linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
To: dri-devel@lists.freedesktop.org, amd-gfx@lists.freedesktop.org,
	linux-pci@vger.kernel.org, ckoenig.leichtzumerken@gmail.com,
	daniel.vetter@ffwll.ch, Harry.Wentland@amd.com
Cc: ppaalanen@gmail.com, Alexander.Deucher@amd.com,
	gregkh@linuxfoundation.org, helgaas@kernel.org,
	Felix.Kuehling@amd.com,
	Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Subject: [PATCH v7 12/16] drm/amdgpu: Fix hang on device removal.
Date: Wed, 12 May 2021 10:26:44 -0400	[thread overview]
Message-ID: <20210512142648.666476-13-andrey.grodzovsky@amd.com> (raw)
In-Reply-To: <20210512142648.666476-1-andrey.grodzovsky@amd.com>

If removing while commands in flight you cannot wait to flush the
HW fences on a ring since the device is gone.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c | 16 ++++++++++------
 1 file changed, 10 insertions(+), 6 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
index 1ffb36bd0b19..fa03702ecbfb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_fence.c
@@ -36,6 +36,7 @@
 #include <linux/firmware.h>
 #include <linux/pm_runtime.h>
 
+#include <drm/drm_drv.h>
 #include "amdgpu.h"
 #include "amdgpu_trace.h"
 
@@ -525,8 +526,7 @@ int amdgpu_fence_driver_init(struct amdgpu_device *adev)
  */
 void amdgpu_fence_driver_fini_hw(struct amdgpu_device *adev)
 {
-	unsigned i, j;
-	int r;
+	int i, r;
 
 	for (i = 0; i < AMDGPU_MAX_RINGS; i++) {
 		struct amdgpu_ring *ring = adev->rings[i];
@@ -535,11 +535,15 @@ void amdgpu_fence_driver_fini_hw(struct amdgpu_device *adev)
 			continue;
 		if (!ring->no_scheduler)
 			drm_sched_fini(&ring->sched);
-		r = amdgpu_fence_wait_empty(ring);
-		if (r) {
-			/* no need to trigger GPU reset as we are unloading */
+		/* You can't wait for HW to signal if it's gone */
+		if (!drm_dev_is_unplugged(&adev->ddev))
+			r = amdgpu_fence_wait_empty(ring);
+		else
+			r = -ENODEV;
+		/* no need to trigger GPU reset as we are unloading */
+		if (r)
 			amdgpu_fence_driver_force_completion(ring);
-		}
+
 		if (ring->fence_drv.irq_src)
 			amdgpu_irq_put(adev, ring->fence_drv.irq_src,
 				       ring->fence_drv.irq_type);
-- 
2.25.1


  parent reply	other threads:[~2021-05-12 14:27 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-12 14:26 [PATCH v7 00/16] RFC Support hot device unplug in amdgpu Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 01/16] drm/ttm: Remap all page faults to per process dummy page Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 02/16] drm/amdgpu: Split amdgpu_device_fini into early and late Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 03/16] drm/amdkfd: Split kfd suspend from device exit Andrey Grodzovsky
2021-05-12 20:33   ` Felix Kuehling
2021-05-12 20:38     ` Andrey Grodzovsky
2021-05-20  3:20     ` [PATCH] drm/amdgpu: Add early fini callback Andrey Grodzovsky
2021-05-20  3:29       ` Felix Kuehling
2021-05-20  3:58         ` Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 04/16] " Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 05/16] drm/amdgpu: Handle IOMMU enabled case Andrey Grodzovsky
2021-05-14 14:41   ` Andrey Grodzovsky
2021-05-14 16:25     ` Felix Kuehling
2021-05-14 16:26       ` Andrey Grodzovsky
2021-05-17 14:38       ` [PATCH] " Andrey Grodzovsky
2021-05-17 14:48         ` Felix Kuehling
2021-05-12 14:26 ` [PATCH v7 06/16] drm/amdgpu: Remap all page faults to per process dummy page Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 07/16] PCI: Add support for dev_groups to struct pci_driver Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 08/16] drm/amdgpu: Convert driver sysfs attributes to static attributes Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 09/16] drm/amdgpu: Guard against write accesses after device removal Andrey Grodzovsky
2021-05-12 20:17   ` Alex Deucher
2021-05-12 20:30     ` Andrey Grodzovsky
2021-05-12 20:50       ` Alex Deucher
2021-05-13 14:47         ` Andrey Grodzovsky
2021-05-13 14:54           ` Alex Deucher
2021-05-12 14:26 ` [PATCH v7 10/16] drm/sched: Make timeout timer rearm conditional Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 11/16] drm/amdgpu: Prevent any job recoveries after device is unplugged Andrey Grodzovsky
2021-05-12 14:26 ` Andrey Grodzovsky [this message]
2021-05-14 14:42   ` [PATCH v7 12/16] drm/amdgpu: Fix hang on device removal Andrey Grodzovsky
2021-05-17 14:40     ` Andrey Grodzovsky
2021-05-17 17:39       ` Alex Deucher
2021-05-17 19:39       ` Christian König
2021-05-17 19:46         ` Andrey Grodzovsky
2021-05-17 19:54           ` Christian König
2021-05-12 14:26 ` [PATCH v7 13/16] drm/scheduler: Fix hang when sched_entity released Andrey Grodzovsky
2021-05-18 14:07   ` Christian König
2021-05-18 15:03     ` Andrey Grodzovsky
2021-05-18 15:15       ` Christian König
2021-05-18 16:17         ` Andrey Grodzovsky
2021-05-18 16:33           ` Christian König
2021-05-18 17:43             ` Andrey Grodzovsky
2021-05-18 18:02               ` Christian König
2021-05-18 18:09                 ` Andrey Grodzovsky
2021-05-18 18:13                   ` Christian König
2021-05-18 18:48                     ` Andrey Grodzovsky
2021-05-18 20:56                       ` Andrey Grodzovsky
2021-05-19 10:57                       ` Christian König
2021-05-19 11:03                         ` Andrey Grodzovsky
2021-05-19 11:46                           ` Christian König
2021-05-19 11:51                             ` Andrey Grodzovsky
2021-05-19 11:56                               ` Christian König
2021-05-19 14:14                                 ` [PATCH] drm/sched: Avoid data corruptions Andrey Grodzovsky
2021-05-19 14:15                                   ` Christian König
2021-05-12 14:26 ` [PATCH v7 14/16] drm/amd/display: Remove superfluous drm_mode_config_cleanup Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 15/16] drm/amdgpu: Verify DMA opearations from device are done Andrey Grodzovsky
2021-05-12 14:26 ` [PATCH v7 16/16] drm/amdgpu: Unmap all MMIO mappings Andrey Grodzovsky
2021-05-14 14:42   ` Andrey Grodzovsky
2021-05-17 14:41     ` Andrey Grodzovsky
2021-05-17 17:43   ` Alex Deucher
2021-05-17 18:46     ` Andrey Grodzovsky
2021-05-17 18:56       ` Alex Deucher
2021-05-17 19:22         ` Andrey Grodzovsky
2021-05-17 19:31     ` [PATCH] " Andrey Grodzovsky
2021-05-18 14:01       ` Andrey Grodzovsky

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=20210512142648.666476-13-andrey.grodzovsky@amd.com \
    --to=andrey.grodzovsky@amd.com \
    --cc=Alexander.Deucher@amd.com \
    --cc=Felix.Kuehling@amd.com \
    --cc=Harry.Wentland@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    --cc=ckoenig.leichtzumerken@gmail.com \
    --cc=daniel.vetter@ffwll.ch \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=helgaas@kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=ppaalanen@gmail.com \
    /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).