summaryrefslogtreecommitdiff
path: root/drivers/gpu
AgeCommit message (Collapse)AuthorFilesLines
2023-06-09Revert "drm/amd/display: Do not set drr on pipe commit"Michel Dänzer2-13/+0
This reverts commit e101bf95ea87ccc03ac2f48dfc0757c6364ff3c7. Caused a regression: Samsung Odyssey Neo G9, running at 5120x1440@240/VRR, connected to Navi 21 via DisplayPort, blanks and the GPU hangs while starting the Steam game Assetto Corsa Competizione (via Proton 7.0). Example dmesg excerpt: amdgpu 0000:0c:00.0: [drm] ERROR [CRTC:82:crtc-0] flip_done timed out NMI watchdog: Watchdog detected hard LOCKUP on cpu 6 [...] RIP: 0010:amdgpu_device_rreg.part.0+0x2f/0xf0 [amdgpu] Code: 41 54 44 8d 24 b5 00 00 00 00 55 89 f5 53 48 89 fb 4c 3b a7 60 0b 00 00 73 6a 83 e2 02 74 29 4c 03 a3 68 0b 00 00 45 8b 24 24 <48> 8b 43 08 0f b7 70 3e 66 90 44 89 e0 5b 5d 41 5c 31 d2 31 c9 31 RSP: 0000:ffffb39a119dfb88 EFLAGS: 00000086 RAX: ffffffffc0eb96a0 RBX: ffff9e7963dc0000 RCX: 0000000000007fff RDX: 0000000000000000 RSI: 0000000000004ff6 RDI: ffff9e7963dc0000 RBP: 0000000000004ff6 R08: ffffb39a119dfc40 R09: 0000000000000010 R10: ffffb39a119dfc40 R11: ffffb39a119dfc44 R12: 00000000000e05ae R13: 0000000000000000 R14: ffff9e7963dc0010 R15: 0000000000000000 FS: 000000001012f6c0(0000) GS:ffff9e805eb80000(0000) knlGS:000000007fd40000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00000000461ca000 CR3: 00000002a8a20000 CR4: 0000000000350ee0 Call Trace: <TASK> dm_read_reg_func+0x37/0xc0 [amdgpu] generic_reg_get2+0x22/0x60 [amdgpu] optc1_get_crtc_scanoutpos+0x6a/0xc0 [amdgpu] dc_stream_get_scanoutpos+0x74/0x90 [amdgpu] dm_crtc_get_scanoutpos+0x82/0xf0 [amdgpu] amdgpu_display_get_crtc_scanoutpos+0x91/0x190 [amdgpu] ? dm_read_reg_func+0x37/0xc0 [amdgpu] amdgpu_get_vblank_counter_kms+0xb4/0x1a0 [amdgpu] dm_pflip_high_irq+0x213/0x2f0 [amdgpu] amdgpu_dm_irq_handler+0x8a/0x200 [amdgpu] amdgpu_irq_dispatch+0xd4/0x220 [amdgpu] amdgpu_ih_process+0x7f/0x110 [amdgpu] amdgpu_irq_handler+0x1f/0x70 [amdgpu] __handle_irq_event_percpu+0x46/0x1b0 handle_irq_event+0x34/0x80 handle_edge_irq+0x9f/0x240 __common_interrupt+0x66/0x110 common_interrupt+0x5c/0xd0 asm_common_interrupt+0x22/0x40 Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com> Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Michel Dänzer <mdaenzer@redhat.com> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/amdgpu: Fix warnings in amdgpu_irq.cSrinivasan Shanmugam1-13/+14
checkpatch reports: WARNING: Prefer 'unsigned int' to bare use of 'unsigned' WARNING: braces {} are not necessary for any arm of this statement + if (nvec <= 0) { [...] + } else { [...] WARNING: Block comments use a trailing */ on a separate line Cc: Christian König <christian.koenig@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09Revert "drm/amd/display: Block optimize on consecutive FAMS enables"Michel Dänzer2-22/+3
This reverts commit 751e17147953bc30036b8fe0eaaf780b6951404c. It depends on its parent commit, which we want to revert. Acked-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Michel Dänzer <mdaenzer@redhat.com> [Hamza: fix a whitespace issue in dcn30_prepare_bandwidth()] Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: Add a low priority scheduler for VRAM clearingMukul Joshi3-12/+34
Add a low priority DRM scheduler for VRAM clearing instead of using the exisiting high priority scheduler. Use the high priority scheduler for migrations and evictions. Signed-off-by: Mukul Joshi <mukul.joshi@amd.com> Acked-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu/vcn: Modify mismatched function nameJiapeng Chong1-2/+2
No functional modification involved. drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c:374: warning: expecting prototype for vcn_v4_0_mc_resume_dpg_mode(). Prototype was for vcn_v4_0_3_mc_resume_dpg_mode() instead. drivers/gpu/drm/amd/amdgpu/vcn_v4_0_3.c:631: warning: expecting prototype for vcn_v4_0_enable_clock_gating(). Prototype was for vcn_v4_0_3_enable_clock_gating() instead. Reported-by: Abaci Robot <abaci@linux.alibaba.com> Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=5284 Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: Modify mismatched function nameJiapeng Chong1-7/+7
No functional modification involved. drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:426: warning: expecting prototype for sdma_v4_4_2_gfx_stop(). Prototype was for sdma_v4_4_2_inst_gfx_stop() instead. drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:457: warning: expecting prototype for sdma_v4_4_2_rlc_stop(). Prototype was for sdma_v4_4_2_inst_rlc_stop() instead. drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:470: warning: expecting prototype for sdma_v4_4_2_page_stop(). Prototype was for sdma_v4_4_2_inst_page_stop() instead. drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:506: warning: expecting prototype for sdma_v4_4_2_ctx_switch_enable(). Prototype was for sdma_v4_4_2_inst_ctx_switch_enable() instead. drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:561: warning: expecting prototype for sdma_v4_4_2_enable(). Prototype was for sdma_v4_4_2_inst_enable() instead. drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:798: warning: expecting prototype for sdma_v4_4_2_rlc_resume(). Prototype was for sdma_v4_4_2_inst_rlc_resume() instead. drivers/gpu/drm/amd/amdgpu/sdma_v4_4_2.c:814: warning: expecting prototype for sdma_v4_4_2_load_microcode(). Prototype was for sdma_v4_4_2_inst_load_microcode() instead. Reported-by: Abaci Robot <abaci@linux.alibaba.com> Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=5283 Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: Remove duplicate includeJiapeng Chong1-1/+0
./drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c: amdgpu_xcp.h is included more than once. Reported-by: Abaci Robot <abaci@linux.alibaba.com> Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=5281 Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: remove unused variables res_create_maximus_funcs and ↵Tom Rix1-23/+0
debug_defaults_diags gcc with W=1 reports drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20_resource.c:1069:43: error: ‘res_create_maximus_funcs’ defined but not used [-Werror=unused-const-variable=] 1069 | static const struct resource_create_funcs res_create_maximus_funcs = { | ^~~~~~~~~~~~~~~~~~~~~~~~ drivers/gpu/drm/amd/amdgpu/../display/dc/dcn20/dcn20_resource.c:727:38: error: ‘debug_defaults_diags’ defined but not used [-Werror=unused-const-variable=] 727 | static const struct dc_debug_options debug_defaults_diags = { | ^~~~~~~~~~~~~~~~~~~~ These variables are not used so remove them. Fixes: 25879d7b4986 ("drm/amd/display: Clean FPGA code in dc") Signed-off-by: Tom Rix <trix@redhat.com> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: avoid calling missing .resync_fifo_dccg_dio()Arnd Bergmann1-0/+2
The .resync_fifo_dccg_dio() callback pointer was added in an #ifdef block, but is called unconditionally: drivers/gpu/drm/amd/amdgpu/../display/dc/dce110/dce110_hw_sequencer.c:2292:31: error: 'struct hwseq_private_funcs' has no member named 'resync_fifo_dccg_dio' Add the same #ifdef around the caller as well. Fixes: 3e8d74cb128f ("drm/amd/display: Trigger DIO FIFO resync on commit streams") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Hamza Mahfooz <hamza.mahfooz@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: Align partition memory size to page sizePhilip Yang1-1/+1
The compute partition memory size calculated from KFD_XCP_MEMORY_SIZE may not align to page size if xcp_mgr->num_xcp_per_mem_partition is 6. Change the KFD_XCP_MEMORY_SIZE macro to return page align size, so KFD node memory size reported in sysfs is page align size, to avoid application VRAM allocation failure because application may use the size directly and Thunk requires the memory allocation size is page size align. Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/pm: reverse mclk and fclk clocks levels for vangoghTim Huang1-4/+6
This patch reverses the DPM clocks levels output of pp_dpm_mclk and pp_dpm_fclk. On dGPUs and older APUs we expose the levels from lowest clocks to highest clocks. But for some APUs, the clocks levels that from the DFPstateTable are given the reversed orders by PMFW. Like the memory DPM clocks that are exposed by pp_dpm_mclk. It's not intuitive that they are reversed on these APUs. All tools and software that talks to the driver then has to know different ways to interpret the data depending on the asic. So we need to reverse them to expose the clocks levels from the driver consistently. Signed-off-by: Tim Huang <Tim.Huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/pm: reverse mclk and fclk clocks levels for yellow carpTim Huang1-2/+3
This patch reverses the DPM clocks levels output of pp_dpm_mclk and pp_dpm_fclk. On dGPUs and older APUs we expose the levels from lowest clocks to highest clocks. But for some APUs, the clocks levels that from the DFPstateTable are given the reversed orders by PMFW. Like the memory DPM clocks that are exposed by pp_dpm_mclk. It's not intuitive that they are reversed on these APUs. All tools and software that talks to the driver then has to know different ways to interpret the data depending on the asic. So we need to reverse them to expose the clocks levels from the driver consistently. Signed-off-by: Tim Huang <Tim.Huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/pm: reverse mclk clocks levels for SMU v13.0.5Tim Huang1-2/+3
This patch reverses the DPM clocks levels output of pp_dpm_mclk. On dGPUs and older APUs we expose the levels from lowest clocks to highest clocks. But for some APUs, the clocks levels that from the DFPstateTable are given the reversed orders by PMFW. Like the memory DPM clocks that are exposed by pp_dpm_mclk. It's not intuitive that they are reversed on these APUs. All tools and software that talks to the driver then has to know different ways to interpret the data depending on the asic. So we need to reverse them to expose the clocks levels from the driver consistently. Signed-off-by: Tim Huang <Tim.Huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/pm: reverse mclk and fclk clocks levels for SMU v13.0.4Tim Huang1-2/+3
This patch reverses the DPM clocks levels output of pp_dpm_mclk and pp_dpm_fclk. On dGPUs and older APUs we expose the levels from lowest clocks to highest clocks. But for some APUs, the clocks levels that from the DFPstateTable are given the reversed orders by PMFW. Like the memory DPM clocks that are exposed by pp_dpm_mclk. It's not intuitive that they are reversed on these APUs. All tools and software that talks to the driver then has to know different ways to interpret the data depending on the asic. So we need to reverse them to expose the clocks levels from the driver consistently. Signed-off-by: Tim Huang <Tim.Huang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: remove unused variable num_xccTom Rix1-3/+0
gcc with W=1 reports drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c:2138:13: error: variable ‘num_xcc’ set but not used [-Werror=unused-but-set-variable] 2138 | int num_xcc; | ^~~~~~~ This variable is not used so remove it. Signed-off-by: Tom Rix <trix@redhat.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: fix acpi build warningsArnd Bergmann1-2/+2
Two newly introduced functions are in the global namespace but have no prototypes or callers outside of amdgpu_acpi.c, another function is static but only has a caller inside of an #ifdef: drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:902:13: error: no previous prototype for 'amdgpu_acpi_get_node_id' [-Werror=missing-prototypes] drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:928:30: error: no previous prototype for 'amdgpu_acpi_get_dev' [-Werror=missing-prototypes] drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:860:33: error: 'amdgpu_acpi_get_numa_info' defined but not used [-Werror=unused-function] Avoid the warnings by marking all of them static and ensuring that the compiler is able to see the callsites. v2: rebase on latest code (Alex) Fixes: fa0497c34eb7 ("drm/amdgpu: Add API to get numa information of XCC") Fixes: 1cc823011a23 ("drm/amdgpu: Store additional numa node information") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: use %pad format string for dma_addr_tArnd Bergmann1-1/+1
DMA addresses can be shorter than u64, which results in a broken debug output: drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c: In function 'amdgpu_gart_table_ram_alloc': drivers/gpu/drm/amd/amdgpu/amdgpu.h:41:22: error: format '%llx' expects argument of type 'long long unsigned int', but argument 4 has type 'dma_addr_t' {aka 'unsigned int'} [-Werror=format=] drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c:146:9: note: in expansion of macro 'dev_info' 146 | dev_info(adev->dev, "%s dma_addr:%llx\n", __func__, dma_addr); Use the special %pad format string and pass the DMA address by reference. Fixes: c9a502e981a9 ("drm/amdgpu: Allocate GART table in RAM for AMD APU") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu:mark aqua_vanjaram_reg_init.c function as staticArnd Bergmann1-4/+4
A few newly added global functions have no prototype, which causes warnings: drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c:169:5: error: no previous prototype for 'aqua_vanjaram_select_scheds' [-Werror=missing-prototypes] drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c:310:5: error: no previous prototype for '__aqua_vanjaram_get_xcc_per_xcp' [-Werror=missing-prototypes] drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c:337:5: error: no previous prototype for '__aqua_vanjaram_get_xcp_ip_info' [-Werror=missing-prototypes] drivers/gpu/drm/amd/amdgpu/aqua_vanjaram_reg_init.c:593:5: error: no previous prototype for 'aqua_vanjaram_get_xcp_ip_details' [-Werror=missing-prototypes] There are no callers from other files, so just mark them as 'static'. Fixes: cd7d8400aa04 ("drm/amdgpu: add partition schedule for GC(9, 4, 3)") Fixes: 9cb18287d8f1 ("drm/amdgpu: Add SOC partition funcs for GC v9.4.3") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdkfd: mark local functions as staticArnd Bergmann1-4/+4
The file was newly added and causes some -Wmissing-prototype warnings: drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c:57:5: error: no previous prototype for 'kgd_gfx_v9_4_3_hqd_sdma_load' [-Werror=missing-prototypes] drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c:126:5: error: no previous prototype for 'kgd_gfx_v9_4_3_hqd_sdma_dump' [-Werror=missing-prototypes] drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c:163:6: error: no previous prototype for 'kgd_gfx_v9_4_3_hqd_sdma_is_occupied' [-Werror=missing-prototypes] drivers/gpu/drm/amd/amdgpu/amdgpu_amdkfd_gc_9_4_3.c:181:5: error: no previous prototype for 'kgd_gfx_v9_4_3_hqd_sdma_destroy' [-Werror=missing-prototypes] Mark these all as 'static' since there are no outside callers. Fixes: a805889a1531 ("drm/amdkfd: Update SDMA queue management for GFX9.4.3") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/pm: mark irq functions as 'static'Arnd Bergmann1-2/+2
Two newly added functions cause a warning because they lack a prototype: drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu13/smu_v13_0_6_ppt.c:1328:5: error: no previous prototype for 'smu_v13_0_6_set_irq_state' [-Werror=missing-prototypes] drivers/gpu/drm/amd/amdgpu/../pm/swsmu/smu13/smu_v13_0_6_ppt.c:1368:5: error: no previous prototype for 'smu_v13_0_6_register_irq_handler' [-Werror=missing-prototypes] They are only used locally, so just mark them static. Fixes: 676915e4108f ("drm/amd/pm: Add ih for SMU v13.0.6 thermal throttling") Signed-off-by: Arnd Bergmann <arnd@arndb.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: Fix unsigned comparison with zero in gmc_v9_0_process_interrupt()Harshit Mogalapalli1-2/+2
Smatch warns: drivers/gpu/drm/amd/amdgpu/gmc_v9_0.c:579: unsigned 'xcc_id' is never less than zero. gfx_v9_4_3_ih_to_xcc_inst() returns negative numbers as well. Fix this by changing type of xcc_id to int. Fixes: 98b2e9cad227 ("drm/amdgpu: correct the vmhub index when page fault occurs") Signed-off-by: Harshit Mogalapalli <harshit.m.mogalapalli@oracle.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: Fix a couple of spelling mistakes in info and debug messagesColin Ian King2-2/+2
There are a couple of spelling mistakes, one in a dev_info message and the other in a dev_debug message. Fix them. Signed-off-by: Colin Ian King <colin.i.king@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/pm: Fix output of pp_od_clk_voltageJonatas Esteves1-7/+5
Printing the other clock types should not be conditioned on being able to print OD_SCLK. Some GPUs currently have limited capability of only printing a subset of these. Since this condition was introduced in v5.18-rc1, reading from `pp_od_clk_voltage` has been returning empty on the Asus ROG Strix G15 (2021). Fixes: 79c65f3fcbb1 ("drm/amd/pm: do not expose power implementation details to amdgpu_pm.c") Reviewed-by: Evan Quan <evan.quan@amd.com> Signed-off-by: Jonatas Esteves <jntesteves@gmail.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: Disable interrupt tracker on NBIOv7.9Lijo Lazar2-0/+4
Enabling nBIF interrupt history tracker prevents LCLK deep sleep, hence disable it Signed-off-by: Lijo Lazar <lijo.lazar@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/pm: add missing NotifyPowerSource message mapping for SMU13.0.7Evan Quan1-0/+1
Otherwise, the power source switching will fail due to message unavailable. Fixes: bf4823267a81 ("drm/amd/pm: fix possible power mode mismatch between driver and PMFW") Signed-off-by: Evan Quan <evan.quan@amd.com> Reviewed-by: Guchun Chen <guchun.chen@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/radeon: fix possible division-by-zero errorsNikita Zhandarovich3-6/+18
Function rv740_get_decoded_reference_divider() may return 0 due to unpredictable reference divider value calculated in radeon_atom_get_clock_dividers(). This will lead to division-by-zero error once that value is used as a divider in calculating 'clk_s'. While unlikely, this issue should nonetheless be prevented so add a sanity check for such cases by testing 'decoded_ref' value against 0. Found by Linux Verification Center (linuxtesting.org) with static analysis tool SVACE. v2: minor coding style fixes (Alex) In practice this should actually happen as the vbios should be properly populated. Fixes: 66229b200598 ("drm/radeon/kms: add dpm support for rv7xx (v4)") Signed-off-by: Nikita Zhandarovich <n.zhandarovich@fintech.ru> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/radeon: stop including swiotlb.hChristoph Hellwig1-1/+0
radeon does not need swiotlb.h, so stop including it. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: init the XCC_DOORBELL_FENCE regsShiwu Zhang1-0/+17
add the the init_registers callback for nbio_v7_9 Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com> Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com> Reviewed-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: remove unused definitionTao Zhou1-8/+0
mmhub_v1_8_mmea_cgtt_clk_cntl_reg is defined but not used. Reported-by: kernel test robot <lkp@intel.com> Signed-off-by: Tao Zhou <tao.zhou1@amd.com> Acked-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: Fix uninitialized variable in gfxhub_v1_2_xcp_resumeSrinivasan Shanmugam1-3/+2
drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c:657:6: error: variable 'ret' is used uninitialized whenever 'if' condition is false [-Werror,-Wsometimes-uninitialized] if (!amdgpu_sriov_vf(adev)) ^~~~~~~~~~~~~~~~~~~~~~ drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c:660:9: note: uninitialized use occurs here return ret; ^~~ drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c:657:2: note: remove the 'if' if its condition is always true if (!amdgpu_sriov_vf(adev)) ^~~~~~~~~~~~~~~~~~~~~~~~~~~ drivers/gpu/drm/amd/amdgpu/gfxhub_v1_2.c:648:9: note: initialize the variable 'ret' to silence this warning int ret; ^ = 0 1 error generated. Cc: Luben Tuikov <luben.tuikov@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Lijo Lazar <lijo.lazar@amd.com> Cc: Hawking Zhang <Hawking.Zhang@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: Fix unused amdgpu_acpi_get_numa_info function in ↵Srinivasan Shanmugam1-0/+2
amdgpu_acpi_get_node_id() Fix the below compiler complaining error: drivers/gpu/drm/amd/amdgpu/amdgpu_acpi.c:860:33: error: unused function 'amdgpu_acpi_get_numa_info' [-Werror,-Wunused-function] static struct amdgpu_numa_info *amdgpu_acpi_get_numa_info(uint32_t pxm) ^ 1 error generated. By guarding amdgpu_acpi_get_numa_info & amdgpu_acpi_get_numa_size function, only when CONFIG_ACPI_NUMA is enabled. Suggested-by: Lijo Lazar <lijo.lazar@amd.com> Cc: Christian König <christian.koenig@amd.com> Cc: Lijo Lazar <lijo.lazar@amd.com> Cc: Luben Tuikov <luben.tuikov@amd.com> Cc: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Srinivasan Shanmugam <srinivasan.shanmugam@amd.com> Reviewed-by: Lijo Lazar <lijo.lazar@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: 3.2.237Aric Cyr1-1/+1
This version brings along following fixes: - Improve the message printed - Disable dcn315 pixel rate crb when scaling - Update SR watermarks for DCN314 - Fix dcn315 pixel rate crb scaling check - Reset CRTC State Before Restore from S2idle - Have Payload Properly Created After Resume - Trigger DIO FIFO resync on commit streams - Revert vblank change that causes null pointer crash - Fix possible underflow for displays with large vblank - Apply 60us prefetch for DCFCLK <= 300Mhz - Update correct DCN314 register header Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Aric Cyr <aric.cyr@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: cache trace buffer sizeJosip Pavic2-0/+3
[Why & How] Cache the trace buffer size retrieved from DMUB FW in the driver Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Reviewed-by: Aric Cyr <aric.cyr@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Josip Pavic <josip.pavic@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Reorganize DCN30 MakefileRodrigo Siqueira1-12/+26
DCN30 has a lot of files in the Makefile, and adding each one next to the other makes it hard to read and can increase the chance of merge conflicts. This commit just reorganize the Makefile to put each file associated with DCN30 in its own line. Reviewed-by: Chris Park <chris.park@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Rodrigo Siqueira <rodrigo.siqueira@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Trigger DIO FIFO resync on commit streams for DCN32Saaem Rizvi6-18/+42
[WHY and HOW] Currently, on DCN32 we have an old workaround to resolve a DIO FIFO speed issue when writing to the OTG DIVIDER register. However, this workaround is not safe as we should be applying the DIO FIFO rampup logic when the OTG re disabled along with the encoders. This new workaround accounts for this. If the workaround sequence is incorrect, like it is was, there is a chance we might hang. this new workaround was first implemented in DCN314. Reviewed-by: Alvin Lee <alvin.lee2@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Saaem Rizvi <syedsaaem.rizvi@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Update correct DCN314 register headerCruise Hung4-2/+100
[Why] The register header for DCN314 is not correct. [How] Update correct DCN314 register header. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Cruise Hung <cruise.hung@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Clean FPGA code in dcQingqing Zhuo63-1292/+770
[Why] Drop dead code for Linux. [How] Remove all IS_FPGA_MAXIMUS_DC and IS_DIAG_DC Reviewed-by: Ariel Bernstein <eric.bernstein@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Qingqing Zhuo <qingqing.zhuo@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Apply 60us prefetch for DCFCLK <= 300MhzAlvin Lee2-3/+3
[Description] - Previously we wanted to apply extra 60us of prefetch for min DCFCLK (200Mhz), but DCFCLK can be calculated to be 201Mhz which underflows also without the extra prefetch - Instead, apply the the extra 60us prefetch for any DCFCLK freq <= 300Mhz Reviewed-by: Nevenko Stupar <nevenko.stupar@amd.com> Reviewed-by: Jun Lei <jun.lei@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Alvin Lee <alvin.lee2@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Fix possible underflow for displays with large vblankDaniel Miess1-12/+7
[Why] Underflow observed when using a display with a large vblank region and low refresh rate [How] Simplify calculation of vblank_nom Increase value for VBlankNomDefaultUS to 800us Fixed a null pointer from previous commit of this change Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Daniel Miess <daniel.miess@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Revert vblank change that causes null pointer crashDaniel Miess1-7/+12
Revert commit 1a4bcdbea431 ("drm/amd/display: Fix possible underflow for displays with large vblank") Because it cause some regression Fixes: 1a4bcdbea431 ("drm/amd/display: Fix possible underflow for displays with large vblank") Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Daniel Miess <daniel.miess@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Trigger DIO FIFO resync on commit streamsSaaem Rizvi9-2/+56
[WHY] Currently, there is an intermittent issue where a screen can either go blank or be corrupted. [HOW] To resolve the issue we trigger the ramping logic for DIO FIFO so that it goes back up to the correct speed. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Saaem Rizvi <syedsaaem.rizvi@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Have Payload Properly Created After ResumeFangzhi Zuo1-2/+2
At drm suspend sequence, MST dc_sink is removed. When commit cached MST stream back in drm resume sequence, the MST stream payload is not properly created and added into the payload table. After resume, topology change is reprobed by removing existing streams first. That leads to no payload is found in the existing payload table as below error "[drm] ERROR No payload for [MST PORT:] found in mst state" 1. In encoder .atomic_check routine, remove check existance of dc_sink 2. Bypass MST by checking existence of MST root port. dc_link_type cannot differentiate MST port before topology is rediscovered. Reviewed-by: Wayne Lin <wayne.lin@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Fangzhi Zuo <jerry.zuo@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: fix dcn315 pixel rate crb scaling checkDmytro Laktyushkin1-4/+4
fix dcn315 pixel rate crb scaling check error Reviewed-by: Charlene Liu <charlene.liu@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: lower dp link training message levelHersen Wu1-4/+13
[Why] some test apps report dp link training waring even dp training pass. there are 4 tries of lt within perform_link_training_with_retries. if lt pass within 4 tries, it will NOT be reated as lt failure. for each try of lt, if lt fails, current driver implementation prints message at warning level. this let people think dp lt does not work properly. [How] for 1st, 2nd and 3rd try of lt, print message at debug level. for the 4th try of lt, print message at warning level. Reviewed-by: Jerry Zuo <jerry.zuo@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Hersen Wu <hersenxs.wu@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Fix warning in disabling vblank irqAlan Liu2-21/+16
[Why] During gpu-reset, we toggle vblank irq by calling dc_interrupt_set() instead of amdgpu_irq_get/put() because we don't want to change the irq source's refcount. However, we see the warning when vblank irq is enabled by dc_interrupt_set() during gpu-reset but disabled by amdgpu_irq_put() after gpu-reset. [How] Only in dm_gpureset_toggle_interrupts() we toggle vblank interrupts by calling dc_interrupt_set(). Apart from this we call dm_set_vblank() which uses amdgpu_irq_get/put() to operate vblank irq. Reviewed-by: Bhawanpreet Lakha <bhawanpreet.lakha@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Alan Liu <haoping.liu@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Update SR watermarks for DCN314Nicholas Kazlauskas1-8/+8
[Why & How] Update parameters for SR watermarks for DCN314 Reviewed-by: Charlene Liu <charlene.liu@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: disable dcn315 pixel rate crb when scalingDmytro Laktyushkin1-0/+13
The rough calculation does not account for scaling. Also, make 2 segments the minimum allowed per surface to avoid potential 0 detile with mpc/odm combine on such outputs. Reviewed-by: Ariel Bernstein <eric.bernstein@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amd/display: Fix DMUB debugging print issueCruise Hung3-68/+49
[Why] The DMUB diagnostic data was not printed out correctly. [How] Print the DMUB diagnostic data line by line. Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com> Acked-by: Tom Chung <chiahsuan.chung@amd.com> Signed-off-by: Cruise Hung <cruise.hung@amd.com> Tested-by: Daniel Wheeler <daniel.wheeler@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/amdgpu: stop including swiotlb.hChristoph Hellwig1-1/+0
amdgpu does not need swiotlb.h, so stop including it. Signed-off-by: Christoph Hellwig <hch@lst.de> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2023-06-09drm/radeon: reintroduce radeon_dp_work_func contentAlex Deucher1-0/+10
Put back the radeon_dp_work_func logic. It seems that handling DP RX interrupts is necessary to make some panels work. This was removed with the MST support, but it regresses some systems so add it back. While we are here, add the proper mutex locking. Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2567 Fixes: 01ad1d9c2888 ("drm/radeon: Drop legacy MST support") Reviewed-by: Lyude Paul <lyude@redhat.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Cc: Lyude Paul <lyude@redhat.com>
generated by cgit v1.2.3 (git 2.39.0) at 2024-09-20 07:02:39 +0300