[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20211221015751.116328-19-sashal@kernel.org>
Date: Mon, 20 Dec 2021 20:57:40 -0500
From: Sasha Levin <sashal@...nel.org>
To: linux-kernel@...r.kernel.org, stable@...r.kernel.org
Cc: Evan Quan <evan.quan@....com>, Guchun Chen <guchun.chen@....com>,
Mario Limonciello <mario.limonciello@....com>,
Alex Deucher <alexander.deucher@....com>,
Sasha Levin <sashal@...nel.org>, christian.koenig@....com,
Xinhui.Pan@....com, airlied@...ux.ie, daniel@...ll.ch,
Hawking.Zhang@....com, john.clements@....com, Oak.Zeng@....com,
rajneesh.bhardwaj@....com, Likun.Gao@....com, alex.sierra@....com,
Harish.Kasiviswanathan@....com, Dennis.Li@....com,
lijo.lazar@....com, darren.powell@....com,
jinhuieric.huang@....com, amd-gfx@...ts.freedesktop.org,
dri-devel@...ts.freedesktop.org
Subject: [PATCH AUTOSEL 5.15 19/29] drm/amdgpu: correct the wrong cached state for GMC on PICASSO
From: Evan Quan <evan.quan@....com>
[ Upstream commit 17c65d6fca844ee72a651944d8ce721e9040bf70 ]
Pair the operations did in GMC ->hw_init and ->hw_fini. That
can help to maintain correct cached state for GMC and avoid
unintention gate operation dropping due to wrong cached state.
BugLink: https://gitlab.freedesktop.org/drm/amd/-/issues/1828
Signed-off-by: Evan Quan <evan.quan@....com>
Acked-by: Guchun Chen <guchun.chen@....com>
Reviewed-by: Mario Limonciello <mario.limonciello@....com>
Signed-off-by: Alex Deucher <alexander.deucher@....com>
Signed-off-by: Sasha Levin <sashal@...nel.org>
---
drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c | 8 ++++++++
drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c | 8 ++++----
drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c | 7 ++++++-
3 files changed, 18 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
index 5551359d5dfdc..398596aac8ca5 100644
--- a/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c
@@ -1802,6 +1802,14 @@ static int gmc_v9_0_hw_fini(void *handle)
return 0;
}
+ /*
+ * Pair the operations did in gmc_v9_0_hw_init and thus maintain
+ * a correct cached state for GMC. Otherwise, the "gate" again
+ * operation on S3 resuming will fail due to wrong cached state.
+ */
+ if (adev->mmhub.funcs->update_power_gating)
+ adev->mmhub.funcs->update_power_gating(adev, false);
+
amdgpu_irq_put(adev, &adev->gmc.ecc_irq, 0);
amdgpu_irq_put(adev, &adev->gmc.vm_fault, 0);
diff --git a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
index a99953833820e..3ef5e368d125d 100644
--- a/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
+++ b/drivers/gpu/drm/amd/amdgpu/mmhub_v1_0.c
@@ -302,10 +302,10 @@ static void mmhub_v1_0_update_power_gating(struct amdgpu_device *adev,
if (amdgpu_sriov_vf(adev))
return;
- if (enable && adev->pg_flags & AMD_PG_SUPPORT_MMHUB) {
- amdgpu_dpm_set_powergating_by_smu(adev, AMD_IP_BLOCK_TYPE_GMC, true);
-
- }
+ if (adev->pg_flags & AMD_PG_SUPPORT_MMHUB)
+ amdgpu_dpm_set_powergating_by_smu(adev,
+ AMD_IP_BLOCK_TYPE_GMC,
+ enable);
}
static int mmhub_v1_0_gart_enable(struct amdgpu_device *adev)
diff --git a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c
index 321215003643b..69f67206b5b5a 100644
--- a/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c
+++ b/drivers/gpu/drm/amd/pm/powerplay/amd_powerplay.c
@@ -1332,7 +1332,12 @@ static int pp_set_powergating_by_smu(void *handle,
pp_dpm_powergate_vce(handle, gate);
break;
case AMD_IP_BLOCK_TYPE_GMC:
- pp_dpm_powergate_mmhub(handle);
+ /*
+ * For now, this is only used on PICASSO.
+ * And only "gate" operation is supported.
+ */
+ if (gate)
+ pp_dpm_powergate_mmhub(handle);
break;
case AMD_IP_BLOCK_TYPE_GFX:
ret = pp_dpm_powergate_gfx(handle, gate);
--
2.34.1
Powered by blists - more mailing lists