* [PATCH] mtd: rawnand: mtk: Fix WAITRDY break condition and timeout
@ 2021-03-09 0:01 ` Hauke Mehrtens
0 siblings, 0 replies; 6+ messages in thread
From: Hauke Mehrtens @ 2021-03-09 0:01 UTC (permalink / raw)
To: miquel.raynal, richard, boris.brezillon
Cc: vigneshr, matthias.bgg, linux-mtd, linux-mediatek, Hauke Mehrtens
This fixes NAND_OP_WAITRDY_INSTR operation in the driver. Without this
change the driver waits till the system is busy, but we should wait till
the busy flag is cleared. The readl_poll_timeout() function gets a break
condition, not a wait condition.
In addition fix the timeout. The timeout_ms is given in ms, but the
readl_poll_timeout() function takes the timeout in us. Multiple the
given timeout by 1000 to convert it.
Without this change, the driver does not work at all, it doesn't even
identify the NAND chip.
Fixes: 5197360f9e09 ("mtd: rawnand: mtk: Convert the driver to exec_op()")
Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
---
drivers/mtd/nand/raw/mtk_nand.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/mtd/nand/raw/mtk_nand.c b/drivers/mtd/nand/raw/mtk_nand.c
index 57f1f1708994..5c5c92132287 100644
--- a/drivers/mtd/nand/raw/mtk_nand.c
+++ b/drivers/mtd/nand/raw/mtk_nand.c
@@ -488,8 +488,8 @@ static int mtk_nfc_exec_instr(struct nand_chip *chip,
return 0;
case NAND_OP_WAITRDY_INSTR:
return readl_poll_timeout(nfc->regs + NFI_STA, status,
- status & STA_BUSY, 20,
- instr->ctx.waitrdy.timeout_ms);
+ !(status & STA_BUSY), 20,
+ instr->ctx.waitrdy.timeout_ms * 1000);
default:
break;
}
--
2.30.1
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply related [flat|nested] 6+ messages in thread* Re: [PATCH] mtd: rawnand: mtk: Fix WAITRDY break condition and timeout
2021-03-09 0:01 ` Hauke Mehrtens
@ 2021-03-09 7:06 ` Boris Brezillon
-1 siblings, 0 replies; 6+ messages in thread
From: Boris Brezillon @ 2021-03-09 7:06 UTC (permalink / raw)
To: Hauke Mehrtens
Cc: miquel.raynal, richard, vigneshr, matthias.bgg, linux-mtd,
linux-mediatek
On Tue, 9 Mar 2021 01:01:07 +0100
Hauke Mehrtens <hauke@hauke-m.de> wrote:
> This fixes NAND_OP_WAITRDY_INSTR operation in the driver. Without this
> change the driver waits till the system is busy, but we should wait till
> the busy flag is cleared. The readl_poll_timeout() function gets a break
> condition, not a wait condition.
>
> In addition fix the timeout. The timeout_ms is given in ms, but the
> readl_poll_timeout() function takes the timeout in us. Multiple the
> given timeout by 1000 to convert it.
>
> Without this change, the driver does not work at all, it doesn't even
> identify the NAND chip.
>
> Fixes: 5197360f9e09 ("mtd: rawnand: mtk: Convert the driver to exec_op()")
> Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
> ---
> drivers/mtd/nand/raw/mtk_nand.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mtd/nand/raw/mtk_nand.c b/drivers/mtd/nand/raw/mtk_nand.c
> index 57f1f1708994..5c5c92132287 100644
> --- a/drivers/mtd/nand/raw/mtk_nand.c
> +++ b/drivers/mtd/nand/raw/mtk_nand.c
> @@ -488,8 +488,8 @@ static int mtk_nfc_exec_instr(struct nand_chip *chip,
> return 0;
> case NAND_OP_WAITRDY_INSTR:
> return readl_poll_timeout(nfc->regs + NFI_STA, status,
> - status & STA_BUSY, 20,
> - instr->ctx.waitrdy.timeout_ms);
> + !(status & STA_BUSY), 20,
> + instr->ctx.waitrdy.timeout_ms * 1000);
> default:
> break;
> }
_______________________________________________
Linux-mediatek mailing list
Linux-mediatek@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-mediatek
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH] mtd: rawnand: mtk: Fix WAITRDY break condition and timeout
@ 2021-03-09 7:06 ` Boris Brezillon
0 siblings, 0 replies; 6+ messages in thread
From: Boris Brezillon @ 2021-03-09 7:06 UTC (permalink / raw)
To: Hauke Mehrtens
Cc: miquel.raynal, richard, vigneshr, matthias.bgg, linux-mtd,
linux-mediatek
On Tue, 9 Mar 2021 01:01:07 +0100
Hauke Mehrtens <hauke@hauke-m.de> wrote:
> This fixes NAND_OP_WAITRDY_INSTR operation in the driver. Without this
> change the driver waits till the system is busy, but we should wait till
> the busy flag is cleared. The readl_poll_timeout() function gets a break
> condition, not a wait condition.
>
> In addition fix the timeout. The timeout_ms is given in ms, but the
> readl_poll_timeout() function takes the timeout in us. Multiple the
> given timeout by 1000 to convert it.
>
> Without this change, the driver does not work at all, it doesn't even
> identify the NAND chip.
>
> Fixes: 5197360f9e09 ("mtd: rawnand: mtk: Convert the driver to exec_op()")
> Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
> ---
> drivers/mtd/nand/raw/mtk_nand.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mtd/nand/raw/mtk_nand.c b/drivers/mtd/nand/raw/mtk_nand.c
> index 57f1f1708994..5c5c92132287 100644
> --- a/drivers/mtd/nand/raw/mtk_nand.c
> +++ b/drivers/mtd/nand/raw/mtk_nand.c
> @@ -488,8 +488,8 @@ static int mtk_nfc_exec_instr(struct nand_chip *chip,
> return 0;
> case NAND_OP_WAITRDY_INSTR:
> return readl_poll_timeout(nfc->regs + NFI_STA, status,
> - status & STA_BUSY, 20,
> - instr->ctx.waitrdy.timeout_ms);
> + !(status & STA_BUSY), 20,
> + instr->ctx.waitrdy.timeout_ms * 1000);
> default:
> break;
> }
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [PATCH] mtd: rawnand: mtk: Fix WAITRDY break condition and timeout
2021-03-09 0:01 ` Hauke Mehrtens
@ 2021-03-11 11:39 ` Miquel Raynal
-1 siblings, 0 replies; 6+ messages in thread
From: Miquel Raynal @ 2021-03-11 11:39 UTC (permalink / raw)
To: Hauke Mehrtens
Cc: richard, boris.brezillon, vigneshr, matthias.bgg, linux-mtd,
linux-mediatek
Hi Hauke,
Hauke Mehrtens <hauke@hauke-m.de> wrote on Tue, 9 Mar 2021 01:01:07
+0100:
> This fixes NAND_OP_WAITRDY_INSTR operation in the driver. Without this
> change the driver waits till the system is busy, but we should wait till
> the busy flag is cleared. The readl_poll_timeout() function gets a break
> condition, not a wait condition.
>
> In addition fix the timeout. The timeout_ms is given in ms, but the
> readl_poll_timeout() function takes the timeout in us. Multiple the
> given timeout by 1000 to convert it.
>
> Without this change, the driver does not work at all, it doesn't even
> identify the NAND chip.
>
> Fixes: 5197360f9e09 ("mtd: rawnand: mtk: Convert the driver to exec_op()")
> Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
> ---
> drivers/mtd/nand/raw/mtk_nand.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mtd/nand/raw/mtk_nand.c b/drivers/mtd/nand/raw/mtk_nand.c
> index 57f1f1708994..5c5c92132287 100644
> --- a/drivers/mtd/nand/raw/mtk_nand.c
> +++ b/drivers/mtd/nand/raw/mtk_nand.c
> @@ -488,8 +488,8 @@ static int mtk_nfc_exec_instr(struct nand_chip *chip,
> return 0;
> case NAND_OP_WAITRDY_INSTR:
> return readl_poll_timeout(nfc->regs + NFI_STA, status,
> - status & STA_BUSY, 20,
> - instr->ctx.waitrdy.timeout_ms);
> + !(status & STA_BUSY), 20,
> + instr->ctx.waitrdy.timeout_ms * 1000);
> default:
> break;
> }
Applied to mtd/fixes.
Thanks,
Miquèl
_______________________________________________
Linux-mediatek mailing list
Linux-mediatek@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-mediatek
^ permalink raw reply [flat|nested] 6+ messages in thread* Re: [PATCH] mtd: rawnand: mtk: Fix WAITRDY break condition and timeout
@ 2021-03-11 11:39 ` Miquel Raynal
0 siblings, 0 replies; 6+ messages in thread
From: Miquel Raynal @ 2021-03-11 11:39 UTC (permalink / raw)
To: Hauke Mehrtens
Cc: richard, boris.brezillon, vigneshr, matthias.bgg, linux-mtd,
linux-mediatek
Hi Hauke,
Hauke Mehrtens <hauke@hauke-m.de> wrote on Tue, 9 Mar 2021 01:01:07
+0100:
> This fixes NAND_OP_WAITRDY_INSTR operation in the driver. Without this
> change the driver waits till the system is busy, but we should wait till
> the busy flag is cleared. The readl_poll_timeout() function gets a break
> condition, not a wait condition.
>
> In addition fix the timeout. The timeout_ms is given in ms, but the
> readl_poll_timeout() function takes the timeout in us. Multiple the
> given timeout by 1000 to convert it.
>
> Without this change, the driver does not work at all, it doesn't even
> identify the NAND chip.
>
> Fixes: 5197360f9e09 ("mtd: rawnand: mtk: Convert the driver to exec_op()")
> Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
> ---
> drivers/mtd/nand/raw/mtk_nand.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/mtd/nand/raw/mtk_nand.c b/drivers/mtd/nand/raw/mtk_nand.c
> index 57f1f1708994..5c5c92132287 100644
> --- a/drivers/mtd/nand/raw/mtk_nand.c
> +++ b/drivers/mtd/nand/raw/mtk_nand.c
> @@ -488,8 +488,8 @@ static int mtk_nfc_exec_instr(struct nand_chip *chip,
> return 0;
> case NAND_OP_WAITRDY_INSTR:
> return readl_poll_timeout(nfc->regs + NFI_STA, status,
> - status & STA_BUSY, 20,
> - instr->ctx.waitrdy.timeout_ms);
> + !(status & STA_BUSY), 20,
> + instr->ctx.waitrdy.timeout_ms * 1000);
> default:
> break;
> }
Applied to mtd/fixes.
Thanks,
Miquèl
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply [flat|nested] 6+ messages in thread