 0;
}
-static int sirfsoc_dma_terminate_all(struct sirfsoc_dma_chan *schan)
+static int sirfsoc_dma_terminate_all(struct dma_chan *chan)
{
+ struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
struct sirfsoc_dma *sdma = dma_chan_to_sirfsoc_dma(&schan->chan);
int cid = schan->chan.chan_id;
unsigned long flags;
@@ -327,8 +329,9 @@ static int sirfsoc_dma_terminate_all(struct sirfsoc_dma_chan *schan)
return 0;
}
-static int sirfsoc_dma_pause_chan(struct sirfsoc_dma_chan *schan)
+static int sirfsoc_dma_pause_chan(struct dma_chan *chan)
{
+ struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
struct sirfsoc_dma *sdma = dma_chan_to_sirfsoc_dma(&schan->chan);
int cid = schan->chan.chan_id;
unsigned long flags;
@@ -348,8 +351,9 @@ static int sirfsoc_dma_pause_chan(struct sirfsoc_dma_chan *schan)
return 0;
}
-static int sirfsoc_dma_resume_chan(struct sirfsoc_dma_chan *schan)
+static int sirfsoc_dma_resume_chan(struct dma_chan *chan)
{
+ struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
struct sirfsoc_dma *sdma = dma_chan_to_sirfsoc_dma(&schan->chan);
int cid = schan->chan.chan_id;
unsigned long flags;
@@ -369,30 +373,6 @@ static int sirfsoc_dma_resume_chan(struct sirfsoc_dma_chan *schan)
return 0;
}
-static int sirfsoc_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct dma_slave_config *config;
- struct sirfsoc_dma_chan *schan = dma_chan_to_sirfsoc_dma_chan(chan);
-
- switch (cmd) {
- case DMA_PAUSE:
- return sirfsoc_dma_pause_chan(schan);
- case DMA_RESUME:
- return sirfsoc_dma_resume_chan(schan);
- case DMA_TERMINATE_ALL:
- return sirfsoc_dma_terminate_all(schan);
- case DMA_SLAVE_CONFIG:
- config = (struct dma_slave_config *)arg;
- return sirfsoc_dma_slave_config(schan, config);
-
- default:
- break;
- }
-
- return -ENOSYS;
-}
-
/* Alloc channel resources */
static int sirfsoc_dma_alloc_chan_resources(struct dma_chan *chan)
{
@@ -739,7 +719,10 @@ static int sirfsoc_dma_probe(struct platform_device *op)
dma->device_alloc_chan_resources = sirfsoc_dma_alloc_chan_resources;
dma->device_free_chan_resources = sirfsoc_dma_free_chan_resources;
dma->device_issue_pending = sirfsoc_dma_issue_pending;
- dma->device_control = sirfsoc_dma_control;
+ dma->device_config = sirfsoc_dma_slave_config;
+ dma->device_pause = sirfsoc_dma_pause_chan;
+ dma->device_resume = sirfsoc_dma_resume_chan;
+ dma->device_terminate_all = sirfsoc_dma_terminate_all;
dma->device_tx_status = sirfsoc_dma_tx_status;
dma->device_prep_interleaved_dma = sirfsoc_dma_prep_interleaved;
dma->device_prep_dma_cyclic = sirfsoc_dma_prep_cyclic;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 41/61] dmaengine: sun6i: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Allwinner A31 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sun6i-dma.c | 150 +++++++++++++++++++++++++-----------------------
1 file changed, 78 insertions(+), 72 deletions(-)
diff --git a/drivers/dma/sun6i-dma.c b/drivers/dma/sun6i-dma.c
index 77fb124dad67..4ef76bc22da4 100644
--- a/drivers/dma/sun6i-dma.c
+++ b/drivers/dma/sun6i-dma.c
@@ -355,38 +355,6 @@ static void sun6i_dma_free_desc(struct virt_dma_desc *vd)
kfree(txd);
}
-static int sun6i_dma_terminate_all(struct sun6i_vchan *vchan)
-{
- struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(vchan->vc.chan.device);
- struct sun6i_pchan *pchan = vchan->phy;
- unsigned long flags;
- LIST_HEAD(head);
-
- spin_lock(&sdev->lock);
- list_del_init(&vchan->node);
- spin_unlock(&sdev->lock);
-
- spin_lock_irqsave(&vchan->vc.lock, flags);
-
- vchan_get_all_descriptors(&vchan->vc, &head);
-
- if (pchan) {
- writel(DMA_CHAN_ENABLE_STOP, pchan->base + DMA_CHAN_ENABLE);
- writel(DMA_CHAN_PAUSE_RESUME, pchan->base + DMA_CHAN_PAUSE);
-
- vchan->phy = NULL;
- pchan->vchan = NULL;
- pchan->desc = NULL;
- pchan->done = NULL;
- }
-
- spin_unlock_irqrestore(&vchan->vc.lock, flags);
-
- vchan_dma_desc_free_list(&vchan->vc, &head);
-
- return 0;
-}
-
static int sun6i_dma_start_desc(struct sun6i_vchan *vchan)
{
struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(vchan->vc.chan.device);
@@ -675,57 +643,92 @@ err_lli_free:
return NULL;
}
-static int sun6i_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int sun6i_dma_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
+{
+ struct sun6i_vchan *vchan = to_sun6i_vchan(chan);
+
+ memcpy(&vchan->cfg, config, sizeof(*config));
+
+ return 0;
+}
+
+static int sun6i_dma_pause(struct dma_chan *chan)
+{
+ struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(chan->device);
+ struct sun6i_vchan *vchan = to_sun6i_vchan(chan);
+ struct sun6i_pchan *pchan = vchan->phy;
+
+ dev_dbg(chan2dev(chan), "vchan %p: pause\n", &vchan->vc);
+
+ if (pchan) {
+ writel(DMA_CHAN_PAUSE_PAUSE,
+ pchan->base + DMA_CHAN_PAUSE);
+ } else {
+ spin_lock(&sdev->lock);
+ list_del_init(&vchan->node);
+ spin_unlock(&sdev->lock);
+ }
+
+ return 0;
+}
+
+static int sun6i_dma_resume(struct dma_chan *chan)
{
struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(chan->device);
struct sun6i_vchan *vchan = to_sun6i_vchan(chan);
struct sun6i_pchan *pchan = vchan->phy;
unsigned long flags;
- int ret = 0;
- switch (cmd) {
- case DMA_RESUME:
- dev_dbg(chan2dev(chan), "vchan %p: resume\n", &vchan->vc);
+ dev_dbg(chan2dev(chan), "vchan %p: resume\n", &vchan->vc);
- spin_lock_irqsave(&vchan->vc.lock, flags);
+ spin_lock_irqsave(&vchan->vc.lock, flags);
- if (pchan) {
- writel(DMA_CHAN_PAUSE_RESUME,
- pchan->base + DMA_CHAN_PAUSE);
- } else if (!list_empty(&vchan->vc.desc_issued)) {
- spin_lock(&sdev->lock);
- list_add_tail(&vchan->node, &sdev->pending);
- spin_unlock(&sdev->lock);
- }
+ if (pchan) {
+ writel(DMA_CHAN_PAUSE_RESUME,
+ pchan->base + DMA_CHAN_PAUSE);
+ } else if (!list_empty(&vchan->vc.desc_issued)) {
+ spin_lock(&sdev->lock);
+ list_add_tail(&vchan->node, &sdev->pending);
+ spin_unlock(&sdev->lock);
+ }
- spin_unlock_irqrestore(&vchan->vc.lock, flags);
- break;
+ spin_unlock_irqrestore(&vchan->vc.lock, flags);
- case DMA_PAUSE:
- dev_dbg(chan2dev(chan), "vchan %p: pause\n", &vchan->vc);
+ return 0;
+}
- if (pchan) {
- writel(DMA_CHAN_PAUSE_PAUSE,
- pchan->base + DMA_CHAN_PAUSE);
- } else {
- spin_lock(&sdev->lock);
- list_del_init(&vchan->node);
- spin_unlock(&sdev->lock);
- }
- break;
-
- case DMA_TERMINATE_ALL:
- ret = sun6i_dma_terminate_all(vchan);
- break;
- case DMA_SLAVE_CONFIG:
- memcpy(&vchan->cfg, (void *)arg, sizeof(struct dma_slave_config));
- break;
- default:
- ret = -ENXIO;
- break;
+static int sun6i_dma_terminate_all(struct dma_chan *chan)
+{
+ struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(chan->device);
+ struct sun6i_vchan *vchan = to_sun6i_vchan(chan);
+ struct sun6i_pchan *pchan = vchan->phy;
+ unsigned long flags;
+ LIST_HEAD(head);
+
+ spin_lock(&sdev->lock);
+ list_del_init(&vchan->node);
+ spin_unlock(&sdev->lock);
+
+ spin_lock_irqsave(&vchan->vc.lock, flags);
+
+ vchan_get_all_descriptors(&vchan->vc, &head);
+
+ if (pchan) {
+ writel(DMA_CHAN_ENABLE_STOP, pchan->base + DMA_CHAN_ENABLE);
+ writel(DMA_CHAN_PAUSE_RESUME, pchan->base + DMA_CHAN_PAUSE);
+
+ vchan->phy = NULL;
+ pchan->vchan = NULL;
+ pchan->desc = NULL;
+ pchan->done = NULL;
}
- return ret;
+
+ spin_unlock_irqrestore(&vchan->vc.lock, flags);
+
+ vchan_dma_desc_free_list(&vchan->vc, &head);
+
+ return 0;
}
static enum dma_status sun6i_dma_tx_status(struct dma_chan *chan,
@@ -960,7 +963,10 @@ static int sun6i_dma_probe(struct platform_device *pdev)
sdc->slave.device_issue_pending = sun6i_dma_issue_pending;
sdc->slave.device_prep_slave_sg = sun6i_dma_prep_slave_sg;
sdc->slave.device_prep_dma_memcpy = sun6i_dma_prep_dma_memcpy;
- sdc->slave.device_control = sun6i_dma_control;
+ sdc->slave.device_config = sun6i_dma_config;
+ sdc->slave.device_pause = sun6i_dma_pause;
+ sdc->slave.device_resume = sun6i_dma_resume;
+ sdc->slave.device_terminate_all = sun6i_dma_terminate_all;
sdc->slave.copy_align = 4;
sdc->slave.dev = &pdev->dev;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 41/61] dmaengine: sun6i: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Allwinner A31 DMA driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sun6i-dma.c | 150 +++++++++++++++++++++++++-----------------------
1 file changed, 78 insertions(+), 72 deletions(-)
diff --git a/drivers/dma/sun6i-dma.c b/drivers/dma/sun6i-dma.c
index 77fb124dad67..4ef76bc22da4 100644
--- a/drivers/dma/sun6i-dma.c
+++ b/drivers/dma/sun6i-dma.c
@@ -355,38 +355,6 @@ static void sun6i_dma_free_desc(struct virt_dma_desc *vd)
kfree(txd);
}
-static int sun6i_dma_terminate_all(struct sun6i_vchan *vchan)
-{
- struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(vchan->vc.chan.device);
- struct sun6i_pchan *pchan = vchan->phy;
- unsigned long flags;
- LIST_HEAD(head);
-
- spin_lock(&sdev->lock);
- list_del_init(&vchan->node);
- spin_unlock(&sdev->lock);
-
- spin_lock_irqsave(&vchan->vc.lock, flags);
-
- vchan_get_all_descriptors(&vchan->vc, &head);
-
- if (pchan) {
- writel(DMA_CHAN_ENABLE_STOP, pchan->base + DMA_CHAN_ENABLE);
- writel(DMA_CHAN_PAUSE_RESUME, pchan->base + DMA_CHAN_PAUSE);
-
- vchan->phy = NULL;
- pchan->vchan = NULL;
- pchan->desc = NULL;
- pchan->done = NULL;
- }
-
- spin_unlock_irqrestore(&vchan->vc.lock, flags);
-
- vchan_dma_desc_free_list(&vchan->vc, &head);
-
- return 0;
-}
-
static int sun6i_dma_start_desc(struct sun6i_vchan *vchan)
{
struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(vchan->vc.chan.device);
@@ -675,57 +643,92 @@ err_lli_free:
return NULL;
}
-static int sun6i_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int sun6i_dma_config(struct dma_chan *chan,
+ struct dma_slave_config *config)
+{
+ struct sun6i_vchan *vchan = to_sun6i_vchan(chan);
+
+ memcpy(&vchan->cfg, config, sizeof(*config));
+
+ return 0;
+}
+
+static int sun6i_dma_pause(struct dma_chan *chan)
+{
+ struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(chan->device);
+ struct sun6i_vchan *vchan = to_sun6i_vchan(chan);
+ struct sun6i_pchan *pchan = vchan->phy;
+
+ dev_dbg(chan2dev(chan), "vchan %p: pause\n", &vchan->vc);
+
+ if (pchan) {
+ writel(DMA_CHAN_PAUSE_PAUSE,
+ pchan->base + DMA_CHAN_PAUSE);
+ } else {
+ spin_lock(&sdev->lock);
+ list_del_init(&vchan->node);
+ spin_unlock(&sdev->lock);
+ }
+
+ return 0;
+}
+
+static int sun6i_dma_resume(struct dma_chan *chan)
{
struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(chan->device);
struct sun6i_vchan *vchan = to_sun6i_vchan(chan);
struct sun6i_pchan *pchan = vchan->phy;
unsigned long flags;
- int ret = 0;
- switch (cmd) {
- case DMA_RESUME:
- dev_dbg(chan2dev(chan), "vchan %p: resume\n", &vchan->vc);
+ dev_dbg(chan2dev(chan), "vchan %p: resume\n", &vchan->vc);
- spin_lock_irqsave(&vchan->vc.lock, flags);
+ spin_lock_irqsave(&vchan->vc.lock, flags);
- if (pchan) {
- writel(DMA_CHAN_PAUSE_RESUME,
- pchan->base + DMA_CHAN_PAUSE);
- } else if (!list_empty(&vchan->vc.desc_issued)) {
- spin_lock(&sdev->lock);
- list_add_tail(&vchan->node, &sdev->pending);
- spin_unlock(&sdev->lock);
- }
+ if (pchan) {
+ writel(DMA_CHAN_PAUSE_RESUME,
+ pchan->base + DMA_CHAN_PAUSE);
+ } else if (!list_empty(&vchan->vc.desc_issued)) {
+ spin_lock(&sdev->lock);
+ list_add_tail(&vchan->node, &sdev->pending);
+ spin_unlock(&sdev->lock);
+ }
- spin_unlock_irqrestore(&vchan->vc.lock, flags);
- break;
+ spin_unlock_irqrestore(&vchan->vc.lock, flags);
- case DMA_PAUSE:
- dev_dbg(chan2dev(chan), "vchan %p: pause\n", &vchan->vc);
+ return 0;
+}
- if (pchan) {
- writel(DMA_CHAN_PAUSE_PAUSE,
- pchan->base + DMA_CHAN_PAUSE);
- } else {
- spin_lock(&sdev->lock);
- list_del_init(&vchan->node);
- spin_unlock(&sdev->lock);
- }
- break;
-
- case DMA_TERMINATE_ALL:
- ret = sun6i_dma_terminate_all(vchan);
- break;
- case DMA_SLAVE_CONFIG:
- memcpy(&vchan->cfg, (void *)arg, sizeof(struct dma_slave_config));
- break;
- default:
- ret = -ENXIO;
- break;
+static int sun6i_dma_terminate_all(struct dma_chan *chan)
+{
+ struct sun6i_dma_dev *sdev = to_sun6i_dma_dev(chan->device);
+ struct sun6i_vchan *vchan = to_sun6i_vchan(chan);
+ struct sun6i_pchan *pchan = vchan->phy;
+ unsigned long flags;
+ LIST_HEAD(head);
+
+ spin_lock(&sdev->lock);
+ list_del_init(&vchan->node);
+ spin_unlock(&sdev->lock);
+
+ spin_lock_irqsave(&vchan->vc.lock, flags);
+
+ vchan_get_all_descriptors(&vchan->vc, &head);
+
+ if (pchan) {
+ writel(DMA_CHAN_ENABLE_STOP, pchan->base + DMA_CHAN_ENABLE);
+ writel(DMA_CHAN_PAUSE_RESUME, pchan->base + DMA_CHAN_PAUSE);
+
+ vchan->phy = NULL;
+ pchan->vchan = NULL;
+ pchan->desc = NULL;
+ pchan->done = NULL;
}
- return ret;
+
+ spin_unlock_irqrestore(&vchan->vc.lock, flags);
+
+ vchan_dma_desc_free_list(&vchan->vc, &head);
+
+ return 0;
}
static enum dma_status sun6i_dma_tx_status(struct dma_chan *chan,
@@ -960,7 +963,10 @@ static int sun6i_dma_probe(struct platform_device *pdev)
sdc->slave.device_issue_pending = sun6i_dma_issue_pending;
sdc->slave.device_prep_slave_sg = sun6i_dma_prep_slave_sg;
sdc->slave.device_prep_dma_memcpy = sun6i_dma_prep_dma_memcpy;
- sdc->slave.device_control = sun6i_dma_control;
+ sdc->slave.device_config = sun6i_dma_config;
+ sdc->slave.device_pause = sun6i_dma_pause;
+ sdc->slave.device_resume = sun6i_dma_resume;
+ sdc->slave.device_terminate_all = sun6i_dma_terminate_all;
sdc->slave.copy_align = 4;
sdc->slave.dev = &pdev->dev;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 42/61] dmaengine: d40: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the ST-Ericsson DMA 40 driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
---
drivers/dma/ste_dma40.c | 60 +++++++++++++++++++++++--------------------------
1 file changed, 28 insertions(+), 32 deletions(-)
diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c
index 5fe59335e247..b781c988b592 100644
--- a/drivers/dma/ste_dma40.c
+++ b/drivers/dma/ste_dma40.c
@@ -1429,11 +1429,17 @@ static bool d40_tx_is_linked(struct d40_chan *d40c)
return is_link;
}
-static int d40_pause(struct d40_chan *d40c)
+static int d40_pause(struct dma_chan *chan)
{
+ struct d40_chan *d40c = container_of(chan, struct d40_chan, chan);
int res = 0;
unsigned long flags;
+ if (d40c->phy_chan == NULL) {
+ chan_err(d40c, "Channel is not allocated!\n");
+ return -EINVAL;
+ }
+
if (!d40c->busy)
return 0;
@@ -1448,11 +1454,17 @@ static int d40_pause(struct d40_chan *d40c)
return res;
}
-static int d40_resume(struct d40_chan *d40c)
+static int d40_resume(struct dma_chan *chan)
{
+ struct d40_chan *d40c = container_of(chan, struct d40_chan, chan);
int res = 0;
unsigned long flags;
+ if (d40c->phy_chan == NULL) {
+ chan_err(d40c, "Channel is not allocated!\n");
+ return -EINVAL;
+ }
+
if (!d40c->busy)
return 0;
@@ -2610,6 +2622,11 @@ static void d40_terminate_all(struct dma_chan *chan)
struct d40_chan *d40c = container_of(chan, struct d40_chan, chan);
int ret;
+ if (d40c->phy_chan == NULL) {
+ chan_err(d40c, "Channel is not allocated!\n");
+ return -EINVAL;
+ }
+
spin_lock_irqsave(&d40c->lock, flags);
pm_runtime_get_sync(d40c->base->dev);
@@ -2673,6 +2690,11 @@ static int d40_set_runtime_config(struct dma_chan *chan,
u32 src_maxburst, dst_maxburst;
int ret;
+ if (d40c->phy_chan == NULL) {
+ chan_err(d40c, "Channel is not allocated!\n");
+ return -EINVAL;
+ }
+
src_addr_width = config->src_addr_width;
src_maxburst = config->src_maxburst;
dst_addr_width = config->dst_addr_width;
@@ -2781,35 +2803,6 @@ static int d40_set_runtime_config(struct dma_chan *chan,
return 0;
}
-static int d40_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct d40_chan *d40c = container_of(chan, struct d40_chan, chan);
-
- if (d40c->phy_chan == NULL) {
- chan_err(d40c, "Channel is not allocated!\n");
- return -EINVAL;
- }
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- d40_terminate_all(chan);
- return 0;
- case DMA_PAUSE:
- return d40_pause(d40c);
- case DMA_RESUME:
- return d40_resume(d40c);
- case DMA_SLAVE_CONFIG:
- return d40_set_runtime_config(chan,
- (struct dma_slave_config *) arg);
- default:
- break;
- }
-
- /* Other commands are unimplemented */
- return -ENXIO;
-}
-
/* Initialization functions */
static void __init d40_chan_init(struct d40_base *base, struct dma_device *dma,
@@ -2870,7 +2863,10 @@ static void d40_ops_init(struct d40_base *base, struct dma_device *dev)
dev->device_free_chan_resources = d40_free_chan_resources;
dev->device_issue_pending = d40_issue_pending;
dev->device_tx_status = d40_tx_status;
- dev->device_control = d40_control;
+ dev->device_config = d40_set_runtime_config;
+ dev->device_pause = d40_pause;
+ dev->device_resume = d40_resume;
+ dev->device_terminate_all = d40_terminate_all;
dev->dev = base->dev;
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 42/61] dmaengine: d40: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the ST-Ericsson DMA 40 driver to make use
of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
---
drivers/dma/ste_dma40.c | 60 +++++++++++++++++++++++--------------------------
1 file changed, 28 insertions(+), 32 deletions(-)
diff --git a/drivers/dma/ste_dma40.c b/drivers/dma/ste_dma40.c
index 5fe59335e247..b781c988b592 100644
--- a/drivers/dma/ste_dma40.c
+++ b/drivers/dma/ste_dma40.c
@@ -1429,11 +1429,17 @@ static bool d40_tx_is_linked(struct d40_chan *d40c)
return is_link;
}
-static int d40_pause(struct d40_chan *d40c)
+static int d40_pause(struct dma_chan *chan)
{
+ struct d40_chan *d40c = container_of(chan, struct d40_chan, chan);
int res = 0;
unsigned long flags;
+ if (d40c->phy_chan == NULL) {
+ chan_err(d40c, "Channel is not allocated!\n");
+ return -EINVAL;
+ }
+
if (!d40c->busy)
return 0;
@@ -1448,11 +1454,17 @@ static int d40_pause(struct d40_chan *d40c)
return res;
}
-static int d40_resume(struct d40_chan *d40c)
+static int d40_resume(struct dma_chan *chan)
{
+ struct d40_chan *d40c = container_of(chan, struct d40_chan, chan);
int res = 0;
unsigned long flags;
+ if (d40c->phy_chan == NULL) {
+ chan_err(d40c, "Channel is not allocated!\n");
+ return -EINVAL;
+ }
+
if (!d40c->busy)
return 0;
@@ -2610,6 +2622,11 @@ static void d40_terminate_all(struct dma_chan *chan)
struct d40_chan *d40c = container_of(chan, struct d40_chan, chan);
int ret;
+ if (d40c->phy_chan == NULL) {
+ chan_err(d40c, "Channel is not allocated!\n");
+ return -EINVAL;
+ }
+
spin_lock_irqsave(&d40c->lock, flags);
pm_runtime_get_sync(d40c->base->dev);
@@ -2673,6 +2690,11 @@ static int d40_set_runtime_config(struct dma_chan *chan,
u32 src_maxburst, dst_maxburst;
int ret;
+ if (d40c->phy_chan == NULL) {
+ chan_err(d40c, "Channel is not allocated!\n");
+ return -EINVAL;
+ }
+
src_addr_width = config->src_addr_width;
src_maxburst = config->src_maxburst;
dst_addr_width = config->dst_addr_width;
@@ -2781,35 +2803,6 @@ static int d40_set_runtime_config(struct dma_chan *chan,
return 0;
}
-static int d40_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- struct d40_chan *d40c = container_of(chan, struct d40_chan, chan);
-
- if (d40c->phy_chan == NULL) {
- chan_err(d40c, "Channel is not allocated!\n");
- return -EINVAL;
- }
-
- switch (cmd) {
- case DMA_TERMINATE_ALL:
- d40_terminate_all(chan);
- return 0;
- case DMA_PAUSE:
- return d40_pause(d40c);
- case DMA_RESUME:
- return d40_resume(d40c);
- case DMA_SLAVE_CONFIG:
- return d40_set_runtime_config(chan,
- (struct dma_slave_config *) arg);
- default:
- break;
- }
-
- /* Other commands are unimplemented */
- return -ENXIO;
-}
-
/* Initialization functions */
static void __init d40_chan_init(struct d40_base *base, struct dma_device *dma,
@@ -2870,7 +2863,10 @@ static void d40_ops_init(struct d40_base *base, struct dma_device *dev)
dev->device_free_chan_resources = d40_free_chan_resources;
dev->device_issue_pending = d40_issue_pending;
dev->device_tx_status = d40_tx_status;
- dev->device_control = d40_control;
+ dev->device_config = d40_set_runtime_config;
+ dev->device_pause = d40_pause;
+ dev->device_resume = d40_resume;
+ dev->device_terminate_all = d40_terminate_all;
dev->dev = base->dev;
}
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 43/61] dmaengine: tegra20: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the NVidia Tegra20 APB DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/tegra20-apb-dma.c | 22 ++--------------------
1 file changed, 2 insertions(+), 20 deletions(-)
diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c
index a6c0b3f68393..053b44f3aef9 100644
--- a/drivers/dma/tegra20-apb-dma.c
+++ b/drivers/dma/tegra20-apb-dma.c
@@ -827,25 +827,6 @@ static enum dma_status tegra_dma_tx_status(struct dma_chan *dc,
return ret;
}
-static int tegra_dma_device_control(struct dma_chan *dc, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- return tegra_dma_slave_config(dc,
- (struct dma_slave_config *)arg);
-
- case DMA_TERMINATE_ALL:
- tegra_dma_terminate_all(dc);
- return 0;
-
- default:
- break;
- }
-
- return -ENXIO;
-}
-
static inline int get_bus_width(struct tegra_dma_channel *tdc,
enum dma_slave_buswidth slave_bw)
{
@@ -1443,7 +1424,8 @@ static int tegra_dma_probe(struct platform_device *pdev)
tegra_dma_free_chan_resources;
tdma->dma_dev.device_prep_slave_sg = tegra_dma_prep_slave_sg;
tdma->dma_dev.device_prep_dma_cyclic = tegra_dma_prep_dma_cyclic;
- tdma->dma_dev.device_control = tegra_dma_device_control;
+ tdma->dma_dev.device_config = tegra_dma_slave_config;
+ tdma->dma_dev.device_terminate_all = tegra_dma_terminate_all;
tdma->dma_dev.device_tx_status = tegra_dma_tx_status;
tdma->dma_dev.device_issue_pending = tegra_dma_issue_pending;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 43/61] dmaengine: tegra20: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the NVidia Tegra20 APB DMA driver to make
use of the newly introduced callbacks, that will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/tegra20-apb-dma.c | 22 ++--------------------
1 file changed, 2 insertions(+), 20 deletions(-)
diff --git a/drivers/dma/tegra20-apb-dma.c b/drivers/dma/tegra20-apb-dma.c
index a6c0b3f68393..053b44f3aef9 100644
--- a/drivers/dma/tegra20-apb-dma.c
+++ b/drivers/dma/tegra20-apb-dma.c
@@ -827,25 +827,6 @@ static enum dma_status tegra_dma_tx_status(struct dma_chan *dc,
return ret;
}
-static int tegra_dma_device_control(struct dma_chan *dc, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- switch (cmd) {
- case DMA_SLAVE_CONFIG:
- return tegra_dma_slave_config(dc,
- (struct dma_slave_config *)arg);
-
- case DMA_TERMINATE_ALL:
- tegra_dma_terminate_all(dc);
- return 0;
-
- default:
- break;
- }
-
- return -ENXIO;
-}
-
static inline int get_bus_width(struct tegra_dma_channel *tdc,
enum dma_slave_buswidth slave_bw)
{
@@ -1443,7 +1424,8 @@ static int tegra_dma_probe(struct platform_device *pdev)
tegra_dma_free_chan_resources;
tdma->dma_dev.device_prep_slave_sg = tegra_dma_prep_slave_sg;
tdma->dma_dev.device_prep_dma_cyclic = tegra_dma_prep_dma_cyclic;
- tdma->dma_dev.device_control = tegra_dma_device_control;
+ tdma->dma_dev.device_config = tegra_dma_slave_config;
+ tdma->dma_dev.device_terminate_all = tegra_dma_terminate_all;
tdma->dma_dev.device_tx_status = tegra_dma_tx_status;
tdma->dma_dev.device_issue_pending = tegra_dma_issue_pending;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 44/61] dmaengine: xilinx: Split device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Split the device_control callback of the Xilinx VDMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/xilinx/xilinx_vdma.c | 29 ++++++-----------------------
1 file changed, 6 insertions(+), 23 deletions(-)
diff --git a/drivers/dma/xilinx/xilinx_vdma.c b/drivers/dma/xilinx/xilinx_vdma.c
index 4a3a8f3137b3..bdd2a5dd7220 100644
--- a/drivers/dma/xilinx/xilinx_vdma.c
+++ b/drivers/dma/xilinx/xilinx_vdma.c
@@ -1001,13 +1001,17 @@ error:
* xilinx_vdma_terminate_all - Halt the channel and free descriptors
* @chan: Driver specific VDMA Channel pointer
*/
-static void xilinx_vdma_terminate_all(struct xilinx_vdma_chan *chan)
+static int xilinx_vdma_terminate_all(struct dma_chan *dchan)
{
+ struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
+
/* Halt the DMA engine */
xilinx_vdma_halt(chan);
/* Remove and free all of the descriptors in the lists */
xilinx_vdma_free_descriptors(chan);
+
+ return 0;
}
/**
@@ -1075,27 +1079,6 @@ int xilinx_vdma_channel_set_config(struct dma_chan *dchan,
}
EXPORT_SYMBOL(xilinx_vdma_channel_set_config);
-/**
- * xilinx_vdma_device_control - Configure DMA channel of the device
- * @dchan: DMA Channel pointer
- * @cmd: DMA control command
- * @arg: Channel configuration
- *
- * Return: '0' on success and failure value on error
- */
-static int xilinx_vdma_device_control(struct dma_chan *dchan,
- enum dma_ctrl_cmd cmd, unsigned long arg)
-{
- struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
-
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
- xilinx_vdma_terminate_all(chan);
-
- return 0;
-}
-
/* -----------------------------------------------------------------------------
* Probe and remove
*/
@@ -1300,7 +1283,7 @@ static int xilinx_vdma_probe(struct platform_device *pdev)
xilinx_vdma_free_chan_resources;
xdev->common.device_prep_interleaved_dma =
xilinx_vdma_dma_prep_interleaved;
- xdev->common.device_control = xilinx_vdma_device_control;
+ xdev->common.device_terminate_all = xilinx_vdma_terminate_all;
xdev->common.device_tx_status = xilinx_vdma_tx_status;
xdev->common.device_issue_pending = xilinx_vdma_issue_pending;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 44/61] dmaengine: xilinx: Split device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Split the device_control callback of the Xilinx VDMA driver to make use of the
newly introduced callbacks, that will eventually be used to retrieve slave
capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/xilinx/xilinx_vdma.c | 29 ++++++-----------------------
1 file changed, 6 insertions(+), 23 deletions(-)
diff --git a/drivers/dma/xilinx/xilinx_vdma.c b/drivers/dma/xilinx/xilinx_vdma.c
index 4a3a8f3137b3..bdd2a5dd7220 100644
--- a/drivers/dma/xilinx/xilinx_vdma.c
+++ b/drivers/dma/xilinx/xilinx_vdma.c
@@ -1001,13 +1001,17 @@ error:
* xilinx_vdma_terminate_all - Halt the channel and free descriptors
* @chan: Driver specific VDMA Channel pointer
*/
-static void xilinx_vdma_terminate_all(struct xilinx_vdma_chan *chan)
+static int xilinx_vdma_terminate_all(struct dma_chan *dchan)
{
+ struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
+
/* Halt the DMA engine */
xilinx_vdma_halt(chan);
/* Remove and free all of the descriptors in the lists */
xilinx_vdma_free_descriptors(chan);
+
+ return 0;
}
/**
@@ -1075,27 +1079,6 @@ int xilinx_vdma_channel_set_config(struct dma_chan *dchan,
}
EXPORT_SYMBOL(xilinx_vdma_channel_set_config);
-/**
- * xilinx_vdma_device_control - Configure DMA channel of the device
- * @dchan: DMA Channel pointer
- * @cmd: DMA control command
- * @arg: Channel configuration
- *
- * Return: '0' on success and failure value on error
- */
-static int xilinx_vdma_device_control(struct dma_chan *dchan,
- enum dma_ctrl_cmd cmd, unsigned long arg)
-{
- struct xilinx_vdma_chan *chan = to_xilinx_chan(dchan);
-
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
- xilinx_vdma_terminate_all(chan);
-
- return 0;
-}
-
/* -----------------------------------------------------------------------------
* Probe and remove
*/
@@ -1300,7 +1283,7 @@ static int xilinx_vdma_probe(struct platform_device *pdev)
xilinx_vdma_free_chan_resources;
xdev->common.device_prep_interleaved_dma =
xilinx_vdma_dma_prep_interleaved;
- xdev->common.device_control = xilinx_vdma_device_control;
+ xdev->common.device_terminate_all = xilinx_vdma_terminate_all;
xdev->common.device_tx_status = xilinx_vdma_tx_status;
xdev->common.device_issue_pending = xilinx_vdma_issue_pending;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 45/61] dmaengine: mv_xor: Remove device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
The Marvell XOR engine doesn't allow any operations that use to be defined in
device_control, it shouldn't need to be defined. Since it's going to be
deprecated, remove it altogether.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mv_xor.c | 9 ---------
1 file changed, 9 deletions(-)
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index a63837ca1410..10d6dd8b8ee4 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -928,14 +928,6 @@ out:
return err;
}
-/* This driver does not implement any of the optional DMA operations. */
-static int
-mv_xor_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- return -ENOSYS;
-}
-
static int mv_xor_channel_remove(struct mv_xor_chan *mv_chan)
{
struct dma_chan *chan, *_chan;
@@ -1008,7 +1000,6 @@ mv_xor_channel_add(struct mv_xor_device *xordev,
dma_dev->device_free_chan_resources = mv_xor_free_chan_resources;
dma_dev->device_tx_status = mv_xor_status;
dma_dev->device_issue_pending = mv_xor_issue_pending;
- dma_dev->device_control = mv_xor_control;
dma_dev->dev = &pdev->dev;
/* set prep routines based on capability */
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 45/61] dmaengine: mv_xor: Remove device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
The Marvell XOR engine doesn't allow any operations that use to be defined in
device_control, it shouldn't need to be defined. Since it's going to be
deprecated, remove it altogether.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/mv_xor.c | 9 ---------
1 file changed, 9 deletions(-)
diff --git a/drivers/dma/mv_xor.c b/drivers/dma/mv_xor.c
index a63837ca1410..10d6dd8b8ee4 100644
--- a/drivers/dma/mv_xor.c
+++ b/drivers/dma/mv_xor.c
@@ -928,14 +928,6 @@ out:
return err;
}
-/* This driver does not implement any of the optional DMA operations. */
-static int
-mv_xor_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- return -ENOSYS;
-}
-
static int mv_xor_channel_remove(struct mv_xor_chan *mv_chan)
{
struct dma_chan *chan, *_chan;
@@ -1008,7 +1000,6 @@ mv_xor_channel_add(struct mv_xor_device *xordev,
dma_dev->device_free_chan_resources = mv_xor_free_chan_resources;
dma_dev->device_tx_status = mv_xor_status;
dma_dev->device_issue_pending = mv_xor_issue_pending;
- dma_dev->device_control = mv_xor_control;
dma_dev->dev = &pdev->dev;
/* set prep routines based on capability */
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 46/61] dmaengine: pch-dma: Rename device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Rename the device_control callback of the Intel PCH DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/pch_dma.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/drivers/dma/pch_dma.c b/drivers/dma/pch_dma.c
index 9f9ca9fe5ce6..d03b9c93ccab 100644
--- a/drivers/dma/pch_dma.c
+++ b/drivers/dma/pch_dma.c
@@ -665,16 +665,12 @@ err_desc_get:
return NULL;
}
-static int pd_device_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int pd_device_terminate_all(struct dma_chan *chan)
{
struct pch_dma_chan *pd_chan = to_pd_chan(chan);
struct pch_dma_desc *desc, *_d;
LIST_HEAD(list);
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
spin_lock_irq(&pd_chan->lock);
pdc_set_mode(&pd_chan->chan, DMA_CTL0_DISABLE);
@@ -932,7 +928,7 @@ static int pch_dma_probe(struct pci_dev *pdev,
pd->dma.device_tx_status = pd_tx_status;
pd->dma.device_issue_pending = pd_issue_pending;
pd->dma.device_prep_slave_sg = pd_prep_slave_sg;
- pd->dma.device_control = pd_device_control;
+ pd->dma.device_terminate_all = pd_device_terminate_all;
err = dma_async_device_register(&pd->dma);
if (err) {
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 46/61] dmaengine: pch-dma: Rename device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Rename the device_control callback of the Intel PCH DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/pch_dma.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/drivers/dma/pch_dma.c b/drivers/dma/pch_dma.c
index 9f9ca9fe5ce6..d03b9c93ccab 100644
--- a/drivers/dma/pch_dma.c
+++ b/drivers/dma/pch_dma.c
@@ -665,16 +665,12 @@ err_desc_get:
return NULL;
}
-static int pd_device_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int pd_device_terminate_all(struct dma_chan *chan)
{
struct pch_dma_chan *pd_chan = to_pd_chan(chan);
struct pch_dma_desc *desc, *_d;
LIST_HEAD(list);
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
spin_lock_irq(&pd_chan->lock);
pdc_set_mode(&pd_chan->chan, DMA_CTL0_DISABLE);
@@ -932,7 +928,7 @@ static int pch_dma_probe(struct pci_dev *pdev,
pd->dma.device_tx_status = pd_tx_status;
pd->dma.device_issue_pending = pd_issue_pending;
pd->dma.device_prep_slave_sg = pd_prep_slave_sg;
- pd->dma.device_control = pd_device_control;
+ pd->dma.device_terminate_all = pd_device_terminate_all;
err = dma_async_device_register(&pd->dma);
if (err) {
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 47/61] dmaengine: td: Rename device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Rename the device_control callback of the Timberdal DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/timb_dma.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/drivers/dma/timb_dma.c b/drivers/dma/timb_dma.c
index 2407ccf1a64b..c4c3d93fdd1b 100644
--- a/drivers/dma/timb_dma.c
+++ b/drivers/dma/timb_dma.c
@@ -561,8 +561,7 @@ static struct dma_async_tx_descriptor *td_prep_slave_sg(struct dma_chan *chan,
return &td_desc->txd;
}
-static int td_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int td_terminate_all(struct dma_chan *chan)
{
struct timb_dma_chan *td_chan =
container_of(chan, struct timb_dma_chan, chan);
@@ -570,9 +569,6 @@ static int td_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
dev_dbg(chan2dev(chan), "%s: Entry\n", __func__);
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
/* first the easy part, put the queue into the free list */
spin_lock_bh(&td_chan->lock);
list_for_each_entry_safe(td_desc, _td_desc, &td_chan->queue,
@@ -697,7 +693,7 @@ static int td_probe(struct platform_device *pdev)
dma_cap_set(DMA_SLAVE, td->dma.cap_mask);
dma_cap_set(DMA_PRIVATE, td->dma.cap_mask);
td->dma.device_prep_slave_sg = td_prep_slave_sg;
- td->dma.device_control = td_control;
+ td->dma.device_terminate_all = td_terminate_all;
td->dma.dev = &pdev->dev;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 47/61] dmaengine: td: Rename device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Rename the device_control callback of the Timberdal DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/timb_dma.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/drivers/dma/timb_dma.c b/drivers/dma/timb_dma.c
index 2407ccf1a64b..c4c3d93fdd1b 100644
--- a/drivers/dma/timb_dma.c
+++ b/drivers/dma/timb_dma.c
@@ -561,8 +561,7 @@ static struct dma_async_tx_descriptor *td_prep_slave_sg(struct dma_chan *chan,
return &td_desc->txd;
}
-static int td_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int td_terminate_all(struct dma_chan *chan)
{
struct timb_dma_chan *td_chan =
container_of(chan, struct timb_dma_chan, chan);
@@ -570,9 +569,6 @@ static int td_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
dev_dbg(chan2dev(chan), "%s: Entry\n", __func__);
- if (cmd != DMA_TERMINATE_ALL)
- return -ENXIO;
-
/* first the easy part, put the queue into the free list */
spin_lock_bh(&td_chan->lock);
list_for_each_entry_safe(td_desc, _td_desc, &td_chan->queue,
@@ -697,7 +693,7 @@ static int td_probe(struct platform_device *pdev)
dma_cap_set(DMA_SLAVE, td->dma.cap_mask);
dma_cap_set(DMA_PRIVATE, td->dma.cap_mask);
td->dma.device_prep_slave_sg = td_prep_slave_sg;
- td->dma.device_control = td_control;
+ td->dma.device_terminate_all = td_terminate_all;
td->dma.dev = &pdev->dev;
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 48/61] dmaengine: txx9: Rename device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Rename the device_control callback of the TXX9 DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/txx9dmac.c | 9 ++-------
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/drivers/dma/txx9dmac.c b/drivers/dma/txx9dmac.c
index 17686caf64d5..cb24d3efe3a2 100644
--- a/drivers/dma/txx9dmac.c
+++ b/drivers/dma/txx9dmac.c
@@ -901,17 +901,12 @@ txx9dmac_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
return &first->txd;
}
-static int txx9dmac_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int txx9dmac_terminate_all(struct dma_chan *chan)
{
struct txx9dmac_chan *dc = to_txx9dmac_chan(chan);
struct txx9dmac_desc *desc, *_desc;
LIST_HEAD(list);
- /* Only supports DMA_TERMINATE_ALL */
- if (cmd != DMA_TERMINATE_ALL)
- return -EINVAL;
-
dev_vdbg(chan2dev(chan), "terminate_all\n");
spin_lock_bh(&dc->lock);
@@ -1109,7 +1104,7 @@ static int __init txx9dmac_chan_probe(struct platform_device *pdev)
dc->dma.dev = &pdev->dev;
dc->dma.device_alloc_chan_resources = txx9dmac_alloc_chan_resources;
dc->dma.device_free_chan_resources = txx9dmac_free_chan_resources;
- dc->dma.device_control = txx9dmac_control;
+ dc->dma.device_terminate_all = txx9dmac_terminate_all;
dc->dma.device_tx_status = txx9dmac_tx_status;
dc->dma.device_issue_pending = txx9dmac_issue_pending;
if (pdata && pdata->memcpy_chan == ch) {
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 48/61] dmaengine: txx9: Rename device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Rename the device_control callback of the TXX9 DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/txx9dmac.c | 9 ++-------
1 file changed, 2 insertions(+), 7 deletions(-)
diff --git a/drivers/dma/txx9dmac.c b/drivers/dma/txx9dmac.c
index 17686caf64d5..cb24d3efe3a2 100644
--- a/drivers/dma/txx9dmac.c
+++ b/drivers/dma/txx9dmac.c
@@ -901,17 +901,12 @@ txx9dmac_prep_slave_sg(struct dma_chan *chan, struct scatterlist *sgl,
return &first->txd;
}
-static int txx9dmac_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int txx9dmac_terminate_all(struct dma_chan *chan)
{
struct txx9dmac_chan *dc = to_txx9dmac_chan(chan);
struct txx9dmac_desc *desc, *_desc;
LIST_HEAD(list);
- /* Only supports DMA_TERMINATE_ALL */
- if (cmd != DMA_TERMINATE_ALL)
- return -EINVAL;
-
dev_vdbg(chan2dev(chan), "terminate_all\n");
spin_lock_bh(&dc->lock);
@@ -1109,7 +1104,7 @@ static int __init txx9dmac_chan_probe(struct platform_device *pdev)
dc->dma.dev = &pdev->dev;
dc->dma.device_alloc_chan_resources = txx9dmac_alloc_chan_resources;
dc->dma.device_free_chan_resources = txx9dmac_free_chan_resources;
- dc->dma.device_control = txx9dmac_control;
+ dc->dma.device_terminate_all = txx9dmac_terminate_all;
dc->dma.device_tx_status = txx9dmac_tx_status;
dc->dma.device_issue_pending = txx9dmac_issue_pending;
if (pdata && pdata->memcpy_chan == ch) {
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 49/61] dmaengine: rapidio: tsi721: Rename device_control
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Rename the device_control callback of the TXX9 DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/rapidio/devices/tsi721_dma.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/drivers/rapidio/devices/tsi721_dma.c b/drivers/rapidio/devices/tsi721_dma.c
index f64c5decb747..47295940a868 100644
--- a/drivers/rapidio/devices/tsi721_dma.c
+++ b/drivers/rapidio/devices/tsi721_dma.c
@@ -815,8 +815,7 @@ struct dma_async_tx_descriptor *tsi721_prep_rio_sg(struct dma_chan *dchan,
return txd;
}
-static int tsi721_device_control(struct dma_chan *dchan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int tsi721_terminate_all(struct dma_chan *dchan)
{
struct tsi721_bdma_chan *bdma_chan = to_tsi721_chan(dchan);
struct tsi721_tx_desc *desc, *_d;
@@ -825,9 +824,6 @@ static int tsi721_device_control(struct dma_chan *dchan, enum dma_ctrl_cmd cmd,
dev_dbg(dchan->device->dev, "%s: Entry\n", __func__);
- if (cmd != DMA_TERMINATE_ALL)
- return -ENOSYS;
-
spin_lock_bh(&bdma_chan->lock);
bdma_chan->active = false;
@@ -901,7 +897,7 @@ int tsi721_register_dma(struct tsi721_device *priv)
mport->dma.device_tx_status = tsi721_tx_status;
mport->dma.device_issue_pending = tsi721_issue_pending;
mport->dma.device_prep_slave_sg = tsi721_prep_rio_sg;
- mport->dma.device_control = tsi721_device_control;
+ mport->dma.device_terminate_all = tsi721_terminate_all;
err = dma_async_device_register(&mport->dma);
if (err)
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 49/61] dmaengine: rapidio: tsi721: Rename device_control
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Rename the device_control callback of the TXX9 DMA driver to terminate_all
since it's all it's really doing. That will eventually be used to retrieve
slave capabilities.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/rapidio/devices/tsi721_dma.c | 8 ++------
1 file changed, 2 insertions(+), 6 deletions(-)
diff --git a/drivers/rapidio/devices/tsi721_dma.c b/drivers/rapidio/devices/tsi721_dma.c
index f64c5decb747..47295940a868 100644
--- a/drivers/rapidio/devices/tsi721_dma.c
+++ b/drivers/rapidio/devices/tsi721_dma.c
@@ -815,8 +815,7 @@ struct dma_async_tx_descriptor *tsi721_prep_rio_sg(struct dma_chan *dchan,
return txd;
}
-static int tsi721_device_control(struct dma_chan *dchan, enum dma_ctrl_cmd cmd,
- unsigned long arg)
+static int tsi721_terminate_all(struct dma_chan *dchan)
{
struct tsi721_bdma_chan *bdma_chan = to_tsi721_chan(dchan);
struct tsi721_tx_desc *desc, *_d;
@@ -825,9 +824,6 @@ static int tsi721_device_control(struct dma_chan *dchan, enum dma_ctrl_cmd cmd,
dev_dbg(dchan->device->dev, "%s: Entry\n", __func__);
- if (cmd != DMA_TERMINATE_ALL)
- return -ENOSYS;
-
spin_lock_bh(&bdma_chan->lock);
bdma_chan->active = false;
@@ -901,7 +897,7 @@ int tsi721_register_dma(struct tsi721_device *priv)
mport->dma.device_tx_status = tsi721_tx_status;
mport->dma.device_issue_pending = tsi721_issue_pending;
mport->dma.device_prep_slave_sg = tsi721_prep_rio_sg;
- mport->dma.device_control = tsi721_device_control;
+ mport->dma.device_terminate_all = tsi721_terminate_all;
err = dma_async_device_register(&mport->dma);
if (err)
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 50/61] dmaengine: at_xdmac: Declare slave capabilities for the generic code
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Ludovic Desroches,
Maxime Ripard
From: Ludovic Desroches <ludovic.desroches@atmel.com>
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Ludovic Desroches <ludovic.desroches@atmel.com>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/at_xdmac.c | 33 +++++++++++----------------------
1 file changed, 11 insertions(+), 22 deletions(-)
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index 9906e7cfbe37..3951ba8982de 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -174,6 +174,13 @@
#define AT_XDMAC_MAX_CHAN 0x20
+#define AT_XDMAC_DMA_BUSWIDTHS\
+ (BIT(DMA_SLAVE_BUSWIDTH_UNDEFINED) |\
+ BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |\
+ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |\
+ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |\
+ BIT(DMA_SLAVE_BUSWIDTH_8_BYTES))
+
enum atc_status {
AT_XDMAC_CHAN_IS_CYCLIC = 0,
AT_XDMAC_CHAN_IS_PAUSED,
@@ -1224,27 +1231,6 @@ static void at_xdmac_free_chan_resources(struct dma_chan *chan)
return;
}
-#define AT_XDMAC_DMA_BUSWIDTHS\
- (BIT(DMA_SLAVE_BUSWIDTH_UNDEFINED) |\
- BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |\
- BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |\
- BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |\
- BIT(DMA_SLAVE_BUSWIDTH_8_BYTES))
-
-static int at_xdmac_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
-
- caps->src_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
- caps->dst_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
- caps->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
-
- return 0;
-}
-
#ifdef CONFIG_PM
static int atmel_xdmac_prepare(struct device *dev)
{
@@ -1414,7 +1400,10 @@ static int at_xdmac_probe(struct platform_device *pdev)
atxdmac->dma.device_resume = at_xdmac_device_resume;
atxdmac->dma.device_terminate_all = at_xdmac_device_terminate_all;
atxdmac->dma.chancnt = nr_channels;
- atxdmac->dma.device_slave_caps = at_xdmac_device_slave_caps;
+ atxdmac->dma.src_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
+ atxdmac->dma.dst_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
+ atxdmac->dma.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ atxdmac->dma.residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
/* Disable all chans and interrupts. */
at_xdmac_off(atxdmac);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 50/61] dmaengine: at_xdmac: Declare slave capabilities for the generic code
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
From: Ludovic Desroches <ludovic.desroches@atmel.com>
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Ludovic Desroches <ludovic.desroches@atmel.com>
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/at_xdmac.c | 33 +++++++++++----------------------
1 file changed, 11 insertions(+), 22 deletions(-)
diff --git a/drivers/dma/at_xdmac.c b/drivers/dma/at_xdmac.c
index 9906e7cfbe37..3951ba8982de 100644
--- a/drivers/dma/at_xdmac.c
+++ b/drivers/dma/at_xdmac.c
@@ -174,6 +174,13 @@
#define AT_XDMAC_MAX_CHAN 0x20
+#define AT_XDMAC_DMA_BUSWIDTHS\
+ (BIT(DMA_SLAVE_BUSWIDTH_UNDEFINED) |\
+ BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |\
+ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |\
+ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |\
+ BIT(DMA_SLAVE_BUSWIDTH_8_BYTES))
+
enum atc_status {
AT_XDMAC_CHAN_IS_CYCLIC = 0,
AT_XDMAC_CHAN_IS_PAUSED,
@@ -1224,27 +1231,6 @@ static void at_xdmac_free_chan_resources(struct dma_chan *chan)
return;
}
-#define AT_XDMAC_DMA_BUSWIDTHS\
- (BIT(DMA_SLAVE_BUSWIDTH_UNDEFINED) |\
- BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |\
- BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |\
- BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) |\
- BIT(DMA_SLAVE_BUSWIDTH_8_BYTES))
-
-static int at_xdmac_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
-
- caps->src_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
- caps->dst_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
- caps->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
-
- return 0;
-}
-
#ifdef CONFIG_PM
static int atmel_xdmac_prepare(struct device *dev)
{
@@ -1414,7 +1400,10 @@ static int at_xdmac_probe(struct platform_device *pdev)
atxdmac->dma.device_resume = at_xdmac_device_resume;
atxdmac->dma.device_terminate_all = at_xdmac_device_terminate_all;
atxdmac->dma.chancnt = nr_channels;
- atxdmac->dma.device_slave_caps = at_xdmac_device_slave_caps;
+ atxdmac->dma.src_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
+ atxdmac->dma.dst_addr_widths = AT_XDMAC_DMA_BUSWIDTHS;
+ atxdmac->dma.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ atxdmac->dma.residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
/* Disable all chans and interrupts. */
at_xdmac_off(atxdmac);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 51/61] dmaengine: bcm2835: Declare slave capabilities for the generic code
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
---
drivers/dma/bcm2835-dma.c | 15 +++------------
1 file changed, 3 insertions(+), 12 deletions(-)
diff --git a/drivers/dma/bcm2835-dma.c b/drivers/dma/bcm2835-dma.c
index 13b05c1673b1..0723096fb50a 100644
--- a/drivers/dma/bcm2835-dma.c
+++ b/drivers/dma/bcm2835-dma.c
@@ -550,18 +550,6 @@ static struct dma_chan *bcm2835_dma_xlate(struct of_phandle_args *spec,
return chan;
}
-static int bcm2835_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
- caps->dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = false;
- caps->cmd_terminate = true;
-
- return 0;
-}
-
static int bcm2835_dma_probe(struct platform_device *pdev)
{
struct bcm2835_dmadev *od;
@@ -603,6 +591,9 @@ static int bcm2835_dma_probe(struct platform_device *pdev)
od->ddev.device_prep_dma_cyclic = bcm2835_dma_prep_dma_cyclic;
od->ddev.device_config = bcm2835_dma_slave_config;
od->ddev.device_terminate_all = bcm2835_dma_terminate_all;
+ od->ddev.src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ od->ddev.dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ od->ddev.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
od->ddev.dev = &pdev->dev;
INIT_LIST_HEAD(&od->ddev.channels);
spin_lock_init(&od->lock);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 51/61] dmaengine: bcm2835: Declare slave capabilities for the generic code
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Stephen Warren <swarren@wwwdotorg.org>
---
drivers/dma/bcm2835-dma.c | 15 +++------------
1 file changed, 3 insertions(+), 12 deletions(-)
diff --git a/drivers/dma/bcm2835-dma.c b/drivers/dma/bcm2835-dma.c
index 13b05c1673b1..0723096fb50a 100644
--- a/drivers/dma/bcm2835-dma.c
+++ b/drivers/dma/bcm2835-dma.c
@@ -550,18 +550,6 @@ static struct dma_chan *bcm2835_dma_xlate(struct of_phandle_args *spec,
return chan;
}
-static int bcm2835_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
- caps->dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = false;
- caps->cmd_terminate = true;
-
- return 0;
-}
-
static int bcm2835_dma_probe(struct platform_device *pdev)
{
struct bcm2835_dmadev *od;
@@ -603,6 +591,9 @@ static int bcm2835_dma_probe(struct platform_device *pdev)
od->ddev.device_prep_dma_cyclic = bcm2835_dma_prep_dma_cyclic;
od->ddev.device_config = bcm2835_dma_slave_config;
od->ddev.device_terminate_all = bcm2835_dma_terminate_all;
+ od->ddev.src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ od->ddev.dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ od->ddev.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
od->ddev.dev = &pdev->dev;
INIT_LIST_HEAD(&od->ddev.channels);
spin_lock_init(&od->lock);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 52/61] dmaengine: fsl-edma: Declare slave capabilities for the generic code
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/fsl-edma.c | 17 ++++-------------
1 file changed, 4 insertions(+), 13 deletions(-)
diff --git a/drivers/dma/fsl-edma.c b/drivers/dma/fsl-edma.c
index b5b7474f8859..1382332e7e49 100644
--- a/drivers/dma/fsl-edma.c
+++ b/drivers/dma/fsl-edma.c
@@ -784,18 +784,6 @@ static void fsl_edma_free_chan_resources(struct dma_chan *chan)
fsl_chan->tcd_pool = NULL;
}
-static int fsl_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = FSL_EDMA_BUSWIDTHS;
- caps->dst_addr_widths = FSL_EDMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
-
- return 0;
-}
-
static int
fsl_edma_irq_init(struct platform_device *pdev, struct fsl_edma_engine *fsl_edma)
{
@@ -926,7 +914,10 @@ static int fsl_edma_probe(struct platform_device *pdev)
fsl_edma->dma_dev.device_resume = fsl_edma_resume;
fsl_edma->dma_dev.device_terminate_all = fsl_edma_terminate_all;
fsl_edma->dma_dev.device_issue_pending = fsl_edma_issue_pending;
- fsl_edma->dma_dev.device_slave_caps = fsl_dma_device_slave_caps;
+
+ fsl_edma->dma_dev.src_addr_widths = FSL_EDMA_BUSWIDTHS;
+ fsl_edma->dma_dev.dst_addr_widths = FSL_EDMA_BUSWIDTHS;
+ fsl_edma->dma_dev.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
platform_set_drvdata(pdev, fsl_edma);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 52/61] dmaengine: fsl-edma: Declare slave capabilities for the generic code
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/fsl-edma.c | 17 ++++-------------
1 file changed, 4 insertions(+), 13 deletions(-)
diff --git a/drivers/dma/fsl-edma.c b/drivers/dma/fsl-edma.c
index b5b7474f8859..1382332e7e49 100644
--- a/drivers/dma/fsl-edma.c
+++ b/drivers/dma/fsl-edma.c
@@ -784,18 +784,6 @@ static void fsl_edma_free_chan_resources(struct dma_chan *chan)
fsl_chan->tcd_pool = NULL;
}
-static int fsl_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = FSL_EDMA_BUSWIDTHS;
- caps->dst_addr_widths = FSL_EDMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
-
- return 0;
-}
-
static int
fsl_edma_irq_init(struct platform_device *pdev, struct fsl_edma_engine *fsl_edma)
{
@@ -926,7 +914,10 @@ static int fsl_edma_probe(struct platform_device *pdev)
fsl_edma->dma_dev.device_resume = fsl_edma_resume;
fsl_edma->dma_dev.device_terminate_all = fsl_edma_terminate_all;
fsl_edma->dma_dev.device_issue_pending = fsl_edma_issue_pending;
- fsl_edma->dma_dev.device_slave_caps = fsl_dma_device_slave_caps;
+
+ fsl_edma->dma_dev.src_addr_widths = FSL_EDMA_BUSWIDTHS;
+ fsl_edma->dma_dev.dst_addr_widths = FSL_EDMA_BUSWIDTHS;
+ fsl_edma->dma_dev.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
platform_set_drvdata(pdev, fsl_edma);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 53/61] dmaengine: edma: Declare slave capabilities for the generic code
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/edma.c | 20 ++++++--------------
1 file changed, 6 insertions(+), 14 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 56735a0aa3a8..7aaa59048e2d 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -971,19 +971,6 @@ static void __init edma_chan_init(struct edma_cc *ecc,
BIT(DMA_SLAVE_BUSWIDTH_3_BYTES) | \
BIT(DMA_SLAVE_BUSWIDTH_4_BYTES))
-static int edma_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = EDMA_DMA_BUSWIDTHS;
- caps->dst_addr_widths = EDMA_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
- caps->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
-
- return 0;
-}
-
static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
struct device *dev)
{
@@ -998,7 +985,12 @@ static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
dma->device_pause = edma_dma_pause;
dma->device_resume = edma_dma_resume;
dma->device_terminate_all = edma_terminate_all;
- dma->device_slave_caps = edma_dma_device_slave_caps;
+
+ dma->src_addr_widths = EDMA_DMA_BUSWIDTHS;
+ dma->dst_addr_widths = EDMA_DMA_BUSWIDTHS;
+ dma->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ dma->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
+
dma->dev = dev;
/*
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 53/61] dmaengine: edma: Declare slave capabilities for the generic code
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/edma.c | 20 ++++++--------------
1 file changed, 6 insertions(+), 14 deletions(-)
diff --git a/drivers/dma/edma.c b/drivers/dma/edma.c
index 56735a0aa3a8..7aaa59048e2d 100644
--- a/drivers/dma/edma.c
+++ b/drivers/dma/edma.c
@@ -971,19 +971,6 @@ static void __init edma_chan_init(struct edma_cc *ecc,
BIT(DMA_SLAVE_BUSWIDTH_3_BYTES) | \
BIT(DMA_SLAVE_BUSWIDTH_4_BYTES))
-static int edma_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = EDMA_DMA_BUSWIDTHS;
- caps->dst_addr_widths = EDMA_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
- caps->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
-
- return 0;
-}
-
static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
struct device *dev)
{
@@ -998,7 +985,12 @@ static void edma_dma_init(struct edma_cc *ecc, struct dma_device *dma,
dma->device_pause = edma_dma_pause;
dma->device_resume = edma_dma_resume;
dma->device_terminate_all = edma_terminate_all;
- dma->device_slave_caps = edma_dma_device_slave_caps;
+
+ dma->src_addr_widths = EDMA_DMA_BUSWIDTHS;
+ dma->dst_addr_widths = EDMA_DMA_BUSWIDTHS;
+ dma->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ dma->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
+
dma->dev = dev;
/*
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 54/61] dmaengine: nbpfaxi: Declare slave capabilities for the generic code
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/nbpfaxi.c | 17 ++++-------------
1 file changed, 4 insertions(+), 13 deletions(-)
diff --git a/drivers/dma/nbpfaxi.c b/drivers/dma/nbpfaxi.c
index 9f151dcd4f73..6a963928e958 100644
--- a/drivers/dma/nbpfaxi.c
+++ b/drivers/dma/nbpfaxi.c
@@ -1069,18 +1069,6 @@ static void nbpf_free_chan_resources(struct dma_chan *dchan)
}
}
-static int nbpf_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = NBPF_DMA_BUSWIDTHS;
- caps->dst_addr_widths = NBPF_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = false;
- caps->cmd_terminate = true;
-
- return 0;
-}
-
static struct dma_chan *nbpf_of_xlate(struct of_phandle_args *dma_spec,
struct of_dma *ofdma)
{
@@ -1411,7 +1399,6 @@ static int nbpf_probe(struct platform_device *pdev)
dma_dev->device_prep_dma_memcpy = nbpf_prep_memcpy;
dma_dev->device_tx_status = nbpf_tx_status;
dma_dev->device_issue_pending = nbpf_issue_pending;
- dma_dev->device_slave_caps = nbpf_slave_caps;
/*
* If we drop support for unaligned MEMCPY buffer addresses and / or
@@ -1427,6 +1414,10 @@ static int nbpf_probe(struct platform_device *pdev)
dma_dev->device_pause = nbpf_pause;
dma_dev->device_terminate_all = nbpf_terminate_all;
+ dma_dev->src_addr_widths = NBPF_DMA_BUSWIDTHS;
+ dma_dev->dst_addr_widths = NBPF_DMA_BUSWIDTHS;
+ dma_dev->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+
platform_set_drvdata(pdev, nbpf);
ret = clk_prepare_enable(nbpf->clk);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 54/61] dmaengine: nbpfaxi: Declare slave capabilities for the generic code
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/nbpfaxi.c | 17 ++++-------------
1 file changed, 4 insertions(+), 13 deletions(-)
diff --git a/drivers/dma/nbpfaxi.c b/drivers/dma/nbpfaxi.c
index 9f151dcd4f73..6a963928e958 100644
--- a/drivers/dma/nbpfaxi.c
+++ b/drivers/dma/nbpfaxi.c
@@ -1069,18 +1069,6 @@ static void nbpf_free_chan_resources(struct dma_chan *dchan)
}
}
-static int nbpf_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = NBPF_DMA_BUSWIDTHS;
- caps->dst_addr_widths = NBPF_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = false;
- caps->cmd_terminate = true;
-
- return 0;
-}
-
static struct dma_chan *nbpf_of_xlate(struct of_phandle_args *dma_spec,
struct of_dma *ofdma)
{
@@ -1411,7 +1399,6 @@ static int nbpf_probe(struct platform_device *pdev)
dma_dev->device_prep_dma_memcpy = nbpf_prep_memcpy;
dma_dev->device_tx_status = nbpf_tx_status;
dma_dev->device_issue_pending = nbpf_issue_pending;
- dma_dev->device_slave_caps = nbpf_slave_caps;
/*
* If we drop support for unaligned MEMCPY buffer addresses and / or
@@ -1427,6 +1414,10 @@ static int nbpf_probe(struct platform_device *pdev)
dma_dev->device_pause = nbpf_pause;
dma_dev->device_terminate_all = nbpf_terminate_all;
+ dma_dev->src_addr_widths = NBPF_DMA_BUSWIDTHS;
+ dma_dev->dst_addr_widths = NBPF_DMA_BUSWIDTHS;
+ dma_dev->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+
platform_set_drvdata(pdev, nbpf);
ret = clk_prepare_enable(nbpf->clk);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 55/61] dmaengine: omap: Declare slave capabilities for the generic code
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/omap-dma.c | 18 ++++--------------
1 file changed, 4 insertions(+), 14 deletions(-)
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index 1cadf812f3b3..76bb63d6b223 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c
@@ -1070,19 +1070,6 @@ static void omap_dma_free(struct omap_dmadev *od)
BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) | \
BIT(DMA_SLAVE_BUSWIDTH_4_BYTES))
-static int omap_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = OMAP_DMA_BUSWIDTHS;
- caps->dst_addr_widths = OMAP_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
- caps->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
-
- return 0;
-}
-
static int omap_dma_probe(struct platform_device *pdev)
{
struct omap_dmadev *od;
@@ -1116,7 +1103,10 @@ static int omap_dma_probe(struct platform_device *pdev)
od->ddev.device_pause = omap_dma_pause;
od->ddev.device_resume = omap_dma_resume;
od->ddev.device_terminate_all = omap_dma_terminate_all;
- od->ddev.device_slave_caps = omap_dma_device_slave_caps;
+ od->ddev.src_addr_widths = OMAP_DMA_BUSWIDTHS;
+ od->ddev.dst_addr_widths = OMAP_DMA_BUSWIDTHS;
+ od->ddev.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ od->ddev.residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
od->ddev.dev = &pdev->dev;
INIT_LIST_HEAD(&od->ddev.channels);
INIT_LIST_HEAD(&od->pending);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 55/61] dmaengine: omap: Declare slave capabilities for the generic code
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/omap-dma.c | 18 ++++--------------
1 file changed, 4 insertions(+), 14 deletions(-)
diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
index 1cadf812f3b3..76bb63d6b223 100644
--- a/drivers/dma/omap-dma.c
+++ b/drivers/dma/omap-dma.c
@@ -1070,19 +1070,6 @@ static void omap_dma_free(struct omap_dmadev *od)
BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) | \
BIT(DMA_SLAVE_BUSWIDTH_4_BYTES))
-static int omap_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = OMAP_DMA_BUSWIDTHS;
- caps->dst_addr_widths = OMAP_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
- caps->residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
-
- return 0;
-}
-
static int omap_dma_probe(struct platform_device *pdev)
{
struct omap_dmadev *od;
@@ -1116,7 +1103,10 @@ static int omap_dma_probe(struct platform_device *pdev)
od->ddev.device_pause = omap_dma_pause;
od->ddev.device_resume = omap_dma_resume;
od->ddev.device_terminate_all = omap_dma_terminate_all;
- od->ddev.device_slave_caps = omap_dma_device_slave_caps;
+ od->ddev.src_addr_widths = OMAP_DMA_BUSWIDTHS;
+ od->ddev.dst_addr_widths = OMAP_DMA_BUSWIDTHS;
+ od->ddev.directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ od->ddev.residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
od->ddev.dev = &pdev->dev;
INIT_LIST_HEAD(&od->ddev.channels);
INIT_LIST_HEAD(&od->pending);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 56/61] dmaengine: pl330: Declare slave capabilities for the generic code
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/pl330.c | 18 ++++--------------
1 file changed, 4 insertions(+), 14 deletions(-)
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index c2db237e455b..8063563618b3 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2581,19 +2581,6 @@ static irqreturn_t pl330_irq_handler(int irq, void *data)
BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) | \
BIT(DMA_SLAVE_BUSWIDTH_8_BYTES)
-static int pl330_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = PL330_DMA_BUSWIDTHS;
- caps->dst_addr_widths = PL330_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = false;
- caps->cmd_terminate = true;
- caps->residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
-
- return 0;
-}
-
static int
pl330_probe(struct amba_device *adev, const struct amba_id *id)
{
@@ -2714,7 +2701,10 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
pd->device_config = pl330_config;
pd->device_terminate_all = pl330_terminate_all;
pd->device_issue_pending = pl330_issue_pending;
- pd->device_slave_caps = pl330_dma_device_slave_caps;
+ pd->src_addr_widths = PL330_DMA_BUSWIDTHS;
+ pd->dst_addr_widths = PL330_DMA_BUSWIDTHS;
+ pd->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ pd->residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
ret = dma_async_device_register(pd);
if (ret) {
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 56/61] dmaengine: pl330: Declare slave capabilities for the generic code
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/pl330.c | 18 ++++--------------
1 file changed, 4 insertions(+), 14 deletions(-)
diff --git a/drivers/dma/pl330.c b/drivers/dma/pl330.c
index c2db237e455b..8063563618b3 100644
--- a/drivers/dma/pl330.c
+++ b/drivers/dma/pl330.c
@@ -2581,19 +2581,6 @@ static irqreturn_t pl330_irq_handler(int irq, void *data)
BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) | \
BIT(DMA_SLAVE_BUSWIDTH_8_BYTES)
-static int pl330_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = PL330_DMA_BUSWIDTHS;
- caps->dst_addr_widths = PL330_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = false;
- caps->cmd_terminate = true;
- caps->residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
-
- return 0;
-}
-
static int
pl330_probe(struct amba_device *adev, const struct amba_id *id)
{
@@ -2714,7 +2701,10 @@ pl330_probe(struct amba_device *adev, const struct amba_id *id)
pd->device_config = pl330_config;
pd->device_terminate_all = pl330_terminate_all;
pd->device_issue_pending = pl330_issue_pending;
- pd->device_slave_caps = pl330_dma_device_slave_caps;
+ pd->src_addr_widths = PL330_DMA_BUSWIDTHS;
+ pd->dst_addr_widths = PL330_DMA_BUSWIDTHS;
+ pd->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
+ pd->residue_granularity = DMA_RESIDUE_GRANULARITY_DESCRIPTOR;
ret = dma_async_device_register(pd);
if (ret) {
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 57/61] dmaengine: sirf: Declare slave capabilities for the generic code
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sirf-dma.c | 16 +++-------------
1 file changed, 3 insertions(+), 13 deletions(-)
diff --git a/drivers/dma/sirf-dma.c b/drivers/dma/sirf-dma.c
index eeb2af440b9f..2f56d3d66d5d 100644
--- a/drivers/dma/sirf-dma.c
+++ b/drivers/dma/sirf-dma.c
@@ -628,18 +628,6 @@ EXPORT_SYMBOL(sirfsoc_dma_filter_id);
BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) | \
BIT(DMA_SLAVE_BUSWIDTH_8_BYTES))
-static int sirfsoc_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
- caps->dst_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
-
- return 0;
-}
-
static struct dma_chan *of_dma_sirfsoc_xlate(struct of_phandle_args *dma_spec,
struct of_dma *ofdma)
{
@@ -726,7 +714,9 @@ static int sirfsoc_dma_probe(struct platform_device *op)
dma->device_tx_status = sirfsoc_dma_tx_status;
dma->device_prep_interleaved_dma = sirfsoc_dma_prep_interleaved;
dma->device_prep_dma_cyclic = sirfsoc_dma_prep_cyclic;
- dma->device_slave_caps = sirfsoc_dma_device_slave_caps;
+ dma->src_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
+ dma->dst_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
+ dma->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
INIT_LIST_HEAD(&dma->channels);
dma_cap_set(DMA_SLAVE, dma->cap_mask);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 57/61] dmaengine: sirf: Declare slave capabilities for the generic code
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sirf-dma.c | 16 +++-------------
1 file changed, 3 insertions(+), 13 deletions(-)
diff --git a/drivers/dma/sirf-dma.c b/drivers/dma/sirf-dma.c
index eeb2af440b9f..2f56d3d66d5d 100644
--- a/drivers/dma/sirf-dma.c
+++ b/drivers/dma/sirf-dma.c
@@ -628,18 +628,6 @@ EXPORT_SYMBOL(sirfsoc_dma_filter_id);
BIT(DMA_SLAVE_BUSWIDTH_4_BYTES) | \
BIT(DMA_SLAVE_BUSWIDTH_8_BYTES))
-static int sirfsoc_dma_device_slave_caps(struct dma_chan *dchan,
- struct dma_slave_caps *caps)
-{
- caps->src_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
- caps->dst_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
- caps->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
- caps->cmd_pause = true;
- caps->cmd_terminate = true;
-
- return 0;
-}
-
static struct dma_chan *of_dma_sirfsoc_xlate(struct of_phandle_args *dma_spec,
struct of_dma *ofdma)
{
@@ -726,7 +714,9 @@ static int sirfsoc_dma_probe(struct platform_device *op)
dma->device_tx_status = sirfsoc_dma_tx_status;
dma->device_prep_interleaved_dma = sirfsoc_dma_prep_interleaved;
dma->device_prep_dma_cyclic = sirfsoc_dma_prep_cyclic;
- dma->device_slave_caps = sirfsoc_dma_device_slave_caps;
+ dma->src_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
+ dma->dst_addr_widths = SIRFSOC_DMA_BUSWIDTHS;
+ dma->directions = BIT(DMA_DEV_TO_MEM) | BIT(DMA_MEM_TO_DEV);
INIT_LIST_HEAD(&dma->channels);
dma_cap_set(DMA_SLAVE, dma->cap_mask);
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 58/61] dmaengine: sun6i: Declare slave capabilities for the generic code
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sun6i-dma.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/drivers/dma/sun6i-dma.c b/drivers/dma/sun6i-dma.c
index 4ef76bc22da4..c2082d318a11 100644
--- a/drivers/dma/sun6i-dma.c
+++ b/drivers/dma/sun6i-dma.c
@@ -968,6 +968,15 @@ static int sun6i_dma_probe(struct platform_device *pdev)
sdc->slave.device_resume = sun6i_dma_resume;
sdc->slave.device_terminate_all = sun6i_dma_terminate_all;
sdc->slave.copy_align = 4;
+ sdc->slave.src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ sdc->slave.dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ sdc->slave.directions = BIT(DMA_DEV_TO_MEM) |
+ BIT(DMA_MEM_TO_DEV);
+ sdc->slave.residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
sdc->slave.dev = &pdev->dev;
sdc->pchans = devm_kcalloc(&pdev->dev, sdc->cfg->nr_max_channels,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 58/61] dmaengine: sun6i: Declare slave capabilities for the generic code
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that the generic slave caps code can make use of the device assigned
capabilities, instead of relying on a callback to be implemented.
Make use of this code.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
drivers/dma/sun6i-dma.c | 9 +++++++++
1 file changed, 9 insertions(+)
diff --git a/drivers/dma/sun6i-dma.c b/drivers/dma/sun6i-dma.c
index 4ef76bc22da4..c2082d318a11 100644
--- a/drivers/dma/sun6i-dma.c
+++ b/drivers/dma/sun6i-dma.c
@@ -968,6 +968,15 @@ static int sun6i_dma_probe(struct platform_device *pdev)
sdc->slave.device_resume = sun6i_dma_resume;
sdc->slave.device_terminate_all = sun6i_dma_terminate_all;
sdc->slave.copy_align = 4;
+ sdc->slave.src_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ sdc->slave.dst_addr_widths = BIT(DMA_SLAVE_BUSWIDTH_1_BYTE) |
+ BIT(DMA_SLAVE_BUSWIDTH_2_BYTES) |
+ BIT(DMA_SLAVE_BUSWIDTH_4_BYTES);
+ sdc->slave.directions = BIT(DMA_DEV_TO_MEM) |
+ BIT(DMA_MEM_TO_DEV);
+ sdc->slave.residue_granularity = DMA_RESIDUE_GRANULARITY_BURST;
sdc->slave.dev = &pdev->dev;
sdc->pchans = devm_kcalloc(&pdev->dev, sdc->cfg->nr_max_channels,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
For the slave caps retrieval to be really useful, most drivers need to
implement it.
Hence, we need to be slightly more aggressive, and trigger a warning at
registration time for drivers that don't fill their caps infos in order to
encourage them to implement it.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/dmaengine.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 98e9431f85ec..300c8cd2786c 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -827,6 +827,9 @@ int dma_async_device_register(struct dma_device *device)
BUG_ON(!device->device_issue_pending);
BUG_ON(!device->dev);
+ WARN(dma_has_cap(DMA_SLAVE, device->cap_mask) && !device->directions,
+ "this driver doesn't support generic slave capabilities reporting\n");
+
/* note: this only matters in the
* CONFIG_ASYNC_TX_ENABLE_CHANNEL_SWITCH=n case
*/
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
For the slave caps retrieval to be really useful, most drivers need to
implement it.
Hence, we need to be slightly more aggressive, and trigger a warning at
registration time for drivers that don't fill their caps infos in order to
encourage them to implement it.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
drivers/dma/dmaengine.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
index 98e9431f85ec..300c8cd2786c 100644
--- a/drivers/dma/dmaengine.c
+++ b/drivers/dma/dmaengine.c
@@ -827,6 +827,9 @@ int dma_async_device_register(struct dma_device *device)
BUG_ON(!device->device_issue_pending);
BUG_ON(!device->dev);
+ WARN(dma_has_cap(DMA_SLAVE, device->cap_mask) && !device->directions,
+ "this driver doesn't support generic slave capabilities reporting\n");
+
/* note: this only matters in the
* CONFIG_ASYNC_TX_ENABLE_CHANNEL_SWITCH=n case
*/
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 60/61] dmaengine: Remove device_control and device_slave_caps
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that device_control has been split into several functions, and
device_slave_caps rendered useless, we can safely remove them.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 52 ++++++-----------------------------------------
1 file changed, 6 insertions(+), 46 deletions(-)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index be9e60aa8f5e..b01b7a7c581b 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -189,25 +189,6 @@ enum dma_ctrl_flags {
};
/**
- * enum dma_ctrl_cmd - DMA operations that can optionally be exercised
- * on a running channel.
- * @DMA_TERMINATE_ALL: terminate all ongoing transfers
- * @DMA_PAUSE: pause ongoing transfers
- * @DMA_RESUME: resume paused transfer
- * @DMA_SLAVE_CONFIG: this command is only implemented by DMA controllers
- * that need to runtime reconfigure the slave channels (as opposed to passing
- * configuration data in statically from the platform). An additional
- * argument of struct dma_slave_config must be passed in with this
- * command.
- */
-enum dma_ctrl_cmd {
- DMA_TERMINATE_ALL,
- DMA_PAUSE,
- DMA_RESUME,
- DMA_SLAVE_CONFIG,
-};
-
-/**
* enum sum_check_bits - bit position of pq_check_flags
*/
enum sum_check_bits {
@@ -336,9 +317,8 @@ enum dma_slave_buswidth {
* This struct is passed in as configuration data to a DMA engine
* in order to set up a certain channel for DMA transport at runtime.
* The DMA device/engine has to provide support for an additional
- * command in the channel config interface, DMA_SLAVE_CONFIG
- * and this struct will then be passed in as an argument to the
- * DMA engine device_control() function.
+ * callback in the dma_device structure, device_config and this struct
+ * will then be passed in as an argument to the function.
*
* The rationale for adding configuration information to this struct is as
* follows: if it is likely that more than one DMA slave controllers in
@@ -617,8 +597,6 @@ struct dma_tx_state {
* @device_prep_interleaved_dma: Transfer expression in a generic way.
* @device_config: Pushes a new configuration to a channel, return 0 or an error
* code
- * @device_control: manipulate all pending operations on a channel, returns
- * zero or error code
* @device_pause: Pauses any transfer happening on a channel. Returns
* 0 or an error code
* @device_resume: Resumes any transfer on a channel previously
@@ -630,7 +608,6 @@ struct dma_tx_state {
* struct with auxiliary transfer status information, otherwise the call
* will just return a simple status code
* @device_issue_pending: push pending transactions to hardware
- * @device_slave_caps: return the slave channel capabilities
*/
struct dma_device {
@@ -697,8 +674,6 @@ struct dma_device {
int (*device_config)(struct dma_chan *chan,
struct dma_slave_config *config);
- int (*device_control)(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg);
int (*device_pause)(struct dma_chan *chan);
int (*device_resume)(struct dma_chan *chan);
int (*device_terminate_all)(struct dma_chan *chan);
@@ -707,27 +682,15 @@ struct dma_device {
dma_cookie_t cookie,
struct dma_tx_state *txstate);
void (*device_issue_pending)(struct dma_chan *chan);
- int (*device_slave_caps)(struct dma_chan *chan, struct dma_slave_caps *caps);
};
-static inline int dmaengine_device_control(struct dma_chan *chan,
- enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- if (chan->device->device_control)
- return chan->device->device_control(chan, cmd, arg);
-
- return -ENOSYS;
-}
-
static inline int dmaengine_slave_config(struct dma_chan *chan,
struct dma_slave_config *config)
{
if (chan->device->device_config)
return chan->device->device_config(chan, config);
- return dmaengine_device_control(chan, DMA_SLAVE_CONFIG,
- (unsigned long)config);
+ return -ENOSYS;
}
static inline bool is_slave_direction(enum dma_transfer_direction direction)
@@ -807,9 +770,6 @@ static inline int dma_get_slave_caps(struct dma_chan *chan, struct dma_slave_cap
if (!test_bit(DMA_SLAVE, device->cap_mask.bits))
return -ENXIO;
- if (device->device_slave_caps)
- return device->device_slave_caps(chan, caps);
-
/*
* Check whether it reports it uses the generic slave
* capabilities, if not, that means it doesn't support any
@@ -834,7 +794,7 @@ static inline int dmaengine_terminate_all(struct dma_chan *chan)
if (chan->device->device_terminate_all)
return chan->device->device_terminate_all(chan);
- return dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
+ return -ENOSYS;
}
static inline int dmaengine_pause(struct dma_chan *chan)
@@ -842,7 +802,7 @@ static inline int dmaengine_pause(struct dma_chan *chan)
if (chan->device->device_pause)
return chan->device->device_pause(chan);
- return dmaengine_device_control(chan, DMA_PAUSE, 0);
+ return -ENOSYS;
}
static inline int dmaengine_resume(struct dma_chan *chan)
@@ -850,7 +810,7 @@ static inline int dmaengine_resume(struct dma_chan *chan)
if (chan->device->device_resume)
return chan->device->device_resume(chan);
- return dmaengine_device_control(chan, DMA_RESUME, 0);
+ return -ENOSYS;
}
static inline enum dma_status dmaengine_tx_status(struct dma_chan *chan,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 60/61] dmaengine: Remove device_control and device_slave_caps
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that device_control has been split into several functions, and
device_slave_caps rendered useless, we can safely remove them.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
include/linux/dmaengine.h | 52 ++++++-----------------------------------------
1 file changed, 6 insertions(+), 46 deletions(-)
diff --git a/include/linux/dmaengine.h b/include/linux/dmaengine.h
index be9e60aa8f5e..b01b7a7c581b 100644
--- a/include/linux/dmaengine.h
+++ b/include/linux/dmaengine.h
@@ -189,25 +189,6 @@ enum dma_ctrl_flags {
};
/**
- * enum dma_ctrl_cmd - DMA operations that can optionally be exercised
- * on a running channel.
- * @DMA_TERMINATE_ALL: terminate all ongoing transfers
- * @DMA_PAUSE: pause ongoing transfers
- * @DMA_RESUME: resume paused transfer
- * @DMA_SLAVE_CONFIG: this command is only implemented by DMA controllers
- * that need to runtime reconfigure the slave channels (as opposed to passing
- * configuration data in statically from the platform). An additional
- * argument of struct dma_slave_config must be passed in with this
- * command.
- */
-enum dma_ctrl_cmd {
- DMA_TERMINATE_ALL,
- DMA_PAUSE,
- DMA_RESUME,
- DMA_SLAVE_CONFIG,
-};
-
-/**
* enum sum_check_bits - bit position of pq_check_flags
*/
enum sum_check_bits {
@@ -336,9 +317,8 @@ enum dma_slave_buswidth {
* This struct is passed in as configuration data to a DMA engine
* in order to set up a certain channel for DMA transport at runtime.
* The DMA device/engine has to provide support for an additional
- * command in the channel config interface, DMA_SLAVE_CONFIG
- * and this struct will then be passed in as an argument to the
- * DMA engine device_control() function.
+ * callback in the dma_device structure, device_config and this struct
+ * will then be passed in as an argument to the function.
*
* The rationale for adding configuration information to this struct is as
* follows: if it is likely that more than one DMA slave controllers in
@@ -617,8 +597,6 @@ struct dma_tx_state {
* @device_prep_interleaved_dma: Transfer expression in a generic way.
* @device_config: Pushes a new configuration to a channel, return 0 or an error
* code
- * @device_control: manipulate all pending operations on a channel, returns
- * zero or error code
* @device_pause: Pauses any transfer happening on a channel. Returns
* 0 or an error code
* @device_resume: Resumes any transfer on a channel previously
@@ -630,7 +608,6 @@ struct dma_tx_state {
* struct with auxiliary transfer status information, otherwise the call
* will just return a simple status code
* @device_issue_pending: push pending transactions to hardware
- * @device_slave_caps: return the slave channel capabilities
*/
struct dma_device {
@@ -697,8 +674,6 @@ struct dma_device {
int (*device_config)(struct dma_chan *chan,
struct dma_slave_config *config);
- int (*device_control)(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
- unsigned long arg);
int (*device_pause)(struct dma_chan *chan);
int (*device_resume)(struct dma_chan *chan);
int (*device_terminate_all)(struct dma_chan *chan);
@@ -707,27 +682,15 @@ struct dma_device {
dma_cookie_t cookie,
struct dma_tx_state *txstate);
void (*device_issue_pending)(struct dma_chan *chan);
- int (*device_slave_caps)(struct dma_chan *chan, struct dma_slave_caps *caps);
};
-static inline int dmaengine_device_control(struct dma_chan *chan,
- enum dma_ctrl_cmd cmd,
- unsigned long arg)
-{
- if (chan->device->device_control)
- return chan->device->device_control(chan, cmd, arg);
-
- return -ENOSYS;
-}
-
static inline int dmaengine_slave_config(struct dma_chan *chan,
struct dma_slave_config *config)
{
if (chan->device->device_config)
return chan->device->device_config(chan, config);
- return dmaengine_device_control(chan, DMA_SLAVE_CONFIG,
- (unsigned long)config);
+ return -ENOSYS;
}
static inline bool is_slave_direction(enum dma_transfer_direction direction)
@@ -807,9 +770,6 @@ static inline int dma_get_slave_caps(struct dma_chan *chan, struct dma_slave_cap
if (!test_bit(DMA_SLAVE, device->cap_mask.bits))
return -ENXIO;
- if (device->device_slave_caps)
- return device->device_slave_caps(chan, caps);
-
/*
* Check whether it reports it uses the generic slave
* capabilities, if not, that means it doesn't support any
@@ -834,7 +794,7 @@ static inline int dmaengine_terminate_all(struct dma_chan *chan)
if (chan->device->device_terminate_all)
return chan->device->device_terminate_all(chan);
- return dmaengine_device_control(chan, DMA_TERMINATE_ALL, 0);
+ return -ENOSYS;
}
static inline int dmaengine_pause(struct dma_chan *chan)
@@ -842,7 +802,7 @@ static inline int dmaengine_pause(struct dma_chan *chan)
if (chan->device->device_pause)
return chan->device->device_pause(chan);
- return dmaengine_device_control(chan, DMA_PAUSE, 0);
+ return -ENOSYS;
}
static inline int dmaengine_resume(struct dma_chan *chan)
@@ -850,7 +810,7 @@ static inline int dmaengine_resume(struct dma_chan *chan)
if (chan->device->device_resume)
return chan->device->device_resume(chan);
- return dmaengine_device_control(chan, DMA_RESUME, 0);
+ return -ENOSYS;
}
static inline enum dma_status dmaengine_tx_status(struct dma_chan *chan,
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 61/61] Documentation: dmaengine: Update the documentation
2014-11-17 13:41 ` Maxime Ripard
@ 2014-11-17 13:42 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: Vinod Koul, dmaengine
Cc: linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars, Maxime Ripard
Now that we have splitted device_control and removed device_slave_caps in favor
of a few dma_device variables, update the documentation accordingly.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
Documentation/dmaengine/provider.txt | 93 ++++++++++++++++++++----------------
1 file changed, 51 insertions(+), 42 deletions(-)
diff --git a/Documentation/dmaengine/provider.txt b/Documentation/dmaengine/provider.txt
index 766658ccf235..2c391cfe37eb 100644
--- a/Documentation/dmaengine/provider.txt
+++ b/Documentation/dmaengine/provider.txt
@@ -113,6 +113,31 @@ need to initialize a few fields in there:
* channels: should be initialized as a list using the
INIT_LIST_HEAD macro for example
+ * src_addr_widths:
+ - should contain a bitmask of the supported source transfer width
+
+ * dst_addr_widths:
+ - should contain a bitmask of the supported destination transfer
+ width
+
+ * directions:
+ - should contain a bitmask of the supported slave directions
+ (i.e. excluding mem2mem transfers)
+
+ * residue_granularity:
+ - Granularity of the transfer residue reported to dma_set_residue.
+ - This can be either:
+ + Descriptor
+ -> Your device doesn't support any kind of residue
+ reporting. The framework will only know that a particular
+ transaction descriptor is done.
+ + Segment
+ -> Your device is able to report which chunks have been
+ transferred
+ + Burst
+ -> Your device is able to report which burst have been
+ transferred
+
* dev: should hold the pointer to the struct device associated
to your current driver instance.
@@ -274,48 +299,32 @@ supported.
account the current period.
- This function can be called in an interrupt context.
- * device_control
- - Used by client drivers to control and configure the channel it
- has a handle on.
- - Called with a command and an argument
- + The command is one of the values listed by the enum
- dma_ctrl_cmd. The valid commands are:
- + DMA_PAUSE
- + Pauses a transfer on the channel
- + This command should operate synchronously on the channel,
- pausing right away the work of the given channel
- + DMA_RESUME
- + Restarts a transfer on the channel
- + This command should operate synchronously on the channel,
- resuming right away the work of the given channel
- + DMA_TERMINATE_ALL
- + Aborts all the pending and ongoing transfers on the
- channel
- + This command should operate synchronously on the channel,
- terminating right away all the channels
- + DMA_SLAVE_CONFIG
- + Reconfigures the channel with passed configuration
- + This command should NOT perform synchronously, or on any
- currently queued transfers, but only on subsequent ones
- + In this case, the function will receive a
- dma_slave_config structure pointer as an argument, that
- will detail which configuration to use.
- + Even though that structure contains a direction field,
- this field is deprecated in favor of the direction
- argument given to the prep_* functions
- + FSLDMA_EXTERNAL_START
- + TODO: Why does that even exist?
- + The argument is an opaque unsigned long. This actually is a
- pointer to a struct dma_slave_config that should be used only
- in the DMA_SLAVE_CONFIG.
-
- * device_slave_caps
- - Called through the framework by client drivers in order to have
- an idea of what are the properties of the channel allocated to
- them.
- - Such properties are the buswidth, available directions, etc.
- - Required for every generic layer doing DMA transfers, such as
- ASoC.
+ * device_config
+ - Reconfigures the channel with the configuration given as
+ argument
+ - This command should NOT perform synchronously, or on any
+ currently queued transfers, but only on subsequent ones
+ - In this case, the function will receive a dma_slave_config
+ structure pointer as an argument, that will detail which
+ configuration to use.
+ - Even though that structure contains a direction field, this
+ field is deprecated in favor of the direction argument given to
+ the prep_* functions
+
+ * device_pause
+ - Pauses a transfer on the channel
+ - This command should operate synchronously on the channel,
+ pausing right away the work of the given channel
+
+ * device_resume
+ - Resumes a transfer on the channel
+ - This command should operate synchronously on the channel,
+ pausing right away the work of the given channel
+
+ * device_terminate_all
+ - Aborts all the pending and ongoing transfers on the channel
+ - This command should operate synchronously on the channel,
+ terminating right away all the channels
Misc notes (stuff that should be documented, but don't really know
where to put them)
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* [PATCH v5 61/61] Documentation: dmaengine: Update the documentation
@ 2014-11-17 13:42 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-11-17 13:42 UTC (permalink / raw)
To: linux-arm-kernel
Now that we have splitted device_control and removed device_slave_caps in favor
of a few dma_device variables, update the documentation accordingly.
Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
---
Documentation/dmaengine/provider.txt | 93 ++++++++++++++++++++----------------
1 file changed, 51 insertions(+), 42 deletions(-)
diff --git a/Documentation/dmaengine/provider.txt b/Documentation/dmaengine/provider.txt
index 766658ccf235..2c391cfe37eb 100644
--- a/Documentation/dmaengine/provider.txt
+++ b/Documentation/dmaengine/provider.txt
@@ -113,6 +113,31 @@ need to initialize a few fields in there:
* channels: should be initialized as a list using the
INIT_LIST_HEAD macro for example
+ * src_addr_widths:
+ - should contain a bitmask of the supported source transfer width
+
+ * dst_addr_widths:
+ - should contain a bitmask of the supported destination transfer
+ width
+
+ * directions:
+ - should contain a bitmask of the supported slave directions
+ (i.e. excluding mem2mem transfers)
+
+ * residue_granularity:
+ - Granularity of the transfer residue reported to dma_set_residue.
+ - This can be either:
+ + Descriptor
+ -> Your device doesn't support any kind of residue
+ reporting. The framework will only know that a particular
+ transaction descriptor is done.
+ + Segment
+ -> Your device is able to report which chunks have been
+ transferred
+ + Burst
+ -> Your device is able to report which burst have been
+ transferred
+
* dev: should hold the pointer to the struct device associated
to your current driver instance.
@@ -274,48 +299,32 @@ supported.
account the current period.
- This function can be called in an interrupt context.
- * device_control
- - Used by client drivers to control and configure the channel it
- has a handle on.
- - Called with a command and an argument
- + The command is one of the values listed by the enum
- dma_ctrl_cmd. The valid commands are:
- + DMA_PAUSE
- + Pauses a transfer on the channel
- + This command should operate synchronously on the channel,
- pausing right away the work of the given channel
- + DMA_RESUME
- + Restarts a transfer on the channel
- + This command should operate synchronously on the channel,
- resuming right away the work of the given channel
- + DMA_TERMINATE_ALL
- + Aborts all the pending and ongoing transfers on the
- channel
- + This command should operate synchronously on the channel,
- terminating right away all the channels
- + DMA_SLAVE_CONFIG
- + Reconfigures the channel with passed configuration
- + This command should NOT perform synchronously, or on any
- currently queued transfers, but only on subsequent ones
- + In this case, the function will receive a
- dma_slave_config structure pointer as an argument, that
- will detail which configuration to use.
- + Even though that structure contains a direction field,
- this field is deprecated in favor of the direction
- argument given to the prep_* functions
- + FSLDMA_EXTERNAL_START
- + TODO: Why does that even exist?
- + The argument is an opaque unsigned long. This actually is a
- pointer to a struct dma_slave_config that should be used only
- in the DMA_SLAVE_CONFIG.
-
- * device_slave_caps
- - Called through the framework by client drivers in order to have
- an idea of what are the properties of the channel allocated to
- them.
- - Such properties are the buswidth, available directions, etc.
- - Required for every generic layer doing DMA transfers, such as
- ASoC.
+ * device_config
+ - Reconfigures the channel with the configuration given as
+ argument
+ - This command should NOT perform synchronously, or on any
+ currently queued transfers, but only on subsequent ones
+ - In this case, the function will receive a dma_slave_config
+ structure pointer as an argument, that will detail which
+ configuration to use.
+ - Even though that structure contains a direction field, this
+ field is deprecated in favor of the direction argument given to
+ the prep_* functions
+
+ * device_pause
+ - Pauses a transfer on the channel
+ - This command should operate synchronously on the channel,
+ pausing right away the work of the given channel
+
+ * device_resume
+ - Resumes a transfer on the channel
+ - This command should operate synchronously on the channel,
+ pausing right away the work of the given channel
+
+ * device_terminate_all
+ - Aborts all the pending and ongoing transfers on the channel
+ - This command should operate synchronously on the channel,
+ terminating right away all the channels
Misc notes (stuff that should be documented, but don't really know
where to put them)
--
2.1.1
^ permalink raw reply related [flat|nested] 164+ messages in thread
* Re: [PATCH v5 34/61] dmaengine: omap: Split device_control
2014-11-17 13:42 ` Maxime Ripard
@ 2014-12-05 14:27 ` Vinod Koul
-1 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-05 14:27 UTC (permalink / raw)
To: Maxime Ripard
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
On Mon, Nov 17, 2014 at 02:42:28PM +0100, Maxime Ripard wrote:
> Split the device_control callback of the TI OMAP DMA driver to make use
> of the newly introduced callbacks, that will eventually be used to retrieve
> slave capabilities.
>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> ---
> drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
> 1 file changed, 15 insertions(+), 36 deletions(-)
>
> diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
> index a8aa5b6bb5e2..1cadf812f3b3 100644
> --- a/drivers/dma/omap-dma.c
> +++ b/drivers/dma/omap-dma.c
> @@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
> return vchan_tx_prep(&c->vc, &d->vd, flags);
> }
>
> -static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
> +static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
> {
> + struct omap_chan *c = to_omap_dma_chan(chan);
> +
> if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
> cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
> return -EINVAL;
> @@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
> return 0;
> }
>
> -static int omap_dma_terminate_all(struct omap_chan *c)
> +static int omap_dma_terminate_all(struct dma_chan *chan)
> {
> + struct omap_chan *c = to_omap_dma_chan(chan);
> struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
> unsigned long flags;
> LIST_HEAD(head);
> @@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
> return 0;
> }
>
> -static int omap_dma_pause(struct omap_chan *c)
> +static int omap_dma_pause(struct dma_chan *chan)
> {
> + struct omap_chan *c = to_omap_dma_chan(chan);
> +
> /* Pause/Resume only allowed with cyclic mode */
> if (!c->cyclic)
> return -EINVAL;
> @@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
> return 0;
> }
>
> -static int omap_dma_resume(struct omap_chan *c)
> +static int omap_dma_resume(struct dma_chan *chan)
> {
> + struct omap_chan *c = to_omap_dma_chan(chan);
> +
> /* Pause/Resume only allowed with cyclic mode */
> if (!c->cyclic)
> return -EINVAL;
> @@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
> return 0;
> }
>
> -static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> - unsigned long arg)
> -{
> - struct omap_chan *c = to_omap_dma_chan(chan);
> - int ret;
> -
> - switch (cmd) {
> - case DMA_SLAVE_CONFIG:
> - ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
> - break;
> -
> - case DMA_TERMINATE_ALL:
> - ret = omap_dma_terminate_all(c);
> - break;
> -
> - case DMA_PAUSE:
> - ret = omap_dma_pause(c);
> - break;
> -
> - case DMA_RESUME:
> - ret = omap_dma_resume(c);
> - break;
> -
> - default:
> - ret = -ENXIO;
> - break;
> - }
> -
> - return ret;
> -}
> -
> static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
> {
> struct omap_chan *c;
> @@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
> od->ddev.device_issue_pending = omap_dma_issue_pending;
> od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
> od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
> - od->ddev.device_control = omap_dma_control;
this should be assigned to omap_dma_slave_config. omap_dma_control is not
here :(
--
~Vinod
> + od->ddev.device_config = omap_dma_config;
> + od->ddev.device_pause = omap_dma_pause;
> + od->ddev.device_resume = omap_dma_resume;
> + od->ddev.device_terminate_all = omap_dma_terminate_all;
> od->ddev.device_slave_caps = omap_dma_device_slave_caps;
> od->ddev.dev = &pdev->dev;
> INIT_LIST_HEAD(&od->ddev.channels);
> --
> 2.1.1
>
--
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 34/61] dmaengine: omap: Split device_control
@ 2014-12-05 14:27 ` Vinod Koul
0 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-05 14:27 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Nov 17, 2014 at 02:42:28PM +0100, Maxime Ripard wrote:
> Split the device_control callback of the TI OMAP DMA driver to make use
> of the newly introduced callbacks, that will eventually be used to retrieve
> slave capabilities.
>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> ---
> drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
> 1 file changed, 15 insertions(+), 36 deletions(-)
>
> diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
> index a8aa5b6bb5e2..1cadf812f3b3 100644
> --- a/drivers/dma/omap-dma.c
> +++ b/drivers/dma/omap-dma.c
> @@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
> return vchan_tx_prep(&c->vc, &d->vd, flags);
> }
>
> -static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
> +static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
> {
> + struct omap_chan *c = to_omap_dma_chan(chan);
> +
> if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
> cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
> return -EINVAL;
> @@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
> return 0;
> }
>
> -static int omap_dma_terminate_all(struct omap_chan *c)
> +static int omap_dma_terminate_all(struct dma_chan *chan)
> {
> + struct omap_chan *c = to_omap_dma_chan(chan);
> struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
> unsigned long flags;
> LIST_HEAD(head);
> @@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
> return 0;
> }
>
> -static int omap_dma_pause(struct omap_chan *c)
> +static int omap_dma_pause(struct dma_chan *chan)
> {
> + struct omap_chan *c = to_omap_dma_chan(chan);
> +
> /* Pause/Resume only allowed with cyclic mode */
> if (!c->cyclic)
> return -EINVAL;
> @@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
> return 0;
> }
>
> -static int omap_dma_resume(struct omap_chan *c)
> +static int omap_dma_resume(struct dma_chan *chan)
> {
> + struct omap_chan *c = to_omap_dma_chan(chan);
> +
> /* Pause/Resume only allowed with cyclic mode */
> if (!c->cyclic)
> return -EINVAL;
> @@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
> return 0;
> }
>
> -static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> - unsigned long arg)
> -{
> - struct omap_chan *c = to_omap_dma_chan(chan);
> - int ret;
> -
> - switch (cmd) {
> - case DMA_SLAVE_CONFIG:
> - ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
> - break;
> -
> - case DMA_TERMINATE_ALL:
> - ret = omap_dma_terminate_all(c);
> - break;
> -
> - case DMA_PAUSE:
> - ret = omap_dma_pause(c);
> - break;
> -
> - case DMA_RESUME:
> - ret = omap_dma_resume(c);
> - break;
> -
> - default:
> - ret = -ENXIO;
> - break;
> - }
> -
> - return ret;
> -}
> -
> static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
> {
> struct omap_chan *c;
> @@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
> od->ddev.device_issue_pending = omap_dma_issue_pending;
> od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
> od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
> - od->ddev.device_control = omap_dma_control;
this should be assigned to omap_dma_slave_config. omap_dma_control is not
here :(
--
~Vinod
> + od->ddev.device_config = omap_dma_config;
> + od->ddev.device_pause = omap_dma_pause;
> + od->ddev.device_resume = omap_dma_resume;
> + od->ddev.device_terminate_all = omap_dma_terminate_all;
> od->ddev.device_slave_caps = omap_dma_device_slave_caps;
> od->ddev.dev = &pdev->dev;
> INIT_LIST_HEAD(&od->ddev.channels);
> --
> 2.1.1
>
--
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 32/61] dmaengine: mxs: Split device_control
2014-11-17 13:42 ` Maxime Ripard
@ 2014-12-05 14:29 ` Vinod Koul
-1 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-05 14:29 UTC (permalink / raw)
To: Maxime Ripard
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
On Mon, Nov 17, 2014 at 02:42:26PM +0100, Maxime Ripard wrote:
> Split the device_control callback of the Freescale MXS DMA driver to make use
> of the newly introduced callbacks, that will eventually be used to retrieve
> slave capabilities.
>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> ---
> drivers/dma/mxs-dma.c | 59 ++++++++++++++++++++-------------------------------
> 1 file changed, 23 insertions(+), 36 deletions(-)
>
> diff --git a/drivers/dma/mxs-dma.c b/drivers/dma/mxs-dma.c
> index 5ea61201dbf0..834041e5a769 100644
> --- a/drivers/dma/mxs-dma.c
> +++ b/drivers/dma/mxs-dma.c
> @@ -202,8 +202,9 @@ static struct mxs_dma_chan *to_mxs_dma_chan(struct dma_chan *chan)
> return container_of(chan, struct mxs_dma_chan, chan);
> }
>
> -static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_reset_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
> int chan_id = mxs_chan->chan.chan_id;
>
> @@ -250,8 +251,9 @@ static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
> mxs_chan->status = DMA_COMPLETE;
> }
>
> -static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_enable_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
> int chan_id = mxs_chan->chan.chan_id;
>
> @@ -272,13 +274,16 @@ static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
> mxs_chan->reset = false;
> }
>
> -static void mxs_dma_disable_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_disable_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> +
> mxs_chan->status = DMA_COMPLETE;
> }
>
> -static void mxs_dma_pause_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_pause_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
> int chan_id = mxs_chan->chan.chan_id;
>
> @@ -293,8 +298,9 @@ static void mxs_dma_pause_chan(struct mxs_dma_chan *mxs_chan)
> mxs_chan->status = DMA_PAUSED;
> }
>
> -static void mxs_dma_resume_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_resume_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
> int chan_id = mxs_chan->chan.chan_id;
>
> @@ -383,7 +389,7 @@ static irqreturn_t mxs_dma_int_handler(int irq, void *dev_id)
> "%s: error in channel %d\n", __func__,
> chan);
> mxs_chan->status = DMA_ERROR;
> - mxs_dma_reset_chan(mxs_chan);
> + mxs_dma_reset_chan(mxs_chan->chan);
this would fail. chan is struct type not a pointer
--
~Vinod
> } else if (mxs_chan->status != DMA_COMPLETE) {
> if (mxs_chan->flags & MXS_DMA_SG_LOOP) {
> mxs_chan->status = DMA_IN_PROGRESS;
> @@ -432,7 +438,7 @@ static int mxs_dma_alloc_chan_resources(struct dma_chan *chan)
> if (ret)
> goto err_clk;
>
> - mxs_dma_reset_chan(mxs_chan);
> + mxs_dma_reset_chan(chan);
>
> dma_async_tx_descriptor_init(&mxs_chan->desc, chan);
> mxs_chan->desc.tx_submit = mxs_dma_tx_submit;
> @@ -456,7 +462,7 @@ static void mxs_dma_free_chan_resources(struct dma_chan *chan)
> struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
>
> - mxs_dma_disable_chan(mxs_chan);
> + mxs_dma_disable_chan(chan);
>
> free_irq(mxs_chan->chan_irq, mxs_dma);
>
> @@ -651,28 +657,14 @@ err_out:
> return NULL;
> }
>
> -static int mxs_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> - unsigned long arg)
> +static int mxs_dma_terminate_all(struct dma_chan *chan)
> {
> struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> - int ret = 0;
> -
> - switch (cmd) {
> - case DMA_TERMINATE_ALL:
> - mxs_dma_reset_chan(mxs_chan);
> - mxs_dma_disable_chan(mxs_chan);
> - break;
> - case DMA_PAUSE:
> - mxs_dma_pause_chan(mxs_chan);
> - break;
> - case DMA_RESUME:
> - mxs_dma_resume_chan(mxs_chan);
> - break;
> - default:
> - ret = -ENOSYS;
> - }
>
> - return ret;
> + mxs_dma_reset_chan(chan);
> + mxs_dma_disable_chan(chan);
> +
> + return 0;
> }
>
> static enum dma_status mxs_dma_tx_status(struct dma_chan *chan,
> @@ -701,13 +693,6 @@ static enum dma_status mxs_dma_tx_status(struct dma_chan *chan,
> return mxs_chan->status;
> }
>
> -static void mxs_dma_issue_pending(struct dma_chan *chan)
> -{
> - struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> -
> - mxs_dma_enable_chan(mxs_chan);
> -}
> -
> static int __init mxs_dma_init(struct mxs_dma_engine *mxs_dma)
> {
> int ret;
> @@ -860,8 +845,10 @@ static int __init mxs_dma_probe(struct platform_device *pdev)
> mxs_dma->dma_device.device_tx_status = mxs_dma_tx_status;
> mxs_dma->dma_device.device_prep_slave_sg = mxs_dma_prep_slave_sg;
> mxs_dma->dma_device.device_prep_dma_cyclic = mxs_dma_prep_dma_cyclic;
> - mxs_dma->dma_device.device_control = mxs_dma_control;
> - mxs_dma->dma_device.device_issue_pending = mxs_dma_issue_pending;
> + mxs_dma->dma_device.device_pause = mxs_dma_pause_chan;
> + mxs_dma->dma_device.device_resume = mxs_dma_resume_chan;
> + mxs_dma->dma_device.device_terminate_all = mxs_dma_terminate_all;
> + mxs_dma->dma_device.device_issue_pending = mxs_dma_enable_chan;
>
> ret = dma_async_device_register(&mxs_dma->dma_device);
> if (ret) {
> --
> 2.1.1
>
--
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 32/61] dmaengine: mxs: Split device_control
@ 2014-12-05 14:29 ` Vinod Koul
0 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-05 14:29 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Nov 17, 2014 at 02:42:26PM +0100, Maxime Ripard wrote:
> Split the device_control callback of the Freescale MXS DMA driver to make use
> of the newly introduced callbacks, that will eventually be used to retrieve
> slave capabilities.
>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> ---
> drivers/dma/mxs-dma.c | 59 ++++++++++++++++++++-------------------------------
> 1 file changed, 23 insertions(+), 36 deletions(-)
>
> diff --git a/drivers/dma/mxs-dma.c b/drivers/dma/mxs-dma.c
> index 5ea61201dbf0..834041e5a769 100644
> --- a/drivers/dma/mxs-dma.c
> +++ b/drivers/dma/mxs-dma.c
> @@ -202,8 +202,9 @@ static struct mxs_dma_chan *to_mxs_dma_chan(struct dma_chan *chan)
> return container_of(chan, struct mxs_dma_chan, chan);
> }
>
> -static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_reset_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
> int chan_id = mxs_chan->chan.chan_id;
>
> @@ -250,8 +251,9 @@ static void mxs_dma_reset_chan(struct mxs_dma_chan *mxs_chan)
> mxs_chan->status = DMA_COMPLETE;
> }
>
> -static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_enable_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
> int chan_id = mxs_chan->chan.chan_id;
>
> @@ -272,13 +274,16 @@ static void mxs_dma_enable_chan(struct mxs_dma_chan *mxs_chan)
> mxs_chan->reset = false;
> }
>
> -static void mxs_dma_disable_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_disable_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> +
> mxs_chan->status = DMA_COMPLETE;
> }
>
> -static void mxs_dma_pause_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_pause_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
> int chan_id = mxs_chan->chan.chan_id;
>
> @@ -293,8 +298,9 @@ static void mxs_dma_pause_chan(struct mxs_dma_chan *mxs_chan)
> mxs_chan->status = DMA_PAUSED;
> }
>
> -static void mxs_dma_resume_chan(struct mxs_dma_chan *mxs_chan)
> +static void mxs_dma_resume_chan(struct dma_chan *chan)
> {
> + struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
> int chan_id = mxs_chan->chan.chan_id;
>
> @@ -383,7 +389,7 @@ static irqreturn_t mxs_dma_int_handler(int irq, void *dev_id)
> "%s: error in channel %d\n", __func__,
> chan);
> mxs_chan->status = DMA_ERROR;
> - mxs_dma_reset_chan(mxs_chan);
> + mxs_dma_reset_chan(mxs_chan->chan);
this would fail. chan is struct type not a pointer
--
~Vinod
> } else if (mxs_chan->status != DMA_COMPLETE) {
> if (mxs_chan->flags & MXS_DMA_SG_LOOP) {
> mxs_chan->status = DMA_IN_PROGRESS;
> @@ -432,7 +438,7 @@ static int mxs_dma_alloc_chan_resources(struct dma_chan *chan)
> if (ret)
> goto err_clk;
>
> - mxs_dma_reset_chan(mxs_chan);
> + mxs_dma_reset_chan(chan);
>
> dma_async_tx_descriptor_init(&mxs_chan->desc, chan);
> mxs_chan->desc.tx_submit = mxs_dma_tx_submit;
> @@ -456,7 +462,7 @@ static void mxs_dma_free_chan_resources(struct dma_chan *chan)
> struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> struct mxs_dma_engine *mxs_dma = mxs_chan->mxs_dma;
>
> - mxs_dma_disable_chan(mxs_chan);
> + mxs_dma_disable_chan(chan);
>
> free_irq(mxs_chan->chan_irq, mxs_dma);
>
> @@ -651,28 +657,14 @@ err_out:
> return NULL;
> }
>
> -static int mxs_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> - unsigned long arg)
> +static int mxs_dma_terminate_all(struct dma_chan *chan)
> {
> struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> - int ret = 0;
> -
> - switch (cmd) {
> - case DMA_TERMINATE_ALL:
> - mxs_dma_reset_chan(mxs_chan);
> - mxs_dma_disable_chan(mxs_chan);
> - break;
> - case DMA_PAUSE:
> - mxs_dma_pause_chan(mxs_chan);
> - break;
> - case DMA_RESUME:
> - mxs_dma_resume_chan(mxs_chan);
> - break;
> - default:
> - ret = -ENOSYS;
> - }
>
> - return ret;
> + mxs_dma_reset_chan(chan);
> + mxs_dma_disable_chan(chan);
> +
> + return 0;
> }
>
> static enum dma_status mxs_dma_tx_status(struct dma_chan *chan,
> @@ -701,13 +693,6 @@ static enum dma_status mxs_dma_tx_status(struct dma_chan *chan,
> return mxs_chan->status;
> }
>
> -static void mxs_dma_issue_pending(struct dma_chan *chan)
> -{
> - struct mxs_dma_chan *mxs_chan = to_mxs_dma_chan(chan);
> -
> - mxs_dma_enable_chan(mxs_chan);
> -}
> -
> static int __init mxs_dma_init(struct mxs_dma_engine *mxs_dma)
> {
> int ret;
> @@ -860,8 +845,10 @@ static int __init mxs_dma_probe(struct platform_device *pdev)
> mxs_dma->dma_device.device_tx_status = mxs_dma_tx_status;
> mxs_dma->dma_device.device_prep_slave_sg = mxs_dma_prep_slave_sg;
> mxs_dma->dma_device.device_prep_dma_cyclic = mxs_dma_prep_dma_cyclic;
> - mxs_dma->dma_device.device_control = mxs_dma_control;
> - mxs_dma->dma_device.device_issue_pending = mxs_dma_issue_pending;
> + mxs_dma->dma_device.device_pause = mxs_dma_pause_chan;
> + mxs_dma->dma_device.device_resume = mxs_dma_resume_chan;
> + mxs_dma->dma_device.device_terminate_all = mxs_dma_terminate_all;
> + mxs_dma->dma_device.device_issue_pending = mxs_dma_enable_chan;
>
> ret = dma_async_device_register(&mxs_dma->dma_device);
> if (ret) {
> --
> 2.1.1
>
--
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 34/61] dmaengine: omap: Split device_control
2014-12-05 14:27 ` Vinod Koul
@ 2014-12-05 14:33 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-05 14:33 UTC (permalink / raw)
To: Vinod Koul
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
[-- Attachment #1: Type: text/plain, Size: 4072 bytes --]
Hi Vinod,
On Fri, Dec 05, 2014 at 07:57:02PM +0530, Vinod Koul wrote:
> On Mon, Nov 17, 2014 at 02:42:28PM +0100, Maxime Ripard wrote:
> > Split the device_control callback of the TI OMAP DMA driver to make use
> > of the newly introduced callbacks, that will eventually be used to retrieve
> > slave capabilities.
> >
> > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> > ---
> > drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
> > 1 file changed, 15 insertions(+), 36 deletions(-)
> >
> > diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
> > index a8aa5b6bb5e2..1cadf812f3b3 100644
> > --- a/drivers/dma/omap-dma.c
> > +++ b/drivers/dma/omap-dma.c
> > @@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
> > return vchan_tx_prep(&c->vc, &d->vd, flags);
> > }
> >
> > -static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
> > +static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
> > {
> > + struct omap_chan *c = to_omap_dma_chan(chan);
> > +
> > if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
> > cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
> > return -EINVAL;
> > @@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
> > return 0;
> > }
> >
> > -static int omap_dma_terminate_all(struct omap_chan *c)
> > +static int omap_dma_terminate_all(struct dma_chan *chan)
> > {
> > + struct omap_chan *c = to_omap_dma_chan(chan);
> > struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
> > unsigned long flags;
> > LIST_HEAD(head);
> > @@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
> > return 0;
> > }
> >
> > -static int omap_dma_pause(struct omap_chan *c)
> > +static int omap_dma_pause(struct dma_chan *chan)
> > {
> > + struct omap_chan *c = to_omap_dma_chan(chan);
> > +
> > /* Pause/Resume only allowed with cyclic mode */
> > if (!c->cyclic)
> > return -EINVAL;
> > @@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
> > return 0;
> > }
> >
> > -static int omap_dma_resume(struct omap_chan *c)
> > +static int omap_dma_resume(struct dma_chan *chan)
> > {
> > + struct omap_chan *c = to_omap_dma_chan(chan);
> > +
> > /* Pause/Resume only allowed with cyclic mode */
> > if (!c->cyclic)
> > return -EINVAL;
> > @@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
> > return 0;
> > }
> >
> > -static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> > - unsigned long arg)
> > -{
> > - struct omap_chan *c = to_omap_dma_chan(chan);
> > - int ret;
> > -
> > - switch (cmd) {
> > - case DMA_SLAVE_CONFIG:
> > - ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
> > - break;
> > -
> > - case DMA_TERMINATE_ALL:
> > - ret = omap_dma_terminate_all(c);
> > - break;
> > -
> > - case DMA_PAUSE:
> > - ret = omap_dma_pause(c);
> > - break;
> > -
> > - case DMA_RESUME:
> > - ret = omap_dma_resume(c);
> > - break;
> > -
> > - default:
> > - ret = -ENXIO;
> > - break;
> > - }
> > -
> > - return ret;
> > -}
> > -
> > static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
> > {
> > struct omap_chan *c;
> > @@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
> > od->ddev.device_issue_pending = omap_dma_issue_pending;
> > od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
> > od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
> > - od->ddev.device_control = omap_dma_control;
> this should be assigned to omap_dma_slave_config. omap_dma_control is not
> here :(
I'm not sure what you mean here. I'm removing this line, together with
the whole omap_dma_control function.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 34/61] dmaengine: omap: Split device_control
@ 2014-12-05 14:33 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-05 14:33 UTC (permalink / raw)
To: linux-arm-kernel
Hi Vinod,
On Fri, Dec 05, 2014 at 07:57:02PM +0530, Vinod Koul wrote:
> On Mon, Nov 17, 2014 at 02:42:28PM +0100, Maxime Ripard wrote:
> > Split the device_control callback of the TI OMAP DMA driver to make use
> > of the newly introduced callbacks, that will eventually be used to retrieve
> > slave capabilities.
> >
> > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> > ---
> > drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
> > 1 file changed, 15 insertions(+), 36 deletions(-)
> >
> > diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
> > index a8aa5b6bb5e2..1cadf812f3b3 100644
> > --- a/drivers/dma/omap-dma.c
> > +++ b/drivers/dma/omap-dma.c
> > @@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
> > return vchan_tx_prep(&c->vc, &d->vd, flags);
> > }
> >
> > -static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
> > +static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
> > {
> > + struct omap_chan *c = to_omap_dma_chan(chan);
> > +
> > if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
> > cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
> > return -EINVAL;
> > @@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
> > return 0;
> > }
> >
> > -static int omap_dma_terminate_all(struct omap_chan *c)
> > +static int omap_dma_terminate_all(struct dma_chan *chan)
> > {
> > + struct omap_chan *c = to_omap_dma_chan(chan);
> > struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
> > unsigned long flags;
> > LIST_HEAD(head);
> > @@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
> > return 0;
> > }
> >
> > -static int omap_dma_pause(struct omap_chan *c)
> > +static int omap_dma_pause(struct dma_chan *chan)
> > {
> > + struct omap_chan *c = to_omap_dma_chan(chan);
> > +
> > /* Pause/Resume only allowed with cyclic mode */
> > if (!c->cyclic)
> > return -EINVAL;
> > @@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
> > return 0;
> > }
> >
> > -static int omap_dma_resume(struct omap_chan *c)
> > +static int omap_dma_resume(struct dma_chan *chan)
> > {
> > + struct omap_chan *c = to_omap_dma_chan(chan);
> > +
> > /* Pause/Resume only allowed with cyclic mode */
> > if (!c->cyclic)
> > return -EINVAL;
> > @@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
> > return 0;
> > }
> >
> > -static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> > - unsigned long arg)
> > -{
> > - struct omap_chan *c = to_omap_dma_chan(chan);
> > - int ret;
> > -
> > - switch (cmd) {
> > - case DMA_SLAVE_CONFIG:
> > - ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
> > - break;
> > -
> > - case DMA_TERMINATE_ALL:
> > - ret = omap_dma_terminate_all(c);
> > - break;
> > -
> > - case DMA_PAUSE:
> > - ret = omap_dma_pause(c);
> > - break;
> > -
> > - case DMA_RESUME:
> > - ret = omap_dma_resume(c);
> > - break;
> > -
> > - default:
> > - ret = -ENXIO;
> > - break;
> > - }
> > -
> > - return ret;
> > -}
> > -
> > static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
> > {
> > struct omap_chan *c;
> > @@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
> > od->ddev.device_issue_pending = omap_dma_issue_pending;
> > od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
> > od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
> > - od->ddev.device_control = omap_dma_control;
> this should be assigned to omap_dma_slave_config. omap_dma_control is not
> here :(
I'm not sure what you mean here. I'm removing this line, together with
the whole omap_dma_control function.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141205/790319c0/attachment.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 34/61] dmaengine: omap: Split device_control
2014-12-05 14:33 ` Maxime Ripard
@ 2014-12-05 14:56 ` Vinod Koul
-1 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-05 14:56 UTC (permalink / raw)
To: Maxime Ripard
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
[-- Attachment #1: Type: text/plain, Size: 4866 bytes --]
On Fri, Dec 05, 2014 at 03:33:59PM +0100, Maxime Ripard wrote:
> Hi Vinod,
>
> On Fri, Dec 05, 2014 at 07:57:02PM +0530, Vinod Koul wrote:
> > On Mon, Nov 17, 2014 at 02:42:28PM +0100, Maxime Ripard wrote:
> > > Split the device_control callback of the TI OMAP DMA driver to make use
> > > of the newly introduced callbacks, that will eventually be used to retrieve
> > > slave capabilities.
> > >
> > > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> > > ---
> > > drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
> > > 1 file changed, 15 insertions(+), 36 deletions(-)
> > >
> > > diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
> > > index a8aa5b6bb5e2..1cadf812f3b3 100644
> > > --- a/drivers/dma/omap-dma.c
> > > +++ b/drivers/dma/omap-dma.c
> > > @@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
> > > return vchan_tx_prep(&c->vc, &d->vd, flags);
> > > }
> > >
> > > -static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
> > > +static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
> > > {
> > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > +
> > > if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
> > > cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
> > > return -EINVAL;
> > > @@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
> > > return 0;
> > > }
> > >
> > > -static int omap_dma_terminate_all(struct omap_chan *c)
> > > +static int omap_dma_terminate_all(struct dma_chan *chan)
> > > {
> > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
> > > unsigned long flags;
> > > LIST_HEAD(head);
> > > @@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
> > > return 0;
> > > }
> > >
> > > -static int omap_dma_pause(struct omap_chan *c)
> > > +static int omap_dma_pause(struct dma_chan *chan)
> > > {
> > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > +
> > > /* Pause/Resume only allowed with cyclic mode */
> > > if (!c->cyclic)
> > > return -EINVAL;
> > > @@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
> > > return 0;
> > > }
> > >
> > > -static int omap_dma_resume(struct omap_chan *c)
> > > +static int omap_dma_resume(struct dma_chan *chan)
> > > {
> > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > +
> > > /* Pause/Resume only allowed with cyclic mode */
> > > if (!c->cyclic)
> > > return -EINVAL;
> > > @@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
> > > return 0;
> > > }
> > >
> > > -static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> > > - unsigned long arg)
> > > -{
> > > - struct omap_chan *c = to_omap_dma_chan(chan);
> > > - int ret;
> > > -
> > > - switch (cmd) {
> > > - case DMA_SLAVE_CONFIG:
> > > - ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
> > > - break;
> > > -
> > > - case DMA_TERMINATE_ALL:
> > > - ret = omap_dma_terminate_all(c);
> > > - break;
> > > -
> > > - case DMA_PAUSE:
> > > - ret = omap_dma_pause(c);
> > > - break;
> > > -
> > > - case DMA_RESUME:
> > > - ret = omap_dma_resume(c);
> > > - break;
> > > -
> > > - default:
> > > - ret = -ENXIO;
> > > - break;
> > > - }
> > > -
> > > - return ret;
> > > -}
> > > -
> > > static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
> > > {
> > > struct omap_chan *c;
> > > @@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
> > > od->ddev.device_issue_pending = omap_dma_issue_pending;
> > > od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
> > > od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
> > > - od->ddev.device_control = omap_dma_control;
> > this should be assigned to omap_dma_slave_config. omap_dma_control is not
> > here :(
>
> I'm not sure what you mean here. I'm removing this line, together with
> the whole omap_dma_control function.
Sorry wrong quote and not good explanations. the lines i quoted is
wrong one...
- od->ddev.device_control = omap_dma_control;
+ od->ddev.device_config = omap_dma_config;
this is troubling line here...
we dont have omap_dma_config. It should be omap_dma_slave_config instead
--
~Vinod
+ od->ddev.device_pause = omap_dma_pause;
+ od->ddev.device_resume = omap_dma_resume;
+ od->ddev.device_terminate_all = omap_dma_terminate_all;
>
> Maxime
>
> --
> Maxime Ripard, Free Electrons
> Embedded Linux, Kernel and Android engineering
> http://free-electrons.com
--
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 34/61] dmaengine: omap: Split device_control
@ 2014-12-05 14:56 ` Vinod Koul
0 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-05 14:56 UTC (permalink / raw)
To: linux-arm-kernel
On Fri, Dec 05, 2014 at 03:33:59PM +0100, Maxime Ripard wrote:
> Hi Vinod,
>
> On Fri, Dec 05, 2014 at 07:57:02PM +0530, Vinod Koul wrote:
> > On Mon, Nov 17, 2014 at 02:42:28PM +0100, Maxime Ripard wrote:
> > > Split the device_control callback of the TI OMAP DMA driver to make use
> > > of the newly introduced callbacks, that will eventually be used to retrieve
> > > slave capabilities.
> > >
> > > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> > > ---
> > > drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
> > > 1 file changed, 15 insertions(+), 36 deletions(-)
> > >
> > > diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
> > > index a8aa5b6bb5e2..1cadf812f3b3 100644
> > > --- a/drivers/dma/omap-dma.c
> > > +++ b/drivers/dma/omap-dma.c
> > > @@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
> > > return vchan_tx_prep(&c->vc, &d->vd, flags);
> > > }
> > >
> > > -static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
> > > +static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
> > > {
> > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > +
> > > if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
> > > cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
> > > return -EINVAL;
> > > @@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
> > > return 0;
> > > }
> > >
> > > -static int omap_dma_terminate_all(struct omap_chan *c)
> > > +static int omap_dma_terminate_all(struct dma_chan *chan)
> > > {
> > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
> > > unsigned long flags;
> > > LIST_HEAD(head);
> > > @@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
> > > return 0;
> > > }
> > >
> > > -static int omap_dma_pause(struct omap_chan *c)
> > > +static int omap_dma_pause(struct dma_chan *chan)
> > > {
> > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > +
> > > /* Pause/Resume only allowed with cyclic mode */
> > > if (!c->cyclic)
> > > return -EINVAL;
> > > @@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
> > > return 0;
> > > }
> > >
> > > -static int omap_dma_resume(struct omap_chan *c)
> > > +static int omap_dma_resume(struct dma_chan *chan)
> > > {
> > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > +
> > > /* Pause/Resume only allowed with cyclic mode */
> > > if (!c->cyclic)
> > > return -EINVAL;
> > > @@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
> > > return 0;
> > > }
> > >
> > > -static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> > > - unsigned long arg)
> > > -{
> > > - struct omap_chan *c = to_omap_dma_chan(chan);
> > > - int ret;
> > > -
> > > - switch (cmd) {
> > > - case DMA_SLAVE_CONFIG:
> > > - ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
> > > - break;
> > > -
> > > - case DMA_TERMINATE_ALL:
> > > - ret = omap_dma_terminate_all(c);
> > > - break;
> > > -
> > > - case DMA_PAUSE:
> > > - ret = omap_dma_pause(c);
> > > - break;
> > > -
> > > - case DMA_RESUME:
> > > - ret = omap_dma_resume(c);
> > > - break;
> > > -
> > > - default:
> > > - ret = -ENXIO;
> > > - break;
> > > - }
> > > -
> > > - return ret;
> > > -}
> > > -
> > > static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
> > > {
> > > struct omap_chan *c;
> > > @@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
> > > od->ddev.device_issue_pending = omap_dma_issue_pending;
> > > od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
> > > od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
> > > - od->ddev.device_control = omap_dma_control;
> > this should be assigned to omap_dma_slave_config. omap_dma_control is not
> > here :(
>
> I'm not sure what you mean here. I'm removing this line, together with
> the whole omap_dma_control function.
Sorry wrong quote and not good explanations. the lines i quoted is
wrong one...
- od->ddev.device_control = omap_dma_control;
+ od->ddev.device_config = omap_dma_config;
this is troubling line here...
we dont have omap_dma_config. It should be omap_dma_slave_config instead
--
~Vinod
+ od->ddev.device_pause = omap_dma_pause;
+ od->ddev.device_resume = omap_dma_resume;
+ od->ddev.device_terminate_all = omap_dma_terminate_all;
>
> Maxime
>
> --
> Maxime Ripard, Free Electrons
> Embedded Linux, Kernel and Android engineering
> http://free-electrons.com
--
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141205/70f9c321/attachment-0001.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 34/61] dmaengine: omap: Split device_control
2014-12-05 14:56 ` Vinod Koul
@ 2014-12-05 17:07 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-05 17:07 UTC (permalink / raw)
To: Vinod Koul
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
[-- Attachment #1: Type: text/plain, Size: 5196 bytes --]
On Fri, Dec 05, 2014 at 08:26:42PM +0530, Vinod Koul wrote:
> On Fri, Dec 05, 2014 at 03:33:59PM +0100, Maxime Ripard wrote:
> > Hi Vinod,
> >
> > On Fri, Dec 05, 2014 at 07:57:02PM +0530, Vinod Koul wrote:
> > > On Mon, Nov 17, 2014 at 02:42:28PM +0100, Maxime Ripard wrote:
> > > > Split the device_control callback of the TI OMAP DMA driver to make use
> > > > of the newly introduced callbacks, that will eventually be used to retrieve
> > > > slave capabilities.
> > > >
> > > > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> > > > ---
> > > > drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
> > > > 1 file changed, 15 insertions(+), 36 deletions(-)
> > > >
> > > > diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
> > > > index a8aa5b6bb5e2..1cadf812f3b3 100644
> > > > --- a/drivers/dma/omap-dma.c
> > > > +++ b/drivers/dma/omap-dma.c
> > > > @@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
> > > > return vchan_tx_prep(&c->vc, &d->vd, flags);
> > > > }
> > > >
> > > > -static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
> > > > +static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
> > > > {
> > > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > > +
> > > > if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
> > > > cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
> > > > return -EINVAL;
> > > > @@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
> > > > return 0;
> > > > }
> > > >
> > > > -static int omap_dma_terminate_all(struct omap_chan *c)
> > > > +static int omap_dma_terminate_all(struct dma_chan *chan)
> > > > {
> > > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > > struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
> > > > unsigned long flags;
> > > > LIST_HEAD(head);
> > > > @@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
> > > > return 0;
> > > > }
> > > >
> > > > -static int omap_dma_pause(struct omap_chan *c)
> > > > +static int omap_dma_pause(struct dma_chan *chan)
> > > > {
> > > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > > +
> > > > /* Pause/Resume only allowed with cyclic mode */
> > > > if (!c->cyclic)
> > > > return -EINVAL;
> > > > @@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
> > > > return 0;
> > > > }
> > > >
> > > > -static int omap_dma_resume(struct omap_chan *c)
> > > > +static int omap_dma_resume(struct dma_chan *chan)
> > > > {
> > > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > > +
> > > > /* Pause/Resume only allowed with cyclic mode */
> > > > if (!c->cyclic)
> > > > return -EINVAL;
> > > > @@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
> > > > return 0;
> > > > }
> > > >
> > > > -static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> > > > - unsigned long arg)
> > > > -{
> > > > - struct omap_chan *c = to_omap_dma_chan(chan);
> > > > - int ret;
> > > > -
> > > > - switch (cmd) {
> > > > - case DMA_SLAVE_CONFIG:
> > > > - ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
> > > > - break;
> > > > -
> > > > - case DMA_TERMINATE_ALL:
> > > > - ret = omap_dma_terminate_all(c);
> > > > - break;
> > > > -
> > > > - case DMA_PAUSE:
> > > > - ret = omap_dma_pause(c);
> > > > - break;
> > > > -
> > > > - case DMA_RESUME:
> > > > - ret = omap_dma_resume(c);
> > > > - break;
> > > > -
> > > > - default:
> > > > - ret = -ENXIO;
> > > > - break;
> > > > - }
> > > > -
> > > > - return ret;
> > > > -}
> > > > -
> > > > static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
> > > > {
> > > > struct omap_chan *c;
> > > > @@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
> > > > od->ddev.device_issue_pending = omap_dma_issue_pending;
> > > > od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
> > > > od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
> > > > - od->ddev.device_control = omap_dma_control;
> > > this should be assigned to omap_dma_slave_config. omap_dma_control is not
> > > here :(
> >
> > I'm not sure what you mean here. I'm removing this line, together with
> > the whole omap_dma_control function.
> Sorry wrong quote and not good explanations. the lines i quoted is
> wrong one...
>
> - od->ddev.device_control = omap_dma_control;
> + od->ddev.device_config = omap_dma_config;
> this is troubling line here...
>
> we dont have omap_dma_config. It should be omap_dma_slave_config instead
Ah, right.
I have no idea why it haven't been picked up by intel's build bot
thuough.
Anyway. Do you want me to send a new version of these patches for this
trivial fix and the mxs one, or do you want me to send a follow-up?
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 34/61] dmaengine: omap: Split device_control
@ 2014-12-05 17:07 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-05 17:07 UTC (permalink / raw)
To: linux-arm-kernel
On Fri, Dec 05, 2014 at 08:26:42PM +0530, Vinod Koul wrote:
> On Fri, Dec 05, 2014 at 03:33:59PM +0100, Maxime Ripard wrote:
> > Hi Vinod,
> >
> > On Fri, Dec 05, 2014 at 07:57:02PM +0530, Vinod Koul wrote:
> > > On Mon, Nov 17, 2014 at 02:42:28PM +0100, Maxime Ripard wrote:
> > > > Split the device_control callback of the TI OMAP DMA driver to make use
> > > > of the newly introduced callbacks, that will eventually be used to retrieve
> > > > slave capabilities.
> > > >
> > > > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> > > > ---
> > > > drivers/dma/omap-dma.c | 51 +++++++++++++++-----------------------------------
> > > > 1 file changed, 15 insertions(+), 36 deletions(-)
> > > >
> > > > diff --git a/drivers/dma/omap-dma.c b/drivers/dma/omap-dma.c
> > > > index a8aa5b6bb5e2..1cadf812f3b3 100644
> > > > --- a/drivers/dma/omap-dma.c
> > > > +++ b/drivers/dma/omap-dma.c
> > > > @@ -948,8 +948,10 @@ static struct dma_async_tx_descriptor *omap_dma_prep_dma_cyclic(
> > > > return vchan_tx_prep(&c->vc, &d->vd, flags);
> > > > }
> > > >
> > > > -static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *cfg)
> > > > +static int omap_dma_slave_config(struct dma_chan *chan, struct dma_slave_config *cfg)
> > > > {
> > > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > > +
> > > > if (cfg->src_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES ||
> > > > cfg->dst_addr_width == DMA_SLAVE_BUSWIDTH_8_BYTES)
> > > > return -EINVAL;
> > > > @@ -959,8 +961,9 @@ static int omap_dma_slave_config(struct omap_chan *c, struct dma_slave_config *c
> > > > return 0;
> > > > }
> > > >
> > > > -static int omap_dma_terminate_all(struct omap_chan *c)
> > > > +static int omap_dma_terminate_all(struct dma_chan *chan)
> > > > {
> > > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > > struct omap_dmadev *d = to_omap_dma_dev(c->vc.chan.device);
> > > > unsigned long flags;
> > > > LIST_HEAD(head);
> > > > @@ -996,8 +999,10 @@ static int omap_dma_terminate_all(struct omap_chan *c)
> > > > return 0;
> > > > }
> > > >
> > > > -static int omap_dma_pause(struct omap_chan *c)
> > > > +static int omap_dma_pause(struct dma_chan *chan)
> > > > {
> > > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > > +
> > > > /* Pause/Resume only allowed with cyclic mode */
> > > > if (!c->cyclic)
> > > > return -EINVAL;
> > > > @@ -1010,8 +1015,10 @@ static int omap_dma_pause(struct omap_chan *c)
> > > > return 0;
> > > > }
> > > >
> > > > -static int omap_dma_resume(struct omap_chan *c)
> > > > +static int omap_dma_resume(struct dma_chan *chan)
> > > > {
> > > > + struct omap_chan *c = to_omap_dma_chan(chan);
> > > > +
> > > > /* Pause/Resume only allowed with cyclic mode */
> > > > if (!c->cyclic)
> > > > return -EINVAL;
> > > > @@ -1029,37 +1036,6 @@ static int omap_dma_resume(struct omap_chan *c)
> > > > return 0;
> > > > }
> > > >
> > > > -static int omap_dma_control(struct dma_chan *chan, enum dma_ctrl_cmd cmd,
> > > > - unsigned long arg)
> > > > -{
> > > > - struct omap_chan *c = to_omap_dma_chan(chan);
> > > > - int ret;
> > > > -
> > > > - switch (cmd) {
> > > > - case DMA_SLAVE_CONFIG:
> > > > - ret = omap_dma_slave_config(c, (struct dma_slave_config *)arg);
> > > > - break;
> > > > -
> > > > - case DMA_TERMINATE_ALL:
> > > > - ret = omap_dma_terminate_all(c);
> > > > - break;
> > > > -
> > > > - case DMA_PAUSE:
> > > > - ret = omap_dma_pause(c);
> > > > - break;
> > > > -
> > > > - case DMA_RESUME:
> > > > - ret = omap_dma_resume(c);
> > > > - break;
> > > > -
> > > > - default:
> > > > - ret = -ENXIO;
> > > > - break;
> > > > - }
> > > > -
> > > > - return ret;
> > > > -}
> > > > -
> > > > static int omap_dma_chan_init(struct omap_dmadev *od, int dma_sig)
> > > > {
> > > > struct omap_chan *c;
> > > > @@ -1136,7 +1112,10 @@ static int omap_dma_probe(struct platform_device *pdev)
> > > > od->ddev.device_issue_pending = omap_dma_issue_pending;
> > > > od->ddev.device_prep_slave_sg = omap_dma_prep_slave_sg;
> > > > od->ddev.device_prep_dma_cyclic = omap_dma_prep_dma_cyclic;
> > > > - od->ddev.device_control = omap_dma_control;
> > > this should be assigned to omap_dma_slave_config. omap_dma_control is not
> > > here :(
> >
> > I'm not sure what you mean here. I'm removing this line, together with
> > the whole omap_dma_control function.
> Sorry wrong quote and not good explanations. the lines i quoted is
> wrong one...
>
> - od->ddev.device_control = omap_dma_control;
> + od->ddev.device_config = omap_dma_config;
> this is troubling line here...
>
> we dont have omap_dma_config. It should be omap_dma_slave_config instead
Ah, right.
I have no idea why it haven't been picked up by intel's build bot
thuough.
Anyway. Do you want me to send a new version of these patches for this
trivial fix and the mxs one, or do you want me to send a follow-up?
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141205/65142da4/attachment.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 34/61] dmaengine: omap: Split device_control
2014-12-05 17:07 ` Maxime Ripard
@ 2014-12-05 17:39 ` Vinod Koul
-1 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-05 17:39 UTC (permalink / raw)
To: Maxime Ripard
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
[-- Attachment #1: Type: text/plain, Size: 747 bytes --]
On Fri, Dec 05, 2014 at 06:07:44PM +0100, Maxime Ripard wrote:
> On Fri, Dec 05, 2014 at 08:26:42PM +0530, Vinod Koul wrote:
> > On Fri, Dec 05, 2014 at 03:33:59PM +0100, Maxime Ripard wrote:
> Ah, right.
>
> I have no idea why it haven't been picked up by intel's build bot
> thuough.
possibly no defconfig enables this
>
> Anyway. Do you want me to send a new version of these patches for this
> trivial fix and the mxs one, or do you want me to send a follow-up?
I am halfway thru series, if I dont find anything serious I will fix it up and
push over the weekend
--
~Vinod
>
> Maxime
>
> --
> Maxime Ripard, Free Electrons
> Embedded Linux, Kernel and Android engineering
> http://free-electrons.com
--
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 34/61] dmaengine: omap: Split device_control
@ 2014-12-05 17:39 ` Vinod Koul
0 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-05 17:39 UTC (permalink / raw)
To: linux-arm-kernel
On Fri, Dec 05, 2014 at 06:07:44PM +0100, Maxime Ripard wrote:
> On Fri, Dec 05, 2014 at 08:26:42PM +0530, Vinod Koul wrote:
> > On Fri, Dec 05, 2014 at 03:33:59PM +0100, Maxime Ripard wrote:
> Ah, right.
>
> I have no idea why it haven't been picked up by intel's build bot
> thuough.
possibly no defconfig enables this
>
> Anyway. Do you want me to send a new version of these patches for this
> trivial fix and the mxs one, or do you want me to send a follow-up?
I am halfway thru series, if I dont find anything serious I will fix it up and
push over the weekend
--
~Vinod
>
> Maxime
>
> --
> Maxime Ripard, Free Electrons
> Embedded Linux, Kernel and Android engineering
> http://free-electrons.com
--
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141205/cbe819e7/attachment.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-11-17 13:41 ` Maxime Ripard
@ 2014-12-08 6:17 ` Vinod Koul
-1 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-08 6:17 UTC (permalink / raw)
To: Maxime Ripard
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> Hi,
>
> As we discussed a couple of weeks ago, this is the third attempt at
> creating a generic behaviour for slave capabilities retrieval so that
> generic layers using dmaengine can actually rely on that.
>
> That has been done mostly through two steps: by moving out the
> sub-commands of the device_control callback, so that the dmaengine
> core can then infer from that wether a sub-command is implemented, and
> then by moving the slave properties, such as the supported buswidth,
> to the structure dma_device itself.
Okay managed to get this done. Apart from the two issues identified didn't
find anything so applied and pushed to a branch
"topic/slave_caps_device_control_fix"
Today did some compile tests and found few warnings, were trivial but I
am worried about the testing of this code. Has anyone tested this, if so
which platforms are covered Since I pushed base branch last night, Feng's
bot covered it and all was OK. Looks like Feng's bot doesn't have wide
coverage of arm platforms, wasn't there one run by arm guys which tries to
test and boot, if so can we get this tested there please.
So bit sceptical for merging now. I will send the patches which I have
applied on top of this
Thanks
--
~Vinod
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-08 6:17 ` Vinod Koul
0 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-08 6:17 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> Hi,
>
> As we discussed a couple of weeks ago, this is the third attempt at
> creating a generic behaviour for slave capabilities retrieval so that
> generic layers using dmaengine can actually rely on that.
>
> That has been done mostly through two steps: by moving out the
> sub-commands of the device_control callback, so that the dmaengine
> core can then infer from that wether a sub-command is implemented, and
> then by moving the slave properties, such as the supported buswidth,
> to the structure dma_device itself.
Okay managed to get this done. Apart from the two issues identified didn't
find anything so applied and pushed to a branch
"topic/slave_caps_device_control_fix"
Today did some compile tests and found few warnings, were trivial but I
am worried about the testing of this code. Has anyone tested this, if so
which platforms are covered Since I pushed base branch last night, Feng's
bot covered it and all was OK. Looks like Feng's bot doesn't have wide
coverage of arm platforms, wasn't there one run by arm guys which tries to
test and boot, if so can we get this tested there please.
So bit sceptical for merging now. I will send the patches which I have
applied on top of this
Thanks
--
~Vinod
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-12-08 6:17 ` Vinod Koul
@ 2014-12-08 9:32 ` Ludovic Desroches
-1 siblings, 0 replies; 164+ messages in thread
From: Ludovic Desroches @ 2014-12-08 9:32 UTC (permalink / raw)
To: Vinod Koul
Cc: Maxime Ripard, dmaengine, linux-kernel, linux-arm-kernel,
Laurent Pinchart, Antoine Ténart, Russell King, lars
On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > Hi,
> >
> > As we discussed a couple of weeks ago, this is the third attempt at
> > creating a generic behaviour for slave capabilities retrieval so that
> > generic layers using dmaengine can actually rely on that.
> >
> > That has been done mostly through two steps: by moving out the
> > sub-commands of the device_control callback, so that the dmaengine
> > core can then infer from that wether a sub-command is implemented, and
> > then by moving the slave properties, such as the supported buswidth,
> > to the structure dma_device itself.
>
> Okay managed to get this done. Apart from the two issues identified didn't
> find anything so applied and pushed to a branch
> "topic/slave_caps_device_control_fix"
>
> Today did some compile tests and found few warnings, were trivial but I
> am worried about the testing of this code. Has anyone tested this, if so
> which platforms are covered Since I pushed base branch last night, Feng's
> bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> coverage of arm platforms, wasn't there one run by arm guys which tries to
> test and boot, if so can we get this tested there please.
Tested with your branch on sama5d3 and sama5d4 so at_hdmac and at xdmac.
I have also sent some patches based on Maxime's series for at_xdmac.
Ludovic
>
> So bit sceptical for merging now. I will send the patches which I have
> applied on top of this
>
> Thanks
> --
> ~Vinod
>
> --
> To unsubscribe from this list: send the line "unsubscribe dmaengine" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-08 9:32 ` Ludovic Desroches
0 siblings, 0 replies; 164+ messages in thread
From: Ludovic Desroches @ 2014-12-08 9:32 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > Hi,
> >
> > As we discussed a couple of weeks ago, this is the third attempt at
> > creating a generic behaviour for slave capabilities retrieval so that
> > generic layers using dmaengine can actually rely on that.
> >
> > That has been done mostly through two steps: by moving out the
> > sub-commands of the device_control callback, so that the dmaengine
> > core can then infer from that wether a sub-command is implemented, and
> > then by moving the slave properties, such as the supported buswidth,
> > to the structure dma_device itself.
>
> Okay managed to get this done. Apart from the two issues identified didn't
> find anything so applied and pushed to a branch
> "topic/slave_caps_device_control_fix"
>
> Today did some compile tests and found few warnings, were trivial but I
> am worried about the testing of this code. Has anyone tested this, if so
> which platforms are covered Since I pushed base branch last night, Feng's
> bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> coverage of arm platforms, wasn't there one run by arm guys which tries to
> test and boot, if so can we get this tested there please.
Tested with your branch on sama5d3 and sama5d4 so at_hdmac and at xdmac.
I have also sent some patches based on Maxime's series for at_xdmac.
Ludovic
>
> So bit sceptical for merging now. I will send the patches which I have
> applied on top of this
>
> Thanks
> --
> ~Vinod
>
> --
> To unsubscribe from this list: send the line "unsubscribe dmaengine" in
> the body of a message to majordomo at vger.kernel.org
> More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-12-08 9:32 ` Ludovic Desroches
@ 2014-12-08 13:30 ` Vinod Koul
-1 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-08 13:30 UTC (permalink / raw)
To: Maxime Ripard, dmaengine, linux-kernel, linux-arm-kernel,
Laurent Pinchart, Antoine Ténart, Russell King, lars
On Mon, Dec 08, 2014 at 10:32:47AM +0100, Ludovic Desroches wrote:
> On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> > On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > > Hi,
> > >
> > > As we discussed a couple of weeks ago, this is the third attempt at
> > > creating a generic behaviour for slave capabilities retrieval so that
> > > generic layers using dmaengine can actually rely on that.
> > >
> > > That has been done mostly through two steps: by moving out the
> > > sub-commands of the device_control callback, so that the dmaengine
> > > core can then infer from that wether a sub-command is implemented, and
> > > then by moving the slave properties, such as the supported buswidth,
> > > to the structure dma_device itself.
> >
> > Okay managed to get this done. Apart from the two issues identified didn't
> > find anything so applied and pushed to a branch
> > "topic/slave_caps_device_control_fix"
> >
> > Today did some compile tests and found few warnings, were trivial but I
> > am worried about the testing of this code. Has anyone tested this, if so
> > which platforms are covered Since I pushed base branch last night, Feng's
> > bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> > coverage of arm platforms, wasn't there one run by arm guys which tries to
> > test and boot, if so can we get this tested there please.
>
> Tested with your branch on sama5d3 and sama5d4 so at_hdmac and at xdmac.
>
> I have also sent some patches based on Maxime's series for at_xdmac.
Thank You!
Anymore voluteers? I am looking at omap, mxs, tegra ones (these had issues)
--
~Vinod
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-08 13:30 ` Vinod Koul
0 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-08 13:30 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Dec 08, 2014 at 10:32:47AM +0100, Ludovic Desroches wrote:
> On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> > On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > > Hi,
> > >
> > > As we discussed a couple of weeks ago, this is the third attempt at
> > > creating a generic behaviour for slave capabilities retrieval so that
> > > generic layers using dmaengine can actually rely on that.
> > >
> > > That has been done mostly through two steps: by moving out the
> > > sub-commands of the device_control callback, so that the dmaengine
> > > core can then infer from that wether a sub-command is implemented, and
> > > then by moving the slave properties, such as the supported buswidth,
> > > to the structure dma_device itself.
> >
> > Okay managed to get this done. Apart from the two issues identified didn't
> > find anything so applied and pushed to a branch
> > "topic/slave_caps_device_control_fix"
> >
> > Today did some compile tests and found few warnings, were trivial but I
> > am worried about the testing of this code. Has anyone tested this, if so
> > which platforms are covered Since I pushed base branch last night, Feng's
> > bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> > coverage of arm platforms, wasn't there one run by arm guys which tries to
> > test and boot, if so can we get this tested there please.
>
> Tested with your branch on sama5d3 and sama5d4 so at_hdmac and at xdmac.
>
> I have also sent some patches based on Maxime's series for at_xdmac.
Thank You!
Anymore voluteers? I am looking at omap, mxs, tegra ones (these had issues)
--
~Vinod
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-12-08 6:17 ` Vinod Koul
@ 2014-12-08 14:18 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-08 14:18 UTC (permalink / raw)
To: Vinod Koul
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
[-- Attachment #1: Type: text/plain, Size: 2072 bytes --]
On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > Hi,
> >
> > As we discussed a couple of weeks ago, this is the third attempt at
> > creating a generic behaviour for slave capabilities retrieval so that
> > generic layers using dmaengine can actually rely on that.
> >
> > That has been done mostly through two steps: by moving out the
> > sub-commands of the device_control callback, so that the dmaengine
> > core can then infer from that wether a sub-command is implemented, and
> > then by moving the slave properties, such as the supported buswidth,
> > to the structure dma_device itself.
>
> Okay managed to get this done. Apart from the two issues identified didn't
> find anything so applied and pushed to a branch
> "topic/slave_caps_device_control_fix"
>
> Today did some compile tests and found few warnings, were trivial but I
> am worried about the testing of this code. Has anyone tested this, if so
> which platforms are covered Since I pushed base branch last night, Feng's
> bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> coverage of arm platforms, wasn't there one run by arm guys which tries to
> test and boot, if so can we get this tested there please.
>
> So bit sceptical for merging now. I will send the patches which I have
> applied on top of this
Which is why I wanted to merge this at the *beginning* of the
development cycle in the first place....
These patches have been sent more than 2 weeks ago, and were exactly
the same as the ones send at the end of October, rebased and updated
to take into account the drivers that were merged in between.
In short, these patches have been hanging around since 6 weeks. I
relied probably too much on the intel's build bot. This is not a
mistake I'll repeat. But blaming it on me because they came too late
is *very* unfair.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-08 14:18 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-08 14:18 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > Hi,
> >
> > As we discussed a couple of weeks ago, this is the third attempt at
> > creating a generic behaviour for slave capabilities retrieval so that
> > generic layers using dmaengine can actually rely on that.
> >
> > That has been done mostly through two steps: by moving out the
> > sub-commands of the device_control callback, so that the dmaengine
> > core can then infer from that wether a sub-command is implemented, and
> > then by moving the slave properties, such as the supported buswidth,
> > to the structure dma_device itself.
>
> Okay managed to get this done. Apart from the two issues identified didn't
> find anything so applied and pushed to a branch
> "topic/slave_caps_device_control_fix"
>
> Today did some compile tests and found few warnings, were trivial but I
> am worried about the testing of this code. Has anyone tested this, if so
> which platforms are covered Since I pushed base branch last night, Feng's
> bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> coverage of arm platforms, wasn't there one run by arm guys which tries to
> test and boot, if so can we get this tested there please.
>
> So bit sceptical for merging now. I will send the patches which I have
> applied on top of this
Which is why I wanted to merge this at the *beginning* of the
development cycle in the first place....
These patches have been sent more than 2 weeks ago, and were exactly
the same as the ones send at the end of October, rebased and updated
to take into account the drivers that were merged in between.
In short, these patches have been hanging around since 6 weeks. I
relied probably too much on the intel's build bot. This is not a
mistake I'll repeat. But blaming it on me because they came too late
is *very* unfair.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141208/e9e02940/attachment.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-12-08 13:30 ` Vinod Koul
@ 2014-12-08 14:31 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-08 14:31 UTC (permalink / raw)
To: Vinod Koul
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
[-- Attachment #1: Type: text/plain, Size: 2263 bytes --]
On Mon, Dec 08, 2014 at 07:00:47PM +0530, Vinod Koul wrote:
> On Mon, Dec 08, 2014 at 10:32:47AM +0100, Ludovic Desroches wrote:
> > On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> > > On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > > > Hi,
> > > >
> > > > As we discussed a couple of weeks ago, this is the third attempt at
> > > > creating a generic behaviour for slave capabilities retrieval so that
> > > > generic layers using dmaengine can actually rely on that.
> > > >
> > > > That has been done mostly through two steps: by moving out the
> > > > sub-commands of the device_control callback, so that the dmaengine
> > > > core can then infer from that wether a sub-command is implemented, and
> > > > then by moving the slave properties, such as the supported buswidth,
> > > > to the structure dma_device itself.
> > >
> > > Okay managed to get this done. Apart from the two issues identified didn't
> > > find anything so applied and pushed to a branch
> > > "topic/slave_caps_device_control_fix"
> > >
> > > Today did some compile tests and found few warnings, were trivial but I
> > > am worried about the testing of this code. Has anyone tested this, if so
> > > which platforms are covered Since I pushed base branch last night, Feng's
> > > bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> > > coverage of arm platforms, wasn't there one run by arm guys which tries to
> > > test and boot, if so can we get this tested there please.
> >
> > Tested with your branch on sama5d3 and sama5d4 so at_hdmac and at xdmac.
> >
> > I have also sent some patches based on Maxime's series for at_xdmac.
I did the development on the Allwinner A31 (sun6i driver). So that's
tested too.
Thanks to Ludovic for his testing.
> Anymore voluteers? I am looking at omap, mxs, tegra ones (these had
> issues)
Since these were compile time issues (and trivial one). From a
functional point of view, beside the warning introduced, this won't
change anything.
And even if it does, we have a full bug-fixing window to take care of
that.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-08 14:31 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-08 14:31 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Dec 08, 2014 at 07:00:47PM +0530, Vinod Koul wrote:
> On Mon, Dec 08, 2014 at 10:32:47AM +0100, Ludovic Desroches wrote:
> > On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> > > On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > > > Hi,
> > > >
> > > > As we discussed a couple of weeks ago, this is the third attempt at
> > > > creating a generic behaviour for slave capabilities retrieval so that
> > > > generic layers using dmaengine can actually rely on that.
> > > >
> > > > That has been done mostly through two steps: by moving out the
> > > > sub-commands of the device_control callback, so that the dmaengine
> > > > core can then infer from that wether a sub-command is implemented, and
> > > > then by moving the slave properties, such as the supported buswidth,
> > > > to the structure dma_device itself.
> > >
> > > Okay managed to get this done. Apart from the two issues identified didn't
> > > find anything so applied and pushed to a branch
> > > "topic/slave_caps_device_control_fix"
> > >
> > > Today did some compile tests and found few warnings, were trivial but I
> > > am worried about the testing of this code. Has anyone tested this, if so
> > > which platforms are covered Since I pushed base branch last night, Feng's
> > > bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> > > coverage of arm platforms, wasn't there one run by arm guys which tries to
> > > test and boot, if so can we get this tested there please.
> >
> > Tested with your branch on sama5d3 and sama5d4 so at_hdmac and at xdmac.
> >
> > I have also sent some patches based on Maxime's series for at_xdmac.
I did the development on the Allwinner A31 (sun6i driver). So that's
tested too.
Thanks to Ludovic for his testing.
> Anymore voluteers? I am looking at omap, mxs, tegra ones (these had
> issues)
Since these were compile time issues (and trivial one). From a
functional point of view, beside the warning introduced, this won't
change anything.
And even if it does, we have a full bug-fixing window to take care of
that.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141208/f1a4ca95/attachment.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-12-08 14:18 ` Maxime Ripard
@ 2014-12-08 16:28 ` Vinod Koul
-1 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-08 16:28 UTC (permalink / raw)
To: Maxime Ripard
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
[-- Attachment #1: Type: text/plain, Size: 2512 bytes --]
On Mon, Dec 08, 2014 at 03:18:07PM +0100, Maxime Ripard wrote:
> On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> > On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > > Hi,
> > >
> > > As we discussed a couple of weeks ago, this is the third attempt at
> > > creating a generic behaviour for slave capabilities retrieval so that
> > > generic layers using dmaengine can actually rely on that.
> > >
> > > That has been done mostly through two steps: by moving out the
> > > sub-commands of the device_control callback, so that the dmaengine
> > > core can then infer from that wether a sub-command is implemented, and
> > > then by moving the slave properties, such as the supported buswidth,
> > > to the structure dma_device itself.
> >
> > Okay managed to get this done. Apart from the two issues identified didn't
> > find anything so applied and pushed to a branch
> > "topic/slave_caps_device_control_fix"
> >
> > Today did some compile tests and found few warnings, were trivial but I
> > am worried about the testing of this code. Has anyone tested this, if so
> > which platforms are covered Since I pushed base branch last night, Feng's
> > bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> > coverage of arm platforms, wasn't there one run by arm guys which tries to
> > test and boot, if so can we get this tested there please.
> >
> > So bit sceptical for merging now. I will send the patches which I have
> > applied on top of this
>
> Which is why I wanted to merge this at the *beginning* of the
> development cycle in the first place....
>
> These patches have been sent more than 2 weeks ago, and were exactly
> the same as the ones send at the end of October, rebased and updated
> to take into account the drivers that were merged in between.
>
> In short, these patches have been hanging around since 6 weeks. I
> relied probably too much on the intel's build bot. This is not a
> mistake I'll repeat. But blaming it on me because they came too late
> is *very* unfair.
as a first step one is expected to compile the patches we send. That is very
basic stuff which should not be avoided. given that we had build failures on
v5 rev of the series and serious failures flagged off by compilers doesn't
inspire a lot of confidence. And lastly noone blamed you for being late, if
things were rosy they would have been merged over the weekend and been in
today's next, but...
--
~Vinod
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-08 16:28 ` Vinod Koul
0 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-08 16:28 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Dec 08, 2014 at 03:18:07PM +0100, Maxime Ripard wrote:
> On Mon, Dec 08, 2014 at 11:47:46AM +0530, Vinod Koul wrote:
> > On Mon, Nov 17, 2014 at 02:41:54PM +0100, Maxime Ripard wrote:
> > > Hi,
> > >
> > > As we discussed a couple of weeks ago, this is the third attempt at
> > > creating a generic behaviour for slave capabilities retrieval so that
> > > generic layers using dmaengine can actually rely on that.
> > >
> > > That has been done mostly through two steps: by moving out the
> > > sub-commands of the device_control callback, so that the dmaengine
> > > core can then infer from that wether a sub-command is implemented, and
> > > then by moving the slave properties, such as the supported buswidth,
> > > to the structure dma_device itself.
> >
> > Okay managed to get this done. Apart from the two issues identified didn't
> > find anything so applied and pushed to a branch
> > "topic/slave_caps_device_control_fix"
> >
> > Today did some compile tests and found few warnings, were trivial but I
> > am worried about the testing of this code. Has anyone tested this, if so
> > which platforms are covered Since I pushed base branch last night, Feng's
> > bot covered it and all was OK. Looks like Feng's bot doesn't have wide
> > coverage of arm platforms, wasn't there one run by arm guys which tries to
> > test and boot, if so can we get this tested there please.
> >
> > So bit sceptical for merging now. I will send the patches which I have
> > applied on top of this
>
> Which is why I wanted to merge this at the *beginning* of the
> development cycle in the first place....
>
> These patches have been sent more than 2 weeks ago, and were exactly
> the same as the ones send at the end of October, rebased and updated
> to take into account the drivers that were merged in between.
>
> In short, these patches have been hanging around since 6 weeks. I
> relied probably too much on the intel's build bot. This is not a
> mistake I'll repeat. But blaming it on me because they came too late
> is *very* unfair.
as a first step one is expected to compile the patches we send. That is very
basic stuff which should not be avoided. given that we had build failures on
v5 rev of the series and serious failures flagged off by compilers doesn't
inspire a lot of confidence. And lastly noone blamed you for being late, if
things were rosy they would have been merged over the weekend and been in
today's next, but...
--
~Vinod
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141208/d617fd04/attachment.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-12-08 16:28 ` Vinod Koul
@ 2014-12-08 18:38 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-08 18:38 UTC (permalink / raw)
To: Vinod Koul
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
[-- Attachment #1: Type: text/plain, Size: 1928 bytes --]
On Mon, Dec 08, 2014 at 09:58:43PM +0530, Vinod Koul wrote:
> > > So bit sceptical for merging now. I will send the patches which I have
> > > applied on top of this
> >
> > Which is why I wanted to merge this at the *beginning* of the
> > development cycle in the first place....
> >
> > These patches have been sent more than 2 weeks ago, and were exactly
> > the same as the ones send at the end of October, rebased and updated
> > to take into account the drivers that were merged in between.
> >
> > In short, these patches have been hanging around since 6 weeks. I
> > relied probably too much on the intel's build bot. This is not a
> > mistake I'll repeat. But blaming it on me because they came too late
> > is *very* unfair.
>
> as a first step one is expected to compile the patches we send. That is very
> basic stuff which should not be avoided.
Indeed. My mistake, it won't happen again.
> given that we had build failures on v5 rev of the series and serious
> failures flagged off by compilers doesn't inspire a lot of
> confidence.
v5 that you never reviewed, just like the v4, v3 and v2 before it.
> And lastly noone blamed you for being late, if things were rosy they
> would have been merged over the weekend and been in today's next,
> but...
I totally understand your point. And I actually am a bit uncomfortable
merging this so late too, and I'd actually prefer to have it merged
for 3.20. But this is a huge patchset, and I'd really like to avoid
rebasing it forever.
I'll send a v6, with your patches, as soon as 3.19-rc1 is out. This
will be my last rebase of this serie. If these patches are not merged
then, then I will just give up on this cleanup, and probably any
subsequent ones.
I'm sorry, but I just don't care enough.
Maxime.
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-08 18:38 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-08 18:38 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Dec 08, 2014 at 09:58:43PM +0530, Vinod Koul wrote:
> > > So bit sceptical for merging now. I will send the patches which I have
> > > applied on top of this
> >
> > Which is why I wanted to merge this at the *beginning* of the
> > development cycle in the first place....
> >
> > These patches have been sent more than 2 weeks ago, and were exactly
> > the same as the ones send at the end of October, rebased and updated
> > to take into account the drivers that were merged in between.
> >
> > In short, these patches have been hanging around since 6 weeks. I
> > relied probably too much on the intel's build bot. This is not a
> > mistake I'll repeat. But blaming it on me because they came too late
> > is *very* unfair.
>
> as a first step one is expected to compile the patches we send. That is very
> basic stuff which should not be avoided.
Indeed. My mistake, it won't happen again.
> given that we had build failures on v5 rev of the series and serious
> failures flagged off by compilers doesn't inspire a lot of
> confidence.
v5 that you never reviewed, just like the v4, v3 and v2 before it.
> And lastly noone blamed you for being late, if things were rosy they
> would have been merged over the weekend and been in today's next,
> but...
I totally understand your point. And I actually am a bit uncomfortable
merging this so late too, and I'd actually prefer to have it merged
for 3.20. But this is a huge patchset, and I'd really like to avoid
rebasing it forever.
I'll send a v6, with your patches, as soon as 3.19-rc1 is out. This
will be my last rebase of this serie. If these patches are not merged
then, then I will just give up on this cleanup, and probably any
subsequent ones.
I'm sorry, but I just don't care enough.
Maxime.
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141208/cc8866b3/attachment.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-12-08 18:38 ` Maxime Ripard
@ 2014-12-08 21:49 ` Lars-Peter Clausen
-1 siblings, 0 replies; 164+ messages in thread
From: Lars-Peter Clausen @ 2014-12-08 21:49 UTC (permalink / raw)
To: Maxime Ripard, Vinod Koul
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King
On 12/08/2014 07:38 PM, Maxime Ripard wrote:
[...]
>> And lastly noone blamed you for being late, if things were rosy they
>> would have been merged over the weekend and been in today's next,
>> but...
>
> I totally understand your point. And I actually am a bit uncomfortable
> merging this so late too, and I'd actually prefer to have it merged
> for 3.20. But this is a huge patchset, and I'd really like to avoid
> rebasing it forever.
We don't have to pull in the whole series at once. I think everybody agrees
that the approach is the right one, so the first couple of patches that work
on the framework should be good to go in. Same goes for the drivers that
have been acked and tested.
- Lars
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-08 21:49 ` Lars-Peter Clausen
0 siblings, 0 replies; 164+ messages in thread
From: Lars-Peter Clausen @ 2014-12-08 21:49 UTC (permalink / raw)
To: linux-arm-kernel
On 12/08/2014 07:38 PM, Maxime Ripard wrote:
[...]
>> And lastly noone blamed you for being late, if things were rosy they
>> would have been merged over the weekend and been in today's next,
>> but...
>
> I totally understand your point. And I actually am a bit uncomfortable
> merging this so late too, and I'd actually prefer to have it merged
> for 3.20. But this is a huge patchset, and I'd really like to avoid
> rebasing it forever.
We don't have to pull in the whole series at once. I think everybody agrees
that the approach is the right one, so the first couple of patches that work
on the framework should be good to go in. Same goes for the drivers that
have been acked and tested.
- Lars
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-12-08 18:38 ` Maxime Ripard
@ 2014-12-09 4:43 ` Vinod Koul
-1 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-09 4:43 UTC (permalink / raw)
To: Maxime Ripard
Cc: dmaengine, linux-kernel, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, lars
[-- Attachment #1: Type: text/plain, Size: 1145 bytes --]
On Mon, Dec 08, 2014 at 07:38:53PM +0100, Maxime Ripard wrote:
> On Mon, Dec 08, 2014 at 09:58:43PM +0530, Vinod Koul wrote:
> I totally understand your point. And I actually am a bit uncomfortable
> merging this so late too, and I'd actually prefer to have it merged
> for 3.20. But this is a huge patchset, and I'd really like to avoid
> rebasing it forever.
>
> I'll send a v6, with your patches, as soon as 3.19-rc1 is out. This
> will be my last rebase of this serie. If these patches are not merged
> then, then I will just give up on this cleanup, and probably any
> subsequent ones.
Since I have applied them to a topic branch, I don't think we you to rebase
and resend. Git does a pretty good job at rebasing so I will rebase
this branch on top of -rc1 and merge. Fixes can keep coming on top of this
one.
So these patches are applied, and will show up in -next after -rc1. Thank
you for the work you put in this.
Last night, I thought of giving it a try so temporarily merged this to my
next. SFR found bunch of issues with it (mostly due to dependencies in
users) so lets keep fixing those..
--
~Vinod
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 836 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-09 4:43 ` Vinod Koul
0 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2014-12-09 4:43 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Dec 08, 2014 at 07:38:53PM +0100, Maxime Ripard wrote:
> On Mon, Dec 08, 2014 at 09:58:43PM +0530, Vinod Koul wrote:
> I totally understand your point. And I actually am a bit uncomfortable
> merging this so late too, and I'd actually prefer to have it merged
> for 3.20. But this is a huge patchset, and I'd really like to avoid
> rebasing it forever.
>
> I'll send a v6, with your patches, as soon as 3.19-rc1 is out. This
> will be my last rebase of this serie. If these patches are not merged
> then, then I will just give up on this cleanup, and probably any
> subsequent ones.
Since I have applied them to a topic branch, I don't think we you to rebase
and resend. Git does a pretty good job at rebasing so I will rebase
this branch on top of -rc1 and merge. Fixes can keep coming on top of this
one.
So these patches are applied, and will show up in -next after -rc1. Thank
you for the work you put in this.
Last night, I thought of giving it a try so temporarily merged this to my
next. SFR found bunch of issues with it (mostly due to dependencies in
users) so lets keep fixing those..
--
~Vinod
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 836 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141209/2ca8a18e/attachment-0001.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
2014-12-08 21:49 ` Lars-Peter Clausen
@ 2014-12-11 17:03 ` Maxime Ripard
-1 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-11 17:03 UTC (permalink / raw)
To: Lars-Peter Clausen
Cc: Vinod Koul, dmaengine, linux-kernel, linux-arm-kernel,
Laurent Pinchart, Antoine Ténart, Russell King
[-- Attachment #1: Type: text/plain, Size: 1275 bytes --]
On Mon, Dec 08, 2014 at 10:49:56PM +0100, Lars-Peter Clausen wrote:
> On 12/08/2014 07:38 PM, Maxime Ripard wrote:
> [...]
> >>And lastly noone blamed you for being late, if things were rosy they
> >>would have been merged over the weekend and been in today's next,
> >>but...
> >
> >I totally understand your point. And I actually am a bit uncomfortable
> >merging this so late too, and I'd actually prefer to have it merged
> >for 3.20. But this is a huge patchset, and I'd really like to avoid
> >rebasing it forever.
>
> We don't have to pull in the whole series at once. I think everybody agrees
> that the approach is the right one, so the first couple of patches that work
> on the framework should be good to go in. Same goes for the drivers that
> have been acked and tested.
I'm a bit skeptical that it is the right approach. These patches have
been here for monthes, and even you failed to ack and test these
patches on the drivers you wrote. As a matter of fact, only Laurent,
Linus, Ludovic and Stephen looked at their respective changes. After 3
monthes and 6 versions.
At some point, cleanups have to come in.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 819 bytes --]
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval
@ 2014-12-11 17:03 ` Maxime Ripard
0 siblings, 0 replies; 164+ messages in thread
From: Maxime Ripard @ 2014-12-11 17:03 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Dec 08, 2014 at 10:49:56PM +0100, Lars-Peter Clausen wrote:
> On 12/08/2014 07:38 PM, Maxime Ripard wrote:
> [...]
> >>And lastly noone blamed you for being late, if things were rosy they
> >>would have been merged over the weekend and been in today's next,
> >>but...
> >
> >I totally understand your point. And I actually am a bit uncomfortable
> >merging this so late too, and I'd actually prefer to have it merged
> >for 3.20. But this is a huge patchset, and I'd really like to avoid
> >rebasing it forever.
>
> We don't have to pull in the whole series at once. I think everybody agrees
> that the approach is the right one, so the first couple of patches that work
> on the framework should be good to go in. Same goes for the drivers that
> have been acked and tested.
I'm a bit skeptical that it is the right approach. These patches have
been here for monthes, and even you failed to ack and test these
patches on the drivers you wrote. As a matter of fact, only Laurent,
Linus, Ludovic and Stephen looked at their respective changes. After 3
monthes and 6 versions.
At some point, cleanups have to come in.
Maxime
--
Maxime Ripard, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20141211/08971499/attachment.sig>
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
2014-11-17 13:42 ` Maxime Ripard
@ 2015-02-19 7:57 ` Viresh Kumar
-1 siblings, 0 replies; 164+ messages in thread
From: Viresh Kumar @ 2015-02-19 7:57 UTC (permalink / raw)
To: Maxime Ripard, bloften80, jdelvare
Cc: Vinod Koul, dmaengine, linux-kernel, linux-arm-kernel,
Laurent Pinchart, Antoine Ténart, Russell King,
Lars-Peter Clausen
On Mon, Nov 17, 2014 at 9:42 PM, Maxime Ripard
<maxime.ripard@free-electrons.com> wrote:
> For the slave caps retrieval to be really useful, most drivers need to
> implement it.
>
> Hence, we need to be slightly more aggressive, and trigger a warning at
> registration time for drivers that don't fill their caps infos in order to
> encourage them to implement it.
>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> ---
> drivers/dma/dmaengine.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
> index 98e9431f85ec..300c8cd2786c 100644
> --- a/drivers/dma/dmaengine.c
> +++ b/drivers/dma/dmaengine.c
> @@ -827,6 +827,9 @@ int dma_async_device_register(struct dma_device *device)
> BUG_ON(!device->device_issue_pending);
> BUG_ON(!device->dev);
>
> + WARN(dma_has_cap(DMA_SLAVE, device->cap_mask) && !device->directions,
> + "this driver doesn't support generic slave capabilities reporting\n");
> +
> /* note: this only matters in the
> * CONFIG_ASYNC_TX_ENABLE_CHANNEL_SWITCH=n case
> */
Probably you need to fix this ?
https://bugzilla.kernel.org/show_bug.cgi?id=90801
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
@ 2015-02-19 7:57 ` Viresh Kumar
0 siblings, 0 replies; 164+ messages in thread
From: Viresh Kumar @ 2015-02-19 7:57 UTC (permalink / raw)
To: linux-arm-kernel
On Mon, Nov 17, 2014 at 9:42 PM, Maxime Ripard
<maxime.ripard@free-electrons.com> wrote:
> For the slave caps retrieval to be really useful, most drivers need to
> implement it.
>
> Hence, we need to be slightly more aggressive, and trigger a warning at
> registration time for drivers that don't fill their caps infos in order to
> encourage them to implement it.
>
> Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> ---
> drivers/dma/dmaengine.c | 3 +++
> 1 file changed, 3 insertions(+)
>
> diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
> index 98e9431f85ec..300c8cd2786c 100644
> --- a/drivers/dma/dmaengine.c
> +++ b/drivers/dma/dmaengine.c
> @@ -827,6 +827,9 @@ int dma_async_device_register(struct dma_device *device)
> BUG_ON(!device->device_issue_pending);
> BUG_ON(!device->dev);
>
> + WARN(dma_has_cap(DMA_SLAVE, device->cap_mask) && !device->directions,
> + "this driver doesn't support generic slave capabilities reporting\n");
> +
> /* note: this only matters in the
> * CONFIG_ASYNC_TX_ENABLE_CHANNEL_SWITCH=n case
> */
Probably you need to fix this ?
https://bugzilla.kernel.org/show_bug.cgi?id=90801
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
2015-02-19 7:57 ` Viresh Kumar
@ 2015-02-19 8:09 ` Jean Delvare
-1 siblings, 0 replies; 164+ messages in thread
From: Jean Delvare @ 2015-02-19 8:09 UTC (permalink / raw)
To: Viresh Kumar
Cc: Maxime Ripard, Brian Loften, Vinod Koul, dmaengine, linux-kernel,
linux-arm-kernel, Laurent Pinchart, Antoine Ténart,
Russell King, Lars-Peter Clausen, Andy Shevchenko
Hi Viresh,
On Thu, 19 Feb 2015 15:57:27 +0800, Viresh Kumar wrote:
> On Mon, Nov 17, 2014 at 9:42 PM, Maxime Ripard
> <maxime.ripard@free-electrons.com> wrote:
> > For the slave caps retrieval to be really useful, most drivers need to
> > implement it.
> >
> > Hence, we need to be slightly more aggressive, and trigger a warning at
> > registration time for drivers that don't fill their caps infos in order to
> > encourage them to implement it.
> >
> > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> > Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > ---
> > drivers/dma/dmaengine.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
> > index 98e9431f85ec..300c8cd2786c 100644
> > --- a/drivers/dma/dmaengine.c
> > +++ b/drivers/dma/dmaengine.c
> > @@ -827,6 +827,9 @@ int dma_async_device_register(struct dma_device *device)
> > BUG_ON(!device->device_issue_pending);
> > BUG_ON(!device->dev);
> >
> > + WARN(dma_has_cap(DMA_SLAVE, device->cap_mask) && !device->directions,
> > + "this driver doesn't support generic slave capabilities reporting\n");
> > +
> > /* note: this only matters in the
> > * CONFIG_ASYNC_TX_ENABLE_CHANNEL_SWITCH=n case
> > */
>
> Probably you need to fix this ?
>
> https://bugzilla.kernel.org/show_bug.cgi?id=90801
My understanding is that the warning message is here on purpose to
instruct DMA driver authors to fix their code. So if anyone need to fix
anything, that would be you and Andy. That being said, I suspect Andy
already did that:
commit 029a40e97d0db269f4a7fc02b0f29f627f628309
Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Fri Jan 2 16:17:24 2015 +0200
dmaengine: dw: provide DMA capabilities
Brian, can you please try the latest Linus or linux-next kernel (or
backport the above commit to your current kernel) and confirm that the
backtraces are no longer printed?
Thanks,
--
Jean Delvare
SUSE L3 Support
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
@ 2015-02-19 8:09 ` Jean Delvare
0 siblings, 0 replies; 164+ messages in thread
From: Jean Delvare @ 2015-02-19 8:09 UTC (permalink / raw)
To: linux-arm-kernel
Hi Viresh,
On Thu, 19 Feb 2015 15:57:27 +0800, Viresh Kumar wrote:
> On Mon, Nov 17, 2014 at 9:42 PM, Maxime Ripard
> <maxime.ripard@free-electrons.com> wrote:
> > For the slave caps retrieval to be really useful, most drivers need to
> > implement it.
> >
> > Hence, we need to be slightly more aggressive, and trigger a warning at
> > registration time for drivers that don't fill their caps infos in order to
> > encourage them to implement it.
> >
> > Signed-off-by: Maxime Ripard <maxime.ripard@free-electrons.com>
> > Acked-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
> > ---
> > drivers/dma/dmaengine.c | 3 +++
> > 1 file changed, 3 insertions(+)
> >
> > diff --git a/drivers/dma/dmaengine.c b/drivers/dma/dmaengine.c
> > index 98e9431f85ec..300c8cd2786c 100644
> > --- a/drivers/dma/dmaengine.c
> > +++ b/drivers/dma/dmaengine.c
> > @@ -827,6 +827,9 @@ int dma_async_device_register(struct dma_device *device)
> > BUG_ON(!device->device_issue_pending);
> > BUG_ON(!device->dev);
> >
> > + WARN(dma_has_cap(DMA_SLAVE, device->cap_mask) && !device->directions,
> > + "this driver doesn't support generic slave capabilities reporting\n");
> > +
> > /* note: this only matters in the
> > * CONFIG_ASYNC_TX_ENABLE_CHANNEL_SWITCH=n case
> > */
>
> Probably you need to fix this ?
>
> https://bugzilla.kernel.org/show_bug.cgi?id=90801
My understanding is that the warning message is here on purpose to
instruct DMA driver authors to fix their code. So if anyone need to fix
anything, that would be you and Andy. That being said, I suspect Andy
already did that:
commit 029a40e97d0db269f4a7fc02b0f29f627f628309
Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Date: Fri Jan 2 16:17:24 2015 +0200
dmaengine: dw: provide DMA capabilities
Brian, can you please try the latest Linus or linux-next kernel (or
backport the above commit to your current kernel) and confirm that the
backtraces are no longer printed?
Thanks,
--
Jean Delvare
SUSE L3 Support
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
2015-02-19 8:09 ` Jean Delvare
@ 2015-02-19 8:53 ` Viresh Kumar
-1 siblings, 0 replies; 164+ messages in thread
From: Viresh Kumar @ 2015-02-19 8:53 UTC (permalink / raw)
To: Jean Delvare
Cc: Maxime Ripard, Brian Loften, Vinod Koul, dmaengine,
Linux Kernel Mailing List, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, Lars-Peter Clausen,
Andy Shevchenko
On 19 February 2015 at 13:39, Jean Delvare <jdelvare@suse.de> wrote:
> My understanding is that the warning message is here on purpose to
> instruct DMA driver authors to fix their code. So if anyone need to fix
> anything, that would be you and Andy. That being said, I suspect Andy
To some extent yes, but Andy has already tried to fix that for most of
the drivers and so I just pointed out that something is missed.
> already did that:
>
> commit 029a40e97d0db269f4a7fc02b0f29f627f628309
> Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> Date: Fri Jan 2 16:17:24 2015 +0200
>
> dmaengine: dw: provide DMA capabilities
>
> Brian, can you please try the latest Linus or linux-next kernel (or
> backport the above commit to your current kernel) and confirm that the
> backtraces are no longer printed?
Maybe yes.
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
@ 2015-02-19 8:53 ` Viresh Kumar
0 siblings, 0 replies; 164+ messages in thread
From: Viresh Kumar @ 2015-02-19 8:53 UTC (permalink / raw)
To: linux-arm-kernel
On 19 February 2015 at 13:39, Jean Delvare <jdelvare@suse.de> wrote:
> My understanding is that the warning message is here on purpose to
> instruct DMA driver authors to fix their code. So if anyone need to fix
> anything, that would be you and Andy. That being said, I suspect Andy
To some extent yes, but Andy has already tried to fix that for most of
the drivers and so I just pointed out that something is missed.
> already did that:
>
> commit 029a40e97d0db269f4a7fc02b0f29f627f628309
> Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> Date: Fri Jan 2 16:17:24 2015 +0200
>
> dmaengine: dw: provide DMA capabilities
>
> Brian, can you please try the latest Linus or linux-next kernel (or
> backport the above commit to your current kernel) and confirm that the
> backtraces are no longer printed?
Maybe yes.
^ permalink raw reply [flat|nested] 164+ messages in thread
* Re: [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
2015-02-19 8:53 ` Viresh Kumar
@ 2015-02-20 6:24 ` Vinod Koul
-1 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2015-02-20 6:24 UTC (permalink / raw)
To: Viresh Kumar
Cc: Jean Delvare, Maxime Ripard, Brian Loften, dmaengine,
Linux Kernel Mailing List, linux-arm-kernel, Laurent Pinchart,
Antoine Ténart, Russell King, Lars-Peter Clausen,
Andy Shevchenko
On Thu, Feb 19, 2015 at 02:23:27PM +0530, Viresh Kumar wrote:
> On 19 February 2015 at 13:39, Jean Delvare <jdelvare@suse.de> wrote:
> > My understanding is that the warning message is here on purpose to
> > instruct DMA driver authors to fix their code. So if anyone need to fix
> > anything, that would be you and Andy. That being said, I suspect Andy
>
> To some extent yes, but Andy has already tried to fix that for most of
> the drivers and so I just pointed out that something is missed.
>
> > already did that:
> >
> > commit 029a40e97d0db269f4a7fc02b0f29f627f628309
> > Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > Date: Fri Jan 2 16:17:24 2015 +0200
> >
> > dmaengine: dw: provide DMA capabilities
> >
> > Brian, can you please try the latest Linus or linux-next kernel (or
> > backport the above commit to your current kernel) and confirm that the
> > backtraces are no longer printed?
>
> Maybe yes.
This warning was intentionally added to help move driver use the right
method. And I have applied patch from Andy for this, so this should be no
longer seen on Linus's tree
--
~Vinod
^ permalink raw reply [flat|nested] 164+ messages in thread
* [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval
@ 2015-02-20 6:24 ` Vinod Koul
0 siblings, 0 replies; 164+ messages in thread
From: Vinod Koul @ 2015-02-20 6:24 UTC (permalink / raw)
To: linux-arm-kernel
On Thu, Feb 19, 2015 at 02:23:27PM +0530, Viresh Kumar wrote:
> On 19 February 2015 at 13:39, Jean Delvare <jdelvare@suse.de> wrote:
> > My understanding is that the warning message is here on purpose to
> > instruct DMA driver authors to fix their code. So if anyone need to fix
> > anything, that would be you and Andy. That being said, I suspect Andy
>
> To some extent yes, but Andy has already tried to fix that for most of
> the drivers and so I just pointed out that something is missed.
>
> > already did that:
> >
> > commit 029a40e97d0db269f4a7fc02b0f29f627f628309
> > Author: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> > Date: Fri Jan 2 16:17:24 2015 +0200
> >
> > dmaengine: dw: provide DMA capabilities
> >
> > Brian, can you please try the latest Linus or linux-next kernel (or
> > backport the above commit to your current kernel) and confirm that the
> > backtraces are no longer printed?
>
> Maybe yes.
This warning was intentionally added to help move driver use the right
method. And I have applied patch from Andy for this, so this should be no
longer seen on Linus's tree
--
~Vinod
^ permalink raw reply [flat|nested] 164+ messages in thread
end of thread, other threads:[~2015-02-20 6:27 UTC | newest]
Thread overview: 164+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2014-11-17 13:41 [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval Maxime Ripard
2014-11-17 13:41 ` Maxime Ripard
2014-11-17 13:41 ` [PATCH v5 01/61] serial: at91: Use dmaengine_slave_config API Maxime Ripard
2014-11-17 13:41 ` Maxime Ripard
2014-11-17 13:41 ` [PATCH v5 02/61] crypto: ux500: Use dmaengine_terminate_all API Maxime Ripard
2014-11-17 13:41 ` Maxime Ripard
2014-11-17 13:41 ` [PATCH v5 03/61] dmaengine: Make the destination abbreviation coherent Maxime Ripard
2014-11-17 13:41 ` Maxime Ripard
2014-11-17 13:41 ` [PATCH v5 04/61] dmaengine: Rework dma_chan_get Maxime Ripard
2014-11-17 13:41 ` Maxime Ripard
2014-11-17 13:41 ` [PATCH v5 05/61] dmaengine: Make channel allocation callbacks optional Maxime Ripard
2014-11-17 13:41 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 06/61] dmaengine: Introduce a device_config callback Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 07/61] dmaengine: split out pause/resume operations from device_control Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 08/61] dmaengine: Add device_terminate_all callback Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 09/61] dmaengine: Remove the need to declare device_control Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 10/61] dmaengine: Create a generic dma_slave_caps callback Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 11/61] dmaengine: pl08x: Split device_control Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 12/61] dmaengine: hdmac: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 13/61] dmaengine: at_xdmac: split device_control Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 14/61] dmaengine: bcm2835: Split device_control Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 15/61] dmaengine: coh901318: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 16/61] dmaengine: cppi41: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 17/61] dmaengine: jz4740: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 18/61] dmaengine: dw: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 19/61] dmaengine: edma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 20/61] dmaengine: ep93xx: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 21/61] dmaengine: fsl-edma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 22/61] dmaengine: imx: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 23/61] dmaengine: imx-sdma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 24/61] dmaengine: intel-mid-dma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 25/61] dmaengine: ipu-idmac: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 26/61] dmaengine: k3: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 27/61] dmaengine: mmp-pdma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 28/61] dmaengine: mmp-tdma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 29/61] dmaengine: moxart: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 30/61] dmaengine: fsl-dma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 31/61] dmaengine: mpc512x: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 32/61] dmaengine: mxs: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-12-05 14:29 ` Vinod Koul
2014-12-05 14:29 ` Vinod Koul
2014-11-17 13:42 ` [PATCH v5 33/61] dmaengine: nbpfaxi: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 34/61] dmaengine: omap: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-12-05 14:27 ` Vinod Koul
2014-12-05 14:27 ` Vinod Koul
2014-12-05 14:33 ` Maxime Ripard
2014-12-05 14:33 ` Maxime Ripard
2014-12-05 14:56 ` Vinod Koul
2014-12-05 14:56 ` Vinod Koul
2014-12-05 17:07 ` Maxime Ripard
2014-12-05 17:07 ` Maxime Ripard
2014-12-05 17:39 ` Vinod Koul
2014-12-05 17:39 ` Vinod Koul
2014-11-17 13:42 ` [PATCH v5 35/61] dmaengine: pl330: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 36/61] dmaengine: bam-dma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 37/61] dmaengine: s3c24xx: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 38/61] dmaengine: sa11x0: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 39/61] dmaengine: sh: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 40/61] dmaengine: sirf: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 41/61] dmaengine: sun6i: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 42/61] dmaengine: d40: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 43/61] dmaengine: tegra20: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 44/61] dmaengine: xilinx: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 45/61] dmaengine: mv_xor: Remove device_control Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 46/61] dmaengine: pch-dma: Rename device_control Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 47/61] dmaengine: td: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 48/61] dmaengine: txx9: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 49/61] dmaengine: rapidio: tsi721: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 50/61] dmaengine: at_xdmac: Declare slave capabilities for the generic code Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 51/61] dmaengine: bcm2835: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 52/61] dmaengine: fsl-edma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 53/61] dmaengine: edma: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 54/61] dmaengine: nbpfaxi: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 55/61] dmaengine: omap: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 56/61] dmaengine: pl330: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 57/61] dmaengine: sirf: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 58/61] dmaengine: sun6i: " Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 59/61] dmaengine: Add a warning for drivers not using the generic slave caps retrieval Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2015-02-19 7:57 ` Viresh Kumar
2015-02-19 7:57 ` Viresh Kumar
2015-02-19 8:09 ` Jean Delvare
2015-02-19 8:09 ` Jean Delvare
2015-02-19 8:53 ` Viresh Kumar
2015-02-19 8:53 ` Viresh Kumar
2015-02-20 6:24 ` Vinod Koul
2015-02-20 6:24 ` Vinod Koul
2014-11-17 13:42 ` [PATCH v5 60/61] dmaengine: Remove device_control and device_slave_caps Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-11-17 13:42 ` [PATCH v5 61/61] Documentation: dmaengine: Update the documentation Maxime Ripard
2014-11-17 13:42 ` Maxime Ripard
2014-12-08 6:17 ` [PATCH v5 00/61] dmaengine: Implement generic slave capabilities retrieval Vinod Koul
2014-12-08 6:17 ` Vinod Koul
2014-12-08 9:32 ` Ludovic Desroches
2014-12-08 9:32 ` Ludovic Desroches
2014-12-08 13:30 ` Vinod Koul
2014-12-08 13:30 ` Vinod Koul
2014-12-08 14:31 ` Maxime Ripard
2014-12-08 14:31 ` Maxime Ripard
2014-12-08 14:18 ` Maxime Ripard
2014-12-08 14:18 ` Maxime Ripard
2014-12-08 16:28 ` Vinod Koul
2014-12-08 16:28 ` Vinod Koul
2014-12-08 18:38 ` Maxime Ripard
2014-12-08 18:38 ` Maxime Ripard
2014-12-08 21:49 ` Lars-Peter Clausen
2014-12-08 21:49 ` Lars-Peter Clausen
2014-12-11 17:03 ` Maxime Ripard
2014-12-11 17:03 ` Maxime Ripard
2014-12-09 4:43 ` Vinod Koul
2014-12-09 4:43 ` Vinod Koul
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.