Skip to contentSkip to conversation

Minor fixes to sdhci-msm

PATCH Superseded Veerabhadrarao Badiganti 3 participants 14 messages Apr 12, 2020 – Apr 16, 2020

Subsystems (3)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

cover Minor fixes to sdhci-msm

Enable a couple of CAPS that qcom sd host controller supports. Set a quirk for enabling support for auto cmd12.
And enable adma length mismatch error interrupt.

Veerabhadrarao Badiganti (4):
mmc: sdhci-msm: Enable MMC_CAP_WAIT_WHILE_BUSY host capability
mmc: sdhci-msm: Enable MMC_CAP_NEED_RSP_BUSY host capability
mmc: sdhci-msm: Set SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12 quirk
mmc: sdhci-msm: Enable ADMA length mismatch error interrupt

 drivers/mmc/host/sdhci-msm.c | 17 +++++++++++------
 1 file changed, 11 insertions(+), 6 deletions(-)
Signature · 1 line
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc., is a member of Code Aurora Forum, a Linux Foundation Collaborative Project
#1 lore raw

PATCH 1/4 mmc: sdhci-msm: Enable MMC_CAP_WAIT_WHILE_BUSY host capability

MSM sd host controller is capable of HW busy detection of device busy singaling over DAT0 line.

So set MMC_CAP_WAIT_WHILE_BUSY capability for qcom sdhc.

---
 drivers/mmc/host/sdhci-msm.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/mmc/host/sdhci-msm.c +2 -0
index 09ff731..013dcea 100644
@@ -2087,6 +2087,8 @@ static int sdhci_msm_probe(struct platform_device *pdev)
 		goto clk_disable;
 	}
 
+	msm_host->mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY;
+
 	pm_runtime_get_noresume(&pdev->dev);
 	pm_runtime_set_active(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
Signature · 1 line
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc., is a member of Code Aurora Forum, a Linux Foundation Collaborative Project
maintainer Acked-by on v1 1/4
#2 lore raw

Re: [PATCH V1 1/4] mmc: sdhci-msm: Enable MMC_CAP_WAIT_WHILE_BUSY host capability

Quoted 2 lines · Veerabhadrarao Badiganti

MSM sd host controller is capable of HW busy detection of device busy singaling over DAT0 line.

singaling -> signaling

Quoted 4 lines

So set MMC_CAP_WAIT_WHILE_BUSY capability for qcom sdhc.

Quoted 18 lines
---
 drivers/mmc/host/sdhci-msm.c | 2 ++
 1 file changed, 2 insertions(+)
drivers/mmc/host/sdhci-msm.c +2 -0
index 09ff731..013dcea 100644
@@ -2087,6 +2087,8 @@ static int sdhci_msm_probe(struct platform_device *pdev)
 		goto clk_disable;
 	}
 
+	msm_host->mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY;
+
 	pm_runtime_get_noresume(&pdev->dev);
 	pm_runtime_set_active(&pdev->dev);
 	pm_runtime_enable(&pdev->dev);
#3 lore raw

PATCH 2/4 mmc: sdhci-msm: Enable MMC_CAP_NEED_RSP_BUSY host capability

In reply to Veerabhadrarao Badiganti

sdhci-msm controller requires the R1B response for commands that has this response associated with them.

So enable MMC_CAP_NEED_RSP_BUSY capability.

---
 drivers/mmc/host/sdhci-msm.c | 1 +
 1 file changed, 1 insertion(+)
drivers/mmc/host/sdhci-msm.c +1 -0
index 013dcea..d826e9b 100644
@@ -2088,6 +2088,7 @@ static int sdhci_msm_probe(struct platform_device *pdev)
 	}
 
 	msm_host->mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY;
+	msm_host->mmc->caps |= MMC_CAP_NEED_RSP_BUSY;
 
 	pm_runtime_get_noresume(&pdev->dev);
 	pm_runtime_set_active(&pdev->dev);
Signature · 1 line
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc., is a member of Code Aurora Forum, a Linux Foundation Collaborative Project
maintainer Acked-by on v1 2/4
#4 lore raw

Re: [PATCH V1 2/4] mmc: sdhci-msm: Enable MMC_CAP_NEED_RSP_BUSY host capability

Quoted 6 lines · Veerabhadrarao Badiganti

sdhci-msm controller requires the R1B response for commands that has this response associated with them.

So enable MMC_CAP_NEED_RSP_BUSY capability.

Quoted 17 lines
---
 drivers/mmc/host/sdhci-msm.c | 1 +
 1 file changed, 1 insertion(+)
drivers/mmc/host/sdhci-msm.c +1 -0
index 013dcea..d826e9b 100644
@@ -2088,6 +2088,7 @@ static int sdhci_msm_probe(struct platform_device *pdev)
 	}
 
 	msm_host->mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY;
+	msm_host->mmc->caps |= MMC_CAP_NEED_RSP_BUSY;
 
 	pm_runtime_get_noresume(&pdev->dev);
 	pm_runtime_set_active(&pdev->dev);
#5 lore raw

Re: [PATCH V1 2/4] mmc: sdhci-msm: Enable MMC_CAP_NEED_RSP_BUSY host capability

In reply to Veerabhadrarao Badiganti
Quoted 5 lines · Veerabhadrarao Badiganti

sdhci-msm controller requires the R1B response for commands that has this response associated with them.

So enable MMC_CAP_NEED_RSP_BUSY capability.

I assume this potentially should be considered as fix and tagged for stable?

Another question, if there is there an upper limit of the busy timeout in the HW (cmd->busy_timeout) or does the driver use a software timeout that is adjustable?

Kind regards
Uffe

Quoted 20 lines
---
 drivers/mmc/host/sdhci-msm.c | 1 +
 1 file changed, 1 insertion(+)
drivers/mmc/host/sdhci-msm.c +1 -1
index 013dcea..d826e9b 100644
@@ -2088,6 +2088,7 @@ static int sdhci_msm_probe(struct platform_device *pdev)
        }
 
        msm_host->mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY;
+       msm_host->mmc->caps |= MMC_CAP_NEED_RSP_BUSY;
 
        pm_runtime_get_noresume(&pdev->dev);
        pm_runtime_set_active(&pdev->dev);
--

Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc., is a member of Code Aurora Forum, a Linux Foundation Collaborative Project

#6 lore raw

Re: [PATCH V1 2/4] mmc: sdhci-msm: Enable MMC_CAP_NEED_RSP_BUSY host capability

Hi Ulf,

Quoted 7 lines · Ulf Hansson
Quoted 4 lines · Veerabhadrarao Badiganti

sdhci-msm controller requires the R1B response for commands that has this response associated with them.

So enable MMC_CAP_NEED_RSP_BUSY capability.

I assume this potentially should be considered as fix and tagged for stable?

Yes Stable flag can be applied to this.

Patch with MMC_CAP_WAIT_WHILE_BUSY cap also needed besides this. Shall I push V2 with stable flag?

Quoted 3 lines

Another question, if there is there an upper limit of the busy timeout in the HW (cmd->busy_timeout) or does the driver use a software timeout that is adjustable?

The max supported h.w busy timeout value on qcom h/w 21sec.

Quoted 22 lines

Kind regards
Uffe

Quoted 19 lines
---
  drivers/mmc/host/sdhci-msm.c | 1 +
  1 file changed, 1 insertion(+)
drivers/mmc/host/sdhci-msm.c +1 -1
index 013dcea..d826e9b 100644
@@ -2088,6 +2088,7 @@ static int sdhci_msm_probe(struct platform_device *pdev)
         }
 
         msm_host->mmc->caps |= MMC_CAP_WAIT_WHILE_BUSY;
+       msm_host->mmc->caps |= MMC_CAP_NEED_RSP_BUSY;
 
         pm_runtime_get_noresume(&pdev->dev);
         pm_runtime_set_active(&pdev->dev);
--

Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc., is a member of Code Aurora Forum, a Linux Foundation Collaborative Project

#7 lore raw

Re: [PATCH V1 2/4] mmc: sdhci-msm: Enable MMC_CAP_NEED_RSP_BUSY host capability

Quoted 9 lines · Veerabhadrarao Badiganti

Hi Ulf,

Quoted 5 lines · Ulf Hansson
Quoted 3 lines · Veerabhadrarao Badiganti

sdhci-msm controller requires the R1B response for commands that has this response associated with them.

Quoted 6 lines
Quoted 2 lines
Quoted 1 line

So enable MMC_CAP_NEED_RSP_BUSY capability.

I assume this potentially should be considered as fix and tagged for stable?

Yes Stable flag can be applied to this.

Patch with MMC_CAP_WAIT_WHILE_BUSY cap also needed besides this. Shall I push V2 with stable flag?

Ah, so maybe squash both into one patch to simplify for stable?

Yes, please add a stable tag in v2.

Quoted 6 lines
Quoted 4 lines

Another question, if there is there an upper limit of the busy timeout in the HW (cmd->busy_timeout) or does the driver use a software timeout that is adjustable?

The max supported h.w busy timeout value on qcom h/w 21sec.

Alright, that sounds like it should be sufficient for most cases. But perhaps not for eMMC sanitize.

How do you cope with the situation when the R1B response is passed to the driver and the cmd->busy_timeout exceeds the 21s?

[...]

Kind regards
Uffe

#8 lore raw

PATCH 3/4 mmc: sdhci-msm: Set SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12 quirk

In reply to Veerabhadrarao Badiganti

sdhci-msm can support auto cmd12.
So enable SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12 quirk.

---
 drivers/mmc/host/sdhci-msm.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/mmc/host/sdhci-msm.c +3 -1
index d826e9b..482045b 100644
@@ -1882,7 +1882,9 @@ static void sdhci_msm_reset(struct sdhci_host *host, u8 mask)
 static const struct sdhci_pltfm_data sdhci_msm_pdata = {
 	.quirks = SDHCI_QUIRK_BROKEN_CARD_DETECTION |
 		  SDHCI_QUIRK_SINGLE_POWER_WRITE |
-		  SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN,
+		  SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN |
+		  SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12,
+
 	.quirks2 = SDHCI_QUIRK2_PRESET_VALUE_BROKEN,
 	.ops = &sdhci_msm_ops,
 };
Signature · 1 line
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc., is a member of Code Aurora Forum, a Linux Foundation Collaborative Project
maintainer Acked-by on v1 3/4
#9 lore raw

Re: [PATCH V1 3/4] mmc: sdhci-msm: Set SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12 quirk

Quoted 4 lines · Veerabhadrarao Badiganti

sdhci-msm can support auto cmd12.
So enable SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12 quirk.

Quoted 20 lines
---
 drivers/mmc/host/sdhci-msm.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/mmc/host/sdhci-msm.c +3 -1
index d826e9b..482045b 100644
@@ -1882,7 +1882,9 @@ static void sdhci_msm_reset(struct sdhci_host *host, u8 mask)
 static const struct sdhci_pltfm_data sdhci_msm_pdata = {
 	.quirks = SDHCI_QUIRK_BROKEN_CARD_DETECTION |
 		  SDHCI_QUIRK_SINGLE_POWER_WRITE |
-		  SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN,
+		  SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN |
+		  SDHCI_QUIRK_MULTIBLOCK_READ_ACMD12,
+
 	.quirks2 = SDHCI_QUIRK2_PRESET_VALUE_BROKEN,
 	.ops = &sdhci_msm_ops,
 };
#10 lore raw

PATCH 4/4 mmc: sdhci-msm: Enable ADMA length mismatch error interrupt

In reply to Veerabhadrarao Badiganti

ADMA_ERR_SIZE_EN bit of VENDOR_SPECIFIC_FUNC register controls ADMA length mismatch error interrupt. Enable it by default.

And update all bit shift defines with BIT macro.

---
 drivers/mmc/host/sdhci-msm.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/mmc/host/sdhci-msm.c +5 -5
index 482045b..660e1bc 100644
@@ -56,19 +56,19 @@
 #define CORE_FLL_CYCLE_CNT	BIT(18)
 #define CORE_DLL_CLOCK_DISABLE	BIT(21)
 
-#define CORE_VENDOR_SPEC_POR_VAL 0xa1c
+#define CORE_VENDOR_SPEC_POR_VAL 0xa3c
 #define CORE_CLK_PWRSAVE	BIT(1)
 #define CORE_HC_MCLK_SEL_DFLT	(2 << 8)
 #define CORE_HC_MCLK_SEL_HS400	(3 << 8)
 #define CORE_HC_MCLK_SEL_MASK	(3 << 8)
-#define CORE_IO_PAD_PWR_SWITCH_EN	(1 << 15)
-#define CORE_IO_PAD_PWR_SWITCH  (1 << 16)
+#define CORE_IO_PAD_PWR_SWITCH_EN	BIT(15)
+#define CORE_IO_PAD_PWR_SWITCH	BIT(16)
 #define CORE_HC_SELECT_IN_EN	BIT(18)
 #define CORE_HC_SELECT_IN_HS400	(6 << 19)
 #define CORE_HC_SELECT_IN_MASK	(7 << 19)
 
-#define CORE_3_0V_SUPPORT	(1 << 25)
-#define CORE_1_8V_SUPPORT	(1 << 26)
+#define CORE_3_0V_SUPPORT	BIT(25)
+#define CORE_1_8V_SUPPORT	BIT(26)
 #define CORE_VOLT_SUPPORT	(CORE_3_0V_SUPPORT | CORE_1_8V_SUPPORT)
 
 #define CORE_CSR_CDC_CTLR_CFG0		0x130
Signature · 1 line
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc., is a member of Code Aurora Forum, a Linux Foundation Collaborative Project
#11 lore raw

Re: [PATCH V1 4/4] mmc: sdhci-msm: Enable ADMA length mismatch error interrupt

Quoted 15 lines · Veerabhadrarao Badiganti

ADMA_ERR_SIZE_EN bit of VENDOR_SPECIFIC_FUNC register controls ADMA length mismatch error interrupt. Enable it by default.

And update all bit shift defines with BIT macro.

---
  drivers/mmc/host/sdhci-msm.c | 10 +++++-----
  1 file changed, 5 insertions(+), 5 deletions(-)
drivers/mmc/host/sdhci-msm.c +0 -0
index 482045b..660e1bc 100644

drivers/mmc/host/sdhci-msm.c line 59

Quoted 5 lines
drivers/mmc/host/sdhci-msm.c +1 -1
@@ -56,19 +56,19 @@
  #define CORE_FLL_CYCLE_CNT	BIT(18)
  #define CORE_DLL_CLOCK_DISABLE	BIT(21)
 
-#define CORE_VENDOR_SPEC_POR_VAL 0xa1c
+#define CORE_VENDOR_SPEC_POR_VAL 0xa3c

Sorry . This should be 0xa9c, not 0xa3c. Correcting it.

Quoted 19 lines
drivers/mmc/host/sdhci-msm.c +4 -4
  #define CORE_CLK_PWRSAVE	BIT(1)
  #define CORE_HC_MCLK_SEL_DFLT	(2 << 8)
  #define CORE_HC_MCLK_SEL_HS400	(3 << 8)
  #define CORE_HC_MCLK_SEL_MASK	(3 << 8)
-#define CORE_IO_PAD_PWR_SWITCH_EN	(1 << 15)
-#define CORE_IO_PAD_PWR_SWITCH  (1 << 16)
+#define CORE_IO_PAD_PWR_SWITCH_EN	BIT(15)
+#define CORE_IO_PAD_PWR_SWITCH	BIT(16)
  #define CORE_HC_SELECT_IN_EN	BIT(18)
  #define CORE_HC_SELECT_IN_HS400	(6 << 19)
  #define CORE_HC_SELECT_IN_MASK	(7 << 19)
 
-#define CORE_3_0V_SUPPORT	(1 << 25)
-#define CORE_1_8V_SUPPORT	(1 << 26)
+#define CORE_3_0V_SUPPORT	BIT(25)
+#define CORE_1_8V_SUPPORT	BIT(26)
  #define CORE_VOLT_SUPPORT	(CORE_3_0V_SUPPORT | CORE_1_8V_SUPPORT)
 
  #define CORE_CSR_CDC_CTLR_CFG0		0x130
#12 lore raw

PATCH 1/1 mmc: sdhci-msm: Enable ADMA length mismatch error interrupt

In reply to Veerabhadrarao Badiganti

ADMA_ERR_SIZE_EN bit of VENDOR_SPECIFIC_FUNC register controls ADMA length mismatch error interrupt. Enable it by default.

And update all bit shift defines with BIT macro.

---
 drivers/mmc/host/sdhci-msm.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/mmc/host/sdhci-msm.c +5 -5
index 482045b..7d744f9 100644
@@ -56,19 +56,19 @@
 #define CORE_FLL_CYCLE_CNT	BIT(18)
 #define CORE_DLL_CLOCK_DISABLE	BIT(21)
 
-#define CORE_VENDOR_SPEC_POR_VAL 0xa1c
+#define CORE_VENDOR_SPEC_POR_VAL 0xa9c
 #define CORE_CLK_PWRSAVE	BIT(1)
 #define CORE_HC_MCLK_SEL_DFLT	(2 << 8)
 #define CORE_HC_MCLK_SEL_HS400	(3 << 8)
 #define CORE_HC_MCLK_SEL_MASK	(3 << 8)
-#define CORE_IO_PAD_PWR_SWITCH_EN	(1 << 15)
-#define CORE_IO_PAD_PWR_SWITCH  (1 << 16)
+#define CORE_IO_PAD_PWR_SWITCH_EN	BIT(15)
+#define CORE_IO_PAD_PWR_SWITCH	BIT(16)
 #define CORE_HC_SELECT_IN_EN	BIT(18)
 #define CORE_HC_SELECT_IN_HS400	(6 << 19)
 #define CORE_HC_SELECT_IN_MASK	(7 << 19)
 
-#define CORE_3_0V_SUPPORT	(1 << 25)
-#define CORE_1_8V_SUPPORT	(1 << 26)
+#define CORE_3_0V_SUPPORT	BIT(25)
+#define CORE_1_8V_SUPPORT	BIT(26)
 #define CORE_VOLT_SUPPORT	(CORE_3_0V_SUPPORT | CORE_1_8V_SUPPORT)
 
 #define CORE_CSR_CDC_CTLR_CFG0		0x130
Signature · 1 line
Qualcomm India Private Limited, on behalf of Qualcomm Innovation Center, Inc., is a member of Code Aurora Forum, a Linux Foundation Collaborative Project
maintainer Acked-by on v1 1/1
#13 lore raw

Re: [PATCH V1] mmc: sdhci-msm: Enable ADMA length mismatch error interrupt

Quoted 6 lines · Veerabhadrarao Badiganti

ADMA_ERR_SIZE_EN bit of VENDOR_SPECIFIC_FUNC register controls ADMA length mismatch error interrupt. Enable it by default.

And update all bit shift defines with BIT macro.

Quoted 34 lines
---
 drivers/mmc/host/sdhci-msm.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
drivers/mmc/host/sdhci-msm.c +5 -5
index 482045b..7d744f9 100644
@@ -56,19 +56,19 @@
 #define CORE_FLL_CYCLE_CNT	BIT(18)
 #define CORE_DLL_CLOCK_DISABLE	BIT(21)
 
-#define CORE_VENDOR_SPEC_POR_VAL 0xa1c
+#define CORE_VENDOR_SPEC_POR_VAL 0xa9c
 #define CORE_CLK_PWRSAVE	BIT(1)
 #define CORE_HC_MCLK_SEL_DFLT	(2 << 8)
 #define CORE_HC_MCLK_SEL_HS400	(3 << 8)
 #define CORE_HC_MCLK_SEL_MASK	(3 << 8)
-#define CORE_IO_PAD_PWR_SWITCH_EN	(1 << 15)
-#define CORE_IO_PAD_PWR_SWITCH  (1 << 16)
+#define CORE_IO_PAD_PWR_SWITCH_EN	BIT(15)
+#define CORE_IO_PAD_PWR_SWITCH	BIT(16)
 #define CORE_HC_SELECT_IN_EN	BIT(18)
 #define CORE_HC_SELECT_IN_HS400	(6 << 19)
 #define CORE_HC_SELECT_IN_MASK	(7 << 19)
 
-#define CORE_3_0V_SUPPORT	(1 << 25)
-#define CORE_1_8V_SUPPORT	(1 << 26)
+#define CORE_3_0V_SUPPORT	BIT(25)
+#define CORE_1_8V_SUPPORT	BIT(26)
 #define CORE_VOLT_SUPPORT	(CORE_3_0V_SUPPORT | CORE_1_8V_SUPPORT)
 
 #define CORE_CSR_CDC_CTLR_CFG0		0x130