[PATCH] drm/amdgpu/sriov: For sriov runtime, use kiq to do invalidate tlb
Deng, Emily
Emily.Deng at amd.com
Tue Aug 14 06:58:31 UTC 2018
Hi all,
Please ignore this patch, will send another patch.
Best wishes
Emily Deng
-----Original Message-----
From: amd-gfx <amd-gfx-bounces at lists.freedesktop.org> On Behalf Of Emily Deng
Sent: Tuesday, August 14, 2018 2:32 PM
To: amd-gfx at lists.freedesktop.org
Cc: Deng, Emily <Emily.Deng at amd.com>
Subject: [PATCH] drm/amdgpu/sriov: For sriov runtime, use kiq to do invalidate tlb
To avoid the tlb flush not interrupted by world switch, use kiq and one command to do tlb invalidate.
SWDEV-161497
Signed-off-by: Emily Deng <Emily.Deng at amd.com>
---
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 14 +++++++++++++-
1 file changed, 13 insertions(+), 1 deletion(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 6999042..b66371d 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -330,13 +330,25 @@ static void gmc_v9_0_flush_gpu_tlb(struct amdgpu_device *adev, {
/* Use register 17 for GART */
const unsigned eng = 17;
- unsigned i, j;
+ unsigned i, j, r;
spin_lock(&adev->gmc.invalidate_lock);
for (i = 0; i < AMDGPU_MAX_VMHUBS; ++i) {
struct amdgpu_vmhub *hub = &adev->vmhub[i];
u32 tmp = gmc_v9_0_get_invalidate_req(vmid);
+ struct amdgpu_ring *ring = &(adev->gfx.kiq.ring);
+
+ if (amdgpu_sriov_vf(adev) && amdgpu_sriov_runtime(adev)) {
+ r = amdgpu_ring_alloc(ring, 7);
+ if (r)
+ DRM_ERROR("amdgpu: tlb flush alloc kiq ring fail.\n");
+ amdgpu_ring_emit_reg_write_reg_wait(ring, hub->vm_inv_eng0_req + eng,
+ hub->vm_inv_eng0_ack + eng,
+ tmp, 1 << vmid);
+ amdgpu_ring_commit(ring);
+ continue;
+ }
WREG32_NO_KIQ(hub->vm_inv_eng0_req + eng, tmp);
--
2.7.4
_______________________________________________
amd-gfx mailing list
amd-gfx at lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/amd-gfx
More information about the amd-gfx
mailing list