kernel/linux.git/drivers/gpu/drm/amd/amdkfd, branch v5.4.113

drm/amdkfd: Fix leak in dmabuf import

2020-12-30T10:51:30+00:00

[ Upstream commit c897934da15f182ce99536007f8ef61c4748c07e ] Release dmabuf reference before returning from kfd_ioctl_import_dmabuf. amdgpu_amdkfd_gpuvm_import_dmabuf takes a reference to the underlying GEM BO and doesn't keep the reference to the dmabuf wrapper. Signed-off-by: Felix Kuehling Reviewed-by: Kent Russell Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: Use same SQ prefetch setting as amdgpu

2020-11-05T10:43:34+00:00

commit d56b1980d7efe9ef08469e856fc0703d0cef65e4 upstream. 0 causes instruction fetch stall at cache line boundary under some conditions on Navi10. A non-zero prefetch is the preferred default in any case. Fixes soft hang in Luxmark. Signed-off-by: Jay Cornwall Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

drm/amdkfd: fix a memory leak issue

2020-10-01T11:18:15+00:00

[ Upstream commit 087d764159996ae378b08c0fdd557537adfd6899 ] In the resume stage of GPU recovery, start_cpsch will call pm_init which set pm->allocated as false, cause the next pm_release_ib has no chance to release ib memory. Add pm_release_ib in stop_cpsch which will be called in the suspend stage of GPU recovery. Reviewed-by: Felix Kuehling Signed-off-by: Dennis Li Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: Fix race in gfx10 context restore handler

2020-10-01T11:17:11+00:00

[ Upstream commit c18cc2bb9e064d3a613d8276f2cab3984926a779 ] Missing synchronization with VGPR restore leads to intermittent VGPR trashing in the user shader. Signed-off-by: Jay Cornwall Reviewed-by: Yong Zhao Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: Fix reference count leaks.

2020-09-03T09:26:43+00:00

[ Upstream commit 20eca0123a35305e38b344d571cf32768854168c ] kobject_init_and_add() takes reference even when it fails. If this function returns an error, kobject_put() must be called to properly clean up the memory associated with the object. Signed-off-by: Qiushi Wu Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amd: fix potential memleak in err branch

2020-06-30T19:37:10+00:00

commit b5b78a6c8d8cb9c307bc6b16a754603424459d6e upstream. The function kobject_init_and_add alloc memory like: kobject_init_and_add->kobject_add_varg->kobject_set_name_vargs ->kvasprintf_const->kstrdup_const->kstrdup->kmalloc_track_caller ->kmalloc_slab, in err branch this memory not free. If use kmemleak, this path maybe catched. These changes are to add kobject_put in kobject_init_and_add failed branch, fix potential memleak. Signed-off-by: Bernard Zhao Reviewed-by: Felix Kuehling Signed-off-by: Felix Kuehling Signed-off-by: Alex Deucher Cc: stable@vger.kernel.org Signed-off-by: Greg Kroah-Hartman

drm/amdkfd: kfree the wrong pointer

2020-04-23T08:36:38+00:00

[ 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 Acked-by: Nirmoy Das Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: Fix permissions of hang_hws

2020-02-24T07:36:43+00:00

[ Upstream commit 2bdac179e217a0c0b548a8c60524977586621b19 ] 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 Reviewed-by: shaoyunl Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: Fix a bug in SDMA RLC queue counting under HWS mode

2020-02-24T07:36:32+00:00

[ Upstream commit f38abc15d157b7b31fa7f651dc8bf92858c963f8 ] 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 Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin

drm/amdkfd: Fix MQD size calculation

2019-12-31T15:43:47+00:00

[ Upstream commit 40a9592a26608e16f7545a068ea4165e1869f629 ] On device initialization, a chunk of GTT memory is pre-allocated for HIQ and all SDMA queues mqd. The size of this allocation was wrong. The correct sdma engine number should be PCIe-optimized SDMA engine number plus xgmi SDMA engine number. Reported-by: Jonathan Kim Signed-off-by: Jonathan Kim Signed-off-by: Oak Zeng Reviewed-by: Felix Kuehling Signed-off-by: Alex Deucher Signed-off-by: Sasha Levin