[PATCH] drm/amdgpu: cleanup coding style in amdgpu_kms.c
kernel test robot
lkp at intel.com
Tue Sep 6 11:11:40 UTC 2022
Hi Jingyu,
Thank you for the patch! Yet something to improve:
[auto build test ERROR on e47eb90a0a9ae20b82635b9b99a8d0979b757ad8]
url: https://github.com/intel-lab-lkp/linux/commits/Jingyu-Wang/drm-amdgpu-cleanup-coding-style-in-amdgpu_kms-c/20220906-104802
base: e47eb90a0a9ae20b82635b9b99a8d0979b757ad8
config: mips-allyesconfig (https://download.01.org/0day-ci/archive/20220906/202209061955.bBpvTHp7-lkp@intel.com/config)
compiler: mips-linux-gcc (GCC) 12.1.0
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
chmod +x ~/bin/make.cross
# https://github.com/intel-lab-lkp/linux/commit/778e4a57afd0db6a6a752487e1a1cda253cd9682
git remote add linux-review https://github.com/intel-lab-lkp/linux
git fetch --no-tags linux-review Jingyu-Wang/drm-amdgpu-cleanup-coding-style-in-amdgpu_kms-c/20220906-104802
git checkout 778e4a57afd0db6a6a752487e1a1cda253cd9682
# save the config file
mkdir build_dir && cp config build_dir/.config
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=mips SHELL=/bin/bash
If you fix the issue, kindly add following tag where applicable
Reported-by: kernel test robot <lkp at intel.com>
All errors (new ones prefixed by >>):
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c: In function 'amdgpu_info_ioctl':
>> drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c:554:76: error: macro "min_t" requires 3 arguments, but only 2 given
554 | ret = copy_to_user(out, &ip, min_t((size_t)size, sizeof(ip)));
| ^
In file included from include/linux/kernel.h:26,
from include/linux/cpumask.h:10,
from include/linux/smp.h:13,
from arch/mips/include/asm/cpu-type.h:12,
from arch/mips/include/asm/timex.h:19,
from include/linux/timex.h:67,
from include/linux/time32.h:13,
from include/linux/time.h:60,
from include/linux/ktime.h:24,
from drivers/gpu/drm/amd/amdgpu/amdgpu_ctx.h:26,
from drivers/gpu/drm/amd/amdgpu/amdgpu.h:43,
from drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c:29:
include/linux/minmax.h:104: note: macro "min_t" defined here
104 | #define min_t(type, x, y) __careful_cmp((type)(x), (type)(y), <)
|
>> drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c:554:46: error: 'min_t' undeclared (first use in this function)
554 | ret = copy_to_user(out, &ip, min_t((size_t)size, sizeof(ip)));
| ^~~~~
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c:554:46: note: each undeclared identifier is reported only once for each function it appears in
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c: In function 'amdgpu_debugfs_firmware_info_show':
>> drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c:1384:27: error: expected expression before '[' token
1384 | #define TA_FW_NAME(type) ([TA_FW_TYPE_PSP_##type] = #type)
| ^
drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c:1385:17: note: in expansion of macro 'TA_FW_NAME'
1385 | TA_FW_NAME(XGMI),
| ^~~~~~~~~~
vim +/min_t +554 drivers/gpu/drm/amd/amdgpu/amdgpu_kms.c
494
495 /*
496 * Userspace get information ioctl
497 */
498 /**
499 * amdgpu_info_ioctl - answer a device specific request.
500 *
501 * @dev: drm device pointer
502 * @data: request object
503 * @filp: drm filp
504 *
505 * This function is used to pass device specific parameters to the userspace
506 * drivers. Examples include: pci device id, pipeline parms, tiling params,
507 * etc. (all asics).
508 * Returns 0 on success, -EINVAL on failure.
509 */
510 int amdgpu_info_ioctl(struct drm_device *dev, void *data, struct drm_file *filp)
511 {
512 struct amdgpu_device *adev = drm_to_adev(dev);
513 struct drm_amdgpu_info *info = data;
514 struct amdgpu_mode_info *minfo = &adev->mode_info;
515 void __user *out = (void __user *)(uintptr_t)info->return_pointer;
516 uint32_t size = info->return_size;
517 struct drm_crtc *crtc;
518 uint32_t ui32 = 0;
519 uint64_t ui64 = 0;
520 int i, found;
521 int ui32_size = sizeof(ui32);
522
523 if (!info->return_size || !info->return_pointer)
524 return -EINVAL;
525
526 switch (info->query) {
527 case AMDGPU_INFO_ACCEL_WORKING:
528 ui32 = adev->accel_working;
529 return copy_to_user(out, &ui32, min(size, 4u)) ? -EFAULT : 0;
530 case AMDGPU_INFO_CRTC_FROM_ID:
531 for (i = 0, found = 0; i < adev->mode_info.num_crtc; i++) {
532 crtc = (struct drm_crtc *)minfo->crtcs[i];
533 if (crtc && crtc->base.id == info->mode_crtc.id) {
534 struct amdgpu_crtc *amdgpu_crtc = to_amdgpu_crtc(crtc);
535
536 ui32 = amdgpu_crtc->crtc_id;
537 found = 1;
538 break;
539 }
540 }
541 if (!found) {
542 DRM_DEBUG_KMS("unknown crtc id %d\n", info->mode_crtc.id);
543 return -EINVAL;
544 }
545 return copy_to_user(out, &ui32, min(size, 4u)) ? -EFAULT : 0;
546 case AMDGPU_INFO_HW_IP_INFO: {
547 struct drm_amdgpu_info_hw_ip ip = {};
548 int ret;
549
550 ret = amdgpu_hw_ip_info(adev, info, &ip);
551 if (ret)
552 return ret;
553
> 554 ret = copy_to_user(out, &ip, min_t((size_t)size, sizeof(ip)));
555 return ret ? -EFAULT : 0;
556 }
557 case AMDGPU_INFO_HW_IP_COUNT: {
558 enum amd_ip_block_type type;
559 uint32_t count = 0;
560
561 switch (info->query_hw_ip.type) {
562 case AMDGPU_HW_IP_GFX:
563 type = AMD_IP_BLOCK_TYPE_GFX;
564 break;
565 case AMDGPU_HW_IP_COMPUTE:
566 type = AMD_IP_BLOCK_TYPE_GFX;
567 break;
568 case AMDGPU_HW_IP_DMA:
569 type = AMD_IP_BLOCK_TYPE_SDMA;
570 break;
571 case AMDGPU_HW_IP_UVD:
572 type = AMD_IP_BLOCK_TYPE_UVD;
573 break;
574 case AMDGPU_HW_IP_VCE:
575 type = AMD_IP_BLOCK_TYPE_VCE;
576 break;
577 case AMDGPU_HW_IP_UVD_ENC:
578 type = AMD_IP_BLOCK_TYPE_UVD;
579 break;
580 case AMDGPU_HW_IP_VCN_DEC:
581 case AMDGPU_HW_IP_VCN_ENC:
582 type = AMD_IP_BLOCK_TYPE_VCN;
583 break;
584 case AMDGPU_HW_IP_VCN_JPEG:
585 type = (amdgpu_device_ip_get_ip_block(adev, AMD_IP_BLOCK_TYPE_JPEG)) ?
586 AMD_IP_BLOCK_TYPE_JPEG : AMD_IP_BLOCK_TYPE_VCN;
587 break;
588 default:
589 return -EINVAL;
590 }
591
592 for (i = 0; i < adev->num_ip_blocks; i++)
593 if (adev->ip_blocks[i].version->type == type &&
594 adev->ip_blocks[i].status.valid &&
595 count < AMDGPU_HW_IP_INSTANCE_MAX_COUNT)
596 count++;
597
598 return copy_to_user(out, &count, min(size, 4u)) ? -EFAULT : 0;
599 }
600 case AMDGPU_INFO_TIMESTAMP:
601 ui64 = amdgpu_gfx_get_gpu_clock_counter(adev);
602 return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
603 case AMDGPU_INFO_FW_VERSION: {
604 struct drm_amdgpu_info_firmware fw_info;
605 int ret;
606
607 /* We only support one instance of each IP block right now. */
608 if (info->query_fw.ip_instance != 0)
609 return -EINVAL;
610
611 ret = amdgpu_firmware_info(&fw_info, &info->query_fw, adev);
612 if (ret)
613 return ret;
614
615 return copy_to_user(out, &fw_info,
616 min((size_t)size, sizeof(fw_info))) ? -EFAULT : 0;
617 }
618 case AMDGPU_INFO_NUM_BYTES_MOVED:
619 ui64 = atomic64_read(&adev->num_bytes_moved);
620 return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
621 case AMDGPU_INFO_NUM_EVICTIONS:
622 ui64 = atomic64_read(&adev->num_evictions);
623 return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
624 case AMDGPU_INFO_NUM_VRAM_CPU_PAGE_FAULTS:
625 ui64 = atomic64_read(&adev->num_vram_cpu_page_faults);
626 return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
627 case AMDGPU_INFO_VRAM_USAGE:
628 ui64 = ttm_resource_manager_usage(&adev->mman.vram_mgr.manager);
629 return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
630 case AMDGPU_INFO_VIS_VRAM_USAGE:
631 ui64 = amdgpu_vram_mgr_vis_usage(&adev->mman.vram_mgr);
632 return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
633 case AMDGPU_INFO_GTT_USAGE:
634 ui64 = ttm_resource_manager_usage(&adev->mman.gtt_mgr.manager);
635 return copy_to_user(out, &ui64, min(size, 8u)) ? -EFAULT : 0;
636 case AMDGPU_INFO_GDS_CONFIG: {
637 struct drm_amdgpu_info_gds gds_info;
638
639 memset(&gds_info, 0, sizeof(gds_info));
640 gds_info.compute_partition_size = adev->gds.gds_size;
641 gds_info.gds_total_size = adev->gds.gds_size;
642 gds_info.gws_per_compute_partition = adev->gds.gws_size;
643 gds_info.oa_per_compute_partition = adev->gds.oa_size;
644 return copy_to_user(out, &gds_info,
645 min((size_t)size, sizeof(gds_info))) ? -EFAULT : 0;
646 }
647 case AMDGPU_INFO_VRAM_GTT: {
648 struct drm_amdgpu_info_vram_gtt vram_gtt;
649
650 vram_gtt.vram_size = adev->gmc.real_vram_size -
651 atomic64_read(&adev->vram_pin_size) -
652 AMDGPU_VM_RESERVED_VRAM;
653 vram_gtt.vram_cpu_accessible_size =
654 min(adev->gmc.visible_vram_size -
655 atomic64_read(&adev->visible_pin_size),
656 vram_gtt.vram_size);
657 vram_gtt.gtt_size = ttm_manager_type(&adev->mman.bdev, TTM_PL_TT)->size;
658 vram_gtt.gtt_size -= atomic64_read(&adev->gart_pin_size);
659 return copy_to_user(out, &vram_gtt,
660 min((size_t)size, sizeof(vram_gtt))) ? -EFAULT : 0;
661 }
662 case AMDGPU_INFO_MEMORY: {
663 struct drm_amdgpu_memory_info mem;
664 struct ttm_resource_manager *gtt_man =
665 &adev->mman.gtt_mgr.manager;
666 struct ttm_resource_manager *vram_man =
667 &adev->mman.vram_mgr.manager;
668
669 memset(&mem, 0, sizeof(mem));
670 mem.vram.total_heap_size = adev->gmc.real_vram_size;
671 mem.vram.usable_heap_size = adev->gmc.real_vram_size -
672 atomic64_read(&adev->vram_pin_size) -
673 AMDGPU_VM_RESERVED_VRAM;
674 mem.vram.heap_usage =
675 ttm_resource_manager_usage(vram_man);
676 mem.vram.max_allocation = mem.vram.usable_heap_size * 3 / 4;
677
678 mem.cpu_accessible_vram.total_heap_size =
679 adev->gmc.visible_vram_size;
680 mem.cpu_accessible_vram.usable_heap_size =
681 min(adev->gmc.visible_vram_size -
682 atomic64_read(&adev->visible_pin_size),
683 mem.vram.usable_heap_size);
684 mem.cpu_accessible_vram.heap_usage =
685 amdgpu_vram_mgr_vis_usage(&adev->mman.vram_mgr);
686 mem.cpu_accessible_vram.max_allocation =
687 mem.cpu_accessible_vram.usable_heap_size * 3 / 4;
688
689 mem.gtt.total_heap_size = gtt_man->size;
690 mem.gtt.usable_heap_size = mem.gtt.total_heap_size -
691 atomic64_read(&adev->gart_pin_size);
692 mem.gtt.heap_usage = ttm_resource_manager_usage(gtt_man);
693 mem.gtt.max_allocation = mem.gtt.usable_heap_size * 3 / 4;
694
695 return copy_to_user(out, &mem,
696 min((size_t)size, sizeof(mem)))
697 ? -EFAULT : 0;
698 }
699 case AMDGPU_INFO_READ_MMR_REG: {
700 unsigned int n, alloc_size;
701 uint32_t *regs;
702 unsigned int se_num = (info->read_mmr_reg.instance >>
703 AMDGPU_INFO_MMR_SE_INDEX_SHIFT) &
704 AMDGPU_INFO_MMR_SE_INDEX_MASK;
705 unsigned int sh_num = (info->read_mmr_reg.instance >>
706 AMDGPU_INFO_MMR_SH_INDEX_SHIFT) &
707 AMDGPU_INFO_MMR_SH_INDEX_MASK;
708
709 /* set full masks if the userspace set all bits
710 * in the bitfields
711 */
712 if (se_num == AMDGPU_INFO_MMR_SE_INDEX_MASK)
713 se_num = 0xffffffff;
714 else if (se_num >= AMDGPU_GFX_MAX_SE)
715 return -EINVAL;
716 if (sh_num == AMDGPU_INFO_MMR_SH_INDEX_MASK)
717 sh_num = 0xffffffff;
718 else if (sh_num >= AMDGPU_GFX_MAX_SH_PER_SE)
719 return -EINVAL;
720
721 if (info->read_mmr_reg.count > 128)
722 return -EINVAL;
723
724 regs = kmalloc_array(info->read_mmr_reg.count, sizeof(*regs), GFP_KERNEL);
725 if (!regs)
726 return -ENOMEM;
727 alloc_size = info->read_mmr_reg.count * sizeof(*regs);
728
729 amdgpu_gfx_off_ctrl(adev, false);
730 for (i = 0; i < info->read_mmr_reg.count; i++) {
731 if (amdgpu_asic_read_register(adev, se_num, sh_num,
732 info->read_mmr_reg.dword_offset + i,
733 ®s[i])) {
734 DRM_DEBUG_KMS("unallowed offset %#x\n",
735 info->read_mmr_reg.dword_offset + i);
736 kfree(regs);
737 amdgpu_gfx_off_ctrl(adev, true);
738 return -EFAULT;
739 }
740 }
741 amdgpu_gfx_off_ctrl(adev, true);
742 n = copy_to_user(out, regs, min(size, alloc_size));
743 kfree(regs);
744 return n ? -EFAULT : 0;
745 }
746 case AMDGPU_INFO_DEV_INFO: {
747 struct drm_amdgpu_info_device *dev_info;
748 uint64_t vm_size;
749 int ret;
750
751 dev_info = kzalloc(sizeof(*dev_info), GFP_KERNEL);
752 if (!dev_info)
753 return -ENOMEM;
754
755 dev_info->device_id = adev->pdev->device;
756 dev_info->chip_rev = adev->rev_id;
757 dev_info->external_rev = adev->external_rev_id;
758 dev_info->pci_rev = adev->pdev->revision;
759 dev_info->family = adev->family;
760 dev_info->num_shader_engines = adev->gfx.config.max_shader_engines;
761 dev_info->num_shader_arrays_per_engine = adev->gfx.config.max_sh_per_se;
762 /* return all clocks in KHz */
763 dev_info->gpu_counter_freq = amdgpu_asic_get_xclk(adev) * 10;
764 if (adev->pm.dpm_enabled) {
765 dev_info->max_engine_clock = amdgpu_dpm_get_sclk(adev, false) * 10;
766 dev_info->max_memory_clock = amdgpu_dpm_get_mclk(adev, false) * 10;
767 } else {
768 dev_info->max_engine_clock = adev->clock.default_sclk * 10;
769 dev_info->max_memory_clock = adev->clock.default_mclk * 10;
770 }
771 dev_info->enabled_rb_pipes_mask = adev->gfx.config.backend_enable_mask;
772 dev_info->num_rb_pipes = adev->gfx.config.max_backends_per_se *
773 adev->gfx.config.max_shader_engines;
774 dev_info->num_hw_gfx_contexts = adev->gfx.config.max_hw_contexts;
775 dev_info->_pad = 0;
776 dev_info->ids_flags = 0;
777 if (adev->flags & AMD_IS_APU)
778 dev_info->ids_flags |= AMDGPU_IDS_FLAGS_FUSION;
779 if (amdgpu_mcbp || amdgpu_sriov_vf(adev))
780 dev_info->ids_flags |= AMDGPU_IDS_FLAGS_PREEMPTION;
781 if (amdgpu_is_tmz(adev))
782 dev_info->ids_flags |= AMDGPU_IDS_FLAGS_TMZ;
783
784 vm_size = adev->vm_manager.max_pfn * AMDGPU_GPU_PAGE_SIZE;
785 vm_size -= AMDGPU_VA_RESERVED_SIZE;
786
787 /* Older VCE FW versions are buggy and can handle only 40bits */
788 if (adev->vce.fw_version &&
789 adev->vce.fw_version < AMDGPU_VCE_FW_53_45)
790 vm_size = min(vm_size, 1ULL << 40);
791
792 dev_info->virtual_address_offset = AMDGPU_VA_RESERVED_SIZE;
793 dev_info->virtual_address_max =
794 min(vm_size, AMDGPU_GMC_HOLE_START);
795
796 if (vm_size > AMDGPU_GMC_HOLE_START) {
797 dev_info->high_va_offset = AMDGPU_GMC_HOLE_END;
798 dev_info->high_va_max = AMDGPU_GMC_HOLE_END | vm_size;
799 }
800 dev_info->virtual_address_alignment = max_t(u32, PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE);
801 dev_info->pte_fragment_size = (1 << adev->vm_manager.fragment_size) * AMDGPU_GPU_PAGE_SIZE;
802 dev_info->gart_page_size = max_t(u32, PAGE_SIZE, AMDGPU_GPU_PAGE_SIZE);
803 dev_info->cu_active_number = adev->gfx.cu_info.number;
804 dev_info->cu_ao_mask = adev->gfx.cu_info.ao_cu_mask;
805 dev_info->ce_ram_size = adev->gfx.ce_ram_size;
806 memcpy(&dev_info->cu_ao_bitmap[0], &adev->gfx.cu_info.ao_cu_bitmap[0],
807 sizeof(adev->gfx.cu_info.ao_cu_bitmap));
808 memcpy(&dev_info->cu_bitmap[0], &adev->gfx.cu_info.bitmap[0],
809 sizeof(adev->gfx.cu_info.bitmap));
810 dev_info->vram_type = adev->gmc.vram_type;
811 dev_info->vram_bit_width = adev->gmc.vram_width;
812 dev_info->vce_harvest_config = adev->vce.harvest_config;
813 dev_info->gc_double_offchip_lds_buf =
814 adev->gfx.config.double_offchip_lds_buf;
815 dev_info->wave_front_size = adev->gfx.cu_info.wave_front_size;
816 dev_info->num_shader_visible_vgprs = adev->gfx.config.max_gprs;
817 dev_info->num_cu_per_sh = adev->gfx.config.max_cu_per_sh;
818 dev_info->num_tcc_blocks = adev->gfx.config.max_texture_channel_caches;
819 dev_info->gs_vgt_table_depth = adev->gfx.config.gs_vgt_table_depth;
820 dev_info->gs_prim_buffer_depth = adev->gfx.config.gs_prim_buffer_depth;
821 dev_info->max_gs_waves_per_vgt = adev->gfx.config.max_gs_threads;
822
823 if (adev->family >= AMDGPU_FAMILY_NV)
824 dev_info->pa_sc_tile_steering_override =
825 adev->gfx.config.pa_sc_tile_steering_override;
826
827 dev_info->tcc_disabled_mask = adev->gfx.config.tcc_disabled_mask;
828
829 ret = copy_to_user(out, dev_info,
830 min((size_t)size, sizeof(*dev_info))) ? -EFAULT : 0;
831 kfree(dev_info);
832 return ret;
833 }
834 case AMDGPU_INFO_VCE_CLOCK_TABLE: {
835 unsigned int i;
836 struct drm_amdgpu_info_vce_clock_table vce_clk_table = {};
837 struct amd_vce_state *vce_state;
838
839 for (i = 0; i < AMDGPU_VCE_CLOCK_TABLE_ENTRIES; i++) {
840 vce_state = amdgpu_dpm_get_vce_clock_state(adev, i);
841 if (vce_state) {
842 vce_clk_table.entries[i].sclk = vce_state->sclk;
843 vce_clk_table.entries[i].mclk = vce_state->mclk;
844 vce_clk_table.entries[i].eclk = vce_state->evclk;
845 vce_clk_table.num_valid_entries++;
846 }
847 }
848
849 return copy_to_user(out, &vce_clk_table,
850 min((size_t)size, sizeof(vce_clk_table))) ? -EFAULT : 0;
851 }
852 case AMDGPU_INFO_VBIOS: {
853 uint32_t bios_size = adev->bios_size;
854
855 switch (info->vbios_info.type) {
856 case AMDGPU_INFO_VBIOS_SIZE:
857 return copy_to_user(out, &bios_size,
858 min((size_t)size, sizeof(bios_size)))
859 ? -EFAULT : 0;
860 case AMDGPU_INFO_VBIOS_IMAGE: {
861 uint8_t *bios;
862 uint32_t bios_offset = info->vbios_info.offset;
863
864 if (bios_offset >= bios_size)
865 return -EINVAL;
866
867 bios = adev->bios + bios_offset;
868 return copy_to_user(out, bios,
869 min((size_t)size, (size_t)(bios_size - bios_offset)))
870 ? -EFAULT : 0;
871 }
872 case AMDGPU_INFO_VBIOS_INFO: {
873 struct drm_amdgpu_info_vbios vbios_info = {};
874 struct atom_context *atom_context;
875
876 atom_context = adev->mode_info.atom_context;
877 memcpy(vbios_info.name, atom_context->name, sizeof(atom_context->name));
878 memcpy(vbios_info.vbios_pn, atom_context->vbios_pn, sizeof(atom_context->vbios_pn));
879 vbios_info.version = atom_context->version;
880 memcpy(vbios_info.vbios_ver_str, atom_context->vbios_ver_str,
881 sizeof(atom_context->vbios_ver_str));
882 memcpy(vbios_info.date, atom_context->date, sizeof(atom_context->date));
883
884 return copy_to_user(out, &vbios_info,
885 min((size_t)size, sizeof(vbios_info))) ? -EFAULT : 0;
886 }
887 default:
888 DRM_DEBUG_KMS("Invalid request %d\n",
889 info->vbios_info.type);
890 return -EINVAL;
891 }
892 }
893 case AMDGPU_INFO_NUM_HANDLES: {
894 struct drm_amdgpu_info_num_handles handle;
895
896 switch (info->query_hw_ip.type) {
897 case AMDGPU_HW_IP_UVD:
898 /* Starting Polaris, we support unlimited UVD handles */
899 if (adev->asic_type < CHIP_POLARIS10) {
900 handle.uvd_max_handles = adev->uvd.max_handles;
901 handle.uvd_used_handles = amdgpu_uvd_used_handles(adev);
902
903 return copy_to_user(out, &handle,
904 min((size_t)size, sizeof(handle))) ? -EFAULT : 0;
905 } else {
906 return -ENODATA;
907 }
908
909 break;
910 default:
911 return -EINVAL;
912 }
913 }
914 case AMDGPU_INFO_SENSOR: {
915 if (!adev->pm.dpm_enabled)
916 return -ENOENT;
917
918 switch (info->sensor_info.type) {
919 case AMDGPU_INFO_SENSOR_GFX_SCLK:
920 /* get sclk in Mhz */
921 if (amdgpu_dpm_read_sensor(adev,
922 AMDGPU_PP_SENSOR_GFX_SCLK,
923 (void *)&ui32, &ui32_size)) {
924 return -EINVAL;
925 }
926 ui32 /= 100;
927 break;
928 case AMDGPU_INFO_SENSOR_GFX_MCLK:
929 /* get mclk in Mhz */
930 if (amdgpu_dpm_read_sensor(adev,
931 AMDGPU_PP_SENSOR_GFX_MCLK,
932 (void *)&ui32, &ui32_size)) {
933 return -EINVAL;
934 }
935 ui32 /= 100;
936 break;
937 case AMDGPU_INFO_SENSOR_GPU_TEMP:
938 /* get temperature in millidegrees C */
939 if (amdgpu_dpm_read_sensor(adev,
940 AMDGPU_PP_SENSOR_GPU_TEMP,
941 (void *)&ui32, &ui32_size)) {
942 return -EINVAL;
943 }
944 break;
945 case AMDGPU_INFO_SENSOR_GPU_LOAD:
946 /* get GPU load */
947 if (amdgpu_dpm_read_sensor(adev,
948 AMDGPU_PP_SENSOR_GPU_LOAD,
949 (void *)&ui32, &ui32_size)) {
950 return -EINVAL;
951 }
952 break;
953 case AMDGPU_INFO_SENSOR_GPU_AVG_POWER:
954 /* get average GPU power */
955 if (amdgpu_dpm_read_sensor(adev,
956 AMDGPU_PP_SENSOR_GPU_POWER,
957 (void *)&ui32, &ui32_size)) {
958 return -EINVAL;
959 }
960 ui32 >>= 8;
961 break;
962 case AMDGPU_INFO_SENSOR_VDDNB:
963 /* get VDDNB in millivolts */
964 if (amdgpu_dpm_read_sensor(adev,
965 AMDGPU_PP_SENSOR_VDDNB,
966 (void *)&ui32, &ui32_size)) {
967 return -EINVAL;
968 }
969 break;
970 case AMDGPU_INFO_SENSOR_VDDGFX:
971 /* get VDDGFX in millivolts */
972 if (amdgpu_dpm_read_sensor(adev,
973 AMDGPU_PP_SENSOR_VDDGFX,
974 (void *)&ui32, &ui32_size)) {
975 return -EINVAL;
976 }
977 break;
978 case AMDGPU_INFO_SENSOR_STABLE_PSTATE_GFX_SCLK:
979 /* get stable pstate sclk in Mhz */
980 if (amdgpu_dpm_read_sensor(adev,
981 AMDGPU_PP_SENSOR_STABLE_PSTATE_SCLK,
982 (void *)&ui32, &ui32_size)) {
983 return -EINVAL;
984 }
985 ui32 /= 100;
986 break;
987 case AMDGPU_INFO_SENSOR_STABLE_PSTATE_GFX_MCLK:
988 /* get stable pstate mclk in Mhz */
989 if (amdgpu_dpm_read_sensor(adev,
990 AMDGPU_PP_SENSOR_STABLE_PSTATE_MCLK,
991 (void *)&ui32, &ui32_size)) {
992 return -EINVAL;
993 }
994 ui32 /= 100;
995 break;
996 default:
997 DRM_DEBUG_KMS("Invalid request %d\n",
998 info->sensor_info.type);
999 return -EINVAL;
1000 }
1001 return copy_to_user(out, &ui32, min(size, 4u)) ? -EFAULT : 0;
1002 }
1003 case AMDGPU_INFO_VRAM_LOST_COUNTER:
1004 ui32 = atomic_read(&adev->vram_lost_counter);
1005 return copy_to_user(out, &ui32, min(size, 4u)) ? -EFAULT : 0;
1006 case AMDGPU_INFO_RAS_ENABLED_FEATURES: {
1007 struct amdgpu_ras *ras = amdgpu_ras_get_context(adev);
1008 uint64_t ras_mask;
1009
1010 if (!ras)
1011 return -EINVAL;
1012 ras_mask = (uint64_t)adev->ras_enabled << 32 | ras->features;
1013
1014 return copy_to_user(out, &ras_mask,
1015 min_t(u64, size, sizeof(ras_mask))) ?
1016 -EFAULT : 0;
1017 }
1018 case AMDGPU_INFO_VIDEO_CAPS: {
1019 const struct amdgpu_video_codecs *codecs;
1020 struct drm_amdgpu_info_video_caps *caps;
1021 int r;
1022
1023 switch (info->video_cap.type) {
1024 case AMDGPU_INFO_VIDEO_CAPS_DECODE:
1025 r = amdgpu_asic_query_video_codecs(adev, false, &codecs);
1026 if (r)
1027 return -EINVAL;
1028 break;
1029 case AMDGPU_INFO_VIDEO_CAPS_ENCODE:
1030 r = amdgpu_asic_query_video_codecs(adev, true, &codecs);
1031 if (r)
1032 return -EINVAL;
1033 break;
1034 default:
1035 DRM_DEBUG_KMS("Invalid request %d\n",
1036 info->video_cap.type);
1037 return -EINVAL;
1038 }
1039
1040 caps = kzalloc(sizeof(*caps), GFP_KERNEL);
1041 if (!caps)
1042 return -ENOMEM;
1043
1044 for (i = 0; i < codecs->codec_count; i++) {
1045 int idx = codecs->codec_array[i].codec_type;
1046
1047 switch (idx) {
1048 case AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG2:
1049 case AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4:
1050 case AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_VC1:
1051 case AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_MPEG4_AVC:
1052 case AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_HEVC:
1053 case AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_JPEG:
1054 case AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_VP9:
1055 case AMDGPU_INFO_VIDEO_CAPS_CODEC_IDX_AV1:
1056 caps->codec_info[idx].valid = 1;
1057 caps->codec_info[idx].max_width =
1058 codecs->codec_array[i].max_width;
1059 caps->codec_info[idx].max_height =
1060 codecs->codec_array[i].max_height;
1061 caps->codec_info[idx].max_pixels_per_frame =
1062 codecs->codec_array[i].max_pixels_per_frame;
1063 caps->codec_info[idx].max_level =
1064 codecs->codec_array[i].max_level;
1065 break;
1066 default:
1067 break;
1068 }
1069 }
1070 r = copy_to_user(out, caps,
1071 min((size_t)size, sizeof(*caps))) ? -EFAULT : 0;
1072 kfree(caps);
1073 return r;
1074 }
1075 default:
1076 DRM_DEBUG_KMS("Invalid request %d\n", info->query);
1077 return -EINVAL;
1078 }
1079 return 0;
1080 }
1081
--
0-DAY CI Kernel Test Service
https://01.org/lkp
More information about the amd-gfx
mailing list