All of lore.kernel.org
 help / color / mirror / Atom feed
* [PATCH v3] drm/ttm: fix error handling if no BO can be swapped out
@ 2021-04-22 10:25 Shiwu Zhang
  2021-04-22 11:10 ` Christian König
  0 siblings, 1 reply; 3+ messages in thread
From: Shiwu Zhang @ 2021-04-22 10:25 UTC (permalink / raw)
  To: dri-devel; +Cc: christian.koenig

In case that all pre-allocated BOs are busy, just continue to populate
BOs since likely half of system memory in total is still free.

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
---
 drivers/gpu/drm/ttm/ttm_device.c | 2 +-
 drivers/gpu/drm/ttm/ttm_tt.c     | 2 ++
 2 files changed, 3 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/ttm/ttm_device.c b/drivers/gpu/drm/ttm/ttm_device.c
index 1f2024164d72..a48fe4dccd61 100644
--- a/drivers/gpu/drm/ttm/ttm_device.c
+++ b/drivers/gpu/drm/ttm/ttm_device.c
@@ -112,7 +112,7 @@ int ttm_global_swapout(struct ttm_operation_ctx *ctx, gfp_t gfp_flags)
 {
 	struct ttm_global *glob = &ttm_glob;
 	struct ttm_device *bdev;
-	int ret = -EBUSY;
+	int ret;
 
 	mutex_lock(&ttm_global_mutex);
 	list_for_each_entry(bdev, &glob->device_list, device_list) {
diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c
index 48c407cff112..539e0232cb3b 100644
--- a/drivers/gpu/drm/ttm/ttm_tt.c
+++ b/drivers/gpu/drm/ttm/ttm_tt.c
@@ -329,6 +329,8 @@ int ttm_tt_populate(struct ttm_device *bdev,
 	       ttm_dma32_pages_limit) {
 
 		ret = ttm_global_swapout(ctx, GFP_KERNEL);
+		if (ret == 0)
+			break;
 		if (ret < 0)
 			goto error;
 	}
-- 
2.17.1

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

^ permalink raw reply related	[flat|nested] 3+ messages in thread
* Re: [PATCH v3] drm/ttm: fix error handling if no BO can be swapped out
  2021-04-22 10:25 [PATCH v3] drm/ttm: fix error handling if no BO can be swapped out Shiwu Zhang
@ 2021-04-22 11:10 ` Christian König
  0 siblings, 0 replies; 3+ messages in thread
From: Christian König @ 2021-04-22 11:10 UTC (permalink / raw)
  To: Shiwu Zhang, dri-devel

Am 22.04.21 um 12:25 schrieb Shiwu Zhang:
> In case that all pre-allocated BOs are busy, just continue to populate

I'm not a native speaker of English either, but I think that should read 
"previously allocated".

> BOs since likely half of system memory in total is still free.
>
> Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>

> ---
>   drivers/gpu/drm/ttm/ttm_device.c | 2 +-
>   drivers/gpu/drm/ttm/ttm_tt.c     | 2 ++
>   2 files changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/gpu/drm/ttm/ttm_device.c b/drivers/gpu/drm/ttm/ttm_device.c
> index 1f2024164d72..a48fe4dccd61 100644
> --- a/drivers/gpu/drm/ttm/ttm_device.c
> +++ b/drivers/gpu/drm/ttm/ttm_device.c
> @@ -112,7 +112,7 @@ int ttm_global_swapout(struct ttm_operation_ctx *ctx, gfp_t gfp_flags)
>   {
>   	struct ttm_global *glob = &ttm_glob;
>   	struct ttm_device *bdev;
> -	int ret = -EBUSY;
> +	int ret;

Oh, we have been inconsistent here?

In this case we should indeed zero initialize the variable and also fix 
the only other user in vmwgfx!

Going to take care of this myself.

Thanks for the help,
Christian.

>   
>   	mutex_lock(&ttm_global_mutex);
>   	list_for_each_entry(bdev, &glob->device_list, device_list) {
> diff --git a/drivers/gpu/drm/ttm/ttm_tt.c b/drivers/gpu/drm/ttm/ttm_tt.c
> index 48c407cff112..539e0232cb3b 100644
> --- a/drivers/gpu/drm/ttm/ttm_tt.c
> +++ b/drivers/gpu/drm/ttm/ttm_tt.c
> @@ -329,6 +329,8 @@ int ttm_tt_populate(struct ttm_device *bdev,
>   	       ttm_dma32_pages_limit) {
>   
>   		ret = ttm_global_swapout(ctx, GFP_KERNEL);
> +		if (ret == 0)
> +			break;
>   		if (ret < 0)
>   			goto error;
>   	}

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

^ permalink raw reply	[flat|nested] 3+ messages in thread
* Re: [PATCH v3] drm/ttm: fix error handling if no BO can be swapped out
@ 2021-04-22 21:34 kernel test robot
  0 siblings, 0 replies; 3+ messages in thread
From: kernel test robot @ 2021-04-22 21:34 UTC (permalink / raw)
  To: kbuild

[-- Attachment #1: Type: text/plain, Size: 3491 bytes --]

CC: kbuild-all(a)lists.01.org
In-Reply-To: <20210422102521.22580-1-shiwu.zhang@amd.com>
References: <20210422102521.22580-1-shiwu.zhang@amd.com>
TO: Shiwu Zhang <shiwu.zhang@amd.com>
TO: dri-devel(a)lists.freedesktop.org
CC: christian.koenig(a)amd.com

Hi Shiwu,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on drm-tip/drm-tip]
[also build test WARNING on next-20210422]
[cannot apply to drm-intel/for-linux-next drm-exynos/exynos-drm-next tegra-drm/drm/tegra/for-next linus/master drm/drm-next v5.12-rc8]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Shiwu-Zhang/drm-ttm-fix-error-handling-if-no-BO-can-be-swapped-out/20210422-182800
base:   git://anongit.freedesktop.org/drm/drm-tip drm-tip
:::::: branch date: 11 hours ago
:::::: commit date: 11 hours ago
config: x86_64-randconfig-m001-20210421 (attached as .config)
compiler: gcc-9 (Debian 9.3.0-22) 9.3.0

If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>

New smatch warnings:
drivers/gpu/drm/ttm/ttm_device.c:126 ttm_global_swapout() error: uninitialized symbol 'ret'.

Old smatch warnings:
drivers/gpu/drm/ttm/ttm_device.c:164 ttm_device_swapout() warn: inconsistent returns '&bdev->lru_lock'.

vim +/ret +126 drivers/gpu/drm/ttm/ttm_device.c

8af8a109b34fa8 Christian König 2020-10-01  106  
ebd59851c796c2 Christian König 2020-10-06  107  /**
ebd59851c796c2 Christian König 2020-10-06  108   * A buffer object shrink method that tries to swap out the first
ebd59851c796c2 Christian König 2020-10-06  109   * buffer object on the global::swap_lru list.
ebd59851c796c2 Christian König 2020-10-06  110   */
ebd59851c796c2 Christian König 2020-10-06  111  int ttm_global_swapout(struct ttm_operation_ctx *ctx, gfp_t gfp_flags)
ebd59851c796c2 Christian König 2020-10-06  112  {
ebd59851c796c2 Christian König 2020-10-06  113  	struct ttm_global *glob = &ttm_glob;
f9e2a03e110ad0 Christian König 2020-10-06  114  	struct ttm_device *bdev;
5ac121e880c745 Shiwu Zhang     2021-04-22  115  	int ret;
f9e2a03e110ad0 Christian König 2020-10-06  116  
f9e2a03e110ad0 Christian König 2020-10-06  117  	mutex_lock(&ttm_global_mutex);
f9e2a03e110ad0 Christian König 2020-10-06  118  	list_for_each_entry(bdev, &glob->device_list, device_list) {
f9e2a03e110ad0 Christian König 2020-10-06  119  		ret = ttm_device_swapout(bdev, ctx, gfp_flags);
f9e2a03e110ad0 Christian König 2020-10-06  120  		if (ret > 0) {
f9e2a03e110ad0 Christian König 2020-10-06  121  			list_move_tail(&bdev->device_list, &glob->device_list);
f9e2a03e110ad0 Christian König 2020-10-06  122  			break;
f9e2a03e110ad0 Christian König 2020-10-06  123  		}
f9e2a03e110ad0 Christian König 2020-10-06  124  	}
f9e2a03e110ad0 Christian König 2020-10-06  125  	mutex_unlock(&ttm_global_mutex);
f9e2a03e110ad0 Christian König 2020-10-06 @126  	return ret;
f9e2a03e110ad0 Christian König 2020-10-06  127  }
f9e2a03e110ad0 Christian König 2020-10-06  128  EXPORT_SYMBOL(ttm_global_swapout);
f9e2a03e110ad0 Christian König 2020-10-06  129  

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org

[-- Attachment #2: config.gz --]
[-- Type: application/gzip, Size: 33143 bytes --]

^ permalink raw reply	[flat|nested] 3+ messages in thread
end of thread, other threads:[~2021-04-22 21:34 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-04-22 10:25 [PATCH v3] drm/ttm: fix error handling if no BO can be swapped out Shiwu Zhang
2021-04-22 11:10 ` Christian König
2021-04-22 21:34 kernel test robot

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.