2021-12-21 14:31:42

by Arnaud Pouliquen

Subject: [PATCH] remoteproc: stm32: Improve crash recovery time

When a stop is requested on a crash, it is useless to try to shutdown it
gracefully, it is crashed.

In this case don't send the STM32_MBX_SHUTDOWN mailbox message that
will block the recovery during 500 ms, waiting an answer from the
coprocessor.

Signed-off-by: Arnaud Pouliquen <[email protected]>
---
drivers/remoteproc/stm32_rproc.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/remoteproc/stm32_rproc.c b/drivers/remoteproc/stm32_rproc.c
index b643efcf995a..7d782ed9e589 100644
--- a/drivers/remoteproc/stm32_rproc.c
+++ b/drivers/remoteproc/stm32_rproc.c
@@ -494,7 +494,7 @@ static int stm32_rproc_stop(struct rproc *rproc)
int err, idx;

/* request shutdown of the remote processor */
- if (rproc->state != RPROC_OFFLINE) {
+ if (rproc->state != RPROC_OFFLINE && rproc->state != RPROC_CRASHED) {
idx = stm32_rproc_mbox_idx(rproc, STM32_MBX_SHUTDOWN);
if (idx >= 0 && ddata->mb[idx].chan) {
err = mbox_send_message(ddata->mb[idx].chan, "detach");
--
2.17.1

2022-01-03 18:52:31

by Mathieu Poirier

Subject: Re: [PATCH] remoteproc: stm32: Improve crash recovery time

On Tue, Dec 21, 2021 at 03:31:29PM +0100, Arnaud Pouliquen wrote:
> When a stop is requested on a crash, it is useless to try to shutdown it
> gracefully, it is crashed.
>

Right

> In this case don't send the STM32_MBX_SHUTDOWN mailbox message that
> will block the recovery during 500 ms, waiting an answer from the
> coprocessor.
>
> Signed-off-by: Arnaud Pouliquen <[email protected]>
> ---
> drivers/remoteproc/stm32_rproc.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/drivers/remoteproc/stm32_rproc.c b/drivers/remoteproc/stm32_rproc.c
> index b643efcf995a..7d782ed9e589 100644
> --- a/drivers/remoteproc/stm32_rproc.c
> +++ b/drivers/remoteproc/stm32_rproc.c
> @@ -494,7 +494,7 @@ static int stm32_rproc_stop(struct rproc *rproc)
> int err, idx;
>
> /* request shutdown of the remote processor */
> - if (rproc->state != RPROC_OFFLINE) {
> + if (rproc->state != RPROC_OFFLINE && rproc->state != RPROC_CRASHED) {
> idx = stm32_rproc_mbox_idx(rproc, STM32_MBX_SHUTDOWN);
> if (idx >= 0 && ddata->mb[idx].chan) {
> err = mbox_send_message(ddata->mb[idx].chan, "detach");

I have applied this patch.

Thanks,
Mathieu

> --
> 2.17.1
>