- 16 Jul, 2018 1 commit
-
-
Felix Kuehling authored
This allows automatic switching to the compute power profile depending on compute activity. Signed-off-by:
Felix Kuehling <Felix.Kuehling@amd.com> Reviewed-by:
Eric Huang <JinHuiEric.Huang@amd.com> Reviewed-by:
Alex Deucher <alexander.deucher@amd.com> Signed-off-by:
Oded Gabbay <oded.gabbay@gmail.com>
-
- 14 Jul, 2018 2 commits
-
-
Felix Kuehling authored
CU-masking allows a KFD client to control the set of CUs used by a user mode queue for executing compute dispatches. This can be used for optimizing the partitioning of the GPU and minimize conflicts between concurrent tasks. Signed-off-by:
Flora Cui <flora.cui@amd.com> Signed-off-by:
Kent Russell <kent.russell@amd.com> Signed-off-by:
-
Felix Kuehling authored
Signed-off-by:
-
- 13 Jul, 2018 6 commits
-
-
Yong Zhao authored
Add DID and kfd_device_info for Raven. Signed-off-by:
Yong Zhao <Yong.Zhao@amd.com> Reviewed-by:
-
Yong Zhao authored
memory_exception_data is already initialized for not-present faults. It only needs to be overridden for permission faults. Signed-off-by:
-
Yong Zhao authored
On Raven multiple PPRs can be queued up by the hardware. When the first of those requests is handled by the IOMMU driver, the memory access succeeds. After that the application may be done with the memory and unmap it. At that point the page table entries are invalidated, but there are still outstanding duplicate PPRs for those addresses. When the IOMMU driver processes those duplicate requests, it finds invalid page table entries and triggers an invalid PPR fault. As a workaround, don't signal invalid PPR faults on Raven to avoid segfaulting applications that haven't done anything wrong. As a side effect, real GPU memory access faults may go unnoticed by the application. Signed-off-by:
-
Yong Zhao authored
On Raven Invalid PPRs (peripheral page requests) can be reported because multiple PPRs can be still queued when memory is freed. Apply a rate limit to avoid flooding the log in this case. Signed-off-by:
-
Yong Zhao authored
On Raven there is only one SDMA engine instead of previously assumed two, so we need to adapt our code to this new scenario. Signed-off-by:
-
Yong Zhao authored
If there are several memory banks that has the same properties in CRAT, we aggregate them into one memory bank. This cleans up memory banks on APUs (e.g. Raven) where the CRAT reports each memory channel as a separate bank. This only confuses user mode, which only deals with virtual memory. Signed-off-by:
-
- 12 Jul, 2018 25 commits
-
-
Yong Zhao authored
Signed-off-by:
Christian König <christian.koenig@amd.com> Signed-off-by:
-
Yong Zhao authored
This will make reading code much easier. Signed-off-by:
-
Yong Zhao authored
This makes all module parameters use the same form. Meanwhile clean up the surrounding code. Signed-off-by:
-
Yong Zhao authored
This avoids triggering a GPU reset or otherwise changing the HW state. Instead KFD will hang, which allows HW debugging tools to analyze the problem. Signed-off-by:
-
Shaoyun Liu authored
Signed-off-by:
Shaoyun Liu <Shaoyun.Liu@amd.com> Reviewed-by:
-
Shaoyun Liu authored
Signed-off-by:
-
Shaoyun Liu authored
Signed-off-by:
-
Shaoyun Liu authored
The bitmap index calculation should reverse the logic used on allocation so it will clear the same bit used on allocation Signed-off-by:
-
Shaoyun Liu authored
job could be NULL when amdgpu_device_gpu_recover is called Signed-off-by:
Andrey Grodzovsky <andrey.grodzovsky@amd.com> Signed-off-by:
-
Shaoyun Liu authored
Compute contexts cannot keep going after a GPU reset. Currently the process must terminate. In the future a process may be able recreate its context from scratch. Either way, there is no need to restore the GPUVM page table from shadow BOs. Signed-off-by:
-
Shaoyun Liu authored
The reset will be performed in a new hw_exception work thread to handle HWS hang without blocking the thread that detected the hang. Signed-off-by:
-
Shaoyun Liu authored
Hook up the gpu_recover callback from KFD to amdgpu to enable handling of GPU hangs detected by KFD. Signed-off-by:
-
Shaoyun Liu authored
Lock KFD and evict existing queues on reset. Notify user mode by signaling hw_exception events. Signed-off-by:
-
Shaoyun Liu authored
Signed-off-by:
-
Shaoyun Liu authored
Signed-off-by:
-
Shaoyun Liu authored
Signed-off-by:
-
Shaoyun Liu authored
Signed-off-by:
-
Lan Xiao authored
Upon VM Fault, the VMID and PASID written by HW are zeros in Hawaii. Instead of reading from ih_ring_entry, read directly from the registers. This workaround fix the soft hang issues caused by mishandled VM Fault in Hawaii. Signed-off-by:
Lan Xiao <Lan.Xiao@amd.com> Signed-off-by:
-
shaoyunl authored
1. Pre-GFX9 the amdgpu ISR saves the vm-fault status and address per per-vmid. amdkfd needs to get the information from amdgpu through the new get_vm_fault_info interface. On GFX9 and later, all the required information is in the IH ring 2. amdkfd unmaps all queues from the faulting process and create new run-list without the guilty process 3. amdkfd notifies the runtime of the vm fault trap via EVENT_TYPE_MEMORY Signed-off-by:
-
shaoyunl authored
amdgpu save the vm fault related information for KFD usage and keep the copy until KFD read it. Signed-off-by:
-
Moses Reuben authored
Signed-off-by:
Moses Reuben <moses.reuben@amd.com> Reviewed-by:
-
Wei Lu authored
Return ERR_PTR(-EINVAL) if kfd_get_process fails to find the process. This fixes kernel oopses when a child process calls KFD ioctls with a file descriptor inherited from the parent process. Signed-off-by:
Wei Lu <wei.lu2@amd.com> Reviewed-by:
-
Jay Cornwall authored
The scheduler may raise SQ_WAVE_STATUS.SPI_PRIO via SQ_CMD before context restore has completed. Restoring SPI_PRIO=0 after this point may cause context save to fail as the lower priority wavefronts are not selected for execution among spin-waiting wavefronts. Leave SPI_PRIO at its SPI-initialized or scheduler-raised value. v2: Also fix race with exception handler Signed-off-by:
Jay Cornwall <Jay.Cornwall@amd.com> Reviewed-by:
-
Felix Kuehling authored
This is no longer needed with the memalloc_nofs_save/restore in dqm_lock/unlock. Signed-off-by:
-
Felix Kuehling authored
This is needed to prevent deadlocks when MMU notifiers run in reclaim-FS context and take the DQM lock for userptr evictions. Previously this was done by making all memory allocations under DQM locks GFP_NOIO. This is error prone. Using memalloc_nofs_save/restore will reliably affect all memory allocations anywhere in the kernel while the DQM lock is held. Signed-off-by:
-
- 11 Jul, 2018 1 commit
-
-
Arnd Bergmann authored
getrawmonotonic64() and get_monotonic_boottime64() are deprecated because of the nonstandard naming. The replacement functions ktime_get_raw_ns() and ktime_get_boot_ns() also simplify the callers. Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>. Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
- 27 Jul, 2018 4 commits
-
-
git://people.freedesktop.org/~agd5f/linuxDave Airlie authored
Updates for 4.19. Mostly bug fixes and cleanups. Highlights: - Internal API cleanup in GPU scheduler - Decouple i2c and aux abstractions in DC - Update maintainers - Misc cleanups - Misc bug fixes Signed-off-by:
Dave Airlie <airlied@redhat.com> Link: https://patchwork.freedesktop.org/patch/msgid/20180725215326.2709-1-alexander.deucher@amd.com
-
Dave Airlie authored
Merge tag 'exynos-drm-next-for-v4.19' of git://git.kernel.org/pub/scm/linux/kernel/git/daeinki/drm-exynos into drm-next Cleanups - Change g2d driver to component based driver . g2d driver was last customed sub driver so this patch series changes it to component based driver, which also makes gem handling to be more simplify. - Cleanup of Exynos DRM suspend/resume . Register exynos drm core suspend/resume functions to prepare/complete callbacks of dev_pm_ops instead of suspend/resume callbacks to ensure exynos_drm_suspend() is called before any suspend callback from the real devices to avoid some issues on boards with complex pipelines. . Also Add pm_runtime_furce_suspend/resume as SYSTEM_SLEEP_PM_OPS to ensure that resources of each devices will be released for the system PM suspend/resume cycle. - Remove local value not used. Signed-off-by:
-
git://linux-arm.org/linux-ldDave Airlie authored
I have a couple of small patches for malidp to be applied in drm-next. They have arisen from the decision to switch the writeback connectors to always connected. Signed-off-by:
-
git://git.pengutronix.de/git/pza/linuxDave Airlie authored
drm/imx: cleanup and csi improvements - Remove the unused struct imx_drm_crtc and the unused pipes field from imx_drm_device and replace drm_dev_unref with drm_dev_put. - Extend CSI configuration to support RGB888 and BGR888 capture, as well as 16-bit RGB565 capture via a parallel bus. - Add CPMEM support for negative interlace offsets, which is necessary to support writing captured bottom-top interlaced fields to memory with interleaved lines. Signed-off-by:
-
- 25 Jul, 2018 1 commit
-
-
Jia-Ju Bai authored
cik_pcie_gen3_enable() is only called by cik_common_hw_init(), which is never called in atomic context. cik_pcie_gen3_enable() calls mdelay() to busily wait, which is not necessary. mdelay() can be replaced with msleep(). This is found by a static analysis tool named DCNS written by myself. Signed-off-by:
Jia-Ju Bai <baijiaju1990@gmail.com> Signed-off-by:
-
