<!DOCTYPE html><html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
<p><br>
</p>
<div class="moz-cite-prefix">On 2025-03-04 18:01, Alex Deucher
wrote:<br>
</div>
<blockquote type="cite" cite="mid:20250304230123.71874-1-alexander.deucher@amd.com">
<pre class="moz-quote-pre" wrap="">VCN 2.5 uses the PG callback to enable VCN DPM which is
a global state. As such, we need to make sure all instances
are in the same state.
v2: switch to a ref count (Lijo)
v3: switch to its own idle work handler
Fixes: 4ce4fe27205c ("drm/amdgpu/vcn: use per instance callbacks for idle work handler")
Signed-off-by: Alex Deucher <a class="moz-txt-link-rfc2396E" href="mailto:alexander.deucher@amd.com"><alexander.deucher@amd.com></a></pre>
</blockquote>
<p><br>
</p>
<div>
<p style="margin-top:0;margin-bottom:0;"><span style="color:black;font-size:11pt;font-family:Aptos,Aptos_EmbeddedFont,Aptos_MSFontService,Calibri,Helvetica,sans-serif;"><span data-markjs="true" class="mark7ru6aprxh" style="" data-ogac="" data-ogab="" data-ogsc="" data-ogsb="" data-olk-copy-source="MessageBody">Reviewed-by</span>:
Boyuan Zhang <a style="margin-top:0;margin-bottom:0;" title="mailto:Boyuan.Zhang@amd.com" data-linkindex="0" id="LPlnkOWA0ed3e1cc-cbb8-88b5-8ab6-131c3f0c5066" href="mailto:Boyuan.Zhang@amd.com"><Boyuan.Zhang@amd.com></a></span></p>
</div>
<p></p>
<p><br>
</p>
<blockquote type="cite" cite="mid:20250304230123.71874-1-alexander.deucher@amd.com">
<pre class="moz-quote-pre" wrap="">
---
drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c | 126 +++++++++++++++++++++++++-
1 file changed, 122 insertions(+), 4 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index dff1a88590363..fa66521b940de 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -107,6 +107,121 @@ static int amdgpu_ih_clientid_vcns[] = {
SOC15_IH_CLIENTID_VCN1
};
+static void vcn_v2_5_idle_work_handler(struct work_struct *work)
+{
+ struct amdgpu_vcn_inst *vcn_inst =
+ container_of(work, struct amdgpu_vcn_inst, idle_work.work);
+ struct amdgpu_device *adev = vcn_inst->adev;
+ unsigned int fences = 0, fence[AMDGPU_MAX_VCN_INSTANCES] = {0};
+ unsigned int i, j;
+ int r = 0;
+
+ for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+ struct amdgpu_vcn_inst *v = &adev->vcn.inst[i];
+
+ if (adev->vcn.harvest_config & (1 << i))
+ continue;
+
+ for (j = 0; j < v->num_enc_rings; ++j)
+ fence[i] += amdgpu_fence_count_emitted(&v->ring_enc[j]);
+
+ /* Only set DPG pause for VCN3 or below, VCN4 and above will be handled by FW */
+ if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG &&
+ !v->using_unified_queue) {
+ struct dpg_pause_state new_state;
+
+ if (fence[i] ||
+ unlikely(atomic_read(&v->dpg_enc_submission_cnt)))
+ new_state.fw_based = VCN_DPG_STATE__PAUSE;
+ else
+ new_state.fw_based = VCN_DPG_STATE__UNPAUSE;
+
+ v->pause_dpg_mode(v, &new_state);
+ }
+
+ fence[i] += amdgpu_fence_count_emitted(&v->ring_dec);
+ fences += fence[i];
+
+ }
+
+ if (!fences && !atomic_read(&adev->vcn.inst[0].total_submission_cnt)) {
+ amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN,
+ AMD_PG_STATE_GATE);
+ r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO,
+ false);
+ if (r)
+ dev_warn(adev->dev, "(%d) failed to disable video power profile mode\n", r);
+ } else {
+ schedule_delayed_work(&adev->vcn.inst[0].idle_work, VCN_IDLE_TIMEOUT);
+ }
+}
+
+static void vcn_v2_5_ring_begin_use(struct amdgpu_ring *ring)
+{
+ struct amdgpu_device *adev = ring->adev;
+ int r = 0, i;
+
+ atomic_inc(&adev->vcn.inst[0].total_submission_cnt);
+
+ if (!cancel_delayed_work_sync(&adev->vcn.inst[0].idle_work)) {
+ r = amdgpu_dpm_switch_power_profile(adev, PP_SMC_POWER_PROFILE_VIDEO,
+ true);
+ if (r)
+ dev_warn(adev->dev, "(%d) failed to switch to video power profile mode\n", r);
+ }
+
+ mutex_lock(&adev->vcn.inst[0].vcn_pg_lock);
+ amdgpu_device_ip_set_powergating_state(adev, AMD_IP_BLOCK_TYPE_VCN,
+ AMD_PG_STATE_UNGATE);
+
+ for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+ struct amdgpu_vcn_inst *v = &adev->vcn.inst[i];
+
+ if (adev->vcn.harvest_config & (1 << i))
+ continue;
+ /* Only set DPG pause for VCN3 or below, VCN4 and above will be handled by FW */
+ if (adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG &&
+ !v->using_unified_queue) {
+ struct dpg_pause_state new_state;
+
+ if (ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC) {
+ atomic_inc(&v->dpg_enc_submission_cnt);
+ new_state.fw_based = VCN_DPG_STATE__PAUSE;
+ } else {
+ unsigned int fences = 0;
+ unsigned int i;
+
+ for (i = 0; i < v->num_enc_rings; ++i)
+ fences += amdgpu_fence_count_emitted(&v->ring_enc[i]);
+
+ if (fences || atomic_read(&v->dpg_enc_submission_cnt))
+ new_state.fw_based = VCN_DPG_STATE__PAUSE;
+ else
+ new_state.fw_based = VCN_DPG_STATE__UNPAUSE;
+ }
+
+ v->pause_dpg_mode(v, &new_state);
+ }
+ }
+ mutex_unlock(&adev->vcn.inst[0].vcn_pg_lock);
+}
+
+static void vcn_v2_5_ring_end_use(struct amdgpu_ring *ring)
+{
+ struct amdgpu_device *adev = ring->adev;
+
+ /* Only set DPG pause for VCN3 or below, VCN4 and above will be handled by FW */
+ if (ring->adev->pg_flags & AMD_PG_SUPPORT_VCN_DPG &&
+ ring->funcs->type == AMDGPU_RING_TYPE_VCN_ENC &&
+ !adev->vcn.inst[ring->me].using_unified_queue)
+ atomic_dec(&adev->vcn.inst[ring->me].dpg_enc_submission_cnt);
+
+ atomic_dec(&adev->vcn.inst[0].total_submission_cnt);
+
+ schedule_delayed_work(&adev->vcn.inst[0].idle_work,
+ VCN_IDLE_TIMEOUT);
+}
+
/**
* vcn_v2_5_early_init - set function pointers and load microcode
*
@@ -201,6 +316,9 @@ static int vcn_v2_5_sw_init(struct amdgpu_ip_block *ip_block)
if (r)
return r;
+ /* Override the work func */
+ adev->vcn.inst[j].idle_work.work.func = vcn_v2_5_idle_work_handler;
+
amdgpu_vcn_setup_ucode(adev, j);
r = amdgpu_vcn_resume(adev, j);
@@ -1661,8 +1779,8 @@ static const struct amdgpu_ring_funcs vcn_v2_5_dec_ring_vm_funcs = {
.insert_start = vcn_v2_0_dec_ring_insert_start,
.insert_end = vcn_v2_0_dec_ring_insert_end,
.pad_ib = amdgpu_ring_generic_pad_ib,
- .begin_use = amdgpu_vcn_ring_begin_use,
- .end_use = amdgpu_vcn_ring_end_use,
+ .begin_use = vcn_v2_5_ring_begin_use,
+ .end_use = vcn_v2_5_ring_end_use,
.emit_wreg = vcn_v2_0_dec_ring_emit_wreg,
.emit_reg_wait = vcn_v2_0_dec_ring_emit_reg_wait,
.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
@@ -1759,8 +1877,8 @@ static const struct amdgpu_ring_funcs vcn_v2_5_enc_ring_vm_funcs = {
.insert_nop = amdgpu_ring_insert_nop,
.insert_end = vcn_v2_0_enc_ring_insert_end,
.pad_ib = amdgpu_ring_generic_pad_ib,
- .begin_use = amdgpu_vcn_ring_begin_use,
- .end_use = amdgpu_vcn_ring_end_use,
+ .begin_use = vcn_v2_5_ring_begin_use,
+ .end_use = vcn_v2_5_ring_end_use,
.emit_wreg = vcn_v2_0_enc_ring_emit_wreg,
.emit_reg_wait = vcn_v2_0_enc_ring_emit_reg_wait,
.emit_reg_write_reg_wait = amdgpu_ring_emit_reg_write_reg_wait_helper,
</pre>
</blockquote>
</body>
</html>