Plumb in support for disabling kernel queues in
GFX11. We have to bring up a GFX queue briefly in
order to initialize the clear state. After that
we can disable it.
v2: use ring counts per Felix' suggestion
Signed-off-by: Alex Deucher <alexander.deucher@xxxxxxx>
---
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 99 +++++++++++++++++---------
1 file changed, 65 insertions(+), 34 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
index 95eefd9a40d28..b20624f8cbbbd 100644
--- a/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c
@@ -1145,6 +1145,10 @@ static int gfx_v11_0_gfx_ring_init(struct amdgpu_device *adev, int ring_id,
ring->ring_obj = NULL;
ring->use_doorbell = true;
+ if (adev->gfx.disable_kq) {
+ ring->no_scheduler = true;
+ ring->no_user_submission = true;
+ }
if (!ring_id)
ring->doorbell_index = adev->doorbell_index.gfx_ring0 << 1;
@@ -1577,7 +1581,7 @@ static void gfx_v11_0_alloc_ip_dump(struct amdgpu_device *adev)
static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
{
- int i, j, k, r, ring_id = 0;
+ int i, j, k, r, ring_id;
int xcc_id = 0;
struct amdgpu_device *adev = ip_block->adev;
@@ -1710,37 +1714,42 @@ static int gfx_v11_0_sw_init(struct amdgpu_ip_block *ip_block)
return r;
}
- /* set up the gfx ring */
- for (i = 0; i < adev->gfx.me.num_me; i++) {
- for (j = 0; j < adev->gfx.me.num_queue_per_pipe; j++) {
- for (k = 0; k < adev->gfx.me.num_pipe_per_me; k++) {
- if (!amdgpu_gfx_is_me_queue_enabled(adev, i, k, j))
- continue;
-
- r = gfx_v11_0_gfx_ring_init(adev, ring_id,
- i, k, j);
- if (r)
- return r;
- ring_id++;
+ if (adev->gfx.num_gfx_rings) {
+ ring_id = 0;
+ /* set up the gfx ring */
+ for (i = 0; i < adev->gfx.me.num_me; i++) {
+ for (j = 0; j < adev->gfx.me.num_queue_per_pipe; j++) {
+ for (k = 0; k < adev->gfx.me.num_pipe_per_me; k++) {
+ if (!amdgpu_gfx_is_me_queue_enabled(adev, i, k, j))
+ continue;
+
+ r = gfx_v11_0_gfx_ring_init(adev, ring_id,
+ i, k, j);
+ if (r)
+ return r;
+ ring_id++;
+ }
}
}
}
- ring_id = 0;
- /* set up the compute queues - allocate horizontally across pipes */
- for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
- for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
- for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
- if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
- k, j))
- continue;
+ if (adev->gfx.num_compute_rings) {
+ ring_id = 0;
+ /* set up the compute queues - allocate horizontally across pipes */
+ for (i = 0; i < adev->gfx.mec.num_mec; ++i) {
+ for (j = 0; j < adev->gfx.mec.num_queue_per_pipe; j++) {
+ for (k = 0; k < adev->gfx.mec.num_pipe_per_mec; k++) {
+ if (!amdgpu_gfx_is_mec_queue_enabled(adev, 0, i,
+ k, j))
+ continue;
- r = gfx_v11_0_compute_ring_init(adev, ring_id,
- i, k, j);
- if (r)
- return r;
+ r = gfx_v11_0_compute_ring_init(adev, ring_id,
+ i, k, j);
+ if (r)
+ return r;
- ring_id++;
+ ring_id++;
+ }
}
}
}
@@ -4578,11 +4587,22 @@ static int gfx_v11_0_cp_resume(struct amdgpu_device *adev)
return r;
}
- for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
- ring = &adev->gfx.gfx_ring[i];
- r = amdgpu_ring_test_helper(ring);
- if (r)
- return r;
+ if (adev->gfx.disable_kq) {
+ for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
+ ring = &adev->gfx.gfx_ring[i];
+ /* we don't want to set ring->ready */
+ r = amdgpu_ring_test_ring(ring);
+ if (r)
+ return r;
+ }
+ amdgpu_gfx_disable_kgq(adev, 0);
+ } else {
+ for (i = 0; i < adev->gfx.num_gfx_rings; i++) {
+ ring = &adev->gfx.gfx_ring[i];
+ r = amdgpu_ring_test_helper(ring);
+ if (r)
+ return r;
+ }
}
for (i = 0; i < adev->gfx.num_compute_rings; i++) {
@@ -5128,11 +5148,22 @@ static int gfx_v11_0_early_init(struct amdgpu_ip_block *ip_block)
{
struct amdgpu_device *adev = ip_block->adev;
+ if (amdgpu_disable_kq == 1)
+ adev->gfx.disable_kq = true;
+
adev->gfx.funcs = &gfx_v11_0_gfx_funcs;
- adev->gfx.num_gfx_rings = GFX11_NUM_GFX_RINGS;
- adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
- AMDGPU_MAX_COMPUTE_RINGS);
+ if (adev->gfx.disable_kq) {
+ /* We need one GFX ring temporarily to set up
+ * the clear state.
+ */
+ adev->gfx.num_gfx_rings = 1;
+ adev->gfx.num_compute_rings = 0;
+ } else {
+ adev->gfx.num_gfx_rings = GFX11_NUM_GFX_RINGS;
+ adev->gfx.num_compute_rings = min(amdgpu_gfx_get_num_kcq(adev),
+ AMDGPU_MAX_COMPUTE_RINGS);
+ }
gfx_v11_0_set_kiq_pm4_funcs(adev);
gfx_v11_0_set_ring_funcs(adev);
--
2.48.1
