linux/drivers/gpu/drm/amd/amdgpu
Andrey Grodzovsky 12ffa55da6 drm/amdgpu: Fix bugs in amdgpu_device_gpu_recover in XGMI case.
Issue 1:
In  XGMI case amdgpu_device_lock_adev for other devices in hive
was called to late, after access to their repsective schedulers.
So relocate the lock to the begining of accessing the other devs.

Issue 2:
Using amdgpu_device_ip_need_full_reset to switch the device list from
all devices in hive to the single 'master' device who owns this reset
call is wrong because when stopping schedulers we iterate all the devices
in hive but when restarting we will only reactivate the 'master' device.
Also, in case amdgpu_device_pre_asic_reset conlcudes that full reset IS
needed we then have to stop schedulers for all devices in hive and not
only the 'master' but with amdgpu_device_ip_need_full_reset  we
already missed the opprotunity do to so. So just remove this logic and
always stop and start all schedulers for all devices in hive.

Also minor cleanup and print fix.

v4: Minor coding style fix.

Signed-off-by: Andrey Grodzovsky <andrey.grodzovsky@amd.com>
Acked-by: Felix Kuehling <Felix.Kuehling@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-09-13 17:39:05 -05:00
..
amdgpu.h drm/amdgpu: remove amdgpu_cs_try_evict 2019-09-13 17:38:56 -05:00
amdgpu_acp.c
amdgpu_acp.h
amdgpu_acpi.c
amdgpu_afmt.c
amdgpu_amdkfd.c drm/amdgpu: enable Navi12 kfd support for amdgpu 2019-08-02 10:30:41 -05:00
amdgpu_amdkfd.h drm/amdgpu: Determing PTE flags separately for each mapping (v3) 2019-09-13 17:35:55 -05:00
amdgpu_amdkfd_arcturus.c drm/amdgpu: drop drmP.h in amdgpu_amdkfd_arcturus.c 2019-07-31 14:32:56 -05:00
amdgpu_amdkfd_fence.c
amdgpu_amdkfd_gfx_v7.c
amdgpu_amdkfd_gfx_v8.c
amdgpu_amdkfd_gfx_v9.c drm/amdgpu: Export function to flush TLB of specific vm hub 2019-08-15 10:57:48 -05:00
amdgpu_amdkfd_gfx_v9.h
amdgpu_amdkfd_gfx_v10.c drm/amdkfd/gfx10: Calling amdgpu functions to invalidate TLB 2019-08-15 10:57:55 -05:00
amdgpu_amdkfd_gpuvm.c drm/amdgpu: Remove unnecessary TLB workaround (v2) 2019-09-13 17:36:08 -05:00
amdgpu_atombios.c
amdgpu_atombios.h
amdgpu_atomfirmware.c
amdgpu_atomfirmware.h
amdgpu_atpx_handler.c drm/amdgpu: Add APTX quirk for Dell Latitude 5495 2019-08-27 10:09:12 -05:00
amdgpu_benchmark.c
amdgpu_bios.c
amdgpu_bo_list.c
amdgpu_bo_list.h
amdgpu_cgs.c
amdgpu_connectors.c drm/amdgpu: Provide ddc symlink in connector sysfs directory 2019-07-31 16:35:37 +02:00
amdgpu_connectors.h
amdgpu_cs.c drm/amdgpu: remove amdgpu_cs_try_evict 2019-09-13 17:38:56 -05:00
amdgpu_csa.c
amdgpu_csa.h
amdgpu_ctx.c drm/amdgpu: correct ras error count type 2019-08-23 11:30:32 -05:00
amdgpu_ctx.h drm/amdgpu: correct ras error count type 2019-08-23 11:30:32 -05:00
amdgpu_debugfs.c drm/amdgpu: fix a potential information leaking bug 2019-07-31 01:26:09 -05:00
amdgpu_debugfs.h
amdgpu_device.c drm/amdgpu: Fix bugs in amdgpu_device_gpu_recover in XGMI case. 2019-09-13 17:39:05 -05:00
amdgpu_discovery.c
amdgpu_discovery.h
amdgpu_display.c drm-misc-next for 5.4: 2019-08-21 16:44:41 +10:00
amdgpu_display.h drm/amdgpu: Fix amdgpu_display_supported_domains logic. 2019-07-30 23:48:32 -05:00
amdgpu_dma_buf.c drm-misc-next for 5.4: 2019-08-21 16:44:41 +10:00
amdgpu_dma_buf.h drm/amdgpu: Fill out gem_object->resv 2019-07-31 10:19:23 +02:00
amdgpu_doorbell.h
amdgpu_dpm.c
amdgpu_dpm.h
amdgpu_drv.c amd/amdgpu: add Arcturus vf DID support 2019-08-22 17:17:12 -05:00
amdgpu_drv.h
amdgpu_encoders.c
amdgpu_fb.c drm/amdgpu: Fix amdgpu_display_supported_domains logic. 2019-07-30 23:48:32 -05:00
amdgpu_fence.c
amdgpu_gart.c dmr/amdgpu: Fix compile error with CONFIG_DRM_AMDGPU_GART_DEBUGFS 2019-08-15 10:59:17 -05:00
amdgpu_gart.h
amdgpu_gds.h Revert "drm/amdgpu: fix transform feedback GDS hang on gfx10 (v2)" 2019-08-12 12:47:47 -05:00
amdgpu_gem.c drm-misc-next for 5.4: 2019-08-21 16:44:41 +10:00
amdgpu_gem.h dma-buf: rename reservation_object to dma_resv 2019-08-13 09:09:30 +02:00
amdgpu_gfx.c
amdgpu_gfx.h drm/amdgpu: add RAS callback for gfx 2019-07-31 14:51:08 -05:00
amdgpu_gmc.c drm/amdgpu: disable agp for sriov 2019-08-22 17:15:06 -05:00
amdgpu_gmc.h drm/amdgpu: Export function to flush TLB of specific vm hub 2019-08-15 10:57:48 -05:00
amdgpu_gtt_mgr.c
amdgpu_i2c.c
amdgpu_i2c.h
amdgpu_ib.c
amdgpu_ids.c drm-misc-next for 5.4: 2019-08-21 16:44:41 +10:00
amdgpu_ids.h dma-buf: rename reservation_object to dma_resv 2019-08-13 09:09:30 +02:00
amdgpu_ih.c
amdgpu_ih.h
amdgpu_ioc32.c
amdgpu_irq.c drm/amdgpu: poll ras_controller_irq and err_event_athub_irq status 2019-09-13 17:11:04 -05:00
amdgpu_irq.h
amdgpu_job.c drm-misc-next for v5.3: 2019-06-14 11:44:24 +02:00
amdgpu_job.h drm/amdgpu: add ib preemption status in amdgpu_job (v2) 2019-06-21 18:57:40 -05:00
amdgpu_kms.c drm/amdgpu: reserve at least 4MB of VRAM for page tables v2 2019-09-13 17:38:47 -05:00
amdgpu_mes.h
amdgpu_mmhub.h drm/amdgpu: add mmhub ras_late_init callback function (v2) 2019-09-13 17:11:05 -05:00
amdgpu_mn.c dma-buf: rename reservation_object to dma_resv 2019-08-13 09:09:30 +02:00
amdgpu_mn.h
amdgpu_mode.h
amdgpu_nbio.h drm/amdgpu: add ras_late_init callback function for nbio v7_4 (v3) 2019-09-13 17:11:05 -05:00
amdgpu_object.c Merge tag 'drm-next-5.4-2019-08-23' of git://people.freedesktop.org/~agd5f/linux into drm-next 2019-08-27 17:22:15 +10:00
amdgpu_object.h drm-misc-next for 5.4: 2019-08-21 16:44:41 +10:00
amdgpu_pll.c
amdgpu_pll.h
amdgpu_pm.c drm/amd/amdgpu: hide voltage and power sensors on SI and KV parts 2019-08-29 15:52:32 -05:00
amdgpu_pm.h
amdgpu_pmu.c
amdgpu_pmu.h
amdgpu_psp.c drm/amdgpu/psp: keep TMR in visible vram region for SRIOV 2019-08-29 15:52:32 -05:00
amdgpu_psp.h drm/amdgpu/psp: move TMR to cpu invisible vram region 2019-08-21 22:16:45 -05:00
amdgpu_ras.c drm/amdgpu: add helper function to do common ras_late_init/fini (v3) 2019-09-13 17:11:04 -05:00
amdgpu_ras.h drm/amdgpu: add helper function to do common ras_late_init/fini (v3) 2019-09-13 17:11:04 -05:00
amdgpu_ras_eeprom.c drm/amdgpu: fix spelling mistake "jumpimng" -> "jumping" 2019-08-29 15:52:32 -05:00
amdgpu_ras_eeprom.h drm/amdgpu: Add RAS EEPROM table. 2019-08-27 08:17:14 -05:00
amdgpu_ring.c
amdgpu_ring.h
amdgpu_rlc.c
amdgpu_rlc.h
amdgpu_sa.c
amdgpu_sched.c
amdgpu_sched.h
amdgpu_sdma.c
amdgpu_sdma.h
amdgpu_socbb.h
amdgpu_sync.c dma-buf: rename reservation_object to dma_resv 2019-08-13 09:09:30 +02:00
amdgpu_sync.h dma-buf: rename reservation_object to dma_resv 2019-08-13 09:09:30 +02:00
amdgpu_test.c
amdgpu_trace.h
amdgpu_trace_points.c
amdgpu_ttm.c Merge tag 'drm-next-5.4-2019-08-30' of git://people.freedesktop.org/~agd5f/linux into drm-next 2019-09-06 16:40:28 +10:00
amdgpu_ttm.h drm-misc-next for 5.4: 2019-08-21 16:44:41 +10:00
amdgpu_ucode.c drm/amdgpu: fix debug level for ppt offset/size 2019-08-21 22:15:28 -05:00
amdgpu_ucode.h drm/amdgpu: extend PSP FW loading support to 8 SDMA instances 2019-08-02 10:30:39 -05:00
amdgpu_umc.h drm/amdgpu: implement UMC 64 bits REG operations 2019-08-09 11:17:10 -05:00
amdgpu_uvd.c dma-buf: rename reservation_object to dma_resv 2019-08-13 09:09:30 +02:00
amdgpu_uvd.h
amdgpu_vce.c
amdgpu_vce.h
amdgpu_vcn.c Revert "drm/amdgpu: use direct loading on renoir vcn for the moment" 2019-08-22 17:48:46 -05:00
amdgpu_vcn.h drm/amd/amdgpu/vcn_v2_0: Mark RB commands as KMD commands 2019-07-30 23:48:32 -05:00
amdgpu_vf_error.c
amdgpu_vf_error.h
amdgpu_virt.c drm/amdgpu: cleanup vega10 SRIOV code path 2019-08-02 10:17:21 -05:00
amdgpu_virt.h drm/amdgpu: cleanup vega10 SRIOV code path 2019-08-02 10:17:21 -05:00
amdgpu_vm.c drm/amdgpu: use moving fence instead of exclusive for VM updates 2019-09-13 17:38:38 -05:00
amdgpu_vm.h drm/amdgpu: reserve at least 4MB of VRAM for page tables v2 2019-09-13 17:38:47 -05:00
amdgpu_vm_cpu.c
amdgpu_vm_sdma.c drm/amdgpu: switch driver from bo->resv to bo->base.resv 2019-08-06 08:21:54 +02:00
amdgpu_vram_mgr.c drm/amdgpu: reserve at least 4MB of VRAM for page tables v2 2019-09-13 17:38:47 -05:00
amdgpu_xgmi.c drm/amdgpu: adding xgmi error monitoring 2019-07-30 23:22:34 -05:00
amdgpu_xgmi.h
arct_reg_init.c drm/amd/powerplay: initialize arcturus MP1 and THM base address 2019-07-30 23:48:33 -05:00
athub_v1_0.c drm/amdgpu: split athub clock gating from mmhub 2019-08-12 12:47:48 -05:00
athub_v1_0.h drm/amdgpu: split athub clock gating from mmhub 2019-08-12 12:47:48 -05:00
athub_v2_0.c drm/amdgpu/athub2: set clock gating for navi12 2019-08-12 12:47:47 -05:00
athub_v2_0.h
atom.c
atom.h
atombios_crtc.c
atombios_crtc.h
atombios_dp.c
atombios_dp.h
atombios_encoders.c
atombios_encoders.h
atombios_i2c.c
atombios_i2c.h
cik.c drm/amdgpu: add reset_method asic callback for cik 2019-07-30 23:24:06 -05:00
cik.h
cik_dpm.h
cik_ih.c
cik_ih.h
cik_sdma.c
cik_sdma.h
cikd.h
clearstate_ci.h
clearstate_defs.h
clearstate_gfx9.h
clearstate_gfx10.h
clearstate_si.h
clearstate_vi.h
cz_ih.c
cz_ih.h
dce_v6_0.c drm/amdgpu: Update pitch on page flips without DC as well 2019-08-12 12:47:47 -05:00
dce_v6_0.h
dce_v8_0.c drm/amdgpu: Update pitch on page flips without DC as well 2019-08-12 12:47:47 -05:00
dce_v8_0.h
dce_v10_0.c drm/amdgpu: Update pitch on page flips without DC as well 2019-08-12 12:47:47 -05:00
dce_v10_0.h
dce_v11_0.c drm/amdgpu: Update pitch on page flips without DC as well 2019-08-12 12:47:47 -05:00
dce_v11_0.h
dce_virtual.c drm/amdgpu/virtual_dce: drop error message in hw_init 2019-08-29 15:52:32 -05:00
dce_virtual.h
df_v1_7.c
df_v1_7.h
df_v3_6.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
df_v3_6.h
emu_soc.c
gfx_v6_0.c
gfx_v6_0.h
gfx_v7_0.c Linux 5.3-rc3 2019-08-09 13:07:28 -05:00
gfx_v7_0.h
gfx_v8_0.c Linux 5.3-rc3 2019-08-09 13:07:28 -05:00
gfx_v8_0.h
gfx_v9_0.c drm/amdgpu: only apply gds clearing workaround when ras is supported 2019-09-13 17:36:29 -05:00
gfx_v9_0.h
gfx_v10_0.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
gfx_v10_0.h
gfxhub_v1_0.c
gfxhub_v1_0.h
gfxhub_v1_1.c
gfxhub_v1_1.h
gfxhub_v2_0.c drm/amdgpu: Set VM_L2_CNTL.PDE_FAULT_CLASSIFICATION to 0 for GFX10 2019-08-15 10:58:14 -05:00
gfxhub_v2_0.h
gmc_v6_0.c drm/amdgpu: set adev->num_vmhubs for gmc6,7,8 2019-08-23 11:35:25 -05:00
gmc_v6_0.h
gmc_v7_0.c drm/amdgpu: set adev->num_vmhubs for gmc6,7,8 2019-08-23 11:35:25 -05:00
gmc_v7_0.h
gmc_v8_0.c drm/amdgpu: set adev->num_vmhubs for gmc6,7,8 2019-08-23 11:35:25 -05:00
gmc_v8_0.h
gmc_v9_0.c drm/amdgpu: fix memory leak when ras is not supported on specific ip block 2019-09-13 17:36:22 -05:00
gmc_v9_0.h
gmc_v10_0.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
gmc_v10_0.h
iceland_ih.c
iceland_ih.h
iceland_sdma_pkt_open.h
Kconfig
kv_dpm.c
kv_dpm.h
kv_smc.c
Makefile drm/amdgpu: Vega20 SMU I2C HW engine controller. 2019-08-27 09:17:35 -05:00
mes_v10_1.c
mes_v10_1.h
mmhub_v1_0.c drm/amdgpu: fix memory leak when ras is not supported on specific ip block 2019-09-13 17:36:22 -05:00
mmhub_v1_0.h drm/amdgpu: add amdgpu_mmhub_funcs definition 2019-08-12 12:47:48 -05:00
mmhub_v2_0.c drm/amdgpu: Set VM_L2_CNTL.PDE_FAULT_CLASSIFICATION to 0 for GFX10 2019-08-15 10:58:14 -05:00
mmhub_v2_0.h
mmhub_v9_4.c drm/amdgpu: enable mmhub clock gating for Arcturus 2019-08-12 12:47:49 -05:00
mmhub_v9_4.h drm/amdgpu: add mmhub clock gating for Arcturus 2019-08-12 12:47:49 -05:00
mmsch_v1_0.h
mxgpu_ai.c drm/amdgpu: cleanup vega10 SRIOV code path 2019-08-02 10:17:21 -05:00
mxgpu_ai.h
mxgpu_vi.c
mxgpu_vi.h
navi10_ih.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
navi10_ih.h
navi10_reg_init.c drm/amdgpu/discovery: move common discovery code out of navi1*_reg_base_init() 2019-08-06 13:53:05 -05:00
navi10_sdma_pkt_open.h
navi12_reg_init.c drm/amdgpu: initialize reg base for navi12 2019-08-02 10:30:39 -05:00
navi14_reg_init.c drm/amdgpu/discovery: move common discovery code out of navi1*_reg_base_init() 2019-08-06 13:53:05 -05:00
nbio_v2_3.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
nbio_v2_3.h drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
nbio_v6_1.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
nbio_v6_1.h drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
nbio_v7_0.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
nbio_v7_0.h drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
nbio_v7_4.c drm/amdgpu: add ras_late_init callback function for nbio v7_4 (v3) 2019-09-13 17:11:05 -05:00
nbio_v7_4.h drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
nv.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
nv.h drm/amdgpu: initialize reg base for navi12 2019-08-02 10:30:39 -05:00
nvd.h
ObjectID.h
ppsmc.h
psp_gfx_if.h drm/amdgpu: extend PSP FW loading support to 8 SDMA instances 2019-08-02 10:30:39 -05:00
psp_v3_1.c drm/amdgpu: remove redundant argument for psp_funcs::cmd_submit callback 2019-08-21 22:16:37 -05:00
psp_v3_1.h
psp_v10_0.c drm/amdgpu: remove redundant argument for psp_funcs::cmd_submit callback 2019-08-21 22:16:37 -05:00
psp_v10_0.h
psp_v11_0.c drm/amdgpu: remove redundant argument for psp_funcs::cmd_submit callback 2019-08-21 22:16:37 -05:00
psp_v11_0.h
psp_v12_0.c drm/amdgpu: remove redundant argument for psp_funcs::cmd_submit callback 2019-08-21 22:16:37 -05:00
psp_v12_0.h drm/amdgpu: add psp_v12_0 for renoir (v2) 2019-08-12 12:47:50 -05:00
r600_dpm.h
sdma_v2_4.c
sdma_v2_4.h
sdma_v3_0.c
sdma_v3_0.h
sdma_v4_0.c drm/amdgpu: fix memory leak when ras is not supported on specific ip block 2019-09-13 17:36:22 -05:00
sdma_v4_0.h
sdma_v5_0.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
sdma_v5_0.h
si.c drm/amdgpu/si: fix ASIC tests 2019-08-29 15:52:32 -05:00
si.h
si_dma.c
si_dma.h
si_dpm.c
si_dpm.h
si_enums.h
si_ih.c
si_ih.h
si_smc.c
sid.h
sislands_smc.h
smu_v11_0_i2c.c drm/amdgpu: Vega20 SMU I2C HW engine controller. 2019-08-27 09:17:35 -05:00
smu_v11_0_i2c.h drm/amdgpu: Vega20 SMU I2C HW engine controller. 2019-08-27 09:17:35 -05:00
soc15.c drm/amdgpu: switch to amdgpu_ras_late_init for nbio v7_4 (v2) 2019-09-13 17:11:05 -05:00
soc15.h
soc15_common.h drm/amdgpu: cleanup vega10 SRIOV code path 2019-08-02 10:17:21 -05:00
soc15d.h
ta_ras_if.h
ta_xgmi_if.h
tonga_ih.c
tonga_ih.h
tonga_sdma_pkt_open.h
umc_v6_1.c drm/amdgpu: implement UMC 64 bits REG operations 2019-08-09 11:17:10 -05:00
umc_v6_1.h drm/amdgpu: implement umc ras init function 2019-08-02 10:30:38 -05:00
uvd_v4_2.c
uvd_v4_2.h
uvd_v5_0.c
uvd_v5_0.h
uvd_v6_0.c
uvd_v6_0.h
uvd_v7_0.c
uvd_v7_0.h
vce_v2_0.c
vce_v2_0.h
vce_v3_0.c
vce_v3_0.h
vce_v4_0.c
vce_v4_0.h
vcn_v1_0.c
vcn_v1_0.h
vcn_v2_0.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
vcn_v2_0.h
vcn_v2_5.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
vcn_v2_5.h
vega10_ih.c drm/amdgpu: switch to new amdgpu_nbio structure 2019-09-13 17:11:03 -05:00
vega10_ih.h
vega10_reg_init.c drm/amdgpu: enable Doorbell support for Renoir (v2) 2019-08-12 12:47:50 -05:00
vega10_sdma_pkt_open.h
vega20_reg_init.c drm/amdgpu: init RSMU and UMC ip base address for vega20 2019-07-31 14:48:51 -05:00
vi.c drm/amdgpu: add reset_method asic callback for vi 2019-07-30 23:24:10 -05:00
vi.h
vid.h