AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Alex Deucher <alexander.deucher@amd.com>
To: <amd-gfx@lists.freedesktop.org>
Cc: Alex Deucher <alexander.deucher@amd.com>
Subject: [PATCH] drm/amdgpu/vcn: fix idle work handler for VCN 2.5
Date: Tue, 4 Mar 2025 11:22:42 -0500	[thread overview]
Message-ID: <20250304162242.4029795-1-alexander.deucher@amd.com> (raw)

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)

Fixes: 4ce4fe27205c ("drm/amdgpu/vcn: use per instance callbacks for idle work handler")
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h |  4 +++
 drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c   | 41 +++++++++++++++++++------
 2 files changed, 36 insertions(+), 9 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
index 26c9c2d90f455..3bc4fe4aeb481 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vcn.h
@@ -358,6 +358,10 @@ struct amdgpu_vcn {
 
 	bool			per_inst_fw;
 	unsigned		fw_version;
+	/* VCN 2.5 global PG handling */
+	struct mutex		global_pg_lock;
+	unsigned int		global_pg_count;
+	enum amd_powergating_state global_pg_state;
 };
 
 struct amdgpu_fw_shared_rb_ptrs_struct {
diff --git a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
index dff1a88590363..972f0842ea47b 100644
--- a/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
+++ b/drivers/gpu/drm/amd/amdgpu/vcn_v2_5.c
@@ -172,6 +172,8 @@ static int vcn_v2_5_sw_init(struct amdgpu_ip_block *ip_block)
 	uint32_t *ptr;
 	struct amdgpu_device *adev = ip_block->adev;
 
+	mutex_init(&adev->vcn.global_pg_lock);
+
 	for (j = 0; j < adev->vcn.num_vcn_inst; j++) {
 		volatile struct amdgpu_fw_shared *fw_shared;
 
@@ -1853,21 +1855,42 @@ static int vcn_v2_5_set_pg_state(struct amdgpu_vcn_inst *vinst,
 				 enum amd_powergating_state state)
 {
 	struct amdgpu_device *adev = vinst->adev;
-	int ret;
+	struct amdgpu_vcn_inst *v;
+	int ret = 0, i;
 
 	if (amdgpu_sriov_vf(adev))
 		return 0;
 
-	if (state == vinst->cur_state)
-		return 0;
+	mutex_lock(&adev->vcn.global_pg_lock);
+	if (state == AMD_PG_STATE_GATE) {
+		if (adev->vcn.global_pg_count == 0)
+			goto unlock;
+		adev->vcn.global_pg_count--;
+		if (adev->vcn.global_pg_count == 0 &&
+		    adev->vcn.global_pg_state == AMD_PG_STATE_UNGATE) {
+			for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+				v = &adev->vcn.inst[i];
+
+				ret = vcn_v2_5_stop(v);
+			}
+			adev->vcn.global_pg_state = AMD_PG_STATE_GATE;
+		}
+	} else {
+		if (adev->vcn.global_pg_count == 0 &&
+		    adev->vcn.global_pg_state == AMD_PG_STATE_GATE) {
+			for (i = 0; i < adev->vcn.num_vcn_inst; ++i) {
+				v = &adev->vcn.inst[i];
 
-	if (state == AMD_PG_STATE_GATE)
-		ret = vcn_v2_5_stop(vinst);
-	else
-		ret = vcn_v2_5_start(vinst);
+				ret = vcn_v2_5_start(v);
+			}
+			adev->vcn.global_pg_state = AMD_PG_STATE_UNGATE;
+		}
+		adev->vcn.global_pg_count++;
+	}
 
-	if (!ret)
-		vinst->cur_state = state;
+unlock:
+	vinst->cur_state = state;
+	mutex_unlock(&adev->vcn.global_pg_lock);
 
 	return ret;
 }
-- 
2.48.1


             reply	other threads:[~2025-03-04 16:23 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-03-04 16:22 Alex Deucher [this message]
2025-03-04 20:52 ` [PATCH] drm/amdgpu/vcn: fix idle work handler for VCN 2.5 Boyuan Zhang
2025-03-04 21:05   ` Alex Deucher
2025-03-04 23:12 ` Alex Deucher
  -- strict thread matches above, loose matches on Subject: below --
2025-03-05 19:17 Alex Deucher
2025-03-06 15:05 ` Alex Deucher
2025-03-07 15:22   ` Alex Deucher
2025-03-07 21:14     ` Zhang, Boyuan
2025-03-04 23:01 Alex Deucher
2025-03-05 14:20 ` Boyuan Zhang
2025-03-04 14:20 Alex Deucher
2025-03-04 14:37 ` Lazar, Lijo

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=20250304162242.4029795-1-alexander.deucher@amd.com \
    --to=alexander.deucher@amd.com \
    --cc=amd-gfx@lists.freedesktop.org \
    /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