drm/amdgpu: gfx_v11_0_cp_ecc_error_irq_funcs is not needed any more
Subsystems (2)
PATCH v2 1/3 drm/amdgpu: gfx_v11_0_cp_ecc_error_irq_funcs is not needed any more
The gfx.cp_ecc_error_irq is retired in gfx11. In gfx_v11_0_hw_fini still use amdgpu_irq_put to disable this interrupt, which caused the call trace in this function.
[ 102.873958] Call Trace: [ 102.873959] <TASK> [ 102.873961] gfx_v11_0_hw_fini+0x23/0x1e0 [amdgpu] [ 102.874019] gfx_v11_0_suspend+0xe/0x20 [amdgpu] [ 102.874072] amdgpu_device_ip_suspend_phase2+0x240/0x460 [amdgpu] [ 102.874122] amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu] [ 102.874172] amdgpu_device_pre_asic_reset+0xd9/0x490 [amdgpu] [ 102.874223] amdgpu_device_gpu_recover.cold+0x548/0xce6 [amdgpu] [ 102.874321] amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu] [ 102.874375] process_one_work+0x21f/0x3f0 [ 102.874377] worker_thread+0x200/0x3e0 [ 102.874378] ? process_one_work+0x3f0/0x3f0 [ 102.874379] kthread+0xfd/0x130 [ 102.874380] ? kthread_complete_and_exit+0x20/0x20 [ 102.874381] ret_from_fork+0x22/0x30
--- drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 38 -------------------------- 1 file changed, 38 deletions(-)
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0 -38
@@ -1355,13 +1355,6 @@ static int gfx_v11_0_sw_init(void *handle)if (r)return r; -/* ECC error */ - r =amdgpu_irq_add_id (adev,SOC21_IH_CLIENTID_GRBM_CP , -GFX_11_0_0__SRCID__CP_ECC_ERROR , - &adev->gfx.cp_ecc_error_irq); -if (r) -return r; -/* FED error */ r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_GFX, GFX_11_0_0__SRCID__RLC_GC_FED_INTERRUPT,
@@ -4483,7 +4476,6 @@ static int gfx_v11_0_hw_fini(void *handle)struct amdgpu_device *adev = (struct amdgpu_device *)handle;int r; -amdgpu_irq_put (adev, &adev->gfx.cp_ecc_error_irq,0 );amdgpu_irq_put (adev, &adev->gfx.priv_reg_irq,0 );amdgpu_irq_put (adev, &adev->gfx.priv_inst_irq,0 );
@@ -5970,28 +5962,6 @@ static void gfx_v11_0_set_compute_eop_interrupt_state(struct amdgpu_device *adevWREG32_SOC15_IP (GC , reg_addr, tmp); \ }while (0 ) -static int gfx_v11_0_set_cp_ecc_error_state(struct amdgpu_device *adev, -struct amdgpu_irq_src *source, -unsigned type, -enum amdgpu_interrupt_state state) -{ -uint32_t ecc_irq_state =0 ; -uint32_t pipe0_int_cntl_addr =0 ; -int i =0 ; - - ecc_irq_state = (state ==AMDGPU_IRQ_STATE_ENABLE ) ?1 :0 ; - - pipe0_int_cntl_addr =SOC15_REG_OFFSET (GC ,0 , regCP_ME1_PIPE0_INT_CNTL); - -WREG32_FIELD15_PREREG (GC ,0 ,CP_INT_CNTL_RING0 ,CP_ECC_ERROR_INT_ENABLE , ecc_irq_state); - -for (i =0 ; i < adev->gfx.mec.num_pipe_per_mec; i++) -SET_ECC_ME_PIPE_STATE (pipe0_int_cntl_addr + i *CP_ME1_PIPE_INST_ADDR_INTERVAL , - ecc_irq_state); - -return 0 ; -} - static int gfx_v11_0_set_eop_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, unsigned type,
@@ -6408,11 +6378,6 @@ static const struct amdgpu_irq_src_funcs gfx_v11_0_priv_inst_irq_funcs = { .process = gfx_v11_0_priv_inst_irq, }; -static const struct amdgpu_irq_src_funcs gfx_v11_0_cp_ecc_error_irq_funcs = { - .set = gfx_v11_0_set_cp_ecc_error_state, - .process = amdgpu_gfx_cp_ecc_error_irq, -}; -static const struct amdgpu_irq_src_funcs gfx_v11_0_rlc_gc_fed_irq_funcs = { .process = gfx_v11_0_rlc_gc_fed_irq, };
@@ -6428,9 +6393,6 @@ static void gfx_v11_0_set_irq_funcs(struct amdgpu_device *adev) adev->gfx.priv_inst_irq.num_types =1 ; adev->gfx.priv_inst_irq.funcs = &gfx_v11_0_priv_inst_irq_funcs; - adev->gfx.cp_ecc_error_irq.num_types =1 ;/* CP ECC error */ - adev->gfx.cp_ecc_error_irq.funcs = &gfx_v11_0_cp_ecc_error_irq_funcs; - adev->gfx.rlc_gc_fed_irq.num_types =1 ;/* 0x80 FED error */ adev->gfx.rlc_gc_fed_irq.funcs = &gfx_v11_0_rlc_gc_fed_irq_funcs;
Signature · 1 line
2.34.1
PATCH v2 2/3 drm/amdgpu: fix amdgpu_irq_put call trace in gmc_v11_0_hw_fini
The gmc.ecc_irq is enabled by firmware per IFWI setting, and the host driver is not privileged to enable/disable the interrupt. So, it is meaningless to use the amdgpu_irq_put function in gmc_v11_0_hw_fini, which also leads to the call trace.
[ 102.980303] Call Trace: [ 102.980303] <TASK> [ 102.980304] gmc_v11_0_hw_fini+0x54/0x90 [amdgpu] [ 102.980357] gmc_v11_0_suspend+0xe/0x20 [amdgpu] [ 102.980409] amdgpu_device_ip_suspend_phase2+0x240/0x460 [amdgpu] [ 102.980459] amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu] [ 102.980520] amdgpu_device_pre_asic_reset+0xd9/0x490 [amdgpu] [ 102.980573] amdgpu_device_gpu_recover.cold+0x548/0xce6 [amdgpu] [ 102.980687] amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu] [ 102.980740] process_one_work+0x21f/0x3f0 [ 102.980741] worker_thread+0x200/0x3e0 [ 102.980742] ? process_one_work+0x3f0/0x3f0 [ 102.980743] kthread+0xfd/0x130 [ 102.980743] ? kthread_complete_and_exit+0x20/0x20 [ 102.980744] ret_from_fork+0x22/0x30
--- drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c | 1 - 1 file changed, 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/gmc_v11_0.c +0 -1
@@ -951,7 +951,6 @@ static int gmc_v11_0_hw_fini(void *handle)return 0 ; } -amdgpu_irq_put (adev, &adev->gmc.ecc_irq,0 );amdgpu_irq_put (adev, &adev->gmc.vm_fault,0 );gmc_v11_0_gart_disable (adev);
Signature · 1 line
2.34.1
PATCH v2 3/3 drm/amdgpu: fix amdgpu_irq_put call trace in gmc_v10_0_hw_fini
In reply to Horatio ZhangThe gmc.ecc_irq is enabled by firmware per IFWI setting, and the host driver is not privileged to enable/disable the interrupt. So, it is meaningless to use the amdgpu_irq_put function in gmc_v10_0_hw_fini, which also leads to the call trace.
[ 82.340264] Call Trace: [ 82.340265] <TASK> [ 82.340269] gmc_v10_0_hw_fini+0x83/0xa0 [amdgpu] [ 82.340447] gmc_v10_0_suspend+0xe/0x20 [amdgpu] [ 82.340623] amdgpu_device_ip_suspend_phase2+0x127/0x1c0 [amdgpu] [ 82.340789] amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu] [ 82.340955] amdgpu_device_pre_asic_reset+0xdd/0x2b0 [amdgpu] [ 82.341122] amdgpu_device_gpu_recover.cold+0x4dd/0xbb2 [amdgpu] [ 82.341359] amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu] [ 82.341529] process_one_work+0x21d/0x3f0 [ 82.341535] worker_thread+0x1fa/0x3c0 [ 82.341538] ? process_one_work+0x3f0/0x3f0 [ 82.341540] kthread+0xff/0x130 [ 82.341544] ? kthread_complete_and_exit+0x20/0x20 [ 82.341547] ret_from_fork+0x22/0x30
--- drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 1 - 1 file changed, 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +0 -1
@@ -1143,7 +1143,6 @@ static int gmc_v10_0_hw_fini(void *handle)return 0 ; } -amdgpu_irq_put (adev, &adev->gmc.ecc_irq,0 );amdgpu_irq_put (adev, &adev->gmc.vm_fault,0 );return 0 ;
Signature · 1 line
2.34.1
Re: [PATCH v2 3/3] drm/amdgpu: fix amdgpu_irq_put call trace in gmc_v10_0_hw_fini
[AMD Official Use Only - General]
Series is
Regards,
Hawking
From: Horatio Zhang <[email protected]>
Date: Tuesday, April 25, 2023 at 14:38
To: Zhang, Hawking <[email protected]>, Koenig, Christian <[email protected]>, [email protected] <[email protected]>
Cc: Chen, Guchun <[email protected]>, Xu, Feifei <[email protected]>, Yao, Longlong <[email protected]>, Zhang, Horatio <[email protected]>
Subject: [PATCH v2 3/3] drm/amdgpu: fix amdgpu_irq_put call trace in gmc_v10_0_hw_fini
The gmc.ecc_irq is enabled by firmware per IFWI setting,
and the host driver is not privileged to enable/disable
the interrupt. So, it is meaningless to use the amdgpu_irq_put
function in gmc_v10_0_hw_fini, which also leads to the call
trace.
[ 82.340264] Call Trace: [ 82.340265] <TASK> [ 82.340269] gmc_v10_0_hw_fini+0x83/0xa0 [amdgpu] [ 82.340447] gmc_v10_0_suspend+0xe/0x20 [amdgpu] [ 82.340623] amdgpu_device_ip_suspend_phase2+0x127/0x1c0 [amdgpu] [ 82.340789] amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu] [ 82.340955] amdgpu_device_pre_asic_reset+0xdd/0x2b0 [amdgpu] [ 82.341122] amdgpu_device_gpu_recover.cold+0x4dd/0xbb2 [amdgpu] [ 82.341359] amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu] [ 82.341529] process_one_work+0x21d/0x3f0 [ 82.341535] worker_thread+0x1fa/0x3c0 [ 82.341538] ? process_one_work+0x3f0/0x3f0 [ 82.341540] kthread+0xff/0x130 [ 82.341544] ? kthread_complete_and_exit+0x20/0x20 [ 82.341547] ret_from_fork+0x22/0x30
--- drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c | 1 - 1 file changed, 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/gmc_v10_0.c +0 -1
@@ -1143,7 +1143,6 @@ static int gmc_v10_0_hw_fini(void *handle)return 0 ; } -amdgpu_irq_put (adev, &adev->gmc.ecc_irq,0 );amdgpu_irq_put (adev, &adev->gmc.vm_fault,0 );return 0 ;
--
2.34.1
Christian König
maintainerAm 25.04.23 um 08:38 schrieb Horatio Zhang:
Quoted 20 lines
The gfx.cp_ecc_error_irq is retired in gfx11. In
gfx_v11_0_hw_finistill useamdgpu_irq_putto disable this interrupt, which caused the call trace in this function.[ 102.873958] Call Trace: [ 102.873959] <TASK> [ 102.873961] gfx_v11_0_hw_fini+0x23/0x1e0 [amdgpu] [ 102.874019] gfx_v11_0_suspend+0xe/0x20 [amdgpu] [ 102.874072] amdgpu_device_ip_suspend_phase2+0x240/0x460 [amdgpu] [ 102.874122] amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu] [ 102.874172] amdgpu_device_pre_asic_reset+0xd9/0x490 [amdgpu] [ 102.874223] amdgpu_device_gpu_recover.cold+0x548/0xce6 [amdgpu] [ 102.874321] amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu] [ 102.874375] process_one_work+0x21f/0x3f0 [ 102.874377] worker_thread+0x200/0x3e0 [ 102.874378] ? process_one_work+0x3f0/0x3f0 [ 102.874379] kthread+0xfd/0x130 [ 102.874380] ? kthread_complete_and_exit+0x20/0x20 [ 102.874381] ret_from_fork+0x22/0x30
Quoted 1 line
Signed-off-by Horatio Zhang <[email protected]>
This goes deeper than my understanding of the hw so I can't fully judge if this is correct or not.
But from the general idea looks good to me, so feel free to add my Acked-by.
Regards,
Christian.
Quoted 81 lines
--- drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 38 -------------------------- 1 file changed, 38 deletions(-)drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-38@@ -1355,13 +1355,6 @@ static int gfx_v11_0_sw_init(void *handle)if (r)return r; -/* ECC error */ - r =amdgpu_irq_add_id (adev,SOC21_IH_CLIENTID_GRBM_CP , -GFX_11_0_0__SRCID__CP_ECC_ERROR , - &adev->gfx.cp_ecc_error_irq); -if (r) -return r; -/* FED error */ r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_GFX, GFX_11_0_0__SRCID__RLC_GC_FED_INTERRUPT,@@ -4483,7 +4476,6 @@ static int gfx_v11_0_hw_fini(void *handle)struct amdgpu_device *adev = (struct amdgpu_device *)handle;int r; -amdgpu_irq_put (adev, &adev->gfx.cp_ecc_error_irq,0 );amdgpu_irq_put (adev, &adev->gfx.priv_reg_irq,0 );amdgpu_irq_put (adev, &adev->gfx.priv_inst_irq,0 );@@ -5970,28 +5962,6 @@ static void gfx_v11_0_set_compute_eop_interrupt_state(struct amdgpu_device *adevWREG32_SOC15_IP (GC , reg_addr, tmp); \ }while (0 ) -static int gfx_v11_0_set_cp_ecc_error_state(struct amdgpu_device *adev, -struct amdgpu_irq_src *source, -unsigned type, -enum amdgpu_interrupt_state state) -{ -uint32_t ecc_irq_state =0 ; -uint32_t pipe0_int_cntl_addr =0 ; -int i =0 ; - - ecc_irq_state = (state ==AMDGPU_IRQ_STATE_ENABLE ) ?1 :0 ; - - pipe0_int_cntl_addr =SOC15_REG_OFFSET (GC ,0 , regCP_ME1_PIPE0_INT_CNTL); - -WREG32_FIELD15_PREREG (GC ,0 ,CP_INT_CNTL_RING0 ,CP_ECC_ERROR_INT_ENABLE , ecc_irq_state); - -for (i =0 ; i < adev->gfx.mec.num_pipe_per_mec; i++) -SET_ECC_ME_PIPE_STATE (pipe0_int_cntl_addr + i *CP_ME1_PIPE_INST_ADDR_INTERVAL , - ecc_irq_state); - -return 0 ; -} - static int gfx_v11_0_set_eop_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, unsigned type,@@ -6408,11 +6378,6 @@ static const struct amdgpu_irq_src_funcs gfx_v11_0_priv_inst_irq_funcs = { .process = gfx_v11_0_priv_inst_irq, }; -static const struct amdgpu_irq_src_funcs gfx_v11_0_cp_ecc_error_irq_funcs = { - .set = gfx_v11_0_set_cp_ecc_error_state, - .process = amdgpu_gfx_cp_ecc_error_irq, -}; -static const struct amdgpu_irq_src_funcs gfx_v11_0_rlc_gc_fed_irq_funcs = { .process = gfx_v11_0_rlc_gc_fed_irq, };@@ -6428,9 +6393,6 @@ static void gfx_v11_0_set_irq_funcs(struct amdgpu_device *adev) adev->gfx.priv_inst_irq.num_types =1 ; adev->gfx.priv_inst_irq.funcs = &gfx_v11_0_priv_inst_irq_funcs; - adev->gfx.cp_ecc_error_irq.num_types =1 ;/* CP ECC error */ - adev->gfx.cp_ecc_error_irq.funcs = &gfx_v11_0_cp_ecc_error_irq_funcs; - adev->gfx.rlc_gc_fed_irq.num_types =1 ;/* 0x80 FED error */ adev->gfx.rlc_gc_fed_irq.funcs = &gfx_v11_0_rlc_gc_fed_irq_funcs;
I guess it's more simple if updating the subject to "drm/amdgpu: drop gfx_v11_0_cp_ecc_error_irq_funcs"
With this improved, the series are:
Regards,
Guchun
Quoted 126 lines
-----Original Message-----
From: Horatio Zhang <[email protected]>
Sent: Tuesday, April 25, 2023 2:39 PM
To: Zhang, Hawking <[email protected]>; Koenig, Christian <[email protected]>; [email protected]
Cc: Chen, Guchun <[email protected]>; Xu, Feifei <[email protected]>; Yao, Longlong <[email protected]>; Zhang, Horatio <[email protected]>
Subject: [PATCH v2 1/3] drm/amdgpu:gfx_v11_0_cp_ecc_error_irq_funcsis not needed any moreThe gfx.cp_ecc_error_irq is retired in gfx11. In
gfx_v11_0_hw_finistill useamdgpu_irq_putto disable this interrupt, which caused the call trace in this function.[ 102.873958] Call Trace: [ 102.873959] <TASK> [ 102.873961] gfx_v11_0_hw_fini+0x23/0x1e0 [amdgpu] [ 102.874019] gfx_v11_0_suspend+0xe/0x20 [amdgpu] [ 102.874072] amdgpu_device_ip_suspend_phase2+0x240/0x460 [amdgpu] [ 102.874122] amdgpu_device_ip_suspend+0x3d/0x80 [amdgpu] [ 102.874172] amdgpu_device_pre_asic_reset+0xd9/0x490 [amdgpu] [ 102.874223] amdgpu_device_gpu_recover.cold+0x548/0xce6 [amdgpu] [ 102.874321] amdgpu_debugfs_reset_work+0x4c/0x70 [amdgpu] [ 102.874375] process_one_work+0x21f/0x3f0 [ 102.874377] worker_thread+0x200/0x3e0 [ 102.874378] ? process_one_work+0x3f0/0x3f0 [ 102.874379] kthread+0xfd/0x130 [ 102.874380] ? kthread_complete_and_exit+0x20/0x20 [ 102.874381] ret_from_fork+0x22/0x30Signed-off-by Horatio Zhang <[email protected]>--- drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c | 38 -------------------------- 1 file changed, 38 deletions(-)drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-7@@ -1355,13 +1355,6 @@ static int gfx_v11_0_sw_init(void *handle)if (r)return r; -/* ECC error */ - r =amdgpu_irq_add_id (adev,SOC21_IH_CLIENTID_GRBM_CP , -GFX_11_0_0__SRCID__CP_ECC_ERROR , - &adev->gfx.cp_ecc_error_irq); -if (r) -return r; -/* FED error */ r = amdgpu_irq_add_id(adev, SOC21_IH_CLIENTID_GFX,
GFX_11_0_0__SRCID__RLC_GC_FED_INTERRUPT,drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-1@@ -4483,7 +4476,6 @@ static int gfx_v11_0_hw_fini(void *handle)struct amdgpu_device *adev = (struct amdgpu_device *)handle;int r; -amdgpu_irq_put (adev, &adev->gfx.cp_ecc_error_irq,0 );amdgpu_irq_put (adev, &adev->gfx.priv_reg_irq,0 );amdgpu_irq_put (adev, &adev->gfx.priv_inst_irq,0 );@@ -5970,28 +5962,6 @@ static voidgfx_v11_0_set_compute_eop_interrupt_state(struct
amdgpu_device*adevdrivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-2WREG32_SOC15_IP (GC , reg_addr, tmp); \ }while (0 ) -static int gfx_v11_0_set_cp_ecc_error_state(struct amdgpu_device *adev, - struct
amdgpu_irq_src*source,drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-2-unsigned type, - enum
amdgpu_interrupt_statestate)drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-8-{ -uint32_t ecc_irq_state =0 ; -uint32_t pipe0_int_cntl_addr =0 ; -int i =0 ; - - ecc_irq_state = (state ==AMDGPU_IRQ_STATE_ENABLE ) ?1 :0 ; - - pipe0_int_cntl_addr = SOC15_REG_OFFSET(GC, 0,
regCP_ME1_PIPE0_INT_CNTL);drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-2- - WREG32_FIELD15_PREREG(GC, 0, CP_INT_CNTL_RING0,
CP_ECC_ERROR_INT_ENABLE,ecc_irq_state);drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-3- -for (i =0 ; i < adev->gfx.mec.num_pipe_per_mec; i++) - SET_ECC_ME_PIPE_STATE(pipe0_int_cntl_addr + i *
CP_ME1_PIPE_INST_ADDR_INTERVAL,drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-10- ecc_irq_state); - -return 0 ; -} - static int gfx_v11_0_set_eop_interrupt_state(struct amdgpu_device *adev, struct amdgpu_irq_src *src, unsigned type,@@ -6408,11 +6378,6 @@ static const struct amdgpu_irq_src_funcs gfx_v11_0_priv_inst_irq_funcs = { .process = gfx_v11_0_priv_inst_irq, }; -static const struct amdgpu_irq_src_funcs gfx_v11_0_cp_ecc_error_irq_funcs = { - .set = gfx_v11_0_set_cp_ecc_error_state, - .process = amdgpu_gfx_cp_ecc_error_irq, -}; -static const struct amdgpu_irq_src_funcs gfx_v11_0_rlc_gc_fed_irq_funcs = {drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c +0
-3.process = gfx_v11_0_rlc_gc_fed_irq, };@@ -6428,9 +6393,6 @@ static void gfx_v11_0_set_irq_funcs(struct amdgpu_device *adev) adev->gfx.priv_inst_irq.num_types = 1; adev->gfx.priv_inst_irq.funcs = &gfx_v11_0_priv_inst_irq_funcs; - adev->gfx.cp_ecc_error_irq.num_types =1 ;/* CP ECC error */ - adev->gfx.cp_ecc_error_irq.funcs = &gfx_v11_0_cp_ecc_error_irq_funcs; - adev->gfx.rlc_gc_fed_irq.num_types =1 ;/* 0x80 FED error */
adev->gfx.rlc_gc_fed_irq.funcs=&gfx_v11_0_rlc_gc_fed_irq_funcs;--
2.34.1