Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 8BA37C4345F for ; Wed, 17 Apr 2024 05:40:44 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 0E7A7113174; Wed, 17 Apr 2024 05:40:44 +0000 (UTC) Received: from us-smtp-delivery-44.mimecast.com (us-smtp-delivery-44.mimecast.com [205.139.111.44]) by gabe.freedesktop.org (Postfix) with ESMTPS id 4DEB3113175 for ; Wed, 17 Apr 2024 05:40:43 +0000 (UTC) Received: from mimecast-mx02.redhat.com (mx-ext.redhat.com [66.187.233.73]) by relay.mimecast.com with ESMTP with STARTTLS (version=TLSv1.3, cipher=TLS_AES_256_GCM_SHA384) id us-mta-654-ydoKEAtIPTS-JvCA_0Di7g-1; Wed, 17 Apr 2024 01:40:39 -0400 X-MC-Unique: ydoKEAtIPTS-JvCA_0Di7g-1 Received: from smtp.corp.redhat.com (int-mx06.intmail.prod.int.rdu2.redhat.com [10.11.54.6]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by mimecast-mx02.redhat.com (Postfix) with ESMTPS id 3047828EC114; Wed, 17 Apr 2024 05:40:39 +0000 (UTC) Received: from dreadlord.redhat.com (unknown [10.64.136.67]) by smtp.corp.redhat.com (Postfix) with ESMTP id 27D102166B31; Wed, 17 Apr 2024 05:40:37 +0000 (UTC) From: Dave Airlie To: dri-devel@lists.freedesktop.org Cc: nouveau@lists.freedesktop.org Subject: [PATCH] nouveau: rip out busy fence waits Date: Wed, 17 Apr 2024 15:40:32 +1000 Message-ID: <20240417054032.3145721-1-airlied@gmail.com> MIME-Version: 1.0 X-Scanned-By: MIMEDefang 3.4.1 on 10.11.54.6 X-Mimecast-Spam-Score: 0 X-Mimecast-Originator: gmail.com Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset=WINDOWS-1252; x-default=true X-BeenThere: nouveau@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Nouveau development list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: nouveau-bounces@lists.freedesktop.org Sender: "Nouveau" From: Dave Airlie I'm pretty sure this optimisation is actually not a great idea, and is racy with other things waiting for fences. Just nuke it, there should be no need to do fence waits in a busy CPU loop. Signed-off-by: Dave Airlie --- drivers/gpu/drm/nouveau/nouveau_bo.c | 2 +- drivers/gpu/drm/nouveau/nouveau_chan.c | 2 +- drivers/gpu/drm/nouveau/nouveau_dmem.c | 2 +- drivers/gpu/drm/nouveau/nouveau_fence.c | 30 +------------------------ drivers/gpu/drm/nouveau/nouveau_fence.h | 2 +- drivers/gpu/drm/nouveau/nouveau_gem.c | 2 +- 6 files changed, 6 insertions(+), 34 deletions(-) diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau= /nouveau_bo.c index 8a30f5a0525b..a4e8f625fce6 100644 --- a/drivers/gpu/drm/nouveau/nouveau_bo.c +++ b/drivers/gpu/drm/nouveau/nouveau_bo.c @@ -902,7 +902,7 @@ nouveau_bo_move_m2mf(struct ttm_buffer_object *bo, int = evict, =09 * Without this the operation can timeout and we'll fallback to a =09 * software copy, which might take several minutes to finish. =09 */ -=09nouveau_fence_wait(fence, false, false); +=09nouveau_fence_wait(fence, false); =09ret =3D ttm_bo_move_accel_cleanup(bo, &fence->base, evict, false, =09=09=09=09=09new_reg); =09nouveau_fence_unref(&fence); diff --git a/drivers/gpu/drm/nouveau/nouveau_chan.c b/drivers/gpu/drm/nouve= au/nouveau_chan.c index 7c97b2886807..66fca95c10c7 100644 --- a/drivers/gpu/drm/nouveau/nouveau_chan.c +++ b/drivers/gpu/drm/nouveau/nouveau_chan.c @@ -72,7 +72,7 @@ nouveau_channel_idle(struct nouveau_channel *chan) =20 =09=09ret =3D nouveau_fence_new(&fence, chan); =09=09if (!ret) { -=09=09=09ret =3D nouveau_fence_wait(fence, false, false); +=09=09=09ret =3D nouveau_fence_wait(fence, false); =09=09=09nouveau_fence_unref(&fence); =09=09} =20 diff --git a/drivers/gpu/drm/nouveau/nouveau_dmem.c b/drivers/gpu/drm/nouve= au/nouveau_dmem.c index 12feecf71e75..033a09cd3c8f 100644 --- a/drivers/gpu/drm/nouveau/nouveau_dmem.c +++ b/drivers/gpu/drm/nouveau/nouveau_dmem.c @@ -128,7 +128,7 @@ static void nouveau_dmem_page_free(struct page *page) static void nouveau_dmem_fence_done(struct nouveau_fence **fence) { =09if (fence) { -=09=09nouveau_fence_wait(*fence, true, false); +=09=09nouveau_fence_wait(*fence, false); =09=09nouveau_fence_unref(fence); =09} else { =09=09/* diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.c b/drivers/gpu/drm/nouv= eau/nouveau_fence.c index c3ea3cd933cd..8de941379324 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.c +++ b/drivers/gpu/drm/nouveau/nouveau_fence.c @@ -312,39 +312,11 @@ nouveau_fence_wait_legacy(struct dma_fence *f, bool i= ntr, long wait) =09return timeout - t; } =20 -static int -nouveau_fence_wait_busy(struct nouveau_fence *fence, bool intr) -{ -=09int ret =3D 0; - -=09while (!nouveau_fence_done(fence)) { -=09=09if (time_after_eq(jiffies, fence->timeout)) { -=09=09=09ret =3D -EBUSY; -=09=09=09break; -=09=09} - -=09=09__set_current_state(intr ? -=09=09=09=09 TASK_INTERRUPTIBLE : -=09=09=09=09 TASK_UNINTERRUPTIBLE); - -=09=09if (intr && signal_pending(current)) { -=09=09=09ret =3D -ERESTARTSYS; -=09=09=09break; -=09=09} -=09} - -=09__set_current_state(TASK_RUNNING); -=09return ret; -} - int -nouveau_fence_wait(struct nouveau_fence *fence, bool lazy, bool intr) +nouveau_fence_wait(struct nouveau_fence *fence, bool intr) { =09long ret; =20 -=09if (!lazy) -=09=09return nouveau_fence_wait_busy(fence, intr); - =09ret =3D dma_fence_wait_timeout(&fence->base, intr, 15 * HZ); =09if (ret < 0) =09=09return ret; diff --git a/drivers/gpu/drm/nouveau/nouveau_fence.h b/drivers/gpu/drm/nouv= eau/nouveau_fence.h index bc13110bdfa4..88213014b675 100644 --- a/drivers/gpu/drm/nouveau/nouveau_fence.h +++ b/drivers/gpu/drm/nouveau/nouveau_fence.h @@ -23,7 +23,7 @@ void nouveau_fence_unref(struct nouveau_fence **); =20 int nouveau_fence_emit(struct nouveau_fence *); bool nouveau_fence_done(struct nouveau_fence *); -int nouveau_fence_wait(struct nouveau_fence *, bool lazy, bool intr); +int nouveau_fence_wait(struct nouveau_fence *, bool intr); int nouveau_fence_sync(struct nouveau_bo *, struct nouveau_channel *, boo= l exclusive, bool intr); =20 struct nouveau_fence_chan { diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouvea= u/nouveau_gem.c index 49c2bcbef129..f715e381da69 100644 --- a/drivers/gpu/drm/nouveau/nouveau_gem.c +++ b/drivers/gpu/drm/nouveau/nouveau_gem.c @@ -928,7 +928,7 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void = *data, =09} =20 =09if (sync) { -=09=09if (!(ret =3D nouveau_fence_wait(fence, false, false))) { +=09=09if (!(ret =3D nouveau_fence_wait(fence, false))) { =09=09=09if ((ret =3D dma_fence_get_status(&fence->base)) =3D=3D 1) =09=09=09=09ret =3D 0; =09=09} --=20 2.43.2