Skip to contentSkip to conversation

drivers/spi: fsl_qspi: fix controller busy check

PATCH New Thomas Schaefer 3 participants 6 messages Jun 26, 2019 – Aug 14, 2019 SPI

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 drivers/spi: fsl_qspi: fix controller busy check

Quoted 11 lines · Fabio Estevam
Quoted 7 lines

From: Thomas Schaefer <[email protected]>

During QSPI reads, current is_busy_controller function sporadically fails with -ETIMEDOUT due to fixed number of 5 test loops. This patch fixes this by using the readl_poll_timeout function with 1000 us timeout.

This sounds like two different functionalities into one patch. better create a timeout fix in one patch and another patch to migrate read_poll.

I have split the patch into a series of 2 to differentiate functionality, as follows

Thomas

#1 lore raw

PATCH 1/2 drivers/spi: fsl_qspi: fix controller busy check

During QSPI reads, current is_controller_busy function sporadically fails with -ETIMEDOUT due to fixed number of 5 test loops. Using timer functions to wait 1000 us instead will fix this.

---
 drivers/spi/fsl_qspi.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
drivers/spi/fsl_qspi.c +4 -5
index 1598c4f698..2c5937509f 100644
@@ -152,7 +152,7 @@ static inline int is_controller_busy(const struct fsl_qspi_priv *priv)
 	u32 val;
 	const u32 mask = QSPI_SR_BUSY_MASK | QSPI_SR_AHB_ACC_MASK |
 			 QSPI_SR_IP_ACC_MASK;
-	unsigned int retry = 5;
+	unsigned long timeout = timer_get_us() + 1000;
 
 	do {
 		val = qspi_read32(priv->flags, &priv->regs->sr);
@@ -160,10 +160,9 @@ static inline int is_controller_busy(const struct fsl_qspi_priv *priv)
 		if ((~val & mask) == mask)
 			return 0;
 
-		udelay(1);
-	} while (--retry);
-
-	return -ETIMEDOUT;
+		if (timer_get_us() > timeout )
+			return -ETIMEDOUT;
+	} while (1);
 }
 
 /* QSPI support swapping the flash read/write data
Signature · 1 line
2.11.0
#2 lore raw

PATCH 1/2 drivers/spi: fsl_qspi: fix controller busy check

Hi Thomas,

Quoted 6 lines · Thomas Schaefer

During QSPI reads, current is_controller_busy function sporadically fails with -ETIMEDOUT due to fixed number of 5 test loops. Using timer functions to wait 1000 us instead will fix this.

Your patch series submission did not go well.

It seems you used the same original thread to submit the two patches. They ended up with the same Subject:

https://lists.denx.de/pipermail/u-boot/2019-June/374188.html

and

https://lists.denx.de/pipermail/u-boot/2019-June/374189.html

Please resend, preferably via git send-email.

#3 lore raw

PATCH 2/2 drivers/spi: fsl_qspi: fix controller busy check

In reply to Thomas Schaefer

Use readl_poll_timeout instead of explicit calculation.

---
 drivers/spi/fsl_qspi.c | 17 ++++++-----------
 1 file changed, 6 insertions(+), 11 deletions(-)
drivers/spi/fsl_qspi.c +6 -11
index 2c5937509f..41abe1996f 100644
@@ -10,6 +10,7 @@
 #include <spi.h>
 #include <asm/io.h>
 #include <linux/sizes.h>
+#include <linux/iopoll.h>
 #include <dm.h>
 #include <errno.h>
 #include <watchdog.h>
@@ -150,19 +151,13 @@ static void qspi_write32(u32 flags, u32 *addr, u32 val)
 static inline int is_controller_busy(const struct fsl_qspi_priv *priv)
 {
 	u32 val;
-	const u32 mask = QSPI_SR_BUSY_MASK | QSPI_SR_AHB_ACC_MASK |
-			 QSPI_SR_IP_ACC_MASK;
-	unsigned long timeout = timer_get_us() + 1000;
+	u32 mask = QSPI_SR_BUSY_MASK | QSPI_SR_AHB_ACC_MASK |
+		   QSPI_SR_IP_ACC_MASK;
 
-	do {
-		val = qspi_read32(priv->flags, &priv->regs->sr);
+	if (priv->flags & QSPI_FLAG_REGMAP_ENDIAN_BIG)
+		mask = (u32)cpu_to_be32(mask);
 
-		if ((~val & mask) == mask)
-			return 0;
-
-		if (timer_get_us() > timeout )
-			return -ETIMEDOUT;
-	} while (1);
+	return readl_poll_timeout(&priv->regs->sr, val, !(val & mask), 1000);
 }
 
 /* QSPI support swapping the flash read/write data
Signature · 1 line
2.11.0
#4 lore raw

PATCH 2/2 drivers/spi: fsl_qspi: fix controller busy check

Dear Thomas,

Quoted 9 lines

-----Original Message-----
From: U-Boot <[email protected]> On Behalf Of Thomas Schaefer
Sent: Wednesday, June 26, 2019 2:15 PM
To: jagan at amarulasolutions.com; festevam at gmail.com
Cc: trini at konsulko.com; Thomas Schäfer <[email protected]>; u-boot at lists.denx.de
Subject: [U-Boot] [PATCH 2/2] drivers/spi: fsl_qspi: fix controller busy check

Quoted 4 lines

Use readl_poll_timeout instead of explicit calculation.

---

Both of your patch has same subject. Please fix it.

Delegating Jagan for future review.

--pk

#5 lore raw

PATCH 2/2 drivers/spi: fsl_qspi: fix controller busy check

Hi Prabhakar,

I have already reworked this with different subjects and Jagan has applied patches to u-boot-spi/master on July, 1st.

See

c6d0c5eb30b5d677a6eeca5079a4e9a27630b530

and

733391e84bd9d67582b2ecd719b35585b70fe74c

Best regards,
Thomas

-----Ursprüngliche Nachricht-----
Von: Prabhakar Kushwaha <[email protected]>
Gesendet: Mittwoch, 14. August 2019 08:53
An: Thomas Schaefer <[email protected]>; jagan at amarulasolutions.com; festevam at gmail.com
Cc: trini at konsulko.com; Thomas Schaefer <[email protected]>; u-boot at lists.denx.de
Betreff: RE: [U-Boot] [PATCH 2/2] drivers/spi: fsl_qspi: fix controller busy check

Dear Thomas,

Quoted 10 lines

-----Original Message-----
From: U-Boot <[email protected]> On Behalf Of Thomas Schaefer
Sent: Wednesday, June 26, 2019 2:15 PM
To: jagan at amarulasolutions.com; festevam at gmail.com
Cc: trini at konsulko.com; Thomas Schäfer <[email protected]>; u-boot at lists.denx.de
Subject: [U-Boot] [PATCH 2/2] drivers/spi: fsl_qspi: fix controller busy check

Quoted 4 lines

Use readl_poll_timeout instead of explicit calculation.

---

Both of your patch has same subject. Please fix it.

Delegating Jagan for future review.

--pk