* [PATCH v2 1/4] mmc: mmci: Support any block sizes for ux500v2 and qcom variant
2014-08-19 11:13 ` [PATCH v2 0/4] mmc: mmci: sdio related fixes Srinivas Kandagatla
@ 2014-08-19 11:14 ` Srinivas Kandagatla
2014-08-19 12:37 ` Ulf Hansson
2014-08-19 11:14 ` [PATCH v2 2/4] mmc: mmci: move block size validation under relevant code Srinivas Kandagatla
` (3 subsequent siblings)
4 siblings, 1 reply; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-08-19 11:14 UTC (permalink / raw)
To: linux-mmc
Cc: Linus Walleij, Chris Ball, Ulf Hansson, Russell King,
linux-kernel, linux-arm-msm, Srinivas Kandagatla
From: Ulf Hansson <ulf.hansson@linaro.org>
For the ux500v2 variant of the PL18x block, any block sizes are
supported. This will make it possible to decrease data overhead
for SDIO transfers.
This patch is based on Ulf Hansson patch
http://www.spinics.net/lists/linux-mmc/msg12160.html
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
enabled this support on qcom variant.
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
drivers/mmc/host/mmci.c | 9 +++++++--
1 file changed, 7 insertions(+), 2 deletions(-)
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index c11cb05..3089fba 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -77,6 +77,7 @@ static unsigned int fmax = 515633;
* @qcom_fifo: enables qcom specific fifo pio read logic.
* @reversed_irq_handling: handle data irq before cmd irq.
* @qcom_dml: enables qcom specific dma glue for dma transfers.
+ * @any_blksize: true if block any sizes are supported
*/
struct variant_data {
unsigned int clkreg;
@@ -102,6 +103,7 @@ struct variant_data {
bool qcom_fifo;
bool reversed_irq_handling;
bool qcom_dml;
+ bool any_blksize;
};
static struct variant_data variant_arm = {
@@ -194,6 +196,7 @@ static struct variant_data variant_ux500v2 = {
.pwrreg_clkgate = true,
.busy_detect = true,
.pwrreg_nopower = true,
+ .any_blksize = true,
};
static struct variant_data variant_qcom = {
@@ -212,6 +215,7 @@ static struct variant_data variant_qcom = {
.explicit_mclk_control = true,
.qcom_fifo = true,
.qcom_dml = true,
+ .any_blksize = true,
};
static int mmci_card_busy(struct mmc_host *mmc)
@@ -239,10 +243,11 @@ static int mmci_card_busy(struct mmc_host *mmc)
static int mmci_validate_data(struct mmci_host *host,
struct mmc_data *data)
{
+ struct variant_data *variant = host->variant;
+
if (!data)
return 0;
-
- if (!is_power_of_2(data->blksz)) {
+ if (!is_power_of_2(data->blksz) && !variant->any_blksize) {
dev_err(mmc_dev(host->mmc),
"unsupported block size (%d bytes)\n", data->blksz);
return -EINVAL;
--
1.9.1
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH v2 1/4] mmc: mmci: Support any block sizes for ux500v2 and qcom variant
2014-08-19 11:14 ` [PATCH v2 1/4] mmc: mmci: Support any block sizes for ux500v2 and qcom variant Srinivas Kandagatla
@ 2014-08-19 12:37 ` Ulf Hansson
0 siblings, 0 replies; 34+ messages in thread
From: Ulf Hansson @ 2014-08-19 12:37 UTC (permalink / raw)
To: Srinivas Kandagatla
Cc: linux-mmc, Linus Walleij, Chris Ball, Russell King, linux-kernel,
linux-arm-msm
On 19 August 2014 13:14, Srinivas Kandagatla
<srinivas.kandagatla@linaro.org> wrote:
> From: Ulf Hansson <ulf.hansson@linaro.org>
>
> For the ux500v2 variant of the PL18x block, any block sizes are
> supported. This will make it possible to decrease data overhead
> for SDIO transfers.
>
> This patch is based on Ulf Hansson patch
> http://www.spinics.net/lists/linux-mmc/msg12160.html
>
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> enabled this support on qcom variant.
>
> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> ---
> drivers/mmc/host/mmci.c | 9 +++++++--
> 1 file changed, 7 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
> index c11cb05..3089fba 100644
> --- a/drivers/mmc/host/mmci.c
> +++ b/drivers/mmc/host/mmci.c
> @@ -77,6 +77,7 @@ static unsigned int fmax = 515633;
> * @qcom_fifo: enables qcom specific fifo pio read logic.
> * @reversed_irq_handling: handle data irq before cmd irq.
> * @qcom_dml: enables qcom specific dma glue for dma transfers.
> + * @any_blksize: true if block any sizes are supported
> */
> struct variant_data {
> unsigned int clkreg;
> @@ -102,6 +103,7 @@ struct variant_data {
> bool qcom_fifo;
> bool reversed_irq_handling;
> bool qcom_dml;
> + bool any_blksize;
> };
>
> static struct variant_data variant_arm = {
> @@ -194,6 +196,7 @@ static struct variant_data variant_ux500v2 = {
> .pwrreg_clkgate = true,
> .busy_detect = true,
> .pwrreg_nopower = true,
> + .any_blksize = true,
> };
There are some prerequisites of the data buffers to supports any block
size, at least for ux500.
Try read up on this discussion:
http://marc.info/?t=135005062400002&r=2&w=2
The conclusion from the above is that we need to adopt
mmci_pio_write() to handle corner cases.
Now, I suppose, unless someone objects, we could go ahead and merge
this patch. Then you will have to fixup mmci_pio_write() later on.
Kind regards
Uffe
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH v2 2/4] mmc: mmci: move block size validation under relevant code
2014-08-19 11:13 ` [PATCH v2 0/4] mmc: mmci: sdio related fixes Srinivas Kandagatla
2014-08-19 11:14 ` [PATCH v2 1/4] mmc: mmci: Support any block sizes for ux500v2 and qcom variant Srinivas Kandagatla
@ 2014-08-19 11:14 ` Srinivas Kandagatla
2014-08-19 11:55 ` Ulf Hansson
2014-08-19 11:15 ` [PATCH v2 3/4] mmc: mmci: Add sdio enable mask in variant data Srinivas Kandagatla
` (2 subsequent siblings)
4 siblings, 1 reply; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-08-19 11:14 UTC (permalink / raw)
To: linux-mmc
Cc: Linus Walleij, Chris Ball, Ulf Hansson, Russell King,
linux-kernel, linux-arm-msm, Srinivas Kandagatla
This code moves a BUG_ON condition to relevant if block, this check is
not necessary for IPs which can set any arbitrary block size in a given
range.
This patch is necessary for SDIO which sets block sizes that are exactly
not power of 2.
Original issue detected while testing WLAN ath6kl on Qualcomm APQ8064 SOC.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
---
drivers/mmc/host/mmci.c | 11 ++++++-----
1 file changed, 6 insertions(+), 5 deletions(-)
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index 3089fba..1c99195 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -800,15 +800,16 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
writel(timeout, base + MMCIDATATIMER);
writel(host->size, base + MMCIDATALENGTH);
- blksz_bits = ffs(data->blksz) - 1;
- BUG_ON(1 << blksz_bits != data->blksz);
- if (variant->blksz_datactrl16)
+ if (variant->blksz_datactrl16) {
datactrl = MCI_DPSM_ENABLE | (data->blksz << 16);
- else if (variant->blksz_datactrl4)
+ } else if (variant->blksz_datactrl4) {
datactrl = MCI_DPSM_ENABLE | (data->blksz << 4);
- else
+ } else {
+ blksz_bits = ffs(data->blksz) - 1;
+ BUG_ON(1 << blksz_bits != data->blksz);
datactrl = MCI_DPSM_ENABLE | blksz_bits << 4;
+ }
if (data->flags & MMC_DATA_READ)
datactrl |= MCI_DPSM_DIRECTION;
--
1.9.1
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH v2 2/4] mmc: mmci: move block size validation under relevant code
2014-08-19 11:14 ` [PATCH v2 2/4] mmc: mmci: move block size validation under relevant code Srinivas Kandagatla
@ 2014-08-19 11:55 ` Ulf Hansson
2014-08-19 12:08 ` Srinivas Kandagatla
0 siblings, 1 reply; 34+ messages in thread
From: Ulf Hansson @ 2014-08-19 11:55 UTC (permalink / raw)
To: Srinivas Kandagatla
Cc: linux-mmc, Linus Walleij, Chris Ball, Russell King, linux-kernel,
linux-arm-msm
On 19 August 2014 13:14, Srinivas Kandagatla
<srinivas.kandagatla@linaro.org> wrote:
> This code moves a BUG_ON condition to relevant if block, this check is
> not necessary for IPs which can set any arbitrary block size in a given
> range.
> This patch is necessary for SDIO which sets block sizes that are exactly
> not power of 2.
>
> Original issue detected while testing WLAN ath6kl on Qualcomm APQ8064 SOC.
>
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> ---
> drivers/mmc/host/mmci.c | 11 ++++++-----
> 1 file changed, 6 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
> index 3089fba..1c99195 100644
> --- a/drivers/mmc/host/mmci.c
> +++ b/drivers/mmc/host/mmci.c
> @@ -800,15 +800,16 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
> writel(timeout, base + MMCIDATATIMER);
> writel(host->size, base + MMCIDATALENGTH);
>
> - blksz_bits = ffs(data->blksz) - 1;
> - BUG_ON(1 << blksz_bits != data->blksz);
I don't like this BUG_ON at all, I would prefer if we remove it. The
original patch "mmc: mmci: Support any block sizes for ux500v2", did
so as well.
Now, if we still think removing it is fragile, let additional tests in
mmci_validate_data() and return and error code from there.
Kind regards
Uffe
>
> - if (variant->blksz_datactrl16)
> + if (variant->blksz_datactrl16) {
> datactrl = MCI_DPSM_ENABLE | (data->blksz << 16);
> - else if (variant->blksz_datactrl4)
> + } else if (variant->blksz_datactrl4) {
> datactrl = MCI_DPSM_ENABLE | (data->blksz << 4);
> - else
> + } else {
> + blksz_bits = ffs(data->blksz) - 1;
> + BUG_ON(1 << blksz_bits != data->blksz);
> datactrl = MCI_DPSM_ENABLE | blksz_bits << 4;
> + }
>
> if (data->flags & MMC_DATA_READ)
> datactrl |= MCI_DPSM_DIRECTION;
> --
> 1.9.1
>
^ permalink raw reply [flat|nested] 34+ messages in thread* Re: [PATCH v2 2/4] mmc: mmci: move block size validation under relevant code
2014-08-19 11:55 ` Ulf Hansson
@ 2014-08-19 12:08 ` Srinivas Kandagatla
0 siblings, 0 replies; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-08-19 12:08 UTC (permalink / raw)
To: Ulf Hansson
Cc: linux-mmc, Linus Walleij, Chris Ball, Russell King, linux-kernel,
linux-arm-msm
On 19/08/14 12:55, Ulf Hansson wrote:
>> writel(host->size, base + MMCIDATALENGTH);
>> >
>> >- blksz_bits = ffs(data->blksz) - 1;
>> >- BUG_ON(1 << blksz_bits != data->blksz);
> I don't like this BUG_ON at all, I would prefer if we remove it. The
> original patch "mmc: mmci: Support any block sizes for ux500v2", did
> so as well.
>
Yes, you are right this is redundant check, mmci_validate_data in
mmci_request should catch this error anyway. we can remove this check
totally and use your original patch.
--srini
> Now, if we still think removing it is fragile, let additional tests in
> mmci_validate_data() and return and error code from there.
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH v2 3/4] mmc: mmci: Add sdio enable mask in variant data
2014-08-19 11:13 ` [PATCH v2 0/4] mmc: mmci: sdio related fixes Srinivas Kandagatla
2014-08-19 11:14 ` [PATCH v2 1/4] mmc: mmci: Support any block sizes for ux500v2 and qcom variant Srinivas Kandagatla
2014-08-19 11:14 ` [PATCH v2 2/4] mmc: mmci: move block size validation under relevant code Srinivas Kandagatla
@ 2014-08-19 11:15 ` Srinivas Kandagatla
2014-08-19 11:15 ` [PATCH v2 4/4] mmc: mmci: rename sdio flag in vendor data to st_sdio Srinivas Kandagatla
2014-08-22 4:53 ` [PATCH v3 0/3] mmc: mmci: sdio related fixes Srinivas Kandagatla
4 siblings, 0 replies; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-08-19 11:15 UTC (permalink / raw)
To: linux-mmc
Cc: Linus Walleij, Chris Ball, Ulf Hansson, Russell King,
linux-kernel, linux-arm-msm, Srinivas Kandagatla
This patch adds sdio enable mask in variant data, SOCs like ST have
special bits in datactrl register to enable sdio. Unconditionally setting
this bit in this driver breaks other SOCs like Qualcomm which maps this
bits to something else, so making this enable bit to come from variant
data solves the issue.
Originally the issue is detected while testing WLAN ath6kl on Qualcomm
APQ8064.
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
---
drivers/mmc/host/mmci.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index 1c99195..fc08203 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -67,6 +67,7 @@ static unsigned int fmax = 515633;
* @blksz_datactrl16: true if Block size is at b16..b30 position in datactrl register
* @blksz_datactrl4: true if Block size is at b4..b16 position in datactrl
* register
+ * @datactrl_mask_sdio: SDIO enable mask in datactrl register
* @pwrreg_powerup: power up value for MMCIPOWER register
* @f_max: maximum clk frequency supported by the controller.
* @signal_direction: input/out direction of bus signals can be indicated
@@ -89,6 +90,7 @@ struct variant_data {
unsigned int fifohalfsize;
unsigned int data_cmd_enable;
unsigned int datactrl_mask_ddrmode;
+ unsigned int datactrl_mask_sdio;
bool sdio;
bool st_clkdiv;
bool blksz_datactrl16;
@@ -138,6 +140,7 @@ static struct variant_data variant_u300 = {
.clkreg_enable = MCI_ST_U300_HWFCEN,
.clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
.datalength_bits = 16,
+ .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
.sdio = true,
.pwrreg_powerup = MCI_PWR_ON,
.f_max = 100000000,
@@ -151,6 +154,7 @@ static struct variant_data variant_nomadik = {
.fifohalfsize = 8 * 4,
.clkreg = MCI_CLK_ENABLE,
.datalength_bits = 24,
+ .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
.sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
@@ -168,6 +172,7 @@ static struct variant_data variant_ux500 = {
.clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
.clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
.datalength_bits = 24,
+ .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
.sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
@@ -187,6 +192,7 @@ static struct variant_data variant_ux500v2 = {
.clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
.datactrl_mask_ddrmode = MCI_ST_DPSM_DDRMODE,
.datalength_bits = 24,
+ .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
.sdio = true,
.st_clkdiv = true,
.blksz_datactrl16 = true,
@@ -814,16 +820,10 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
if (data->flags & MMC_DATA_READ)
datactrl |= MCI_DPSM_DIRECTION;
- /* The ST Micro variants has a special bit to enable SDIO */
if (variant->sdio && host->mmc->card)
if (mmc_card_sdio(host->mmc->card)) {
- /*
- * The ST Micro variants has a special bit
- * to enable SDIO.
- */
u32 clk;
-
- datactrl |= MCI_ST_DPSM_SDIOEN;
+ datactrl |= variant->datactrl_mask_sdio;
/*
* The ST Micro variant for SDIO small write transfers
--
1.9.1
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH v2 4/4] mmc: mmci: rename sdio flag in vendor data to st_sdio
2014-08-19 11:13 ` [PATCH v2 0/4] mmc: mmci: sdio related fixes Srinivas Kandagatla
` (2 preceding siblings ...)
2014-08-19 11:15 ` [PATCH v2 3/4] mmc: mmci: Add sdio enable mask in variant data Srinivas Kandagatla
@ 2014-08-19 11:15 ` Srinivas Kandagatla
2014-08-22 4:53 ` [PATCH v3 0/3] mmc: mmci: sdio related fixes Srinivas Kandagatla
4 siblings, 0 replies; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-08-19 11:15 UTC (permalink / raw)
To: linux-mmc
Cc: Linus Walleij, Chris Ball, Ulf Hansson, Russell King,
linux-kernel, linux-arm-msm, Srinivas Kandagatla
This patch renames sdio flag in vendor data to st_sdio, as this flag is
only used to enable ST specific sdio setup. This will also ensure that
the ST specfic setup is not done on other vendor like Qualcomm.
Originally the issue was detected while testing WLAN ath6kl on IFC6410
board with APQ8064 SOC.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
---
drivers/mmc/host/mmci.c | 48 ++++++++++++++++++++++++------------------------
1 file changed, 24 insertions(+), 24 deletions(-)
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index fc08203..26aa443 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -61,7 +61,7 @@ static unsigned int fmax = 515633;
* @fifohalfsize: number of bytes that can be written when MCI_TXFIFOHALFEMPTY
* is asserted (likewise for RX)
* @data_cmd_enable: enable value for data commands.
- * @sdio: variant supports SDIO
+ * @st_sdio: enable ST specific SDIO logic
* @st_clkdiv: true if using a ST-specific clock divider algorithm
* @datactrl_mask_ddrmode: ddr mode mask in datactrl register.
* @blksz_datactrl16: true if Block size is at b16..b30 position in datactrl register
@@ -91,7 +91,7 @@ struct variant_data {
unsigned int data_cmd_enable;
unsigned int datactrl_mask_ddrmode;
unsigned int datactrl_mask_sdio;
- bool sdio;
+ bool st_sdio;
bool st_clkdiv;
bool blksz_datactrl16;
bool blksz_datactrl4;
@@ -141,7 +141,7 @@ static struct variant_data variant_u300 = {
.clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
.datalength_bits = 16,
.datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
- .sdio = true,
+ .st_sdio = true,
.pwrreg_powerup = MCI_PWR_ON,
.f_max = 100000000,
.signal_direction = true,
@@ -155,7 +155,7 @@ static struct variant_data variant_nomadik = {
.clkreg = MCI_CLK_ENABLE,
.datalength_bits = 24,
.datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
- .sdio = true,
+ .st_sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
.f_max = 100000000,
@@ -173,7 +173,7 @@ static struct variant_data variant_ux500 = {
.clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
.datalength_bits = 24,
.datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
- .sdio = true,
+ .st_sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
.f_max = 100000000,
@@ -193,7 +193,7 @@ static struct variant_data variant_ux500v2 = {
.datactrl_mask_ddrmode = MCI_ST_DPSM_DDRMODE,
.datalength_bits = 24,
.datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
- .sdio = true,
+ .st_sdio = true,
.st_clkdiv = true,
.blksz_datactrl16 = true,
.pwrreg_powerup = MCI_PWR_ON,
@@ -820,26 +820,26 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
if (data->flags & MMC_DATA_READ)
datactrl |= MCI_DPSM_DIRECTION;
- if (variant->sdio && host->mmc->card)
- if (mmc_card_sdio(host->mmc->card)) {
- u32 clk;
- datactrl |= variant->datactrl_mask_sdio;
+ if (host->mmc->card && mmc_card_sdio(host->mmc->card)) {
+ u32 clk;
- /*
- * The ST Micro variant for SDIO small write transfers
- * needs to have clock H/W flow control disabled,
- * otherwise the transfer will not start. The threshold
- * depends on the rate of MCLK.
- */
- if (data->flags & MMC_DATA_WRITE &&
- (host->size < 8 ||
- (host->size <= 8 && host->mclk > 50000000)))
- clk = host->clk_reg & ~variant->clkreg_enable;
- else
- clk = host->clk_reg | variant->clkreg_enable;
+ datactrl |= variant->datactrl_mask_sdio;
- mmci_write_clkreg(host, clk);
- }
+ /*
+ * The ST Micro variant for SDIO small write transfers
+ * needs to have clock H/W flow control disabled,
+ * otherwise the transfer will not start. The threshold
+ * depends on the rate of MCLK.
+ */
+ if (variant->st_sdio && data->flags & MMC_DATA_WRITE &&
+ (host->size < 8 ||
+ (host->size <= 8 && host->mclk > 50000000)))
+ clk = host->clk_reg & ~variant->clkreg_enable;
+ else
+ clk = host->clk_reg | variant->clkreg_enable;
+
+ mmci_write_clkreg(host, clk);
+ }
if (host->mmc->ios.timing == MMC_TIMING_UHS_DDR50 ||
host->mmc->ios.timing == MMC_TIMING_MMC_DDR52)
--
1.9.1
^ permalink raw reply related [flat|nested] 34+ messages in thread* [PATCH v3 0/3] mmc: mmci: sdio related fixes
2014-08-19 11:13 ` [PATCH v2 0/4] mmc: mmci: sdio related fixes Srinivas Kandagatla
` (3 preceding siblings ...)
2014-08-19 11:15 ` [PATCH v2 4/4] mmc: mmci: rename sdio flag in vendor data to st_sdio Srinivas Kandagatla
@ 2014-08-22 4:53 ` Srinivas Kandagatla
2014-08-22 4:54 ` [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant Srinivas Kandagatla
` (3 more replies)
4 siblings, 4 replies; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-08-22 4:53 UTC (permalink / raw)
To: linux-mmc
Cc: Linus Walleij, Chris Ball, Ulf Hansson, Russell King,
linux-kernel, linux-arm-msm, Srinivas Kandagatla
This patchset fixes few sdio related issues encountered while testing
WLAN ath6kl via SDIO on IFC6410 board with Qualcomm APQ8064 SOC.
Patch: "mmc: mmci: Support any block sizes for ux500v2 and qcom variant" is
a very old patch by Ulf to support IPs which support any size of block sizes.
http://www.spinics.net/lists/linux-mmc/msg12160.html I modified the subject
line to include qcom.
Patch fixes below issues reported while testing sdio.
The issue was ath6kl driver was issuing 12 bytes and 24 bytes reads
which are caught as part of the error handing in the driver and
resulting in failures.
Patch "mmc: mmci: Add sdio enable mask in variant data" adds extra
variant parameter to enable sdio. This makes mmci driver more flexible.
Patch "mmc: mmci: rename sdio flag in vendor data to st_sdio" renames sdio
flag in vendor data to st_sdio, as this flag is only used to setup st
specific sdio logic.
All these patches are tested on IFC6410 board with ath6kl WLAN via SDIO.
Thanks to Linus W, Ulf annd Russell for comments since RFC.
Changes since v2:
- removed "mmc: mmci: move block size validation under relevant code" patch
as this is already fixed by original patch from Ulf.
Changes since RFC:
- moved sdio flag to st_sdio to simplify the checks.
- use Ulf's patch to address IP's which support anysize blocks.
Thanks,
srini
Srinivas Kandagatla (2):
mmc: mmci: Add sdio enable mask in variant data
mmc: mmci: rename sdio flag in vendor data to st_sdio
Ulf Hansson (1):
mmc: mmci: Support any block sizes for ux500v2 and qcom variant
drivers/mmc/host/mmci.c | 68 ++++++++++++++++++++++++++-----------------------
1 file changed, 36 insertions(+), 32 deletions(-)
--
1.9.1
^ permalink raw reply [flat|nested] 34+ messages in thread* [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant
2014-08-22 4:53 ` [PATCH v3 0/3] mmc: mmci: sdio related fixes Srinivas Kandagatla
@ 2014-08-22 4:54 ` Srinivas Kandagatla
2014-09-10 7:58 ` Ulf Hansson
` (2 more replies)
2014-08-22 4:54 ` [PATCH v3 2/3] mmc: mmci: Add sdio enable mask in variant data Srinivas Kandagatla
` (2 subsequent siblings)
3 siblings, 3 replies; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-08-22 4:54 UTC (permalink / raw)
To: linux-mmc
Cc: Linus Walleij, Chris Ball, Ulf Hansson, Russell King,
linux-kernel, linux-arm-msm, Srinivas Kandagatla
From: Ulf Hansson <ulf.hansson@linaro.org>
For the ux500v2 variant of the PL18x block, any block sizes are
supported. This will make it possible to decrease data overhead
for SDIO transfers.
This patch is based on Ulf Hansson patch
http://www.spinics.net/lists/linux-mmc/msg12160.html
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
enabled this support on qcom variant.
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
---
drivers/mmc/host/mmci.c | 10 +++++++---
1 file changed, 7 insertions(+), 3 deletions(-)
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index c11cb05..533ad2b 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -77,6 +77,7 @@ static unsigned int fmax = 515633;
* @qcom_fifo: enables qcom specific fifo pio read logic.
* @reversed_irq_handling: handle data irq before cmd irq.
* @qcom_dml: enables qcom specific dma glue for dma transfers.
+ * @any_blksize: true if block any sizes are supported
*/
struct variant_data {
unsigned int clkreg;
@@ -102,6 +103,7 @@ struct variant_data {
bool qcom_fifo;
bool reversed_irq_handling;
bool qcom_dml;
+ bool any_blksize;
};
static struct variant_data variant_arm = {
@@ -194,6 +196,7 @@ static struct variant_data variant_ux500v2 = {
.pwrreg_clkgate = true,
.busy_detect = true,
.pwrreg_nopower = true,
+ .any_blksize = true,
};
static struct variant_data variant_qcom = {
@@ -212,6 +215,7 @@ static struct variant_data variant_qcom = {
.explicit_mclk_control = true,
.qcom_fifo = true,
.qcom_dml = true,
+ .any_blksize = true,
};
static int mmci_card_busy(struct mmc_host *mmc)
@@ -239,10 +243,11 @@ static int mmci_card_busy(struct mmc_host *mmc)
static int mmci_validate_data(struct mmci_host *host,
struct mmc_data *data)
{
+ struct variant_data *variant = host->variant;
+
if (!data)
return 0;
-
- if (!is_power_of_2(data->blksz)) {
+ if (!is_power_of_2(data->blksz) && !variant->any_blksize) {
dev_err(mmc_dev(host->mmc),
"unsupported block size (%d bytes)\n", data->blksz);
return -EINVAL;
@@ -796,7 +801,6 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
writel(host->size, base + MMCIDATALENGTH);
blksz_bits = ffs(data->blksz) - 1;
- BUG_ON(1 << blksz_bits != data->blksz);
if (variant->blksz_datactrl16)
datactrl = MCI_DPSM_ENABLE | (data->blksz << 16);
--
1.9.1
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant
2014-08-22 4:54 ` [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant Srinivas Kandagatla
@ 2014-09-10 7:58 ` Ulf Hansson
2014-09-10 9:07 ` Srinivas Kandagatla
2015-07-09 21:40 ` Bjorn Andersson
2019-03-05 22:25 ` Linus Walleij
2 siblings, 1 reply; 34+ messages in thread
From: Ulf Hansson @ 2014-09-10 7:58 UTC (permalink / raw)
To: Srinivas Kandagatla
Cc: linux-mmc, Linus Walleij, Chris Ball, Russell King, linux-kernel,
linux-arm-msm
On 22 August 2014 06:54, Srinivas Kandagatla
<srinivas.kandagatla@linaro.org> wrote:
> From: Ulf Hansson <ulf.hansson@linaro.org>
>
> For the ux500v2 variant of the PL18x block, any block sizes are
> supported. This will make it possible to decrease data overhead
> for SDIO transfers.
>
> This patch is based on Ulf Hansson patch
> http://www.spinics.net/lists/linux-mmc/msg12160.html
>
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> enabled this support on qcom variant.
>
> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
I am not sure how to handle this patch.
It will as you say in the cover letter for this patchset, improve
situations for the ath6kl driver when it's issuing 12 bytes and 24
bytes reads and solve those issues.
On the other hand, as stated earlier mmci_pio_write need to be fixed
to have full support for any block size. That applies to the qcom
variant as well.
For ux500, I am sure this won't cause any regressions since the cw1200
isn't probed. Also, I am not sure the "any block size" support is even
enabled for that driver.
How about, that we add a comment in the pio_write function describing
that we need to fix it for "SDIO any block size" support? And leave
that as a future improvement?
Kind regards
Uffe
> ---
> drivers/mmc/host/mmci.c | 10 +++++++---
> 1 file changed, 7 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
> index c11cb05..533ad2b 100644
> --- a/drivers/mmc/host/mmci.c
> +++ b/drivers/mmc/host/mmci.c
> @@ -77,6 +77,7 @@ static unsigned int fmax = 515633;
> * @qcom_fifo: enables qcom specific fifo pio read logic.
> * @reversed_irq_handling: handle data irq before cmd irq.
> * @qcom_dml: enables qcom specific dma glue for dma transfers.
> + * @any_blksize: true if block any sizes are supported
> */
> struct variant_data {
> unsigned int clkreg;
> @@ -102,6 +103,7 @@ struct variant_data {
> bool qcom_fifo;
> bool reversed_irq_handling;
> bool qcom_dml;
> + bool any_blksize;
> };
>
> static struct variant_data variant_arm = {
> @@ -194,6 +196,7 @@ static struct variant_data variant_ux500v2 = {
> .pwrreg_clkgate = true,
> .busy_detect = true,
> .pwrreg_nopower = true,
> + .any_blksize = true,
> };
>
> static struct variant_data variant_qcom = {
> @@ -212,6 +215,7 @@ static struct variant_data variant_qcom = {
> .explicit_mclk_control = true,
> .qcom_fifo = true,
> .qcom_dml = true,
> + .any_blksize = true,
> };
>
> static int mmci_card_busy(struct mmc_host *mmc)
> @@ -239,10 +243,11 @@ static int mmci_card_busy(struct mmc_host *mmc)
> static int mmci_validate_data(struct mmci_host *host,
> struct mmc_data *data)
> {
> + struct variant_data *variant = host->variant;
> +
> if (!data)
> return 0;
> -
> - if (!is_power_of_2(data->blksz)) {
> + if (!is_power_of_2(data->blksz) && !variant->any_blksize) {
> dev_err(mmc_dev(host->mmc),
> "unsupported block size (%d bytes)\n", data->blksz);
> return -EINVAL;
> @@ -796,7 +801,6 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
> writel(host->size, base + MMCIDATALENGTH);
>
> blksz_bits = ffs(data->blksz) - 1;
> - BUG_ON(1 << blksz_bits != data->blksz);
>
> if (variant->blksz_datactrl16)
> datactrl = MCI_DPSM_ENABLE | (data->blksz << 16);
> --
> 1.9.1
>
^ permalink raw reply [flat|nested] 34+ messages in thread* Re: [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant
2014-09-10 7:58 ` Ulf Hansson
@ 2014-09-10 9:07 ` Srinivas Kandagatla
2014-09-10 10:51 ` Ulf Hansson
0 siblings, 1 reply; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-09-10 9:07 UTC (permalink / raw)
To: Ulf Hansson
Cc: linux-mmc, Linus Walleij, Chris Ball, Russell King, linux-kernel,
linux-arm-msm
Hi Ulf,
On 10/09/14 08:58, Ulf Hansson wrote:
> On 22 August 2014 06:54, Srinivas Kandagatla
> <srinivas.kandagatla@linaro.org> wrote:
>> From: Ulf Hansson <ulf.hansson@linaro.org>
>>
>> For the ux500v2 variant of the PL18x block, any block sizes are
>> supported. This will make it possible to decrease data overhead
>> for SDIO transfers.
>>
>> This patch is based on Ulf Hansson patch
>> http://www.spinics.net/lists/linux-mmc/msg12160.html
>>
>> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
>> enabled this support on qcom variant.
>>
>> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
>> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
>
> I am not sure how to handle this patch.
>
> It will as you say in the cover letter for this patchset, improve
> situations for the ath6kl driver when it's issuing 12 bytes and 24
> bytes reads and solve those issues.
>
> On the other hand, as stated earlier mmci_pio_write need to be fixed
> to have full support for any block size. That applies to the qcom
> variant as well.
looking at current mmci_pio_write, I see it can support any block sizes
as it is. Unless Am missing something obvious.
block size aligned to 4 is taken care in the code and is straight forward.
block sizes not aligned to 4 are also partly taken care in pio_write and
partly by programming blksz in datactrl register. However Am not sure if
it was safe to handle the buffer pointer out of its boundary.
>
> For ux500, I am sure this won't cause any regressions since the cw1200
> isn't probed. Also, I am not sure the "any block size" support is even
> enabled for that driver.
>
> How about, that we add a comment in the pio_write function describing
> that we need to fix it for "SDIO any block size" support? And leave
> that as a future improvement?
Is below patch any good?
With the below patch It should be possible to address the case where
buffer passed the length is handled safely. If you ok with the approach
I can send a patch as RFC.
----cut here----
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index 264c947..8480c02 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -1134,7 +1134,20 @@ static int mmci_pio_write(struct mmci_host *host,
char *buffer, unsigned int rem
* byte become a 32bit write, 7 bytes will be two
* 32bit writes etc.
*/
- iowrite32_rep(base + MMCIFIFO, ptr, (count + 3) >> 2);
+ if (unlikely(count & 0x3)) {
+ unsigned char buf[4] = {0, };
+
+ if (count < 4) {
+ memcpy(buf, ptr, count);
+ iowrite32_rep(base + MMCIFIFO, buf, 1);
+ } else {
+ iowrite32_rep(base + MMCIFIFO, ptr,
count >> 2);
+ memcpy(buf, ptr + (count & ~0x3), count
& 0x3);
+ iowrite32_rep(base + MMCIFIFO, buf, 1);
+ }
+ } else {
+ iowrite32_rep(base + MMCIFIFO, ptr, count >> 2);
+ }
ptr += count;
remain -= count;
----cut here----
thanks,
srini
>
> Kind regards
> Uffe
>
>> ---
>> drivers/mmc/host/mmci.c | 10 +++++++---
>> 1 file changed, 7 insertions(+), 3 deletions(-)
>>
>> diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
>> index c11cb05..533ad2b 100644
>> --- a/drivers/mmc/host/mmci.c
>> +++ b/drivers/mmc/host/mmci.c
>> @@ -77,6 +77,7 @@ static unsigned int fmax = 515633;
>> * @qcom_fifo: enables qcom specific fifo pio read logic.
>> * @reversed_irq_handling: handle data irq before cmd irq.
>> * @qcom_dml: enables qcom specific dma glue for dma transfers.
>> + * @any_blksize: true if block any sizes are supported
>> */
>> struct variant_data {
>> unsigned int clkreg;
>> @@ -102,6 +103,7 @@ struct variant_data {
>> bool qcom_fifo;
>> bool reversed_irq_handling;
>> bool qcom_dml;
>> + bool any_blksize;
>> };
>>
>> static struct variant_data variant_arm = {
>> @@ -194,6 +196,7 @@ static struct variant_data variant_ux500v2 = {
>> .pwrreg_clkgate = true,
>> .busy_detect = true,
>> .pwrreg_nopower = true,
>> + .any_blksize = true,
>> };
>>
>> static struct variant_data variant_qcom = {
>> @@ -212,6 +215,7 @@ static struct variant_data variant_qcom = {
>> .explicit_mclk_control = true,
>> .qcom_fifo = true,
>> .qcom_dml = true,
>> + .any_blksize = true,
>> };
>>
>> static int mmci_card_busy(struct mmc_host *mmc)
>> @@ -239,10 +243,11 @@ static int mmci_card_busy(struct mmc_host *mmc)
>> static int mmci_validate_data(struct mmci_host *host,
>> struct mmc_data *data)
>> {
>> + struct variant_data *variant = host->variant;
>> +
>> if (!data)
>> return 0;
>> -
>> - if (!is_power_of_2(data->blksz)) {
>> + if (!is_power_of_2(data->blksz) && !variant->any_blksize) {
>> dev_err(mmc_dev(host->mmc),
>> "unsupported block size (%d bytes)\n", data->blksz);
>> return -EINVAL;
>> @@ -796,7 +801,6 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
>> writel(host->size, base + MMCIDATALENGTH);
>>
>> blksz_bits = ffs(data->blksz) - 1;
>> - BUG_ON(1 << blksz_bits != data->blksz);
>>
>> if (variant->blksz_datactrl16)
>> datactrl = MCI_DPSM_ENABLE | (data->blksz << 16);
>> --
>> 1.9.1
>>
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant
2014-09-10 9:07 ` Srinivas Kandagatla
@ 2014-09-10 10:51 ` Ulf Hansson
0 siblings, 0 replies; 34+ messages in thread
From: Ulf Hansson @ 2014-09-10 10:51 UTC (permalink / raw)
To: Srinivas Kandagatla
Cc: linux-mmc, Linus Walleij, Chris Ball, Russell King, linux-kernel,
linux-arm-msm
On 10 September 2014 11:07, Srinivas Kandagatla
<srinivas.kandagatla@linaro.org> wrote:
> Hi Ulf,
>
> On 10/09/14 08:58, Ulf Hansson wrote:
>>
>> On 22 August 2014 06:54, Srinivas Kandagatla
>> <srinivas.kandagatla@linaro.org> wrote:
>>>
>>> From: Ulf Hansson <ulf.hansson@linaro.org>
>>>
>>> For the ux500v2 variant of the PL18x block, any block sizes are
>>> supported. This will make it possible to decrease data overhead
>>> for SDIO transfers.
>>>
>>> This patch is based on Ulf Hansson patch
>>> http://www.spinics.net/lists/linux-mmc/msg12160.html
>>>
>>> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
>>> enabled this support on qcom variant.
>>>
>>> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
>>> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
>>
>>
>> I am not sure how to handle this patch.
>>
>> It will as you say in the cover letter for this patchset, improve
>> situations for the ath6kl driver when it's issuing 12 bytes and 24
>> bytes reads and solve those issues.
>>
>> On the other hand, as stated earlier mmci_pio_write need to be fixed
>> to have full support for any block size. That applies to the qcom
>> variant as well.
>
> looking at current mmci_pio_write, I see it can support any block sizes as
> it is. Unless Am missing something obvious.
>
> block size aligned to 4 is taken care in the code and is straight forward.
> block sizes not aligned to 4 are also partly taken care in pio_write and
> partly by programming blksz in datactrl register. However Am not sure if it
> was safe to handle the buffer pointer out of its boundary.
There are some prerequisites of the data buffers to supports any block
size. Make sure you read up on this discussion to really understand the problem.
http://marc.info/?t=135005062400002&r=2&w=2
>>
>>
>> For ux500, I am sure this won't cause any regressions since the cw1200
>> isn't probed. Also, I am not sure the "any block size" support is even
>> enabled for that driver.
>>
>> How about, that we add a comment in the pio_write function describing
>> that we need to fix it for "SDIO any block size" support? And leave
>> that as a future improvement?
>
>
> Is below patch any good?
> With the below patch It should be possible to address the case where buffer
> passed the length is handled safely. If you ok with the approach I can send
> a patch as RFC.
Please go ahead an send a patch :-) This has been on my todo list
forever, I would happily like to remove it from there.
Kind regards
Uffe
^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant
2014-08-22 4:54 ` [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant Srinivas Kandagatla
2014-09-10 7:58 ` Ulf Hansson
@ 2015-07-09 21:40 ` Bjorn Andersson
2019-03-05 22:25 ` Linus Walleij
2 siblings, 0 replies; 34+ messages in thread
From: Bjorn Andersson @ 2015-07-09 21:40 UTC (permalink / raw)
To: Srinivas Kandagatla
Cc: linux-mmc, Linus Walleij, Chris Ball, Ulf Hansson, Russell King,
linux-kernel, linux-arm-msm
On Thu, Aug 21, 2014 at 9:54 PM, Srinivas Kandagatla
<srinivas.kandagatla@linaro.org> wrote:
> From: Ulf Hansson <ulf.hansson@linaro.org>
>
> For the ux500v2 variant of the PL18x block, any block sizes are
> supported. This will make it possible to decrease data overhead
> for SDIO transfers.
>
> This patch is based on Ulf Hansson patch
> http://www.spinics.net/lists/linux-mmc/msg12160.html
>
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> enabled this support on qcom variant.
>
> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Srinivas,
Pulled this in to my tree today and got the bcm4339 in the Sony Xperia
Z3 to join the internet :)
So, did you pursue this further?
Regards,
Bjorn
^ permalink raw reply [flat|nested] 34+ messages in thread
* Re: [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant
2014-08-22 4:54 ` [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant Srinivas Kandagatla
2014-09-10 7:58 ` Ulf Hansson
2015-07-09 21:40 ` Bjorn Andersson
@ 2019-03-05 22:25 ` Linus Walleij
2019-03-06 17:03 ` Bjorn Andersson
2 siblings, 1 reply; 34+ messages in thread
From: Linus Walleij @ 2019-03-05 22:25 UTC (permalink / raw)
To: Srinivas Kandagatla, Bjorn Andersson, Niklas Cassel
Cc: linux-mmc, Chris Ball, Ulf Hansson, Russell King, linux-kernel,
linux-arm-msm
Sorry for top-posting, patch included for context.
I saw today that Brian has the Nexus 5 working with Broadcom/Cypress
4339 over SDIO on the MMCI.
Brian are you managing to use the WLAN without this patch or is this
something you stacked up to make it work? (If you want context I
can forward the rest of the conversation where we discuss how to
fix it properly.)
I discussed this patch several times with Ulf and I think Niklas looked
at it too, but we never got around to fix it up properly.
Yours,
Linus Walleij
On Fri, Aug 22, 2014 at 6:54 AM Srinivas Kandagatla
<srinivas.kandagatla@linaro.org> wrote:
> From: Ulf Hansson <ulf.hansson@linaro.org>
>
> For the ux500v2 variant of the PL18x block, any block sizes are
> supported. This will make it possible to decrease data overhead
> for SDIO transfers.
>
> This patch is based on Ulf Hansson patch
> http://www.spinics.net/lists/linux-mmc/msg12160.html
>
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> enabled this support on qcom variant.
>
> Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> ---
> drivers/mmc/host/mmci.c | 10 +++++++---
> 1 file changed, 7 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
> index c11cb05..533ad2b 100644
> --- a/drivers/mmc/host/mmci.c
> +++ b/drivers/mmc/host/mmci.c
> @@ -77,6 +77,7 @@ static unsigned int fmax = 515633;
> * @qcom_fifo: enables qcom specific fifo pio read logic.
> * @reversed_irq_handling: handle data irq before cmd irq.
> * @qcom_dml: enables qcom specific dma glue for dma transfers.
> + * @any_blksize: true if block any sizes are supported
> */
> struct variant_data {
> unsigned int clkreg;
> @@ -102,6 +103,7 @@ struct variant_data {
> bool qcom_fifo;
> bool reversed_irq_handling;
> bool qcom_dml;
> + bool any_blksize;
> };
>
> static struct variant_data variant_arm = {
> @@ -194,6 +196,7 @@ static struct variant_data variant_ux500v2 = {
> .pwrreg_clkgate = true,
> .busy_detect = true,
> .pwrreg_nopower = true,
> + .any_blksize = true,
> };
>
> static struct variant_data variant_qcom = {
> @@ -212,6 +215,7 @@ static struct variant_data variant_qcom = {
> .explicit_mclk_control = true,
> .qcom_fifo = true,
> .qcom_dml = true,
> + .any_blksize = true,
> };
>
> static int mmci_card_busy(struct mmc_host *mmc)
> @@ -239,10 +243,11 @@ static int mmci_card_busy(struct mmc_host *mmc)
> static int mmci_validate_data(struct mmci_host *host,
> struct mmc_data *data)
> {
> + struct variant_data *variant = host->variant;
> +
> if (!data)
> return 0;
> -
> - if (!is_power_of_2(data->blksz)) {
> + if (!is_power_of_2(data->blksz) && !variant->any_blksize) {
> dev_err(mmc_dev(host->mmc),
> "unsupported block size (%d bytes)\n", data->blksz);
> return -EINVAL;
> @@ -796,7 +801,6 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
> writel(host->size, base + MMCIDATALENGTH);
>
> blksz_bits = ffs(data->blksz) - 1;
> - BUG_ON(1 << blksz_bits != data->blksz);
>
> if (variant->blksz_datactrl16)
> datactrl = MCI_DPSM_ENABLE | (data->blksz << 16);
> --
> 1.9.1
>
^ permalink raw reply [flat|nested] 34+ messages in thread* Re: [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant
2019-03-05 22:25 ` Linus Walleij
@ 2019-03-06 17:03 ` Bjorn Andersson
0 siblings, 0 replies; 34+ messages in thread
From: Bjorn Andersson @ 2019-03-06 17:03 UTC (permalink / raw)
To: Linus Walleij
Cc: Srinivas Kandagatla, Niklas Cassel, linux-mmc, Chris Ball,
Ulf Hansson, Russell King, linux-kernel, linux-arm-msm
On Tue 05 Mar 14:25 PST 2019, Linus Walleij wrote:
> Sorry for top-posting, patch included for context.
>
> I saw today that Brian has the Nexus 5 working with Broadcom/Cypress
> 4339 over SDIO on the MMCI.
>
> Brian are you managing to use the WLAN without this patch or is this
> something you stacked up to make it work? (If you want context I
> can forward the rest of the conversation where we discuss how to
> fix it properly.)
>
Brian's Nexus 5 is on 8974 where the MMCI block has become the sdhci-msm
block, so while being a superset (afaict) of the old MMCI block we use
the sdhci driver instead.
> I discussed this patch several times with Ulf and I think Niklas looked
> at it too, but we never got around to fix it up properly.
>
This is still needed for at least 8064 to have working sdio wifi (and
perhaps your 8060?).
Regards,
Bjorn
> Yours,
> Linus Walleij
>
>
> On Fri, Aug 22, 2014 at 6:54 AM Srinivas Kandagatla
> <srinivas.kandagatla@linaro.org> wrote:
>
> > From: Ulf Hansson <ulf.hansson@linaro.org>
> >
> > For the ux500v2 variant of the PL18x block, any block sizes are
> > supported. This will make it possible to decrease data overhead
> > for SDIO transfers.
> >
> > This patch is based on Ulf Hansson patch
> > http://www.spinics.net/lists/linux-mmc/msg12160.html
> >
> > Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> > enabled this support on qcom variant.
> >
> > Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
> > Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> > ---
> > drivers/mmc/host/mmci.c | 10 +++++++---
> > 1 file changed, 7 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
> > index c11cb05..533ad2b 100644
> > --- a/drivers/mmc/host/mmci.c
> > +++ b/drivers/mmc/host/mmci.c
> > @@ -77,6 +77,7 @@ static unsigned int fmax = 515633;
> > * @qcom_fifo: enables qcom specific fifo pio read logic.
> > * @reversed_irq_handling: handle data irq before cmd irq.
> > * @qcom_dml: enables qcom specific dma glue for dma transfers.
> > + * @any_blksize: true if block any sizes are supported
> > */
> > struct variant_data {
> > unsigned int clkreg;
> > @@ -102,6 +103,7 @@ struct variant_data {
> > bool qcom_fifo;
> > bool reversed_irq_handling;
> > bool qcom_dml;
> > + bool any_blksize;
> > };
> >
> > static struct variant_data variant_arm = {
> > @@ -194,6 +196,7 @@ static struct variant_data variant_ux500v2 = {
> > .pwrreg_clkgate = true,
> > .busy_detect = true,
> > .pwrreg_nopower = true,
> > + .any_blksize = true,
> > };
> >
> > static struct variant_data variant_qcom = {
> > @@ -212,6 +215,7 @@ static struct variant_data variant_qcom = {
> > .explicit_mclk_control = true,
> > .qcom_fifo = true,
> > .qcom_dml = true,
> > + .any_blksize = true,
> > };
> >
> > static int mmci_card_busy(struct mmc_host *mmc)
> > @@ -239,10 +243,11 @@ static int mmci_card_busy(struct mmc_host *mmc)
> > static int mmci_validate_data(struct mmci_host *host,
> > struct mmc_data *data)
> > {
> > + struct variant_data *variant = host->variant;
> > +
> > if (!data)
> > return 0;
> > -
> > - if (!is_power_of_2(data->blksz)) {
> > + if (!is_power_of_2(data->blksz) && !variant->any_blksize) {
> > dev_err(mmc_dev(host->mmc),
> > "unsupported block size (%d bytes)\n", data->blksz);
> > return -EINVAL;
> > @@ -796,7 +801,6 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
> > writel(host->size, base + MMCIDATALENGTH);
> >
> > blksz_bits = ffs(data->blksz) - 1;
> > - BUG_ON(1 << blksz_bits != data->blksz);
> >
> > if (variant->blksz_datactrl16)
> > datactrl = MCI_DPSM_ENABLE | (data->blksz << 16);
> > --
> > 1.9.1
> >
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH v3 2/3] mmc: mmci: Add sdio enable mask in variant data
2014-08-22 4:53 ` [PATCH v3 0/3] mmc: mmci: sdio related fixes Srinivas Kandagatla
2014-08-22 4:54 ` [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant Srinivas Kandagatla
@ 2014-08-22 4:54 ` Srinivas Kandagatla
2014-09-10 7:47 ` Ulf Hansson
2014-08-22 4:55 ` [PATCH v3 3/3] mmc: mmci: rename sdio flag in vendor data to st_sdio Srinivas Kandagatla
2014-09-09 8:34 ` [PATCH] mmc: mmci_qcom_dml: fix is never less than zero warning Srinivas Kandagatla
3 siblings, 1 reply; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-08-22 4:54 UTC (permalink / raw)
To: linux-mmc
Cc: Linus Walleij, Chris Ball, Ulf Hansson, Russell King,
linux-kernel, linux-arm-msm, Srinivas Kandagatla
This patch adds sdio enable mask in variant data, SOCs like ST have
special bits in datactrl register to enable sdio. Unconditionally setting
this bit in this driver breaks other SOCs like Qualcomm which maps this
bits to something else, so making this enable bit to come from variant
data solves the issue.
Originally the issue is detected while testing WLAN ath6kl on Qualcomm
APQ8064.
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
---
drivers/mmc/host/mmci.c | 14 +++++++-------
1 file changed, 7 insertions(+), 7 deletions(-)
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index 533ad2b..a25759e 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -67,6 +67,7 @@ static unsigned int fmax = 515633;
* @blksz_datactrl16: true if Block size is at b16..b30 position in datactrl register
* @blksz_datactrl4: true if Block size is at b4..b16 position in datactrl
* register
+ * @datactrl_mask_sdio: SDIO enable mask in datactrl register
* @pwrreg_powerup: power up value for MMCIPOWER register
* @f_max: maximum clk frequency supported by the controller.
* @signal_direction: input/out direction of bus signals can be indicated
@@ -89,6 +90,7 @@ struct variant_data {
unsigned int fifohalfsize;
unsigned int data_cmd_enable;
unsigned int datactrl_mask_ddrmode;
+ unsigned int datactrl_mask_sdio;
bool sdio;
bool st_clkdiv;
bool blksz_datactrl16;
@@ -138,6 +140,7 @@ static struct variant_data variant_u300 = {
.clkreg_enable = MCI_ST_U300_HWFCEN,
.clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
.datalength_bits = 16,
+ .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
.sdio = true,
.pwrreg_powerup = MCI_PWR_ON,
.f_max = 100000000,
@@ -151,6 +154,7 @@ static struct variant_data variant_nomadik = {
.fifohalfsize = 8 * 4,
.clkreg = MCI_CLK_ENABLE,
.datalength_bits = 24,
+ .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
.sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
@@ -168,6 +172,7 @@ static struct variant_data variant_ux500 = {
.clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
.clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
.datalength_bits = 24,
+ .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
.sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
@@ -187,6 +192,7 @@ static struct variant_data variant_ux500v2 = {
.clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
.datactrl_mask_ddrmode = MCI_ST_DPSM_DDRMODE,
.datalength_bits = 24,
+ .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
.sdio = true,
.st_clkdiv = true,
.blksz_datactrl16 = true,
@@ -812,16 +818,10 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
if (data->flags & MMC_DATA_READ)
datactrl |= MCI_DPSM_DIRECTION;
- /* The ST Micro variants has a special bit to enable SDIO */
if (variant->sdio && host->mmc->card)
if (mmc_card_sdio(host->mmc->card)) {
- /*
- * The ST Micro variants has a special bit
- * to enable SDIO.
- */
u32 clk;
-
- datactrl |= MCI_ST_DPSM_SDIOEN;
+ datactrl |= variant->datactrl_mask_sdio;
/*
* The ST Micro variant for SDIO small write transfers
--
1.9.1
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH v3 2/3] mmc: mmci: Add sdio enable mask in variant data
2014-08-22 4:54 ` [PATCH v3 2/3] mmc: mmci: Add sdio enable mask in variant data Srinivas Kandagatla
@ 2014-09-10 7:47 ` Ulf Hansson
0 siblings, 0 replies; 34+ messages in thread
From: Ulf Hansson @ 2014-09-10 7:47 UTC (permalink / raw)
To: Srinivas Kandagatla
Cc: linux-mmc, Linus Walleij, Chris Ball, Russell King, linux-kernel,
linux-arm-msm
On 22 August 2014 06:54, Srinivas Kandagatla
<srinivas.kandagatla@linaro.org> wrote:
> This patch adds sdio enable mask in variant data, SOCs like ST have
> special bits in datactrl register to enable sdio. Unconditionally setting
> this bit in this driver breaks other SOCs like Qualcomm which maps this
> bits to something else, so making this enable bit to come from variant
> data solves the issue.
>
> Originally the issue is detected while testing WLAN ath6kl on Qualcomm
> APQ8064.
>
> Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Thanks! Applied for next!
Kind regards
Uffe
> ---
> drivers/mmc/host/mmci.c | 14 +++++++-------
> 1 file changed, 7 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
> index 533ad2b..a25759e 100644
> --- a/drivers/mmc/host/mmci.c
> +++ b/drivers/mmc/host/mmci.c
> @@ -67,6 +67,7 @@ static unsigned int fmax = 515633;
> * @blksz_datactrl16: true if Block size is at b16..b30 position in datactrl register
> * @blksz_datactrl4: true if Block size is at b4..b16 position in datactrl
> * register
> + * @datactrl_mask_sdio: SDIO enable mask in datactrl register
> * @pwrreg_powerup: power up value for MMCIPOWER register
> * @f_max: maximum clk frequency supported by the controller.
> * @signal_direction: input/out direction of bus signals can be indicated
> @@ -89,6 +90,7 @@ struct variant_data {
> unsigned int fifohalfsize;
> unsigned int data_cmd_enable;
> unsigned int datactrl_mask_ddrmode;
> + unsigned int datactrl_mask_sdio;
> bool sdio;
> bool st_clkdiv;
> bool blksz_datactrl16;
> @@ -138,6 +140,7 @@ static struct variant_data variant_u300 = {
> .clkreg_enable = MCI_ST_U300_HWFCEN,
> .clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
> .datalength_bits = 16,
> + .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
> .sdio = true,
> .pwrreg_powerup = MCI_PWR_ON,
> .f_max = 100000000,
> @@ -151,6 +154,7 @@ static struct variant_data variant_nomadik = {
> .fifohalfsize = 8 * 4,
> .clkreg = MCI_CLK_ENABLE,
> .datalength_bits = 24,
> + .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
> .sdio = true,
> .st_clkdiv = true,
> .pwrreg_powerup = MCI_PWR_ON,
> @@ -168,6 +172,7 @@ static struct variant_data variant_ux500 = {
> .clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
> .clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
> .datalength_bits = 24,
> + .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
> .sdio = true,
> .st_clkdiv = true,
> .pwrreg_powerup = MCI_PWR_ON,
> @@ -187,6 +192,7 @@ static struct variant_data variant_ux500v2 = {
> .clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
> .datactrl_mask_ddrmode = MCI_ST_DPSM_DDRMODE,
> .datalength_bits = 24,
> + .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
> .sdio = true,
> .st_clkdiv = true,
> .blksz_datactrl16 = true,
> @@ -812,16 +818,10 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
> if (data->flags & MMC_DATA_READ)
> datactrl |= MCI_DPSM_DIRECTION;
>
> - /* The ST Micro variants has a special bit to enable SDIO */
> if (variant->sdio && host->mmc->card)
> if (mmc_card_sdio(host->mmc->card)) {
> - /*
> - * The ST Micro variants has a special bit
> - * to enable SDIO.
> - */
> u32 clk;
> -
> - datactrl |= MCI_ST_DPSM_SDIOEN;
> + datactrl |= variant->datactrl_mask_sdio;
>
> /*
> * The ST Micro variant for SDIO small write transfers
> --
> 1.9.1
>
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH v3 3/3] mmc: mmci: rename sdio flag in vendor data to st_sdio
2014-08-22 4:53 ` [PATCH v3 0/3] mmc: mmci: sdio related fixes Srinivas Kandagatla
2014-08-22 4:54 ` [PATCH v3 1/3] mmc: mmci: Support any block sizes for ux500v2 and qcom variant Srinivas Kandagatla
2014-08-22 4:54 ` [PATCH v3 2/3] mmc: mmci: Add sdio enable mask in variant data Srinivas Kandagatla
@ 2014-08-22 4:55 ` Srinivas Kandagatla
2014-09-10 7:48 ` Ulf Hansson
2014-09-09 8:34 ` [PATCH] mmc: mmci_qcom_dml: fix is never less than zero warning Srinivas Kandagatla
3 siblings, 1 reply; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-08-22 4:55 UTC (permalink / raw)
To: linux-mmc
Cc: Linus Walleij, Chris Ball, Ulf Hansson, Russell King,
linux-kernel, linux-arm-msm, Srinivas Kandagatla
This patch renames sdio flag in vendor data to st_sdio, as this flag is
only used to enable ST specific sdio setup. This will also ensure that
the ST specfic setup is not done on other vendor like Qualcomm.
Originally the issue was detected while testing WLAN ath6kl on IFC6410
board with APQ8064 SOC.
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
---
drivers/mmc/host/mmci.c | 48 ++++++++++++++++++++++++------------------------
1 file changed, 24 insertions(+), 24 deletions(-)
diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
index a25759e..264c947 100644
--- a/drivers/mmc/host/mmci.c
+++ b/drivers/mmc/host/mmci.c
@@ -61,7 +61,7 @@ static unsigned int fmax = 515633;
* @fifohalfsize: number of bytes that can be written when MCI_TXFIFOHALFEMPTY
* is asserted (likewise for RX)
* @data_cmd_enable: enable value for data commands.
- * @sdio: variant supports SDIO
+ * @st_sdio: enable ST specific SDIO logic
* @st_clkdiv: true if using a ST-specific clock divider algorithm
* @datactrl_mask_ddrmode: ddr mode mask in datactrl register.
* @blksz_datactrl16: true if Block size is at b16..b30 position in datactrl register
@@ -91,7 +91,7 @@ struct variant_data {
unsigned int data_cmd_enable;
unsigned int datactrl_mask_ddrmode;
unsigned int datactrl_mask_sdio;
- bool sdio;
+ bool st_sdio;
bool st_clkdiv;
bool blksz_datactrl16;
bool blksz_datactrl4;
@@ -141,7 +141,7 @@ static struct variant_data variant_u300 = {
.clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
.datalength_bits = 16,
.datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
- .sdio = true,
+ .st_sdio = true,
.pwrreg_powerup = MCI_PWR_ON,
.f_max = 100000000,
.signal_direction = true,
@@ -155,7 +155,7 @@ static struct variant_data variant_nomadik = {
.clkreg = MCI_CLK_ENABLE,
.datalength_bits = 24,
.datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
- .sdio = true,
+ .st_sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
.f_max = 100000000,
@@ -173,7 +173,7 @@ static struct variant_data variant_ux500 = {
.clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
.datalength_bits = 24,
.datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
- .sdio = true,
+ .st_sdio = true,
.st_clkdiv = true,
.pwrreg_powerup = MCI_PWR_ON,
.f_max = 100000000,
@@ -193,7 +193,7 @@ static struct variant_data variant_ux500v2 = {
.datactrl_mask_ddrmode = MCI_ST_DPSM_DDRMODE,
.datalength_bits = 24,
.datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
- .sdio = true,
+ .st_sdio = true,
.st_clkdiv = true,
.blksz_datactrl16 = true,
.pwrreg_powerup = MCI_PWR_ON,
@@ -818,26 +818,26 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
if (data->flags & MMC_DATA_READ)
datactrl |= MCI_DPSM_DIRECTION;
- if (variant->sdio && host->mmc->card)
- if (mmc_card_sdio(host->mmc->card)) {
- u32 clk;
- datactrl |= variant->datactrl_mask_sdio;
+ if (host->mmc->card && mmc_card_sdio(host->mmc->card)) {
+ u32 clk;
- /*
- * The ST Micro variant for SDIO small write transfers
- * needs to have clock H/W flow control disabled,
- * otherwise the transfer will not start. The threshold
- * depends on the rate of MCLK.
- */
- if (data->flags & MMC_DATA_WRITE &&
- (host->size < 8 ||
- (host->size <= 8 && host->mclk > 50000000)))
- clk = host->clk_reg & ~variant->clkreg_enable;
- else
- clk = host->clk_reg | variant->clkreg_enable;
+ datactrl |= variant->datactrl_mask_sdio;
- mmci_write_clkreg(host, clk);
- }
+ /*
+ * The ST Micro variant for SDIO small write transfers
+ * needs to have clock H/W flow control disabled,
+ * otherwise the transfer will not start. The threshold
+ * depends on the rate of MCLK.
+ */
+ if (variant->st_sdio && data->flags & MMC_DATA_WRITE &&
+ (host->size < 8 ||
+ (host->size <= 8 && host->mclk > 50000000)))
+ clk = host->clk_reg & ~variant->clkreg_enable;
+ else
+ clk = host->clk_reg | variant->clkreg_enable;
+
+ mmci_write_clkreg(host, clk);
+ }
if (host->mmc->ios.timing == MMC_TIMING_UHS_DDR50 ||
host->mmc->ios.timing == MMC_TIMING_MMC_DDR52)
--
1.9.1
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH v3 3/3] mmc: mmci: rename sdio flag in vendor data to st_sdio
2014-08-22 4:55 ` [PATCH v3 3/3] mmc: mmci: rename sdio flag in vendor data to st_sdio Srinivas Kandagatla
@ 2014-09-10 7:48 ` Ulf Hansson
0 siblings, 0 replies; 34+ messages in thread
From: Ulf Hansson @ 2014-09-10 7:48 UTC (permalink / raw)
To: Srinivas Kandagatla
Cc: linux-mmc, Linus Walleij, Chris Ball, Russell King, linux-kernel,
linux-arm-msm
On 22 August 2014 06:55, Srinivas Kandagatla
<srinivas.kandagatla@linaro.org> wrote:
> This patch renames sdio flag in vendor data to st_sdio, as this flag is
> only used to enable ST specific sdio setup. This will also ensure that
> the ST specfic setup is not done on other vendor like Qualcomm.
>
> Originally the issue was detected while testing WLAN ath6kl on IFC6410
> board with APQ8064 SOC.
>
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Thanks! Applied for next!
Kind regards
Uffe
> ---
> drivers/mmc/host/mmci.c | 48 ++++++++++++++++++++++++------------------------
> 1 file changed, 24 insertions(+), 24 deletions(-)
>
> diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
> index a25759e..264c947 100644
> --- a/drivers/mmc/host/mmci.c
> +++ b/drivers/mmc/host/mmci.c
> @@ -61,7 +61,7 @@ static unsigned int fmax = 515633;
> * @fifohalfsize: number of bytes that can be written when MCI_TXFIFOHALFEMPTY
> * is asserted (likewise for RX)
> * @data_cmd_enable: enable value for data commands.
> - * @sdio: variant supports SDIO
> + * @st_sdio: enable ST specific SDIO logic
> * @st_clkdiv: true if using a ST-specific clock divider algorithm
> * @datactrl_mask_ddrmode: ddr mode mask in datactrl register.
> * @blksz_datactrl16: true if Block size is at b16..b30 position in datactrl register
> @@ -91,7 +91,7 @@ struct variant_data {
> unsigned int data_cmd_enable;
> unsigned int datactrl_mask_ddrmode;
> unsigned int datactrl_mask_sdio;
> - bool sdio;
> + bool st_sdio;
> bool st_clkdiv;
> bool blksz_datactrl16;
> bool blksz_datactrl4;
> @@ -141,7 +141,7 @@ static struct variant_data variant_u300 = {
> .clkreg_8bit_bus_enable = MCI_ST_8BIT_BUS,
> .datalength_bits = 16,
> .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
> - .sdio = true,
> + .st_sdio = true,
> .pwrreg_powerup = MCI_PWR_ON,
> .f_max = 100000000,
> .signal_direction = true,
> @@ -155,7 +155,7 @@ static struct variant_data variant_nomadik = {
> .clkreg = MCI_CLK_ENABLE,
> .datalength_bits = 24,
> .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
> - .sdio = true,
> + .st_sdio = true,
> .st_clkdiv = true,
> .pwrreg_powerup = MCI_PWR_ON,
> .f_max = 100000000,
> @@ -173,7 +173,7 @@ static struct variant_data variant_ux500 = {
> .clkreg_neg_edge_enable = MCI_ST_UX500_NEG_EDGE,
> .datalength_bits = 24,
> .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
> - .sdio = true,
> + .st_sdio = true,
> .st_clkdiv = true,
> .pwrreg_powerup = MCI_PWR_ON,
> .f_max = 100000000,
> @@ -193,7 +193,7 @@ static struct variant_data variant_ux500v2 = {
> .datactrl_mask_ddrmode = MCI_ST_DPSM_DDRMODE,
> .datalength_bits = 24,
> .datactrl_mask_sdio = MCI_ST_DPSM_SDIOEN,
> - .sdio = true,
> + .st_sdio = true,
> .st_clkdiv = true,
> .blksz_datactrl16 = true,
> .pwrreg_powerup = MCI_PWR_ON,
> @@ -818,26 +818,26 @@ static void mmci_start_data(struct mmci_host *host, struct mmc_data *data)
> if (data->flags & MMC_DATA_READ)
> datactrl |= MCI_DPSM_DIRECTION;
>
> - if (variant->sdio && host->mmc->card)
> - if (mmc_card_sdio(host->mmc->card)) {
> - u32 clk;
> - datactrl |= variant->datactrl_mask_sdio;
> + if (host->mmc->card && mmc_card_sdio(host->mmc->card)) {
> + u32 clk;
>
> - /*
> - * The ST Micro variant for SDIO small write transfers
> - * needs to have clock H/W flow control disabled,
> - * otherwise the transfer will not start. The threshold
> - * depends on the rate of MCLK.
> - */
> - if (data->flags & MMC_DATA_WRITE &&
> - (host->size < 8 ||
> - (host->size <= 8 && host->mclk > 50000000)))
> - clk = host->clk_reg & ~variant->clkreg_enable;
> - else
> - clk = host->clk_reg | variant->clkreg_enable;
> + datactrl |= variant->datactrl_mask_sdio;
>
> - mmci_write_clkreg(host, clk);
> - }
> + /*
> + * The ST Micro variant for SDIO small write transfers
> + * needs to have clock H/W flow control disabled,
> + * otherwise the transfer will not start. The threshold
> + * depends on the rate of MCLK.
> + */
> + if (variant->st_sdio && data->flags & MMC_DATA_WRITE &&
> + (host->size < 8 ||
> + (host->size <= 8 && host->mclk > 50000000)))
> + clk = host->clk_reg & ~variant->clkreg_enable;
> + else
> + clk = host->clk_reg | variant->clkreg_enable;
> +
> + mmci_write_clkreg(host, clk);
> + }
>
> if (host->mmc->ios.timing == MMC_TIMING_UHS_DDR50 ||
> host->mmc->ios.timing == MMC_TIMING_MMC_DDR52)
> --
> 1.9.1
>
^ permalink raw reply [flat|nested] 34+ messages in thread
* [PATCH] mmc: mmci_qcom_dml: fix is never less than zero warning.
2014-08-22 4:53 ` [PATCH v3 0/3] mmc: mmci: sdio related fixes Srinivas Kandagatla
` (2 preceding siblings ...)
2014-08-22 4:55 ` [PATCH v3 3/3] mmc: mmci: rename sdio flag in vendor data to st_sdio Srinivas Kandagatla
@ 2014-09-09 8:34 ` Srinivas Kandagatla
2014-09-09 11:53 ` Ulf Hansson
3 siblings, 1 reply; 34+ messages in thread
From: Srinivas Kandagatla @ 2014-09-09 8:34 UTC (permalink / raw)
To: linux-mmc
Cc: Chris Ball, Ulf Hansson, linux-kernel, linux-arm-msm,
Srinivas Kandagatla
This patch fixes a typo which resulted in 'is never less
than zero warning' reported by static checker.
drivers/mmc/host/mmci_qcom_dml.c:131 dml_hw_init()
warn: unsigned 'producer_id' is never less than zero.
drivers/mmc/host/mmci_qcom_dml.c
121 /* Initialize the dml hardware connected to SD Card controller */
122 int dml_hw_init(struct mmci_host *host, struct device_node *np)
123 {
124 u32 config;
125 void __iomem *base;
126 u32 consumer_id, producer_id;
127
128 consumer_id = of_get_dml_pipe_index(np, "tx");
129 producer_id = of_get_dml_pipe_index(np, "rx");
130
131 if (producer_id < 0 || consumer_id < 0)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
132
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
---
Hi Ulf,
This patch is on top of https://git.linaro.org/people/ulf.hansson/mmc.git next branch
to fix up a warning reported by static checker tool.
Are you ok to take this patch as it is or do you think I should resend the
original DML patch with this fix?
thanks,
srini
drivers/mmc/host/mmci_qcom_dml.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/mmc/host/mmci_qcom_dml.c b/drivers/mmc/host/mmci_qcom_dml.c
index 300ef50..2b7fc37 100644
--- a/drivers/mmc/host/mmci_qcom_dml.c
+++ b/drivers/mmc/host/mmci_qcom_dml.c
@@ -123,7 +123,7 @@ int dml_hw_init(struct mmci_host *host, struct device_node *np)
{
u32 config;
void __iomem *base;
- u32 consumer_id, producer_id;
+ int consumer_id, producer_id;
consumer_id = of_get_dml_pipe_index(np, "tx");
producer_id = of_get_dml_pipe_index(np, "rx");
--
1.9.1
^ permalink raw reply related [flat|nested] 34+ messages in thread* Re: [PATCH] mmc: mmci_qcom_dml: fix is never less than zero warning.
2014-09-09 8:34 ` [PATCH] mmc: mmci_qcom_dml: fix is never less than zero warning Srinivas Kandagatla
@ 2014-09-09 11:53 ` Ulf Hansson
0 siblings, 0 replies; 34+ messages in thread
From: Ulf Hansson @ 2014-09-09 11:53 UTC (permalink / raw)
To: Srinivas Kandagatla; +Cc: linux-mmc, Chris Ball, linux-kernel, linux-arm-msm
On 9 September 2014 10:34, Srinivas Kandagatla
<srinivas.kandagatla@linaro.org> wrote:
> This patch fixes a typo which resulted in 'is never less
> than zero warning' reported by static checker.
>
> drivers/mmc/host/mmci_qcom_dml.c:131 dml_hw_init()
> warn: unsigned 'producer_id' is never less than zero.
>
> drivers/mmc/host/mmci_qcom_dml.c
> 121 /* Initialize the dml hardware connected to SD Card controller */
> 122 int dml_hw_init(struct mmci_host *host, struct device_node *np)
> 123 {
> 124 u32 config;
> 125 void __iomem *base;
> 126 u32 consumer_id, producer_id;
> 127
> 128 consumer_id = of_get_dml_pipe_index(np, "tx");
> 129 producer_id = of_get_dml_pipe_index(np, "rx");
> 130
> 131 if (producer_id < 0 || consumer_id < 0)
> ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
> 132
>
> Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
> Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
> ---
> Hi Ulf,
>
> This patch is on top of https://git.linaro.org/people/ulf.hansson/mmc.git next branch
> to fix up a warning reported by static checker tool.
>
> Are you ok to take this patch as it is or do you think I should resend the
> original DML patch with this fix?
>
I am fine with this, I will fold it into the original patch. Applied for next.
Kind regards
Uffe
> thanks,
> srini
>
> drivers/mmc/host/mmci_qcom_dml.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/mmc/host/mmci_qcom_dml.c b/drivers/mmc/host/mmci_qcom_dml.c
> index 300ef50..2b7fc37 100644
> --- a/drivers/mmc/host/mmci_qcom_dml.c
> +++ b/drivers/mmc/host/mmci_qcom_dml.c
> @@ -123,7 +123,7 @@ int dml_hw_init(struct mmci_host *host, struct device_node *np)
> {
> u32 config;
> void __iomem *base;
> - u32 consumer_id, producer_id;
> + int consumer_id, producer_id;
>
> consumer_id = of_get_dml_pipe_index(np, "tx");
> producer_id = of_get_dml_pipe_index(np, "rx");
> --
> 1.9.1
>
^ permalink raw reply [flat|nested] 34+ messages in thread