[RFC PATCH v8 0/10] Add external dma support for Synopsys MSHC

[RFC PATCH v8 0/10] Add external dma support for Synopsys MSHC

[Shawn Lin]

Synopsys DesignWare mobile storage host controller supports three
types of transfer mode: pio, internal dma and external dma. However,
dw_mmc can only supports pio and internal dma now. Thus some platforms
using dw-mshc integrated with generic dma can't work in dma mode. So we
submit this patch to achieve it.

And the config option, CONFIG_MMC_DW_IDMAC, was added by Will Newton
(commit:f95f3850) for the first version of dw_mmc and never be touched since
then. At that time dt-bindings hadn't been introduced into dw_mmc yet means
we should select CONFIG_MMC_DW_IDMAC to enable internal dma mode at compile
time. Nowadays, device-tree helps us to support a variety of boards with one
kernel. That's why we need to remove it and decide the transfer mode by reading
dw_mmc's HCON reg at runtime.

This RFC patch needs lots of ACKs. I know it's hard, but it does need someone
to make the running.

Patch does the following things:
- remove CONFIG_MMC_DW_IDMAC config option
- add bindings for edmac used by synopsys-dw-mshc
  at runtime
- add edmac support for synopsys-dw-mshc

Patch is based on next of git://git.linaro.org/people/ulf.hansson/mmc

Test emmc throughput on my platform with edmac support and without edmac support(pio only)
iozone -L64 -S32 -azecwI -+n -r4k -r64k -r128k -s1g -i0 -i1 -i2 -f datafile -Rb out.xls > /mnt/result.txt
(light cpu loading, Direct IO, fixed line size, all pattern recycle, 1GB data in total)
 ___________________________________________________________
|                   external dma mode                       |
|-----------------------------------------------------------|
|blksz | Random Read | Random Write | Seq Read   | Seq Write|
|-----------------------------------------------------------|
|4kB   |  13953kB/s  |    8602kB/s  | 13672kB/s  |  9785kB/s|
|-----------------------------------------------------------|
|64kB  |  46058kB/s  |   24794kB/s  | 48058kB/s  | 25418kB/s|
|-----------------------------------------------------------|
|128kB |  57026kB/s  |   35117kB/s  | 57375kB/s  | 35183kB/s|
|-----------------------------------------------------------|
                           VS
 ___________________________________________________________
|                          pio mode                         |
|-----------------------------------------------------------|
|blksz | Random Read  | Random Write | Seq Read  | Seq Write|
|-----------------------------------------------------------|
|4kB   |  11720kB/s   |    8644kB/s  | 11549kB/s |  9624kB/s|
|-----------------------------------------------------------|
|64kB  |  21869kB/s   |   24414kB/s  | 22031kB/s | 27986kB/s|
|-----------------------------------------------------------|
|128kB |  23718kB/s   |   34495kB/s  | 24698kB/s | 34637kB/s|
|-----------------------------------------------------------|


Changes in v8:
- remove trans_mode variable
- remove unnecessary dma_ops check
- remove unnecessary comment
- fix coding style based on latest ulf's next
- add Acked-by: Jaehoon Chung <jh80.chung@samsung.com>
  for HCON's changes

Changes in v7:
- rebased on Ulf's next
- combine condition state
- elaborate more about DMA_INTERFACE
- define some macro for DMA_INERFACE value
- spilt HCON ops' changes into another patch

Changes in v6:
- add trans_mode condition for IDMAC initialization
  suggested by Heiko
- re-test my patch on rk3188 platform and update commit msg
- update performance of pio vs edmac in cover letter

Changes in v5:
- add the title of cover letter
- fix typo of comment
- add macro for reading HCON register
- add "Acked-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>" for exynos_defconfig patch
- add "Acked-by: Vineet Gupta <vgupta@synopsys.com>" for axs10x_defconfig patch
- add "Acked-by: Govindraj Raja <govindraj.raja@imgtec.com>" and
  "Acked-by: Ralf Baechle <ralf@linux-mips.org>" for pistachio_defconfig patch
- add "Acked-by: Joachim Eastwood <manabian@gmail.com>" for lpc18xx_defconfig patch
- add "Acked-by: Wei Xu <xuwei5@hisilicon.com>" for hisi_defconfig patch
- rebase on "https://github.com/jh80chung/dw-mmc.git tags/dw-mmc-for-ulf-v4.2" for merging easily

Changes in v4:
- remove "host->trans_mode" and use "host->use_dma" to indicate
  transfer mode.
- remove all bt-bindings' changes since we don't need new properities.
- check transfer mode at runtime by reading HCON reg
- spilt defconfig changes for each sub-architecture
- fix the title of cover letter
- reuse some code for reducing code size

Changes in v3:
- choose transfer mode at runtime
- remove all CONFIG_MMC_DW_IDMAC config option
- add supports-idmac property for some platforms

Changes in v2:
- Fix typo of dev_info msg
- remove unused dmach from declaration of dw_mci_dma_slave

Shawn Lin (10):
  mmc: dw_mmc: Add external dma interface support
  mmc: dw_mmc: use macro for HCON register operations
  Documentation: synopsys-dw-mshc: add bindings for idmac and edmac
  mips: pistachio_defconfig: remove CONFIG_MMC_DW_IDMAC
  arc: axs10x_defconfig: remove CONFIG_MMC_DW_IDMAC
  arm: exynos_defconfig: remove CONFIG_MMC_DW_IDMAC
  arm: hisi_defconfig: remove CONFIG_MMC_DW_IDMAC
  arm: lpc18xx_defconfig: remove CONFIG_MMC_DW_IDMAC
  arm: multi_v7_defconfig: remove CONFIG_MMC_DW_IDMAC
  arm: zx_defconfig: remove CONFIG_MMC_DW_IDMAC

 .../devicetree/bindings/mmc/synopsys-dw-mshc.txt   |  25 ++
 arch/arc/configs/axs101_defconfig                  |   1 -
 arch/arc/configs/axs103_defconfig                  |   1 -
 arch/arc/configs/axs103_smp_defconfig              |   1 -
 arch/arm/configs/exynos_defconfig                  |   1 -
 arch/arm/configs/hisi_defconfig                    |   1 -
 arch/arm/configs/lpc18xx_defconfig                 |   1 -
 arch/arm/configs/multi_v7_defconfig                |   1 -
 arch/arm/configs/zx_defconfig                      |   1 -
 arch/mips/configs/pistachio_defconfig              |   1 -
 drivers/mmc/host/Kconfig                           |  11 +-
 drivers/mmc/host/dw_mmc-pltfm.c                    |   2 +
 drivers/mmc/host/dw_mmc.c                          | 272 +++++++++++++++++----
 drivers/mmc/host/dw_mmc.h                          |   9 +
 include/linux/mmc/dw_mmc.h                         |  23 +-
 15 files changed, 276 insertions(+), 75 deletions(-)

-- 
2.3.7

[RFC PATCH v8 01/10] mmc: dw_mmc: Add external dma interface support

[Shawn Lin]

DesignWare MMC Controller can supports two types of DMA
mode: external dma and internal dma. We get a RK312x platform
integrated dw_mmc and ARM pl330 dma controller. This patch add
edmac ops to support these platforms. I've tested it on RK31xx
platform with edmac mode and RK3288 platform with idmac mode.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>

---

Changes in v8:
- remove trans_mode variable
- remove unnecessary dma_ops check
- remove unnecessary comment
- fix coding style based on latest ulf's next
- add Acked-by: Jaehoon Chung <jh80.chung@samsung.com>
  for HCON's changes

Changes in v7:
- rebased on Ulf's next
- combine condition state
- elaborate more about DMA_INTERFACE
- define some macro for DMA_INERFACE value
- spilt HCON ops' changes into another patch

Changes in v6:
- add trans_mode condition for IDMAC initialization
  suggested by Heiko
- re-test my patch on rk3188 platform and update commit msg
- update performance of pio vs edmac in cover letter

Changes in v5:
- add the title of cover letter
- fix typo of comment
- add macro for reading HCON register
- add "Acked-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>" for exynos_defconfig patch
- add "Acked-by: Vineet Gupta <vgupta@synopsys.com>" for axs10x_defconfig patch
- add "Acked-by: Govindraj Raja <govindraj.raja@imgtec.com>" and
  "Acked-by: Ralf Baechle <ralf@linux-mips.org>" for pistachio_defconfig patch
- add "Acked-by: Joachim Eastwood <manabian@gmail.com>" for lpc18xx_defconfig patch
- add "Acked-by: Wei Xu <xuwei5@hisilicon.com>" for hisi_defconfig patch
- rebase on "https://github.com/jh80chung/dw-mmc.git tags/dw-mmc-for-ulf-v4.2" for merging easily

Changes in v4:
- remove "host->trans_mode" and use "host->use_dma" to indicate
  transfer mode.
- remove all bt-bindings' changes since we don't need new properities.
- check transfer mode at runtime by reading HCON reg
- spilt defconfig changes for each sub-architecture
- fix the title of cover letter
- reuse some code for reducing code size

Changes in v3:
- choose transfer mode at runtime
- remove all CONFIG_MMC_DW_IDMAC config option
- add supports-idmac property for some platforms

Changes in v2:
- Fix typo of dev_info msg
- remove unused dmach from declaration of dw_mci_dma_slave

 drivers/mmc/host/Kconfig        |  11 +-
 drivers/mmc/host/dw_mmc-pltfm.c |   2 +
 drivers/mmc/host/dw_mmc.c       | 268 ++++++++++++++++++++++++++++++++--------
 drivers/mmc/host/dw_mmc.h       |   6 +
 include/linux/mmc/dw_mmc.h      |  23 +++-
 5 files changed, 246 insertions(+), 64 deletions(-)

diff --git a/drivers/mmc/host/Kconfig b/drivers/mmc/host/Kconfig
index 8a1e349..515527e 100644
--- a/drivers/mmc/host/Kconfig
+++ b/drivers/mmc/host/Kconfig
@@ -615,15 +615,7 @@ config MMC_DW
 	help
 	  This selects support for the Synopsys DesignWare Mobile Storage IP
 	  block, this provides host support for SD and MMC interfaces, in both
-	  PIO and external DMA modes.
-
-config MMC_DW_IDMAC
-	bool "Internal DMAC interface"
-	depends on MMC_DW
-	help
-	  This selects support for the internal DMAC block within the Synopsys
-	  Designware Mobile Storage IP block. This disables the external DMA
-	  interface.
+	  PIO, internal DMA mode and external DMA mode.
 
 config MMC_DW_PLTFM
 	tristate "Synopsys Designware MCI Support as platform device"
@@ -652,7 +644,6 @@ config MMC_DW_K3
 	tristate "K3 specific extensions for Synopsys DW Memory Card Interface"
 	depends on MMC_DW
 	select MMC_DW_PLTFM
-	select MMC_DW_IDMAC
 	help
 	  This selects support for Hisilicon K3 SoC specific extensions to the
 	  Synopsys DesignWare Memory Card Interface driver. Select this option
diff --git a/drivers/mmc/host/dw_mmc-pltfm.c b/drivers/mmc/host/dw_mmc-pltfm.c
index ec6dbcd..7e1d13b 100644
--- a/drivers/mmc/host/dw_mmc-pltfm.c
+++ b/drivers/mmc/host/dw_mmc-pltfm.c
@@ -59,6 +59,8 @@ int dw_mci_pltfm_register(struct platform_device *pdev,
 	host->pdata = pdev->dev.platform_data;
 
 	regs = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	/* Get registers' physical base address */
+	host->phy_regs = (void *)(regs->start);
 	host->regs = devm_ioremap_resource(&pdev->dev, regs);
 	if (IS_ERR(host->regs))
 		return PTR_ERR(host->regs);
diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index fcbf552..5d7a8729 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -56,7 +56,6 @@
 #define DW_MCI_FREQ_MAX	200000000	/* unit: HZ */
 #define DW_MCI_FREQ_MIN	400000		/* unit: HZ */
 
-#ifdef CONFIG_MMC_DW_IDMAC
 #define IDMAC_INT_CLR		(SDMMC_IDMAC_INT_AI | SDMMC_IDMAC_INT_NI | \
 				 SDMMC_IDMAC_INT_CES | SDMMC_IDMAC_INT_DU | \
 				 SDMMC_IDMAC_INT_FBE | SDMMC_IDMAC_INT_RI | \
@@ -102,7 +101,6 @@ struct idmac_desc {
 
 /* Each descriptor can transfer up to 4KB of data in chained mode */
 #define DW_MCI_DESC_DATA_LENGTH	0x1000
-#endif /* CONFIG_MMC_DW_IDMAC */
 
 static bool dw_mci_reset(struct dw_mci *host);
 static bool dw_mci_ctrl_reset(struct dw_mci *host, u32 reset);
@@ -407,7 +405,6 @@ static int dw_mci_get_dma_dir(struct mmc_data *data)
 		return DMA_FROM_DEVICE;
 }
 
-#ifdef CONFIG_MMC_DW_IDMAC
 static void dw_mci_dma_cleanup(struct dw_mci *host)
 {
 	struct mmc_data *data = host->data;
@@ -445,12 +442,21 @@ static void dw_mci_idmac_stop_dma(struct dw_mci *host)
 	mci_writel(host, BMOD, temp);
 }
 
-static void dw_mci_idmac_complete_dma(struct dw_mci *host)
+static void dw_mci_dmac_complete_dma(void *arg)
 {
+	struct dw_mci *host = arg;
 	struct mmc_data *data = host->data;
 
 	dev_vdbg(host->dev, "DMA complete\n");
 
+	if ((host->use_dma == TRANS_MODE_EDMAC) &&
+	    data && (data->flags & MMC_DATA_READ))
+		/* Invalidate cache after read */
+		dma_sync_sg_for_cpu(mmc_dev(host->cur_slot->mmc),
+				    data->sg,
+				    data->sg_len,
+				    DMA_FROM_DEVICE);
+
 	host->dma_ops->cleanup(host);
 
 	/*
@@ -564,7 +570,7 @@ static void dw_mci_translate_sglist(struct dw_mci *host, struct mmc_data *data,
 	wmb(); /* drain writebuffer */
 }
 
-static void dw_mci_idmac_start_dma(struct dw_mci *host, unsigned int sg_len)
+static int dw_mci_idmac_start_dma(struct dw_mci *host, unsigned int sg_len)
 {
 	u32 temp;
 
@@ -589,6 +595,8 @@ static void dw_mci_idmac_start_dma(struct dw_mci *host, unsigned int sg_len)
 
 	/* Start it running */
 	mci_writel(host, PLDMND, 1);
+
+	return 0;
 }
 
 static int dw_mci_idmac_init(struct dw_mci *host)
@@ -669,10 +677,112 @@ static const struct dw_mci_dma_ops dw_mci_idmac_ops = {
 	.init = dw_mci_idmac_init,
 	.start = dw_mci_idmac_start_dma,
 	.stop = dw_mci_idmac_stop_dma,
-	.complete = dw_mci_idmac_complete_dma,
+	.complete = dw_mci_dmac_complete_dma,
+	.cleanup = dw_mci_dma_cleanup,
+};
+
+static void dw_mci_edmac_stop_dma(struct dw_mci *host)
+{
+	dmaengine_terminate_all(host->dms->ch);
+}
+
+static int dw_mci_edmac_start_dma(struct dw_mci *host,
+					    unsigned int sg_len)
+{
+	struct dma_slave_config cfg;
+	struct dma_async_tx_descriptor *desc = NULL;
+	struct scatterlist *sgl = host->data->sg;
+	const u32 mszs[] = {1, 4, 8, 16, 32, 64, 128, 256};
+	u32 sg_elems = host->data->sg_len;
+	u32 fifoth_val;
+	u32 fifo_offset = host->fifo_reg - host->regs;
+	int ret = 0;
+
+	/* Set external dma config: burst size, burst width */
+	cfg.dst_addr = (dma_addr_t)(host->phy_regs + fifo_offset);
+	cfg.src_addr = cfg.dst_addr;
+	cfg.dst_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+	cfg.src_addr_width = DMA_SLAVE_BUSWIDTH_4_BYTES;
+
+	/* Match burst msize with external dma config */
+	fifoth_val = mci_readl(host, FIFOTH);
+	cfg.dst_maxburst = mszs[(fifoth_val >> 28) & 0x7];
+	cfg.src_maxburst = cfg.dst_maxburst;
+
+	if (host->data->flags & MMC_DATA_WRITE)
+		cfg.direction = DMA_MEM_TO_DEV;
+	else
+		cfg.direction = DMA_DEV_TO_MEM;
+
+	ret = dmaengine_slave_config(host->dms->ch, &cfg);
+	if (ret) {
+		dev_err(host->dev, "Failed to config edmac.\n");
+		return -EBUSY;
+	}
+
+	desc = dmaengine_prep_slave_sg(host->dms->ch, sgl,
+				       sg_len, cfg.direction,
+				       DMA_PREP_INTERRUPT | DMA_CTRL_ACK);
+	if (!desc) {
+		dev_err(host->dev, "Can't prepare slave sg.\n");
+		return -EBUSY;
+	}
+
+	/* Set dw_mci_dmac_complete_dma as callback */
+	desc->callback = dw_mci_dmac_complete_dma;
+	desc->callback_param = (void *)host;
+	dmaengine_submit(desc);
+
+	/* Flush cache before write */
+	if (host->data->flags & MMC_DATA_WRITE)
+		dma_sync_sg_for_device(mmc_dev(host->cur_slot->mmc), sgl,
+				       sg_elems, DMA_TO_DEVICE);
+
+	dma_async_issue_pending(host->dms->ch);
+
+	return 0;
+}
+
+static int dw_mci_edmac_init(struct dw_mci *host)
+{
+	/* Request external dma channel */
+	host->dms = kzalloc(sizeof(struct dw_mci_dma_slave), GFP_KERNEL);
+	if (!host->dms)
+		return -ENOMEM;
+
+	host->dms->ch = dma_request_slave_channel(host->dev, "rx-tx");
+	if (!host->dms->ch) {
+		dev_err(host->dev,
+			"Failed to get external DMA channel %d\n",
+			host->dms->ch->chan_id);
+		kfree(host->dms);
+		host->dms = NULL;
+		return -ENXIO;
+	}
+
+	return 0;
+}
+
+static void dw_mci_edmac_exit(struct dw_mci *host)
+{
+	if (host->dms) {
+		if (host->dms->ch) {
+			dma_release_channel(host->dms->ch);
+			host->dms->ch = NULL;
+		}
+		kfree(host->dms);
+		host->dms = NULL;
+	}
+}
+
+static const struct dw_mci_dma_ops dw_mci_edmac_ops = {
+	.init = dw_mci_edmac_init,
+	.exit = dw_mci_edmac_exit,
+	.start = dw_mci_edmac_start_dma,
+	.stop = dw_mci_edmac_stop_dma,
+	.complete = dw_mci_dmac_complete_dma,
 	.cleanup = dw_mci_dma_cleanup,
 };
-#endif /* CONFIG_MMC_DW_IDMAC */
 
 static int dw_mci_pre_dma_transfer(struct dw_mci *host,
 				   struct mmc_data *data,
@@ -752,7 +862,6 @@ static void dw_mci_post_req(struct mmc_host *mmc,
 
 static void dw_mci_adjust_fifoth(struct dw_mci *host, struct mmc_data *data)
 {
-#ifdef CONFIG_MMC_DW_IDMAC
 	unsigned int blksz = data->blksz;
 	const u32 mszs[] = {1, 4, 8, 16, 32, 64, 128, 256};
 	u32 fifo_width = 1 << host->data_shift;
@@ -760,6 +869,10 @@ static void dw_mci_adjust_fifoth(struct dw_mci *host, struct mmc_data *data)
 	u32 msize = 0, rx_wmark = 1, tx_wmark, tx_wmark_invers;
 	int idx = ARRAY_SIZE(mszs) - 1;
 
+	/* pio should ship this scenario */
+	if (!host->use_dma)
+		return;
+
 	tx_wmark = (host->fifo_depth) / 2;
 	tx_wmark_invers = host->fifo_depth - tx_wmark;
 
@@ -788,7 +901,6 @@ static void dw_mci_adjust_fifoth(struct dw_mci *host, struct mmc_data *data)
 done:
 	fifoth_val = SDMMC_SET_FIFOTH(msize, rx_wmark, tx_wmark);
 	mci_writel(host, FIFOTH, fifoth_val);
-#endif
 }
 
 static void dw_mci_ctrl_rd_thld(struct dw_mci *host, struct mmc_data *data)
@@ -850,10 +962,12 @@ static int dw_mci_submit_data_dma(struct dw_mci *host, struct mmc_data *data)
 
 	host->using_dma = 1;
 
-	dev_vdbg(host->dev,
-		 "sd sg_cpu: %#lx sg_dma: %#lx sg_len: %d\n",
-		 (unsigned long)host->sg_cpu, (unsigned long)host->sg_dma,
-		 sg_len);
+	if (host->use_dma == TRANS_MODE_IDMAC)
+		dev_vdbg(host->dev,
+			 "sd sg_cpu: %#lx sg_dma: %#lx sg_len: %d\n",
+			 (unsigned long)host->sg_cpu,
+			 (unsigned long)host->sg_dma,
+			 sg_len);
 
 	/*
 	 * Decide the MSIZE and RX/TX Watermark.
@@ -875,7 +989,11 @@ static int dw_mci_submit_data_dma(struct dw_mci *host, struct mmc_data *data)
 	mci_writel(host, INTMASK, temp);
 	spin_unlock_irqrestore(&host->irq_lock, irqflags);
 
-	host->dma_ops->start(host, sg_len);
+	if (host->dma_ops->start(host, sg_len)) {
+		/* We can't do DMA */
+		dev_err(host->dev, "%s: failed to start DMA.\n", __func__);
+		return -ENODEV;
+	}
 
 	return 0;
 }
@@ -2343,15 +2461,17 @@ static irqreturn_t dw_mci_interrupt(int irq, void *dev_id)
 
 	}
 
-#ifdef CONFIG_MMC_DW_IDMAC
-	/* Handle DMA interrupts */
+	if (host->use_dma != TRANS_MODE_IDMAC)
+		return IRQ_HANDLED;
+
+	/* Handle IDMA interrupts */
 	if (host->dma_64bit_address == 1) {
 		pending = mci_readl(host, IDSTS64);
 		if (pending & (SDMMC_IDMAC_INT_TI | SDMMC_IDMAC_INT_RI)) {
 			mci_writel(host, IDSTS64, SDMMC_IDMAC_INT_TI |
 							SDMMC_IDMAC_INT_RI);
 			mci_writel(host, IDSTS64, SDMMC_IDMAC_INT_NI);
-			host->dma_ops->complete(host);
+			host->dma_ops->complete((void *)host);
 		}
 	} else {
 		pending = mci_readl(host, IDSTS);
@@ -2359,10 +2479,9 @@ static irqreturn_t dw_mci_interrupt(int irq, void *dev_id)
 			mci_writel(host, IDSTS, SDMMC_IDMAC_INT_TI |
 							SDMMC_IDMAC_INT_RI);
 			mci_writel(host, IDSTS, SDMMC_IDMAC_INT_NI);
-			host->dma_ops->complete(host);
+			host->dma_ops->complete((void *)host);
 		}
 	}
-#endif
 
 	return IRQ_HANDLED;
 }
@@ -2471,13 +2590,21 @@ static int dw_mci_init_slot(struct dw_mci *host, unsigned int id)
 		goto err_host_allocated;
 
 	/* Useful defaults if platform data is unset. */
-	if (host->use_dma) {
+	if (host->use_dma == TRANS_MODE_IDMAC) {
 		mmc->max_segs = host->ring_size;
 		mmc->max_blk_size = 65536;
 		mmc->max_seg_size = 0x1000;
 		mmc->max_req_size = mmc->max_seg_size * host->ring_size;
 		mmc->max_blk_count = mmc->max_req_size / 512;
+	} else if (host->use_dma == TRANS_MODE_EDMAC) {
+		mmc->max_segs = 64;
+		mmc->max_blk_size = 65536;
+		mmc->max_blk_count = 65535;
+		mmc->max_req_size =
+				mmc->max_blk_size * mmc->max_blk_count;
+		mmc->max_seg_size = mmc->max_req_size;
 	} else {
+		/* TRANS_MODE_PIO */
 		mmc->max_segs = 64;
 		mmc->max_blk_size = 65536; /* BLKSIZ is 16 bits */
 		mmc->max_blk_count = 512;
@@ -2517,38 +2644,74 @@ static void dw_mci_cleanup_slot(struct dw_mci_slot *slot, unsigned int id)
 static void dw_mci_init_dma(struct dw_mci *host)
 {
 	int addr_config;
-	/* Check ADDR_CONFIG bit in HCON to find IDMAC address bus width */
-	addr_config = (mci_readl(host, HCON) >> 27) & 0x01;
-
-	if (addr_config == 1) {
-		/* host supports IDMAC in 64-bit address mode */
-		host->dma_64bit_address = 1;
-		dev_info(host->dev, "IDMAC supports 64-bit address mode.\n");
-		if (!dma_set_mask(host->dev, DMA_BIT_MASK(64)))
-			dma_set_coherent_mask(host->dev, DMA_BIT_MASK(64));
-	} else {
-		/* host supports IDMAC in 32-bit address mode */
-		host->dma_64bit_address = 0;
-		dev_info(host->dev, "IDMAC supports 32-bit address mode.\n");
-	}
+	struct device *dev = host->dev;
+	struct device_node *np = dev->of_node;
 
-	/* Alloc memory for sg translation */
-	host->sg_cpu = dmam_alloc_coherent(host->dev, PAGE_SIZE,
-					  &host->sg_dma, GFP_KERNEL);
-	if (!host->sg_cpu) {
-		dev_err(host->dev, "%s: could not alloc DMA memory\n",
-			__func__);
+	/*
+	* Check tansfer mode from HCON[17:16]
+	* Clear the ambiguous description of dw_mmc databook:
+	* 2b'00: No DMA Interface -> Actually means using Internal DMA block
+	* 2b'01: DesignWare DMA Interface -> Synopsys DW-DMA block
+	* 2b'10: Generic DMA Interface -> non-Synopsys generic DMA block
+	* 2b'11: Non DW DMA Interface -> pio only
+	* Compared to DesignWare DMA Interface, Generic DMA Interface has a
+	* simpler request/acknowledge handshake mechanism and both of them
+	* are regarded as external dma master for dw_mmc.
+	*/
+	host->use_dma = SDMMC_GET_TRANS_MODE(mci_readl(host, HCON));
+	if (host->use_dma == DMA_INTERFACE_IDMA) {
+		host->use_dma = TRANS_MODE_IDMAC;
+	} else if (host->use_dma == DMA_INTERFACE_DWDMA ||
+		   host->use_dma == DMA_INTERFACE_GDMA) {
+		host->use_dma = TRANS_MODE_EDMAC;
+	} else {
 		goto no_dma;
 	}
 
 	/* Determine which DMA interface to use */
-#ifdef CONFIG_MMC_DW_IDMAC
-	host->dma_ops = &dw_mci_idmac_ops;
-	dev_info(host->dev, "Using internal DMA controller.\n");
-#endif
+	if (host->use_dma == TRANS_MODE_IDMAC) {
+		/*
+		* Check ADDR_CONFIG bit in HCON to find
+		* IDMAC address bus width
+		*/
+		addr_config = (mci_readl(host, HCON) >> 27) & 0x01;
+
+		if (addr_config == 1) {
+			/* host supports IDMAC in 64-bit address mode */
+			host->dma_64bit_address = 1;
+			dev_info(host->dev,
+				 "IDMAC supports 64-bit address mode.\n");
+			if (!dma_set_mask(host->dev, DMA_BIT_MASK(64)))
+				dma_set_coherent_mask(host->dev,
+						      DMA_BIT_MASK(64));
+		} else {
+			/* host supports IDMAC in 32-bit address mode */
+			host->dma_64bit_address = 0;
+			dev_info(host->dev,
+				 "IDMAC supports 32-bit address mode.\n");
+		}
 
-	if (!host->dma_ops)
-		goto no_dma;
+		/* Alloc memory for sg translation */
+		host->sg_cpu = dmam_alloc_coherent(host->dev, PAGE_SIZE,
+						   &host->sg_dma, GFP_KERNEL);
+		if (!host->sg_cpu) {
+			dev_err(host->dev,
+				"%s: could not alloc DMA memory\n",
+				__func__);
+			goto no_dma;
+		}
+
+		host->dma_ops = &dw_mci_idmac_ops;
+		dev_info(host->dev, "Using internal DMA controller.\n");
+	} else {
+		/* TRANS_MODE_EDMAC: check dma bindings again */
+		if ((of_property_count_strings(np, "dma-names") < 0) ||
+		    (!of_find_property(np, "dmas", NULL))) {
+			goto no_dma;
+		}
+		host->dma_ops = &dw_mci_edmac_ops;
+		dev_info(host->dev, "Using external DMA controller.\n");
+	}
 
 	if (host->dma_ops->init && host->dma_ops->start &&
 	    host->dma_ops->stop && host->dma_ops->cleanup) {
@@ -2562,12 +2725,11 @@ static void dw_mci_init_dma(struct dw_mci *host)
 		goto no_dma;
 	}
 
-	host->use_dma = 1;
 	return;
 
 no_dma:
 	dev_info(host->dev, "Using PIO mode.\n");
-	host->use_dma = 0;
+	host->use_dma = TRANS_MODE_PIO;
 }
 
 static bool dw_mci_ctrl_reset(struct dw_mci *host, u32 reset)
@@ -2650,10 +2812,9 @@ static bool dw_mci_reset(struct dw_mci *host)
 		}
 	}
 
-#if IS_ENABLED(CONFIG_MMC_DW_IDMAC)
-	/* It is also recommended that we reset and reprogram idmac */
-	dw_mci_idmac_reset(host);
-#endif
+	if (host->use_dma == TRANS_MODE_IDMAC)
+		/* It is also recommended that we reset and reprogram idmac */
+		dw_mci_idmac_reset(host);
 
 	ret = true;
 
@@ -3067,6 +3228,9 @@ EXPORT_SYMBOL(dw_mci_remove);
  */
 int dw_mci_suspend(struct dw_mci *host)
 {
+	if (host->use_dma && host->dma_ops->exit)
+		host->dma_ops->exit(host);
+
 	return 0;
 }
 EXPORT_SYMBOL(dw_mci_suspend);
diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
index 8ce4674..811d467 100644
--- a/drivers/mmc/host/dw_mmc.h
+++ b/drivers/mmc/host/dw_mmc.h
@@ -148,6 +148,12 @@
 #define SDMMC_SET_FIFOTH(m, r, t)	(((m) & 0x7) << 28 | \
 					 ((r) & 0xFFF) << 16 | \
 					 ((t) & 0xFFF))
+/* HCON register defines */
+#define DMA_INTERFACE_IDMA		(0x0)
+#define DMA_INTERFACE_DWDMA		(0x1)
+#define DMA_INTERFACE_GDMA		(0x2)
+#define DMA_INTERFACE_NODMA		(0x3)
+#define SDMMC_GET_TRANS_MODE(x)		(((x)>>16) & 0x3)
 /* Internal DMAC interrupt defines */
 #define SDMMC_IDMAC_INT_AI		BIT(9)
 #define SDMMC_IDMAC_INT_NI		BIT(8)
diff --git a/include/linux/mmc/dw_mmc.h b/include/linux/mmc/dw_mmc.h
index 134c574..f67b2ec 100644
--- a/include/linux/mmc/dw_mmc.h
+++ b/include/linux/mmc/dw_mmc.h
@@ -16,6 +16,7 @@
 
 #include <linux/scatterlist.h>
 #include <linux/mmc/core.h>
+#include <linux/dmaengine.h>
 
 #define MAX_MCI_SLOTS	2
 
@@ -40,6 +41,17 @@ enum {
 
 struct mmc_data;
 
+enum {
+	TRANS_MODE_PIO = 0,
+	TRANS_MODE_IDMAC,
+	TRANS_MODE_EDMAC
+};
+
+struct dw_mci_dma_slave {
+	struct dma_chan *ch;
+	enum dma_transfer_direction direction;
+};
+
 /**
  * struct dw_mci - MMC controller state shared between all slots
  * @lock: Spinlock protecting the queue and associated data.
@@ -154,7 +166,14 @@ struct dw_mci {
 	dma_addr_t		sg_dma;
 	void			*sg_cpu;
 	const struct dw_mci_dma_ops	*dma_ops;
+	/* For idmac */
 	unsigned int		ring_size;
+
+	/* For edmac */
+	struct dw_mci_dma_slave *dms;
+	/* Registers's physical base address */
+	void                    *phy_regs;
+
 	u32			cmd_status;
 	u32			data_status;
 	u32			stop_cmdr;
@@ -208,8 +227,8 @@ struct dw_mci {
 struct dw_mci_dma_ops {
 	/* DMA Ops */
 	int (*init)(struct dw_mci *host);
-	void (*start)(struct dw_mci *host, unsigned int sg_len);
-	void (*complete)(struct dw_mci *host);
+	int (*start)(struct dw_mci *host, unsigned int sg_len);
+	void (*complete)(void *host);
 	void (*stop)(struct dw_mci *host);
 	void (*cleanup)(struct dw_mci *host);
 	void (*exit)(struct dw_mci *host);
-- 
2.3.7

[RFC PATCH v8 02/10] mmc: dw_mmc: use macro for HCON register operations

[Shawn Lin]

This patch add some macros for HCON register operations
to make code more readable.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Acked-by: Jaehoon Chung <jh80.chung@samsung.com>
---

Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 drivers/mmc/host/dw_mmc.c | 6 +++---
 drivers/mmc/host/dw_mmc.h | 3 +++
 2 files changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/mmc/host/dw_mmc.c b/drivers/mmc/host/dw_mmc.c
index 5d7a8729..100f56e 100644
--- a/drivers/mmc/host/dw_mmc.c
+++ b/drivers/mmc/host/dw_mmc.c
@@ -2674,7 +2674,7 @@ static void dw_mci_init_dma(struct dw_mci *host)
 		* Check ADDR_CONFIG bit in HCON to find
 		* IDMAC address bus width
 		*/
-		addr_config = (mci_readl(host, HCON) >> 27) & 0x01;
+		addr_config = SDMMC_GET_ADDR_CONFIG(mci_readl(host, HCON));
 
 		if (addr_config == 1) {
 			/* host supports IDMAC in 64-bit address mode */
@@ -3051,7 +3051,7 @@ int dw_mci_probe(struct dw_mci *host)
 	 * Get the host data width - this assumes that HCON has been set with
 	 * the correct values.
 	 */
-	i = (mci_readl(host, HCON) >> 7) & 0x7;
+	i = SDMMC_GET_HDATA_WIDTH(mci_readl(host, HCON));
 	if (!i) {
 		host->push_data = dw_mci_push_data16;
 		host->pull_data = dw_mci_pull_data16;
@@ -3133,7 +3133,7 @@ int dw_mci_probe(struct dw_mci *host)
 	if (host->pdata->num_slots)
 		host->num_slots = host->pdata->num_slots;
 	else
-		host->num_slots = ((mci_readl(host, HCON) >> 1) & 0x1F) + 1;
+		host->num_slots = SDMMC_GET_SLOT_NUM(mci_readl(host, HCON));
 
 	/*
 	 * Enable interrupts for command done, data over, data empty,
diff --git a/drivers/mmc/host/dw_mmc.h b/drivers/mmc/host/dw_mmc.h
index 811d467..f2a88d4 100644
--- a/drivers/mmc/host/dw_mmc.h
+++ b/drivers/mmc/host/dw_mmc.h
@@ -154,6 +154,9 @@
 #define DMA_INTERFACE_GDMA		(0x2)
 #define DMA_INTERFACE_NODMA		(0x3)
 #define SDMMC_GET_TRANS_MODE(x)		(((x)>>16) & 0x3)
+#define SDMMC_GET_SLOT_NUM(x)		((((x)>>1) & 0x1F) + 1)
+#define SDMMC_GET_HDATA_WIDTH(x)	(((x)>>7) & 0x7)
+#define SDMMC_GET_ADDR_CONFIG(x)	(((x)>>27) & 0x1)
 /* Internal DMAC interrupt defines */
 #define SDMMC_IDMAC_INT_AI		BIT(9)
 #define SDMMC_IDMAC_INT_NI		BIT(8)
-- 
2.3.7

[RFC PATCH v8 03/10] Documentation: synopsys-dw-mshc: add bindings for idmac and edmac

[Shawn Lin]

synopsys-dw-mshc supports three types of transfer mode. We add
bindings and description for how to use them at runtime.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
---

Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 .../devicetree/bindings/mmc/synopsys-dw-mshc.txt   | 25 ++++++++++++++++++++++
 1 file changed, 25 insertions(+)

diff --git a/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt b/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt
index 346c609..8636f5a 100644
--- a/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt
+++ b/Documentation/devicetree/bindings/mmc/synopsys-dw-mshc.txt
@@ -75,6 +75,12 @@ Optional properties:
 * vmmc-supply: The phandle to the regulator to use for vmmc.  If this is
   specified we'll defer probe until we can find this regulator.
 
+* dmas: List of DMA specifiers with the controller specific format as described
+  in the generic DMA client binding. Refer to dma.txt for details.
+
+* dma-names: request names for generic DMA client binding. Must be "rx-tx".
+  Refer to dma.txt for details.
+
 Aliases:
 
 - All the MSHC controller nodes should be represented in the aliases node using
@@ -95,6 +101,23 @@ board specific portions as listed below.
 		#size-cells = <0>;
 	};
 
+[board specific internal DMA resources]
+
+	dwmmc0@12200000 {
+		clock-frequency = <400000000>;
+		clock-freq-min-max = <400000 200000000>;
+		num-slots = <1>;
+		broken-cd;
+		fifo-depth = <0x80>;
+		card-detect-delay = <200>;
+		vmmc-supply = <&buck8>;
+		bus-width = <8>;
+		cap-mmc-highspeed;
+		cap-sd-highspeed;
+	};
+
+[board specific generic DMA request binding]
+
 	dwmmc0@12200000 {
 		clock-frequency = <400000000>;
 		clock-freq-min-max = <400000 200000000>;
@@ -106,4 +129,6 @@ board specific portions as listed below.
 		bus-width = <8>;
 		cap-mmc-highspeed;
 		cap-sd-highspeed;
+		dmas = <&pdma 12>;
+		dma-names = "rx-tx";
 	};
-- 
2.3.7

[RFC PATCH v8 04/10] mips: pistachio_defconfig: remove CONFIG_MMC_DW_IDMAC

[Shawn Lin]

DesignWare MMC Controller's transfer mode should be decided
at runtime instead of compile-time. So we remove this config
option and read dw_mmc's register to select DMA master.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Acked-by: Govindraj Raja <govindraj.raja@imgtec.com>
Acked-by: Ralf Baechle <ralf@linux-mips.org>
---

Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 arch/mips/configs/pistachio_defconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/mips/configs/pistachio_defconfig b/arch/mips/configs/pistachio_defconfig
index 642b509..8b74291 100644
--- a/arch/mips/configs/pistachio_defconfig
+++ b/arch/mips/configs/pistachio_defconfig
@@ -257,7 +257,6 @@ CONFIG_MMC=y
 CONFIG_MMC_BLOCK_MINORS=16
 CONFIG_MMC_TEST=m
 CONFIG_MMC_DW=y
-CONFIG_MMC_DW_IDMAC=y
 CONFIG_NEW_LEDS=y
 CONFIG_LEDS_CLASS=y
 CONFIG_RTC_CLASS=y
-- 
2.3.7

[RFC PATCH v8 05/10] arc: axs10x_defconfig: remove CONFIG_MMC_DW_IDMAC

[Shawn Lin]

DesignWare MMC Controller's transfer mode should be decided
at runtime instead of compile-time. So we remove this config
option and read dw_mmc's register to select DMA master.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Acked-by: Vineet Gupta <vgupta@synopsys.com>
---

Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 arch/arc/configs/axs101_defconfig     | 1 -
 arch/arc/configs/axs103_defconfig     | 1 -
 arch/arc/configs/axs103_smp_defconfig | 1 -
 3 files changed, 3 deletions(-)

diff --git a/arch/arc/configs/axs101_defconfig b/arch/arc/configs/axs101_defconfig
index 562dac6..c92c0ef 100644
--- a/arch/arc/configs/axs101_defconfig
+++ b/arch/arc/configs/axs101_defconfig
@@ -89,7 +89,6 @@ CONFIG_MMC=y
 CONFIG_MMC_SDHCI=y
 CONFIG_MMC_SDHCI_PLTFM=y
 CONFIG_MMC_DW=y
-CONFIG_MMC_DW_IDMAC=y
 # CONFIG_IOMMU_SUPPORT is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT4_FS=y
diff --git a/arch/arc/configs/axs103_defconfig b/arch/arc/configs/axs103_defconfig
index 83a6d8d..cfac24e 100644
--- a/arch/arc/configs/axs103_defconfig
+++ b/arch/arc/configs/axs103_defconfig
@@ -95,7 +95,6 @@ CONFIG_MMC=y
 CONFIG_MMC_SDHCI=y
 CONFIG_MMC_SDHCI_PLTFM=y
 CONFIG_MMC_DW=y
-CONFIG_MMC_DW_IDMAC=y
 # CONFIG_IOMMU_SUPPORT is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT4_FS=y
diff --git a/arch/arc/configs/axs103_smp_defconfig b/arch/arc/configs/axs103_smp_defconfig
index f1e1c84..9922a11 100644
--- a/arch/arc/configs/axs103_smp_defconfig
+++ b/arch/arc/configs/axs103_smp_defconfig
@@ -96,7 +96,6 @@ CONFIG_MMC=y
 CONFIG_MMC_SDHCI=y
 CONFIG_MMC_SDHCI_PLTFM=y
 CONFIG_MMC_DW=y
-CONFIG_MMC_DW_IDMAC=y
 # CONFIG_IOMMU_SUPPORT is not set
 CONFIG_EXT3_FS=y
 CONFIG_EXT4_FS=y
-- 
2.3.7

[RFC PATCH v8 06/10] arm: exynos_defconfig: remove CONFIG_MMC_DW_IDMAC

[Shawn Lin]

DesignWare MMC Controller's transfer mode should be decided
at runtime instead of compile-time. So we remove this config
option and read dw_mmc's register to select DMA master.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Acked-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>
---

Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 arch/arm/configs/exynos_defconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/arm/configs/exynos_defconfig b/arch/arm/configs/exynos_defconfig
index 1ff2bfa..13ba48c 100644
--- a/arch/arm/configs/exynos_defconfig
+++ b/arch/arm/configs/exynos_defconfig
@@ -166,7 +166,6 @@ CONFIG_MMC_SDHCI=y
 CONFIG_MMC_SDHCI_S3C=y
 CONFIG_MMC_SDHCI_S3C_DMA=y
 CONFIG_MMC_DW=y
-CONFIG_MMC_DW_IDMAC=y
 CONFIG_MMC_DW_EXYNOS=y
 CONFIG_RTC_CLASS=y
 CONFIG_RTC_DRV_MAX77686=y
-- 
2.3.7

[RFC PATCH v8 07/10] arm: hisi_defconfig: remove CONFIG_MMC_DW_IDMAC

[Shawn Lin]

DesignWare MMC Controller's transfer mode should be decided
at runtime instead of compile-time. So we remove this config
option and read dw_mmc's register to select DMA master.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Acked-by: Wei Xu <xuwei5@hisilicon.com>
---

Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 arch/arm/configs/hisi_defconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/arm/configs/hisi_defconfig b/arch/arm/configs/hisi_defconfig
index 5997dbc..b2e340b 100644
--- a/arch/arm/configs/hisi_defconfig
+++ b/arch/arm/configs/hisi_defconfig
@@ -69,7 +69,6 @@ CONFIG_NOP_USB_XCEIV=y
 CONFIG_MMC=y
 CONFIG_RTC_CLASS=y
 CONFIG_MMC_DW=y
-CONFIG_MMC_DW_IDMAC=y
 CONFIG_MMC_DW_PLTFM=y
 CONFIG_RTC_DRV_PL031=y
 CONFIG_DMADEVICES=y
-- 
2.3.7

[RFC PATCH v8 08/10] arm: lpc18xx_defconfig: remove CONFIG_MMC_DW_IDMAC

[Shawn Lin]

DesignWare MMC Controller's transfer mode should be decided
at runtime instead of compile-time. So we remove this config
option and read dw_mmc's register to select DMA master.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
Acked-by: Joachim Eastwood <manabian@gmail.com>
---

Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 arch/arm/configs/lpc18xx_defconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/arm/configs/lpc18xx_defconfig b/arch/arm/configs/lpc18xx_defconfig
index 1c47f86..b7e8cda 100644
--- a/arch/arm/configs/lpc18xx_defconfig
+++ b/arch/arm/configs/lpc18xx_defconfig
@@ -119,7 +119,6 @@ CONFIG_USB_EHCI_HCD=y
 CONFIG_USB_EHCI_ROOT_HUB_TT=y
 CONFIG_MMC=y
 CONFIG_MMC_DW=y
-CONFIG_MMC_DW_IDMAC=y
 CONFIG_NEW_LEDS=y
 CONFIG_LEDS_CLASS=y
 CONFIG_LEDS_PCA9532=y
-- 
2.3.7

[RFC PATCH v8 09/10] arm: multi_v7_defconfig: remove CONFIG_MMC_DW_IDMAC

[Shawn Lin]

DesignWare MMC Controller's transfer mode should be decided
at runtime instead of compile-time. So we remove this config
option and read dw_mmc's register to select DMA master.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
---

Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 arch/arm/configs/multi_v7_defconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig
index 03deb7f..ad929ea 100644
--- a/arch/arm/configs/multi_v7_defconfig
+++ b/arch/arm/configs/multi_v7_defconfig
@@ -539,7 +539,6 @@ CONFIG_MMC_ATMELMCI=y
 CONFIG_MMC_MVSDIO=y
 CONFIG_MMC_SDHI=y
 CONFIG_MMC_DW=y
-CONFIG_MMC_DW_IDMAC=y
 CONFIG_MMC_DW_PLTFM=y
 CONFIG_MMC_DW_EXYNOS=y
 CONFIG_MMC_DW_ROCKCHIP=y
-- 
2.3.7

[RFC PATCH v8 10/10] arm: zx_defconfig: remove CONFIG_MMC_DW_IDMAC

[Shawn Lin]

DesignWare MMC Controller's transfer mode should be decided
at runtime instead of compile-time. So we remove this config
option and read dw_mmc's register to select DMA master.

Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
---

Changes in v8: None
Changes in v7: None
Changes in v6: None
Changes in v5: None
Changes in v4: None
Changes in v3: None
Changes in v2: None

 arch/arm/configs/zx_defconfig | 1 -
 1 file changed, 1 deletion(-)

diff --git a/arch/arm/configs/zx_defconfig b/arch/arm/configs/zx_defconfig
index b200bb0..ab683fb 100644
--- a/arch/arm/configs/zx_defconfig
+++ b/arch/arm/configs/zx_defconfig
@@ -83,7 +83,6 @@ CONFIG_MMC=y
 CONFIG_MMC_UNSAFE_RESUME=y
 CONFIG_MMC_BLOCK_MINORS=16
 CONFIG_MMC_DW=y
-CONFIG_MMC_DW_IDMAC=y
 CONFIG_EXT2_FS=y
 CONFIG_EXT4_FS=y
 CONFIG_EXT4_FS_POSIX_ACL=y
-- 
2.3.7

Re: [RFC PATCH v8 0/10] Add external dma support for Synopsys MSHC

[Jaehoon Chung]

Dear, All.

I will apply patch 01-03 at my repository on today.
But i don't know better how i do about other patches relevant to config file.

Best Regards,
Jaehoon Chung

On 09/16/2015 03:40 PM, Shawn Lin wrote:
> Synopsys DesignWare mobile storage host controller supports three
> types of transfer mode: pio, internal dma and external dma. However,
> dw_mmc can only supports pio and internal dma now. Thus some platforms
> using dw-mshc integrated with generic dma can't work in dma mode. So we
> submit this patch to achieve it.
> 
> And the config option, CONFIG_MMC_DW_IDMAC, was added by Will Newton
> (commit:f95f3850) for the first version of dw_mmc and never be touched since
> then. At that time dt-bindings hadn't been introduced into dw_mmc yet means
> we should select CONFIG_MMC_DW_IDMAC to enable internal dma mode at compile
> time. Nowadays, device-tree helps us to support a variety of boards with one
> kernel. That's why we need to remove it and decide the transfer mode by reading
> dw_mmc's HCON reg at runtime.
> 
> This RFC patch needs lots of ACKs. I know it's hard, but it does need someone
> to make the running.
> 
> Patch does the following things:
> - remove CONFIG_MMC_DW_IDMAC config option
> - add bindings for edmac used by synopsys-dw-mshc
>   at runtime
> - add edmac support for synopsys-dw-mshc
> 
> Patch is based on next of git://git.linaro.org/people/ulf.hansson/mmc
> 
> Test emmc throughput on my platform with edmac support and without edmac support(pio only)
> iozone -L64 -S32 -azecwI -+n -r4k -r64k -r128k -s1g -i0 -i1 -i2 -f datafile -Rb out.xls > /mnt/result.txt
> (light cpu loading, Direct IO, fixed line size, all pattern recycle, 1GB data in total)
>  ___________________________________________________________
> |                   external dma mode                       |
> |-----------------------------------------------------------|
> |blksz | Random Read | Random Write | Seq Read   | Seq Write|
> |-----------------------------------------------------------|
> |4kB   |  13953kB/s  |    8602kB/s  | 13672kB/s  |  9785kB/s|
> |-----------------------------------------------------------|
> |64kB  |  46058kB/s  |   24794kB/s  | 48058kB/s  | 25418kB/s|
> |-----------------------------------------------------------|
> |128kB |  57026kB/s  |   35117kB/s  | 57375kB/s  | 35183kB/s|
> |-----------------------------------------------------------|
>                            VS
>  ___________________________________________________________
> |                          pio mode                         |
> |-----------------------------------------------------------|
> |blksz | Random Read  | Random Write | Seq Read  | Seq Write|
> |-----------------------------------------------------------|
> |4kB   |  11720kB/s   |    8644kB/s  | 11549kB/s |  9624kB/s|
> |-----------------------------------------------------------|
> |64kB  |  21869kB/s   |   24414kB/s  | 22031kB/s | 27986kB/s|
> |-----------------------------------------------------------|
> |128kB |  23718kB/s   |   34495kB/s  | 24698kB/s | 34637kB/s|
> |-----------------------------------------------------------|
> 
> 
> Changes in v8:
> - remove trans_mode variable
> - remove unnecessary dma_ops check
> - remove unnecessary comment
> - fix coding style based on latest ulf's next
> - add Acked-by: Jaehoon Chung <jh80.chung@samsung.com>
>   for HCON's changes
> 
> Changes in v7:
> - rebased on Ulf's next
> - combine condition state
> - elaborate more about DMA_INTERFACE
> - define some macro for DMA_INERFACE value
> - spilt HCON ops' changes into another patch
> 
> Changes in v6:
> - add trans_mode condition for IDMAC initialization
>   suggested by Heiko
> - re-test my patch on rk3188 platform and update commit msg
> - update performance of pio vs edmac in cover letter
> 
> Changes in v5:
> - add the title of cover letter
> - fix typo of comment
> - add macro for reading HCON register
> - add "Acked-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>" for exynos_defconfig patch
> - add "Acked-by: Vineet Gupta <vgupta@synopsys.com>" for axs10x_defconfig patch
> - add "Acked-by: Govindraj Raja <govindraj.raja@imgtec.com>" and
>   "Acked-by: Ralf Baechle <ralf@linux-mips.org>" for pistachio_defconfig patch
> - add "Acked-by: Joachim Eastwood <manabian@gmail.com>" for lpc18xx_defconfig patch
> - add "Acked-by: Wei Xu <xuwei5@hisilicon.com>" for hisi_defconfig patch
> - rebase on "https://github.com/jh80chung/dw-mmc.git tags/dw-mmc-for-ulf-v4.2" for merging easily
> 
> Changes in v4:
> - remove "host->trans_mode" and use "host->use_dma" to indicate
>   transfer mode.
> - remove all bt-bindings' changes since we don't need new properities.
> - check transfer mode at runtime by reading HCON reg
> - spilt defconfig changes for each sub-architecture
> - fix the title of cover letter
> - reuse some code for reducing code size
> 
> Changes in v3:
> - choose transfer mode at runtime
> - remove all CONFIG_MMC_DW_IDMAC config option
> - add supports-idmac property for some platforms
> 
> Changes in v2:
> - Fix typo of dev_info msg
> - remove unused dmach from declaration of dw_mci_dma_slave
> 
> Shawn Lin (10):
>   mmc: dw_mmc: Add external dma interface support
>   mmc: dw_mmc: use macro for HCON register operations
>   Documentation: synopsys-dw-mshc: add bindings for idmac and edmac
>   mips: pistachio_defconfig: remove CONFIG_MMC_DW_IDMAC
>   arc: axs10x_defconfig: remove CONFIG_MMC_DW_IDMAC
>   arm: exynos_defconfig: remove CONFIG_MMC_DW_IDMAC
>   arm: hisi_defconfig: remove CONFIG_MMC_DW_IDMAC
>   arm: lpc18xx_defconfig: remove CONFIG_MMC_DW_IDMAC
>   arm: multi_v7_defconfig: remove CONFIG_MMC_DW_IDMAC
>   arm: zx_defconfig: remove CONFIG_MMC_DW_IDMAC
> 
>  .../devicetree/bindings/mmc/synopsys-dw-mshc.txt   |  25 ++
>  arch/arc/configs/axs101_defconfig                  |   1 -
>  arch/arc/configs/axs103_defconfig                  |   1 -
>  arch/arc/configs/axs103_smp_defconfig              |   1 -
>  arch/arm/configs/exynos_defconfig                  |   1 -
>  arch/arm/configs/hisi_defconfig                    |   1 -
>  arch/arm/configs/lpc18xx_defconfig                 |   1 -
>  arch/arm/configs/multi_v7_defconfig                |   1 -
>  arch/arm/configs/zx_defconfig                      |   1 -
>  arch/mips/configs/pistachio_defconfig              |   1 -
>  drivers/mmc/host/Kconfig                           |  11 +-
>  drivers/mmc/host/dw_mmc-pltfm.c                    |   2 +
>  drivers/mmc/host/dw_mmc.c                          | 272 +++++++++++++++++----
>  drivers/mmc/host/dw_mmc.h                          |   9 +
>  include/linux/mmc/dw_mmc.h                         |  23 +-
>  15 files changed, 276 insertions(+), 75 deletions(-)
> 

Re: [RFC PATCH v8 0/10] Add external dma support for Synopsys MSHC

[Shawn Lin]

On 2015/10/1 17:14, Jaehoon Chung wrote:
> Dear, All.
>
> I will apply patch 01-03 at my repository on today.
> But i don't know better how i do about other patches relevant to config file.
>

Thanks, Jaehoon. :)

I guess it would be acceptable to pick the config changes, already they 
were acked by the soc maintainers, via dw_mmc tree when ulf merge dw_mmc 
tree into his mmc-tree.


> Best Regards,
> Jaehoon Chung
>
> On 09/16/2015 03:40 PM, Shawn Lin wrote:
>> Synopsys DesignWare mobile storage host controller supports three
>> types of transfer mode: pio, internal dma and external dma. However,
>> dw_mmc can only supports pio and internal dma now. Thus some platforms
>> using dw-mshc integrated with generic dma can't work in dma mode. So we
>> submit this patch to achieve it.
>>
>> And the config option, CONFIG_MMC_DW_IDMAC, was added by Will Newton
>> (commit:f95f3850) for the first version of dw_mmc and never be touched since
>> then. At that time dt-bindings hadn't been introduced into dw_mmc yet means
>> we should select CONFIG_MMC_DW_IDMAC to enable internal dma mode at compile
>> time. Nowadays, device-tree helps us to support a variety of boards with one
>> kernel. That's why we need to remove it and decide the transfer mode by reading
>> dw_mmc's HCON reg at runtime.
>>
>> This RFC patch needs lots of ACKs. I know it's hard, but it does need someone
>> to make the running.
>>
>> Patch does the following things:
>> - remove CONFIG_MMC_DW_IDMAC config option
>> - add bindings for edmac used by synopsys-dw-mshc
>>    at runtime
>> - add edmac support for synopsys-dw-mshc
>>
>> Patch is based on next of git://git.linaro.org/people/ulf.hansson/mmc
>>
>> Test emmc throughput on my platform with edmac support and without edmac support(pio only)
>> iozone -L64 -S32 -azecwI -+n -r4k -r64k -r128k -s1g -i0 -i1 -i2 -f datafile -Rb out.xls > /mnt/result.txt
>> (light cpu loading, Direct IO, fixed line size, all pattern recycle, 1GB data in total)
>>   ___________________________________________________________
>> |                   external dma mode                       |
>> |-----------------------------------------------------------|
>> |blksz | Random Read | Random Write | Seq Read   | Seq Write|
>> |-----------------------------------------------------------|
>> |4kB   |  13953kB/s  |    8602kB/s  | 13672kB/s  |  9785kB/s|
>> |-----------------------------------------------------------|
>> |64kB  |  46058kB/s  |   24794kB/s  | 48058kB/s  | 25418kB/s|
>> |-----------------------------------------------------------|
>> |128kB |  57026kB/s  |   35117kB/s  | 57375kB/s  | 35183kB/s|
>> |-----------------------------------------------------------|
>>                             VS
>>   ___________________________________________________________
>> |                          pio mode                         |
>> |-----------------------------------------------------------|
>> |blksz | Random Read  | Random Write | Seq Read  | Seq Write|
>> |-----------------------------------------------------------|
>> |4kB   |  11720kB/s   |    8644kB/s  | 11549kB/s |  9624kB/s|
>> |-----------------------------------------------------------|
>> |64kB  |  21869kB/s   |   24414kB/s  | 22031kB/s | 27986kB/s|
>> |-----------------------------------------------------------|
>> |128kB |  23718kB/s   |   34495kB/s  | 24698kB/s | 34637kB/s|
>> |-----------------------------------------------------------|
>>
>>
>> Changes in v8:
>> - remove trans_mode variable
>> - remove unnecessary dma_ops check
>> - remove unnecessary comment
>> - fix coding style based on latest ulf's next
>> - add Acked-by: Jaehoon Chung <jh80.chung@samsung.com>
>>    for HCON's changes
>>
>> Changes in v7:
>> - rebased on Ulf's next
>> - combine condition state
>> - elaborate more about DMA_INTERFACE
>> - define some macro for DMA_INERFACE value
>> - spilt HCON ops' changes into another patch
>>
>> Changes in v6:
>> - add trans_mode condition for IDMAC initialization
>>    suggested by Heiko
>> - re-test my patch on rk3188 platform and update commit msg
>> - update performance of pio vs edmac in cover letter
>>
>> Changes in v5:
>> - add the title of cover letter
>> - fix typo of comment
>> - add macro for reading HCON register
>> - add "Acked-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>" for exynos_defconfig patch
>> - add "Acked-by: Vineet Gupta <vgupta@synopsys.com>" for axs10x_defconfig patch
>> - add "Acked-by: Govindraj Raja <govindraj.raja@imgtec.com>" and
>>    "Acked-by: Ralf Baechle <ralf@linux-mips.org>" for pistachio_defconfig patch
>> - add "Acked-by: Joachim Eastwood <manabian@gmail.com>" for lpc18xx_defconfig patch
>> - add "Acked-by: Wei Xu <xuwei5@hisilicon.com>" for hisi_defconfig patch
>> - rebase on "https://github.com/jh80chung/dw-mmc.git tags/dw-mmc-for-ulf-v4.2" for merging easily
>>
>> Changes in v4:
>> - remove "host->trans_mode" and use "host->use_dma" to indicate
>>    transfer mode.
>> - remove all bt-bindings' changes since we don't need new properities.
>> - check transfer mode at runtime by reading HCON reg
>> - spilt defconfig changes for each sub-architecture
>> - fix the title of cover letter
>> - reuse some code for reducing code size
>>
>> Changes in v3:
>> - choose transfer mode at runtime
>> - remove all CONFIG_MMC_DW_IDMAC config option
>> - add supports-idmac property for some platforms
>>
>> Changes in v2:
>> - Fix typo of dev_info msg
>> - remove unused dmach from declaration of dw_mci_dma_slave
>>
>> Shawn Lin (10):
>>    mmc: dw_mmc: Add external dma interface support
>>    mmc: dw_mmc: use macro for HCON register operations
>>    Documentation: synopsys-dw-mshc: add bindings for idmac and edmac
>>    mips: pistachio_defconfig: remove CONFIG_MMC_DW_IDMAC
>>    arc: axs10x_defconfig: remove CONFIG_MMC_DW_IDMAC
>>    arm: exynos_defconfig: remove CONFIG_MMC_DW_IDMAC
>>    arm: hisi_defconfig: remove CONFIG_MMC_DW_IDMAC
>>    arm: lpc18xx_defconfig: remove CONFIG_MMC_DW_IDMAC
>>    arm: multi_v7_defconfig: remove CONFIG_MMC_DW_IDMAC
>>    arm: zx_defconfig: remove CONFIG_MMC_DW_IDMAC
>>
>>   .../devicetree/bindings/mmc/synopsys-dw-mshc.txt   |  25 ++
>>   arch/arc/configs/axs101_defconfig                  |   1 -
>>   arch/arc/configs/axs103_defconfig                  |   1 -
>>   arch/arc/configs/axs103_smp_defconfig              |   1 -
>>   arch/arm/configs/exynos_defconfig                  |   1 -
>>   arch/arm/configs/hisi_defconfig                    |   1 -
>>   arch/arm/configs/lpc18xx_defconfig                 |   1 -
>>   arch/arm/configs/multi_v7_defconfig                |   1 -
>>   arch/arm/configs/zx_defconfig                      |   1 -
>>   arch/mips/configs/pistachio_defconfig              |   1 -
>>   drivers/mmc/host/Kconfig                           |  11 +-
>>   drivers/mmc/host/dw_mmc-pltfm.c                    |   2 +
>>   drivers/mmc/host/dw_mmc.c                          | 272 +++++++++++++++++----
>>   drivers/mmc/host/dw_mmc.h                          |   9 +
>>   include/linux/mmc/dw_mmc.h                         |  23 +-
>>   15 files changed, 276 insertions(+), 75 deletions(-)
>>
>
>
>
>


-- 
Best Regards
Shawn Lin

Re: [RFC PATCH v8 0/10] Add external dma support for Synopsys MSHC

[Jaehoon Chung]

Hi, Shawn.

On 10/02/2015 06:49 PM, Shawn Lin wrote:
> On 2015/10/1 17:14, Jaehoon Chung wrote:
>> Dear, All.
>>
>> I will apply patch 01-03 at my repository on today.
>> But i don't know better how i do about other patches relevant to config file.
>>
> 
> Thanks, Jaehoon. :)
> 
> I guess it would be acceptable to pick the config changes, already they were acked by the soc maintainers, via dw_mmc tree when ulf merge dw_mmc tree into his mmc-tree.

Some patches didn't get the maintainer's acked-by. 
I will wait for that..until this week. How about?

Best Regards,
Jaehoon Chung

> 
> 
>> Best Regards,
>> Jaehoon Chung
>>
>> On 09/16/2015 03:40 PM, Shawn Lin wrote:
>>> Synopsys DesignWare mobile storage host controller supports three
>>> types of transfer mode: pio, internal dma and external dma. However,
>>> dw_mmc can only supports pio and internal dma now. Thus some platforms
>>> using dw-mshc integrated with generic dma can't work in dma mode. So we
>>> submit this patch to achieve it.
>>>
>>> And the config option, CONFIG_MMC_DW_IDMAC, was added by Will Newton
>>> (commit:f95f3850) for the first version of dw_mmc and never be touched since
>>> then. At that time dt-bindings hadn't been introduced into dw_mmc yet means
>>> we should select CONFIG_MMC_DW_IDMAC to enable internal dma mode at compile
>>> time. Nowadays, device-tree helps us to support a variety of boards with one
>>> kernel. That's why we need to remove it and decide the transfer mode by reading
>>> dw_mmc's HCON reg at runtime.
>>>
>>> This RFC patch needs lots of ACKs. I know it's hard, but it does need someone
>>> to make the running.
>>>
>>> Patch does the following things:
>>> - remove CONFIG_MMC_DW_IDMAC config option
>>> - add bindings for edmac used by synopsys-dw-mshc
>>>    at runtime
>>> - add edmac support for synopsys-dw-mshc
>>>
>>> Patch is based on next of git://git.linaro.org/people/ulf.hansson/mmc
>>>
>>> Test emmc throughput on my platform with edmac support and without edmac support(pio only)
>>> iozone -L64 -S32 -azecwI -+n -r4k -r64k -r128k -s1g -i0 -i1 -i2 -f datafile -Rb out.xls > /mnt/result.txt
>>> (light cpu loading, Direct IO, fixed line size, all pattern recycle, 1GB data in total)
>>>   ___________________________________________________________
>>> |                   external dma mode                       |
>>> |-----------------------------------------------------------|
>>> |blksz | Random Read | Random Write | Seq Read   | Seq Write|
>>> |-----------------------------------------------------------|
>>> |4kB   |  13953kB/s  |    8602kB/s  | 13672kB/s  |  9785kB/s|
>>> |-----------------------------------------------------------|
>>> |64kB  |  46058kB/s  |   24794kB/s  | 48058kB/s  | 25418kB/s|
>>> |-----------------------------------------------------------|
>>> |128kB |  57026kB/s  |   35117kB/s  | 57375kB/s  | 35183kB/s|
>>> |-----------------------------------------------------------|
>>>                             VS
>>>   ___________________________________________________________
>>> |                          pio mode                         |
>>> |-----------------------------------------------------------|
>>> |blksz | Random Read  | Random Write | Seq Read  | Seq Write|
>>> |-----------------------------------------------------------|
>>> |4kB   |  11720kB/s   |    8644kB/s  | 11549kB/s |  9624kB/s|
>>> |-----------------------------------------------------------|
>>> |64kB  |  21869kB/s   |   24414kB/s  | 22031kB/s | 27986kB/s|
>>> |-----------------------------------------------------------|
>>> |128kB |  23718kB/s   |   34495kB/s  | 24698kB/s | 34637kB/s|
>>> |-----------------------------------------------------------|
>>>
>>>
>>> Changes in v8:
>>> - remove trans_mode variable
>>> - remove unnecessary dma_ops check
>>> - remove unnecessary comment
>>> - fix coding style based on latest ulf's next
>>> - add Acked-by: Jaehoon Chung <jh80.chung@samsung.com>
>>>    for HCON's changes
>>>
>>> Changes in v7:
>>> - rebased on Ulf's next
>>> - combine condition state
>>> - elaborate more about DMA_INTERFACE
>>> - define some macro for DMA_INERFACE value
>>> - spilt HCON ops' changes into another patch
>>>
>>> Changes in v6:
>>> - add trans_mode condition for IDMAC initialization
>>>    suggested by Heiko
>>> - re-test my patch on rk3188 platform and update commit msg
>>> - update performance of pio vs edmac in cover letter
>>>
>>> Changes in v5:
>>> - add the title of cover letter
>>> - fix typo of comment
>>> - add macro for reading HCON register
>>> - add "Acked-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>" for exynos_defconfig patch
>>> - add "Acked-by: Vineet Gupta <vgupta@synopsys.com>" for axs10x_defconfig patch
>>> - add "Acked-by: Govindraj Raja <govindraj.raja@imgtec.com>" and
>>>    "Acked-by: Ralf Baechle <ralf@linux-mips.org>" for pistachio_defconfig patch
>>> - add "Acked-by: Joachim Eastwood <manabian@gmail.com>" for lpc18xx_defconfig patch
>>> - add "Acked-by: Wei Xu <xuwei5@hisilicon.com>" for hisi_defconfig patch
>>> - rebase on "https://github.com/jh80chung/dw-mmc.git tags/dw-mmc-for-ulf-v4.2" for merging easily
>>>
>>> Changes in v4:
>>> - remove "host->trans_mode" and use "host->use_dma" to indicate
>>>    transfer mode.
>>> - remove all bt-bindings' changes since we don't need new properities.
>>> - check transfer mode at runtime by reading HCON reg
>>> - spilt defconfig changes for each sub-architecture
>>> - fix the title of cover letter
>>> - reuse some code for reducing code size
>>>
>>> Changes in v3:
>>> - choose transfer mode at runtime
>>> - remove all CONFIG_MMC_DW_IDMAC config option
>>> - add supports-idmac property for some platforms
>>>
>>> Changes in v2:
>>> - Fix typo of dev_info msg
>>> - remove unused dmach from declaration of dw_mci_dma_slave
>>>
>>> Shawn Lin (10):
>>>    mmc: dw_mmc: Add external dma interface support
>>>    mmc: dw_mmc: use macro for HCON register operations
>>>    Documentation: synopsys-dw-mshc: add bindings for idmac and edmac
>>>    mips: pistachio_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>    arc: axs10x_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>    arm: exynos_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>    arm: hisi_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>    arm: lpc18xx_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>    arm: multi_v7_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>    arm: zx_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>
>>>   .../devicetree/bindings/mmc/synopsys-dw-mshc.txt   |  25 ++
>>>   arch/arc/configs/axs101_defconfig                  |   1 -
>>>   arch/arc/configs/axs103_defconfig                  |   1 -
>>>   arch/arc/configs/axs103_smp_defconfig              |   1 -
>>>   arch/arm/configs/exynos_defconfig                  |   1 -
>>>   arch/arm/configs/hisi_defconfig                    |   1 -
>>>   arch/arm/configs/lpc18xx_defconfig                 |   1 -
>>>   arch/arm/configs/multi_v7_defconfig                |   1 -
>>>   arch/arm/configs/zx_defconfig                      |   1 -
>>>   arch/mips/configs/pistachio_defconfig              |   1 -
>>>   drivers/mmc/host/Kconfig                           |  11 +-
>>>   drivers/mmc/host/dw_mmc-pltfm.c                    |   2 +
>>>   drivers/mmc/host/dw_mmc.c                          | 272 +++++++++++++++++----
>>>   drivers/mmc/host/dw_mmc.h                          |   9 +
>>>   include/linux/mmc/dw_mmc.h                         |  23 +-
>>>   15 files changed, 276 insertions(+), 75 deletions(-)
>>>
>>
>>
>>
>>
> 
> 

Re: [RFC PATCH v8 0/10] Add external dma support for Synopsys MSHC

[Shawn Lin]

On 2015/10/5 9:36, Jaehoon Chung wrote:
> Hi, Shawn.
>
> On 10/02/2015 06:49 PM, Shawn Lin wrote:
>> On 2015/10/1 17:14, Jaehoon Chung wrote:
>>> Dear, All.
>>>
>>> I will apply patch 01-03 at my repository on today.
>>> But i don't know better how i do about other patches relevant to config file.
>>>
>>
>> Thanks, Jaehoon. :)
>>
>> I guess it would be acceptable to pick the config changes, already they were acked by the soc maintainers, via dw_mmc tree when ulf merge dw_mmc tree into his mmc-tree.
>
> Some patches didn't get the maintainer's acked-by.
> I will wait for that..until this week. How about?
>

No problem, take your time. ：）

> Best Regards,
> Jaehoon Chung
>
>>
>>
>>> Best Regards,
>>> Jaehoon Chung
>>>
>>> On 09/16/2015 03:40 PM, Shawn Lin wrote:
>>>> Synopsys DesignWare mobile storage host controller supports three
>>>> types of transfer mode: pio, internal dma and external dma. However,
>>>> dw_mmc can only supports pio and internal dma now. Thus some platforms
>>>> using dw-mshc integrated with generic dma can't work in dma mode. So we
>>>> submit this patch to achieve it.
>>>>
>>>> And the config option, CONFIG_MMC_DW_IDMAC, was added by Will Newton
>>>> (commit:f95f3850) for the first version of dw_mmc and never be touched since
>>>> then. At that time dt-bindings hadn't been introduced into dw_mmc yet means
>>>> we should select CONFIG_MMC_DW_IDMAC to enable internal dma mode at compile
>>>> time. Nowadays, device-tree helps us to support a variety of boards with one
>>>> kernel. That's why we need to remove it and decide the transfer mode by reading
>>>> dw_mmc's HCON reg at runtime.
>>>>
>>>> This RFC patch needs lots of ACKs. I know it's hard, but it does need someone
>>>> to make the running.
>>>>
>>>> Patch does the following things:
>>>> - remove CONFIG_MMC_DW_IDMAC config option
>>>> - add bindings for edmac used by synopsys-dw-mshc
>>>>     at runtime
>>>> - add edmac support for synopsys-dw-mshc
>>>>
>>>> Patch is based on next of git://git.linaro.org/people/ulf.hansson/mmc
>>>>
>>>> Test emmc throughput on my platform with edmac support and without edmac support(pio only)
>>>> iozone -L64 -S32 -azecwI -+n -r4k -r64k -r128k -s1g -i0 -i1 -i2 -f datafile -Rb out.xls > /mnt/result.txt
>>>> (light cpu loading, Direct IO, fixed line size, all pattern recycle, 1GB data in total)
>>>>    ___________________________________________________________
>>>> |                   external dma mode                       |
>>>> |-----------------------------------------------------------|
>>>> |blksz | Random Read | Random Write | Seq Read   | Seq Write|
>>>> |-----------------------------------------------------------|
>>>> |4kB   |  13953kB/s  |    8602kB/s  | 13672kB/s  |  9785kB/s|
>>>> |-----------------------------------------------------------|
>>>> |64kB  |  46058kB/s  |   24794kB/s  | 48058kB/s  | 25418kB/s|
>>>> |-----------------------------------------------------------|
>>>> |128kB |  57026kB/s  |   35117kB/s  | 57375kB/s  | 35183kB/s|
>>>> |-----------------------------------------------------------|
>>>>                              VS
>>>>    ___________________________________________________________
>>>> |                          pio mode                         |
>>>> |-----------------------------------------------------------|
>>>> |blksz | Random Read  | Random Write | Seq Read  | Seq Write|
>>>> |-----------------------------------------------------------|
>>>> |4kB   |  11720kB/s   |    8644kB/s  | 11549kB/s |  9624kB/s|
>>>> |-----------------------------------------------------------|
>>>> |64kB  |  21869kB/s   |   24414kB/s  | 22031kB/s | 27986kB/s|
>>>> |-----------------------------------------------------------|
>>>> |128kB |  23718kB/s   |   34495kB/s  | 24698kB/s | 34637kB/s|
>>>> |-----------------------------------------------------------|
>>>>
>>>>
>>>> Changes in v8:
>>>> - remove trans_mode variable
>>>> - remove unnecessary dma_ops check
>>>> - remove unnecessary comment
>>>> - fix coding style based on latest ulf's next
>>>> - add Acked-by: Jaehoon Chung <jh80.chung@samsung.com>
>>>>     for HCON's changes
>>>>
>>>> Changes in v7:
>>>> - rebased on Ulf's next
>>>> - combine condition state
>>>> - elaborate more about DMA_INTERFACE
>>>> - define some macro for DMA_INERFACE value
>>>> - spilt HCON ops' changes into another patch
>>>>
>>>> Changes in v6:
>>>> - add trans_mode condition for IDMAC initialization
>>>>     suggested by Heiko
>>>> - re-test my patch on rk3188 platform and update commit msg
>>>> - update performance of pio vs edmac in cover letter
>>>>
>>>> Changes in v5:
>>>> - add the title of cover letter
>>>> - fix typo of comment
>>>> - add macro for reading HCON register
>>>> - add "Acked-by: Krzysztof Kozlowski <k.kozlowski@samsung.com>" for exynos_defconfig patch
>>>> - add "Acked-by: Vineet Gupta <vgupta@synopsys.com>" for axs10x_defconfig patch
>>>> - add "Acked-by: Govindraj Raja <govindraj.raja@imgtec.com>" and
>>>>     "Acked-by: Ralf Baechle <ralf@linux-mips.org>" for pistachio_defconfig patch
>>>> - add "Acked-by: Joachim Eastwood <manabian@gmail.com>" for lpc18xx_defconfig patch
>>>> - add "Acked-by: Wei Xu <xuwei5@hisilicon.com>" for hisi_defconfig patch
>>>> - rebase on "https://github.com/jh80chung/dw-mmc.git tags/dw-mmc-for-ulf-v4.2" for merging easily
>>>>
>>>> Changes in v4:
>>>> - remove "host->trans_mode" and use "host->use_dma" to indicate
>>>>     transfer mode.
>>>> - remove all bt-bindings' changes since we don't need new properities.
>>>> - check transfer mode at runtime by reading HCON reg
>>>> - spilt defconfig changes for each sub-architecture
>>>> - fix the title of cover letter
>>>> - reuse some code for reducing code size
>>>>
>>>> Changes in v3:
>>>> - choose transfer mode at runtime
>>>> - remove all CONFIG_MMC_DW_IDMAC config option
>>>> - add supports-idmac property for some platforms
>>>>
>>>> Changes in v2:
>>>> - Fix typo of dev_info msg
>>>> - remove unused dmach from declaration of dw_mci_dma_slave
>>>>
>>>> Shawn Lin (10):
>>>>     mmc: dw_mmc: Add external dma interface support
>>>>     mmc: dw_mmc: use macro for HCON register operations
>>>>     Documentation: synopsys-dw-mshc: add bindings for idmac and edmac
>>>>     mips: pistachio_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>>     arc: axs10x_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>>     arm: exynos_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>>     arm: hisi_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>>     arm: lpc18xx_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>>     arm: multi_v7_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>>     arm: zx_defconfig: remove CONFIG_MMC_DW_IDMAC
>>>>
>>>>    .../devicetree/bindings/mmc/synopsys-dw-mshc.txt   |  25 ++
>>>>    arch/arc/configs/axs101_defconfig                  |   1 -
>>>>    arch/arc/configs/axs103_defconfig                  |   1 -
>>>>    arch/arc/configs/axs103_smp_defconfig              |   1 -
>>>>    arch/arm/configs/exynos_defconfig                  |   1 -
>>>>    arch/arm/configs/hisi_defconfig                    |   1 -
>>>>    arch/arm/configs/lpc18xx_defconfig                 |   1 -
>>>>    arch/arm/configs/multi_v7_defconfig                |   1 -
>>>>    arch/arm/configs/zx_defconfig                      |   1 -
>>>>    arch/mips/configs/pistachio_defconfig              |   1 -
>>>>    drivers/mmc/host/Kconfig                           |  11 +-
>>>>    drivers/mmc/host/dw_mmc-pltfm.c                    |   2 +
>>>>    drivers/mmc/host/dw_mmc.c                          | 272 +++++++++++++++++----
>>>>    drivers/mmc/host/dw_mmc.h                          |   9 +
>>>>    include/linux/mmc/dw_mmc.h                         |  23 +-
>>>>    15 files changed, 276 insertions(+), 75 deletions(-)
>>>>
>>>
>>>
>>>
>>>
>>
>>
>
>
>
>


-- 
Best Regards
Shawn Lin

Re: [RFC PATCH v8 09/10] arm: multi_v7_defconfig: remove CONFIG_MMC_DW_IDMAC

[Jaehoon Chung]

Hi, All.

Is there any other opinion about this patch?

Best Regards,
Jaehoon Chung

On 09/16/2015 03:43 PM, Shawn Lin wrote:
> DesignWare MMC Controller's transfer mode should be decided
> at runtime instead of compile-time. So we remove this config
> option and read dw_mmc's register to select DMA master.
> 
> Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
> ---
> 
> Changes in v8: None
> Changes in v7: None
> Changes in v6: None
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
> 
>  arch/arm/configs/multi_v7_defconfig | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/arch/arm/configs/multi_v7_defconfig b/arch/arm/configs/multi_v7_defconfig
> index 03deb7f..ad929ea 100644
> --- a/arch/arm/configs/multi_v7_defconfig
> +++ b/arch/arm/configs/multi_v7_defconfig
> @@ -539,7 +539,6 @@ CONFIG_MMC_ATMELMCI=y
>  CONFIG_MMC_MVSDIO=y
>  CONFIG_MMC_SDHI=y
>  CONFIG_MMC_DW=y
> -CONFIG_MMC_DW_IDMAC=y
>  CONFIG_MMC_DW_PLTFM=y
>  CONFIG_MMC_DW_EXYNOS=y
>  CONFIG_MMC_DW_ROCKCHIP=y
> 

Re: [RFC PATCH v8 10/10] arm: zx_defconfig: remove CONFIG_MMC_DW_IDMAC

[Jaehoon Chung]

Hi, All.

Is there any other opinion about this patch?


Best Regards,
Jaehoon Chung

On 09/16/2015 03:43 PM, Shawn Lin wrote:
> DesignWare MMC Controller's transfer mode should be decided
> at runtime instead of compile-time. So we remove this config
> option and read dw_mmc's register to select DMA master.
> 
> Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
> ---
> 
> Changes in v8: None
> Changes in v7: None
> Changes in v6: None
> Changes in v5: None
> Changes in v4: None
> Changes in v3: None
> Changes in v2: None
> 
>  arch/arm/configs/zx_defconfig | 1 -
>  1 file changed, 1 deletion(-)
> 
> diff --git a/arch/arm/configs/zx_defconfig b/arch/arm/configs/zx_defconfig
> index b200bb0..ab683fb 100644
> --- a/arch/arm/configs/zx_defconfig
> +++ b/arch/arm/configs/zx_defconfig
> @@ -83,7 +83,6 @@ CONFIG_MMC=y
>  CONFIG_MMC_UNSAFE_RESUME=y
>  CONFIG_MMC_BLOCK_MINORS=16
>  CONFIG_MMC_DW=y
> -CONFIG_MMC_DW_IDMAC=y
>  CONFIG_EXT2_FS=y
>  CONFIG_EXT4_FS=y
>  CONFIG_EXT4_FS_POSIX_ACL=y
>