From: Boris Brezillon <boris.brezillon@bootlin.com>
To: Eric Anholt <eric@anholt.net>
Cc: dri-devel@lists.freedesktop.org, linux-kernel@vger.kernel.org,
Daniel Vetter <daniel.vetter@ffwll.ch>
Subject: Re: [PATCH 3/4] drm/v3d: Clean up the reservation object setup.
Date: Tue, 13 Nov 2018 11:22:49 +0100 [thread overview]
Message-ID: <20181113112249.532ed593@bbrezillon> (raw)
In-Reply-To: <20181108161654.19888-4-eric@anholt.net>
On Thu, 8 Nov 2018 08:16:53 -0800
Eric Anholt <eric@anholt.net> wrote:
> The extra to_v3d_bo() calls came from copying this from the vc4
> driver, which stored the cma gem object in the structs.
>
> Signed-off-by: Eric Anholt <eric@anholt.net>
Reviewed-by: Boris Brezillon <boris.brezillon@bootlin.com>
> ---
> drivers/gpu/drm/v3d/v3d_gem.c | 32 +++++++++++---------------------
> 1 file changed, 11 insertions(+), 21 deletions(-)
>
> diff --git a/drivers/gpu/drm/v3d/v3d_gem.c b/drivers/gpu/drm/v3d/v3d_gem.c
> index b88c96911453..d0dfdcbbd42c 100644
> --- a/drivers/gpu/drm/v3d/v3d_gem.c
> +++ b/drivers/gpu/drm/v3d/v3d_gem.c
> @@ -214,10 +214,8 @@ v3d_attach_object_fences(struct v3d_exec_info *exec)
> int i;
>
> for (i = 0; i < exec->bo_count; i++) {
> - bo = to_v3d_bo(&exec->bo[i]->base);
> -
> /* XXX: Use shared fences for read-only objects. */
> - reservation_object_add_excl_fence(bo->resv, out_fence);
> + reservation_object_add_excl_fence(exec->bo[i]->resv, out_fence);
> }
> }
>
> @@ -228,11 +226,8 @@ v3d_unlock_bo_reservations(struct drm_device *dev,
> {
> int i;
>
> - for (i = 0; i < exec->bo_count; i++) {
> - struct v3d_bo *bo = to_v3d_bo(&exec->bo[i]->base);
> -
> - ww_mutex_unlock(&bo->resv->lock);
> - }
> + for (i = 0; i < exec->bo_count; i++)
> + ww_mutex_unlock(&exec->bo[i]->resv->lock);
>
> ww_acquire_fini(acquire_ctx);
> }
> @@ -251,13 +246,13 @@ v3d_lock_bo_reservations(struct drm_device *dev,
> {
> int contended_lock = -1;
> int i, ret;
> - struct v3d_bo *bo;
>
> ww_acquire_init(acquire_ctx, &reservation_ww_class);
>
> retry:
> if (contended_lock != -1) {
> - bo = to_v3d_bo(&exec->bo[contended_lock]->base);
> + struct v3d_bo *bo = exec->bo[contended_lock];
> +
> ret = ww_mutex_lock_slow_interruptible(&bo->resv->lock,
> acquire_ctx);
> if (ret) {
> @@ -270,19 +265,16 @@ v3d_lock_bo_reservations(struct drm_device *dev,
> if (i == contended_lock)
> continue;
>
> - bo = to_v3d_bo(&exec->bo[i]->base);
> -
> - ret = ww_mutex_lock_interruptible(&bo->resv->lock, acquire_ctx);
> + ret = ww_mutex_lock_interruptible(&exec->bo[i]->resv->lock,
> + acquire_ctx);
> if (ret) {
> int j;
>
> - for (j = 0; j < i; j++) {
> - bo = to_v3d_bo(&exec->bo[j]->base);
> - ww_mutex_unlock(&bo->resv->lock);
> - }
> + for (j = 0; j < i; j++)
> + ww_mutex_unlock(&exec->bo[j]->resv->lock);
>
> if (contended_lock != -1 && contended_lock >= i) {
> - bo = to_v3d_bo(&exec->bo[contended_lock]->base);
> + struct v3d_bo *bo = exec->bo[contended_lock];
>
> ww_mutex_unlock(&bo->resv->lock);
> }
> @@ -303,9 +295,7 @@ v3d_lock_bo_reservations(struct drm_device *dev,
> * before we commit the CL to the hardware.
> */
> for (i = 0; i < exec->bo_count; i++) {
> - bo = to_v3d_bo(&exec->bo[i]->base);
> -
> - ret = reservation_object_reserve_shared(bo->resv, 1);
> + ret = reservation_object_reserve_shared(exec->bo[i]->resv, 1);
> if (ret) {
> v3d_unlock_bo_reservations(dev, exec, acquire_ctx);
> return ret;
next prev parent reply other threads:[~2018-11-13 10:22 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-11-08 16:16 [PATCH 0/4] V3D TFU engine support Eric Anholt
2018-11-08 16:16 ` [PATCH 1/4] drm/v3d: Fix whitespace inconsistency in the header Eric Anholt
2018-11-13 10:22 ` Boris Brezillon
2018-11-08 16:16 ` [PATCH 2/4] drm/v3d: Update a comment about what uses v3d_job_dependency() Eric Anholt
2018-11-13 10:22 ` Boris Brezillon
2018-11-08 16:16 ` [PATCH 3/4] drm/v3d: Clean up the reservation object setup Eric Anholt
2018-11-13 10:22 ` Boris Brezillon [this message]
2018-11-08 16:16 ` [PATCH 4/4] drm/v3d: Add support for submitting jobs to the TFU Eric Anholt
2018-11-13 22:03 ` Eric Anholt
2018-11-28 19:45 ` Dave Emett
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20181113112249.532ed593@bbrezillon \
--to=boris.brezillon@bootlin.com \
--cc=daniel.vetter@ffwll.ch \
--cc=dri-devel@lists.freedesktop.org \
--cc=eric@anholt.net \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).