drm/amd/powerplay: add sensor lock support for smu

when multithreading access sysfs of amdgpu_pm_info at the sametime.
the swsmu driver cause smu firmware hang.

eg:
single thread access:
Message A + Param A ==> right
Message B + Param B ==> right
Message C + Param C ==> right
multithreading access:
Message A + Param B ==> error
Message B + Param A ==> error
Message C + Param C ==> right

the patch will add sensor lock(mutex) to avoid this error.

Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

diff --git a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
index 23293e15636bb8bba599c38365c8d480e5464389..df510cb86da507886a8c9cbc6b9c50e1b3f3e7f8 100644 (file)
--- a/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
+++ b/drivers/gpu/drm/amd/powerplay/amdgpu_smu.c
@@ -840,6 +840,8 @@ static int smu_sw_init(void *handle)
        smu->smu_baco.state = SMU_BACO_STATE_EXIT;
        smu->smu_baco.platform_support = false;
 
+       mutex_init(&smu->sensor_lock);
+
        smu->watermarks_bitmap = 0;
        smu->power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;
        smu->default_power_profile_mode = PP_SMC_POWER_PROFILE_BOOTUP_DEFAULT;

diff --git a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c
index a047a7ec3698470fbf6e46e90797e1efa2dcb1e1..b9b7b73354a0e9acad7a73784c80efbe0fe3ceac 100644 (file)
--- a/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/arcturus_ppt.c
@@ -1025,6 +1025,7 @@ static int arcturus_read_sensor(struct smu_context *smu,
        if (!data || !size)
                return -EINVAL;
 
+       mutex_lock(&smu->sensor_lock);
        switch (sensor) {
        case AMDGPU_PP_SENSOR_MAX_FAN_RPM:
                *(uint32_t *)data = pptable->FanMaximumRpm;
@@ -1051,6 +1052,7 @@ static int arcturus_read_sensor(struct smu_context *smu,
        default:
                ret = smu_smc_read_sensor(smu, sensor, data, size);
        }
+       mutex_unlock(&smu->sensor_lock);
 
        return ret;
 }

diff --git a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
index 5c898444ff9743ad4ed0ef67b336645a47346b0d..bc4b73e0718e36b61d39fcfe7f7a23ff470314e1 100644 (file)
--- a/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
+++ b/drivers/gpu/drm/amd/powerplay/inc/amdgpu_smu.h
@@ -350,6 +350,7 @@ struct smu_context
        const struct smu_funcs          *funcs;
        const struct pptable_funcs      *ppt_funcs;
        struct mutex                    mutex;
+       struct mutex                    sensor_lock;
        uint64_t pool_size;
 
        struct smu_table_context        smu_table;

diff --git a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
index 0a44aa81d4353c04f17acdc52f2e407a3cd364f7..a583cf87b514dcba03663c3a054cd88abac9ebe7 100644 (file)
--- a/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/navi10_ppt.c
@@ -1387,6 +1387,7 @@ static int navi10_read_sensor(struct smu_context *smu,
        if(!data || !size)
                return -EINVAL;
 
+       mutex_lock(&smu->sensor_lock);
        switch (sensor) {
        case AMDGPU_PP_SENSOR_MAX_FAN_RPM:
                *(uint32_t *)data = pptable->FanMaximumRpm;
@@ -1410,6 +1411,7 @@ static int navi10_read_sensor(struct smu_context *smu,
        default:
                ret = smu_smc_read_sensor(smu, sensor, data, size);
        }
+       mutex_unlock(&smu->sensor_lock);
 
        return ret;
 }

diff --git a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
index 9082da1578d13ca879286977f6025384c4e97d5d..f655ebd9ba226c66572d2d97ca70c8e8340359ad 100644 (file)
--- a/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
+++ b/drivers/gpu/drm/amd/powerplay/vega20_ppt.c
@@ -3017,6 +3017,7 @@ static int vega20_read_sensor(struct smu_context *smu,
        if(!data || !size)
                return -EINVAL;
 
+       mutex_lock(&smu->sensor_lock);
        switch (sensor) {
        case AMDGPU_PP_SENSOR_MAX_FAN_RPM:
                *(uint32_t *)data = pptable->FanMaximumRpm;
@@ -3042,6 +3043,7 @@ static int vega20_read_sensor(struct smu_context *smu,
        default:
                ret = smu_smc_read_sensor(smu, sensor, data, size);
        }
+       mutex_unlock(&smu->sensor_lock);
 
        return ret;
 }