aboutsummaryrefslogtreecommitdiffstats
path: root/drivers/gpu/drm/amd/amdkfd
AgeCommit message (Collapse)Author
2020-04-23drm/amdkfd: kfree the wrong pointerJack Zhang
[ Upstream commit 3148a6a0ef3cf93570f30a477292768f7eb5d3c3 ] Originally, it kfrees the wrong pointer for mem_obj. It would cause memory leak under stress test. Signed-off-by: Jack Zhang <Jack.Zhang1@amd.com> Acked-by: Nirmoy Das <nirmoy.das@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com> Signed-off-by: Sasha Levin <sashal@kernel.org>
2020-02-04drm/amdkfd: Fix a bug in SDMA RLC queue counting under HWS modeYong Zhao
The sdma_queue_count increment should be done before execute_queues_cpsch(), which calls pm_calc_rlib_size() where sdma_queue_count is used to calculate whether over_subscription is triggered. With the previous code, when a SDMA queue is created, compute_queue_count in pm_calc_rlib_size() is one more than the actual compute queue number, because the queue_count has been incremented while sdma_queue_count has not. This patch fixes that. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-16drm/amdkfd: Add a message when SW scheduler is usedYong Zhao
SW scheduler is previously called non HW scheduler, or non HWS. This message is useful when triaging issues from dmesg. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Acked-by: Huang Rui <ray.huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-16drm/amdkfd: use map_queues for hiq on gfx v10 as wellHuang Rui
To align with gfx v9, we use the map_queues packet to load hiq MQD. Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-16drm/amdkfd: use kiq to load the mqd of hiq queue for gfx v9 (v6)Aaron Liu
There is an issue that CP will check the HIQ queue to be configured and mapped with KIQ ring, otherwise, it will be unable to read back the secure buffer while the gfxoff is enabled even with trusted IP blocks. v1 -> v2: - Fix to remove surplus set_resources packets. - Fill the whole configuration in MQD. - Change the author as Aaron because he addressed the key point of this issue. - Add kiq ring lock. v2 -> v3: - Free the lock while in error return case. - Remove the programming only needed by the queue is unmapped. v3 -> v4: - Remove doorbell programming because it's used for restarting queue. - Remove CP scheduler programming because map_queue packet will handle this. v4 -> v5: - Remove cp_hqd_active because mec ucode will enable it while use map_queues. - Revise goto out_unlock. - Correct the right doorbell offset for HIQ that kfd driver assigned in the packet. v5 -> v6: - Merge Arcturus fix into this patch because it will get oops in Arcturus platform. Reported-by: Lisa Saturday <Lisa.Saturday@amd.com> Signed-off-by: Aaron Liu <aaron.liu@amd.com> Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-and-Tested-by: Aaron Liu <aaron.liu@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-16drm/amdgpu: GPU TLB flush API moved to amdgpu_amdkfdAlex Sierra
[Why] TLB flush method has been deprecated using kfd2kgd interface. This implementation is now on the amdgpu_amdkfd API. [How] TLB flush functions now implemented in amdgpu_amdkfd. Signed-off-by: Alex Sierra <alex.sierra@amd.com> Acked-by: Christian König <christian.koenig@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-09drm/amdkfd: Improve kfd_process lookup in kfd_ioctlFelix Kuehling
Use filep->private_data to store a pointer to the kfd_process data structure. Take an extra reference for that, which gets released in the kfd_release callback. Check that the process calling kfd_ioctl is the same that opened the file descriptor. Return -EBADF if it's not, so that this error can be distinguished in user mode. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: Philip Yang <Philip.Yang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdkfd: Avoid hanging hardware in stop_cpschFelix Kuehling
Don't use the HWS if it's known to be hanging. In a reset also don't try to destroy the HIQ because that may hang on SRIOV if the KIQ is unresponsive. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Tested-by: Emily Deng <Emily.Deng@amd.com> Reviewed-by: shaoyunl <shaoyun.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdkfd: Improve HWS hang detection and handlingFelix Kuehling
Move HWS hang detection into unmap_queues_cpsch to catch hangs in all cases. If this happens during a reset, don't schedule another reset because the reset already in progress is expected to take care of it. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Tested-by: Emily Deng <Emily.Deng@amd.com> Reviewed-by: shaoyunl <shaoyun.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdkfd: Remove unused variableFelix Kuehling
dqm->pipeline_mem wasn't used anywhere. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: shaoyunl <shaoyun.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2020-01-07drm/amdkfd: Fix permissions of hang_hwsFelix Kuehling
Reading from /sys/kernel/debug/kfd/hang_hws would cause a kernel oops because we didn't implement a read callback. Set the permission to write-only to prevent that. Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by: shaoyunl <shaoyun.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-19drm/amdkfd: expose num_cp_queues data field to topology node (v2)Huang Rui
Thunk driver would like to know the num_cp_queues data, however this data relied on different asic specific. So it's better to get it from kfd driver. v2: don't update name size. Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-19drm/amdkfd: expose num_sdma_queues_per_engine data field to topology node (v2)Huang Rui
Thunk driver would like to know the num_sdma_queues_per_engine data, however this data relied on different asic specific. So it's better to get it from kfd driver. v2: don't update the name size. Signed-off-by: Huang Rui <ray.huang@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-18drm/amdkfd: queue kfd interrupt work to different CPUPhilip Yang
Because queue_work schedule the work on the same CPU the interrupt handler is running, if there are many interrupts pending, it takes longer time for work queue to start, or even worse system will hang. v2: queue work to same NUMA node for better cache locality v3: handle cpumask_next wraparound case Signed-off-by: Philip Yang <Philip.Yang@amd.com> Reviewed-by: Eric Huang <JinhuiEric.Huang@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-17Merge tag 'drm-next-5.6-2019-12-11' of ↵Daniel Vetter
git://people.freedesktop.org/~agd5f/linux into drm-next drm-next-5.6-2019-12-11: amdgpu: - Add MST atomic routines - Add support for DMCUB (new helper microengine for displays) - Add OEM i2c support in DC - Use vstartup for vblank events on DCN - Simplify Kconfig for DC - Renoir fixes for DC - Clean up function pointers in DC - Initial support for HDCP 2.x - Misc code cleanups - GFX10 fixes - Rework JPEG engine handling for VCN - Add clock and power gating support for JPEG - BACO support for Arcturus - Cleanup PSP ring handling - Add framework for using BACO with runtime pm to save power - Move core pci state handling out of the driver for pm ops - Allow guest power control in 1 VF case with SR-IOV - SR-IOV fixes - RAS fixes - Support for power metrics on renoir - Golden settings updates for gfx10 - Enable gfxoff on supported navi10 skus - Update MAINTAINERS amdkfd: - Clean up generational gfx code - Fixes for gfx10 - DIQ fixes - Share more code with amdgpu radeon: - PPC DMA fix - Register checker fixes for r1xx/r2xx - Misc cleanups From: Alex Deucher <alexdeucher@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191211223020.7510-1-alexander.deucher@amd.com
2019-12-13Merge tag 'drm-fixes-5.5-2019-12-12' of ↵Dave Airlie
git://people.freedesktop.org/~agd5f/linux into drm-fixes drm-fixes-5.5-2019-12-12: amdgpu: - DC fixes for renoir - Gfx8 fence flush align with mesa - Power profile fix for arcturus - Freesync fix - DC I2c over aux fix - DC aux defer fix - GPU reset fix - GPUVM invalidation semaphore fixes for PCO and SR-IOV - Golden settings updates for gfx10 Signed-off-by: Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexdeucher@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191212223211.8034-1-alexander.deucher@amd.com
2019-12-11drm/amdgpu: fix license on Kconfig and MakefilesAlex Deucher
amdgpu is MIT licensed. Fixes: ec8f24b7faaf3d ("treewide: Add SPDX license identifier - Makefile/Kconfig") Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-11drm/amdgpu: fix license on Kconfig and MakefilesAlex Deucher
amdgpu is MIT licensed. Fixes: ec8f24b7faaf3d ("treewide: Add SPDX license identifier - Makefile/Kconfig") Reviewed-by: Christian König <christian.koenig@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-06Merge tag 'drm-next-2019-12-06' of git://anongit.freedesktop.org/drm/drmLinus Torvalds
Pull more drm updates from Dave Airlie: "Rob pointed out I missed his pull request for msm-next, it's been in next for a while outside of my tree so shouldn't cause any unexpected issues, it has some OCMEM support in drivers/soc that is acked by other maintainers as it's outside my tree. Otherwise it's a usual fixes pull, i915, amdgpu, the main ones, with some tegra, omap, mgag200 and one core fix. Summary: msm-next: - OCMEM support for a3xx and a4xx GPUs. - a510 support + display support core: - mst payload deletion fix i915: - uapi alignment fix - fix for power usage regression due to security fixes - change default preemption timeout to 640ms from 100ms - EHL voltage level display fixes - TGL DGL PHY fix - gvt - MI_ATOMIC cmd parser fix, CFL non-priv warning - CI spotted deadlock fix - EHL port D programming fix amdgpu: - VRAM lost fixes on BACO for CI/VI - navi14 DC fixes - misc SR-IOV, gfx10 fixes - XGMI fixes for arcturus - SRIOV fixes amdkfd: - KFD on ppc64le enabled - page table optimisations radeon: - fix for r1xx/2xx register checker. tegra: - displayport regression fixes - DMA API regression fixes mgag200: - fix devices that can't scanout except at 0 addr omap: - fix dma_addr refcounting" * tag 'drm-next-2019-12-06' of git://anongit.freedesktop.org/drm/drm: (100 commits) drm/dp_mst: Correct the bug in drm_dp_update_payload_part1() drm/omap: fix dma_addr refcounting drm/tegra: Run hub cleanup on ->remove() drm/tegra: sor: Make the +5V HDMI supply optional drm/tegra: Silence expected errors on IOMMU attach drm/tegra: vic: Export module device table drm/tegra: sor: Implement system suspend/resume drm/tegra: Use proper IOVA address for cursor image drm/tegra: gem: Remove premature import restrictions drm/tegra: gem: Properly pin imported buffers drm/tegra: hub: Remove bogus connection mutex check ia64: agp: Replace empty define with do while agp: Add bridge parameter documentation agp: remove unused variable num_segments agp: move AGPGART_MINOR to include/linux/miscdevice.h agp: remove unused variable size in agp_generic_create_gatt_table drm/dp_mst: Fix build on systems with STACKTRACE_SUPPORT=n drm/radeon: fix r1xx/r2xx register checker for POT textures drm/amdgpu: fix GFX10 missing CSIB set(v3) drm/amdgpu: should stop GFX ring in hw_fini ...
2019-12-05drm/amdkfd: Eliminate unnecessary kernel queue function pointersYong Zhao
Up to this point, those functions are all the same for all ASICs, so no need to call them by functions pointers. Removing the function pointers will greatly increase the code readablity. If there is ever need for those function pointers, we can add it back then. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-12-01Merge tag 'compat-ioctl-5.5' of ↵Linus Torvalds
git://git.kernel.org:/pub/scm/linux/kernel/git/arnd/playground Pull removal of most of fs/compat_ioctl.c from Arnd Bergmann: "As part of the cleanup of some remaining y2038 issues, I came to fs/compat_ioctl.c, which still has a couple of commands that need support for time64_t. In completely unrelated work, I spent time on cleaning up parts of this file in the past, moving things out into drivers instead. After Al Viro reviewed an earlier version of this series and did a lot more of that cleanup, I decided to try to completely eliminate the rest of it and move it all into drivers. This series incorporates some of Al's work and many patches of my own, but in the end stops short of actually removing the last part, which is the scsi ioctl handlers. I have patches for those as well, but they need more testing or possibly a rewrite" * tag 'compat-ioctl-5.5' of git://git.kernel.org:/pub/scm/linux/kernel/git/arnd/playground: (42 commits) scsi: sd: enable compat ioctls for sed-opal pktcdvd: add compat_ioctl handler compat_ioctl: move SG_GET_REQUEST_TABLE handling compat_ioctl: ppp: move simple commands into ppp_generic.c compat_ioctl: handle PPPIOCGIDLE for 64-bit time_t compat_ioctl: move PPPIOCSCOMPRESS to ppp_generic compat_ioctl: unify copy-in of ppp filters tty: handle compat PPP ioctls compat_ioctl: move SIOCOUTQ out of compat_ioctl.c compat_ioctl: handle SIOCOUTQNSD af_unix: add compat_ioctl support compat_ioctl: reimplement SG_IO handling compat_ioctl: move WDIOC handling into wdt drivers fs: compat_ioctl: move FITRIM emulation into file systems gfs2: add compat_ioctl support compat_ioctl: remove unused convert_in_user macro compat_ioctl: remove last RAID handling code compat_ioctl: remove /dev/raw ioctl translation compat_ioctl: remove PCI ioctl translation compat_ioctl: remove joystick ioctl translation ...
2019-11-26amdgpu: Enable KFD on POWER systemsTimothy Pearson
KFD has been verified to function on POWER systems (Talos II / Vega 64). It should be available as a kernel configuration option on these systems. Signed-off-by: Timothy Pearson <tpearson@raptorengineering.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-26amdgpu: Enable KFD on POWER systemsTimothy Pearson
KFD has been verified to function on POWER systems (Talos II / Vega 64). It should be available as a kernel configuration option on these systems. Signed-off-by: Timothy Pearson <tpearson@raptorengineering.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-22drm/amdkfd: Remove duplicate functions update_mqd_hiq()Yong Zhao
The functions are the same as update_mqd(). Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Zhan Liu <zhan.liu@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-22drm/amdkfd: DIQ should not use HIQ way to allocate memoryYong Zhao
In the mqd_diq_sdma buffer, there should be only one HIQ mqd. All DIQs should be allocated somewhere else using the regular way. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Oak Zeng <Oak.Zeng@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-22drm/amdkfd: Delete KFD_MQD_TYPE_COMPUTEYong Zhao
It is the same as KFD_MQD_TYPE_CP, so delete it. As a result, we will have one less mqd mananger per device. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-19drm/amdkfd: remove set but not used variable 'top_dev'zhengbin
Fixes gcc '-Wunused-but-set-variable' warning: drivers/gpu/drm/amd/amdkfd/kfd_iommu.c: In function kfd_iommu_device_init: drivers/gpu/drm/amd/amdkfd/kfd_iommu.c:65:30: warning: variable top_dev set but not used [-Wunused-but-set-variable] Reported-by: Hulk Robot <hulkci@huawei.com> Fixes: 1ae99eab34f9 ("drm/amdkfd: Initialize HSA_CAP_ATS_PRESENT capability in topology codes") Signed-off-by: zhengbin <zhengbin13@huawei.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-19drm/amdkfd: Rename kfd_kernel_queue_*.c to kfd_packet_manager_*.cYong Zhao
After the recent cleanup, the functionalities provided by the previous kfd_kernel_queue_*.c are actually all packet manager related. So rename them to reflect that. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-19drm/amdkfd: Eliminate ops_asic_specific in kernel queueYong Zhao
The ops_asic_specific function pointers are actually quite generic after using a simple if condition. Eliminate it by code refactoring. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-19drm/amdkfd: Merge CIK kernel queue functions into VIYong Zhao
The only difference that CIK kernel queue functions are different from VI is avoid allocating eop_mem. We can achieve that by using a if condition. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdgpu: remove set but not used variable 'count'yu kuai
Fixes gcc '-Wunused-but-set-variable' warning: drivers/gpu/drm/amd/amdkfd/kfd_device.c: In function ‘kgd2kfd_post_reset’: drivers/gpu/drm/amd/amdkfd/kfd_device.c:745:11: warning: variable ‘count’ set but not used [-Wunused-but-set-variable] 'count' is never used, so can be removed. Thus 'atomic_dec_return' can be replaced as 'atomic_dec' Fixes: e42051d2133b ("drm/amdkfd: Implement GPU reset handlers in KFD") Signed-off-by: yu kuai <yukuai3@huawei.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Stop using GFP_NOIO explicitly for two placesYong Zhao
Adapt the change from: 1cd106ecfc1f04 ("drm/amdkfd: Stop using GFP_NOIO explicitly") Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Use QUEUE_IS_ACTIVE macro in mqd v10Yong Zhao
This is done for other GFX in commit bb2d2128a54c4. Port it to GFX10. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Fix a bug when calculating save_area_used_sizeYong Zhao
workgroup context data writes from m->cp_hqd_cntl_stack_size, so we should deduct it when calculating the used size. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Update get_wave_state() for GFX10Yong Zhao
Given control stack is now in the userspace context save restore area on GFX10, the same as GFX8, it is not needed to copy it back to userspace. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Implement queue priority controls for gfx10Yong Zhao
Ported from gfx9. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Rename create_cp_queue() to init_user_queue()Yong Zhao
create_cp_queue() could also work with SDMA queues, so we should rename it. It only initialize the data values rather than creating queues. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Avoid using doorbell_off as offset in process doorbell pagesYong Zhao
dorbell_off in the queue properties is mainly used for the doorbell dw offset in pci bar. We should not set it to the doorbell byte offset in process doorbell pages. This makes the code much easier to read. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Use better name to indicate the offset is in dwordsYong Zhao
The doorbell offset could mean the byte offset or the dword offset, and the 0 offset place is also different, sometimes the start of PCI doorbell bar or the start of process doorbell pages. Use better name to avoid confusion. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Simplify the mmap offset related bit operationsYong Zhao
The new code uses straightforward bit shifts and thus has better readability. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Use kernel queue v9 functions for v10Yong Zhao
The kernel queue functions for v9 and v10 are the same except pm_map_process_v* which have small difference, so they should be reused. This eliminates the need of reapplying several patches which were applied on v9 but not on v10, such as bigger GWS and more than 2 SDMA engine support which were introduced on Arcturus. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Only keep release_mem function for HawaiiYong Zhao
release_mem is only used for Hawaii, but because GFX7 and GFX8 share the same function pointer structure, so we only delete release_mem for GFX9 and GFX10. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-13drm/amdkfd: Adjust function sequences to avoid unnecessary declarationsYong Zhao
This is cleaner. Signed-off-by: Yong Zhao <Yong.Zhao@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-11-04Merge tag 'drm-next-5.5-2019-11-01' of ↵Dave Airlie
git://people.freedesktop.org/~agd5f/linux into drm-next drm-next-5.5-2019-11-01: amdgpu: - Add EEPROM support for Arcturus - Enable VCN encode support for Arcturus - Misc PSP fixes - Misc DC fixes - swSMU cleanup amdkfd: - Misc cleanups - Fix typo in cu bitmap parsing Signed-off-by: Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexdeucher@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191101190607.3763-1-alexander.deucher@amd.com
2019-10-30drm/amdkfd: bug fix for out of bounds mem on gpu cache filling infoAlex Sierra
The bitmap in cu_info structure is defined as a 4x4 size array. In Acturus, this matrix is initialized as a 4x2. Based on the 8 shaders. In the gpu cache filling initialization, the access to the bitmap matrix was done as an 8x1 instead of 4x2. Causing an out of bounds memory access error. Due to this, the number of GPU cache entries was inconsistent. Signed-off-by: Alex Sierra <alex.sierra@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-10-25drm/amdkfd: don't use dqm lock during device reset/suspend/resumePhilip Yang
If device reset/suspend/resume failed for some reason, dqm lock is hold forever and this causes deadlock. Below is a kernel backtrace when application open kfd after suspend/resume failed. Instead of holding dqm lock in pre_reset and releasing dqm lock in post_reset, add dqm->sched_running flag which is modified in dqm->ops.start and dqm->ops.stop. The flag doesn't need lock protection because write/read are all inside dqm lock. For HWS case, map_queues_cpsch and unmap_queues_cpsch checks sched_running flag before sending the updated runlist. v2: For no-HWS case, when device is stopped, don't call load/destroy_mqd for eviction, restore and create queue, and avoid debugfs dump hdqs. Backtrace of dqm lock deadlock: [Thu Oct 17 16:43:37 2019] INFO: task rocminfo:3024 blocked for more than 120 seconds. [Thu Oct 17 16:43:37 2019] Not tainted 5.0.0-rc1-kfd-compute-rocm-dkms-no-npi-1131 #1 [Thu Oct 17 16:43:37 2019] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. [Thu Oct 17 16:43:37 2019] rocminfo D 0 3024 2947 0x80000000 [Thu Oct 17 16:43:37 2019] Call Trace: [Thu Oct 17 16:43:37 2019] ? __schedule+0x3d9/0x8a0 [Thu Oct 17 16:43:37 2019] schedule+0x32/0x70 [Thu Oct 17 16:43:37 2019] schedule_preempt_disabled+0xa/0x10 [Thu Oct 17 16:43:37 2019] __mutex_lock.isra.9+0x1e3/0x4e0 [Thu Oct 17 16:43:37 2019] ? __call_srcu+0x264/0x3b0 [Thu Oct 17 16:43:37 2019] ? process_termination_cpsch+0x24/0x2f0 [amdgpu] [Thu Oct 17 16:43:37 2019] process_termination_cpsch+0x24/0x2f0 [amdgpu] [Thu Oct 17 16:43:37 2019] kfd_process_dequeue_from_all_devices+0x42/0x60 [amdgpu] [Thu Oct 17 16:43:37 2019] kfd_process_notifier_release+0x1be/0x220 [amdgpu] [Thu Oct 17 16:43:37 2019] __mmu_notifier_release+0x3e/0xc0 [Thu Oct 17 16:43:37 2019] exit_mmap+0x160/0x1a0 [Thu Oct 17 16:43:37 2019] ? __handle_mm_fault+0xba3/0x1200 [Thu Oct 17 16:43:37 2019] ? exit_robust_list+0x5a/0x110 [Thu Oct 17 16:43:37 2019] mmput+0x4a/0x120 [Thu Oct 17 16:43:37 2019] do_exit+0x284/0xb20 [Thu Oct 17 16:43:37 2019] ? handle_mm_fault+0xfa/0x200 [Thu Oct 17 16:43:37 2019] do_group_exit+0x3a/0xa0 [Thu Oct 17 16:43:37 2019] __x64_sys_exit_group+0x14/0x20 [Thu Oct 17 16:43:37 2019] do_syscall_64+0x4f/0x100 [Thu Oct 17 16:43:37 2019] entry_SYSCALL_64_after_hwframe+0x44/0xa9 Suggested-by: Felix Kuehling <Felix.Kuehling@amd.com> 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>
2019-10-26Merge tag 'drm-next-5.5-2019-10-09' of ↵Dave Airlie
git://people.freedesktop.org/~agd5f/linux into drm-next drm-next-5.5-2019-10-09: amdgpu: - Additional RAS enablement for vega20 - RAS page retirement and bad page storage in EEPROM - No GPU reset with unrecoverable RAS errors - Reserve vram for page tables rather than trying to evict - Fix issues with GPU reset and xgmi hives - DC i2c over aux fixes - Direct submission for clears, PTE/PDE updates - Improvements to help support recoverable GPU page faults - Silence harmless SAD block messages - Clean up code for creating a bo at a fixed location - Initial DC HDCP support - Lots of documentation fixes - GPU reset for renoir - Add IH clockgating support for soc15 asics - Powerplay improvements - DC MST cleanups - Add support for MSI-X - Misc cleanups and bug fixes amdkfd: - Query KFD device info by asic type rather than pci ids - Add navi14 support - Add renoir support - Add navi12 support - gfx10 trap handler improvements - pasid cleanups - Check against device cgroup ttm: - Return -EBUSY with pipelining with no_gpu_wait radeon: - Silence harmless SAD block messages device_cgroup: - Export devcgroup_check_permission Signed-off-by: Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexdeucher@gmail.com> Link: https://patchwork.freedesktop.org/patch/msgid/20191010041713.3412-1-alexander.deucher@amd.com
2019-10-23compat_ioctl: move more drivers to compat_ptr_ioctlArnd Bergmann
The .ioctl and .compat_ioctl file operations have the same prototype so they can both point to the same function, which works great almost all the time when all the commands are compatible. One exception is the s390 architecture, where a compat pointer is only 31 bit wide, and converting it into a 64-bit pointer requires calling compat_ptr(). Most drivers here will never run in s390, but since we now have a generic helper for it, it's easy enough to use it consistently. I double-checked all these drivers to ensure that all ioctl arguments are used as pointers or are ignored, but are not interpreted as integer values. Acked-by: Jason Gunthorpe <jgg@mellanox.com> Acked-by: Daniel Vetter <daniel.vetter@ffwll.ch> Acked-by: Mauro Carvalho Chehab <mchehab+samsung@kernel.org> Acked-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Acked-by: David Sterba <dsterba@suse.com> Acked-by: Darren Hart (VMware) <dvhart@infradead.org> Acked-by: Jonathan Cameron <Jonathan.Cameron@huawei.com> Acked-by: Bjorn Andersson <bjorn.andersson@linaro.org> Acked-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Arnd Bergmann <arnd@arndb.de>
2019-10-09drm/amdkfd: update for drmP.h removalStephen Rothwell
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
2019-10-07drm/amdkfd: Check against device cgroupHarish Kasiviswanathan
Participate in device cgroup. All kfd devices are exposed via /dev/kfd. So use /dev/dri/renderN node. Before exposing the device to a task check if it has permission to access it. If the task (based on its cgroup) can access /dev/dri/renderN then expose the device via kfd node. If the task cannot access /dev/dri/renderN then process device data (pdd) is not created. This will ensure that task cannot use the device. In sysfs topology, all device nodes are visible irrespective of the task cgroup. The sysfs node directories are created at driver load time and cannot be changed dynamically. However, access to information inside nodes is controlled based on the task's cgroup permissions. Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com> Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com> Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
generated by cgit 1.2.3-korg (git 2.39.0) at 2024-05-23 07:08:09 +0000