summaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
diff options
context:
space:
mode:
authorPixel Ding <Pixel.Ding@amd.com>2017-02-23 11:10:33 +0800
committerAlex Deucher <alexander.deucher@amd.com>2017-04-28 17:32:40 -0400
commitee73164a0d8d2fd98f666a5dd35da1d9a19ec009 (patch)
tree9ada31e6635856fb950465546a6f1dee44e25f63 /drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
parentad2fed9ad5907c35e132e43420a0e47ab22350f0 (diff)
downloadop-kernel-dev-ee73164a0d8d2fd98f666a5dd35da1d9a19ec009.zip
op-kernel-dev-ee73164a0d8d2fd98f666a5dd35da1d9a19ec009.tar.gz
drm/amdgpu/virt: don't check VALID bit for FLR completion message
The interrupt after FLR is missed sometimes due to hardware reason, so guest driver get the notification of FLR completion via polling message. Then host doesn't write VALID bit to avoid sending interrupt, otherwise the completion will be handled twice. So there's a valid message without VALID bit for FLR completion, driver should handle it without checking. Signed-off-by: Pixel Ding <Pixel.Ding@amd.com> Reviewed-by: Xiangliang Yu <Xiangliang.Yu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Diffstat (limited to 'drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c')
-rw-r--r--drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c9
1 files changed, 6 insertions, 3 deletions
diff --git a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
index 70a3dd1..7bdc51b 100644
--- a/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
+++ b/drivers/gpu/drm/amd/amdgpu/mxgpu_vi.c
@@ -368,9 +368,12 @@ static int xgpu_vi_mailbox_rcv_msg(struct amdgpu_device *adev,
u32 reg;
u32 mask = REG_FIELD_MASK(MAILBOX_CONTROL, RCV_MSG_VALID);
- reg = RREG32_NO_KIQ(mmMAILBOX_CONTROL);
- if (!(reg & mask))
- return -ENOENT;
+ /* workaround: host driver doesn't set VALID for CMPL now */
+ if (event != IDH_FLR_NOTIFICATION_CMPL) {
+ reg = RREG32_NO_KIQ(mmMAILBOX_CONTROL);
+ if (!(reg & mask))
+ return -ENOENT;
+ }
reg = RREG32_NO_KIQ(mmMAILBOX_MSGBUF_RCV_DW0);
if (reg != event)
OpenPOWER on IntegriCloud