aboutsummaryrefslogtreecommitdiff
path: root/drivers/gpu/drm/amd/amdgpu/amdgpu_device.c
AgeCommit message (Collapse)AuthorFilesLines
2018-03-01drm/amdgpu: cleanups for vram lost handlingMonk Liu1-65/+72
1)create a routine "handle_vram_lost" to do the vram recovery, and put it into amdgpu_device_reset/reset_sriov, this way no need of the extra paramter to hold the VRAM LOST information and the related macros can be removed. 3)show vram_recover failure if time out, and set TMO equal to lockup_timeout if vram_recover is under SRIOV runtime mode. 4)report error if any ip reset failed for SR-IOV Signed-off-by: Monk Liu <[email protected]> Acked-by: Christian König <[email protected]> Acked-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-03-01drm/amdgpu: stop all rings before doing gpu recoverMonk Liu1-25/+15
found recover_vram_from_shadow sometimes get executed in paralle with SDMA scheduler, should stop all schedulers before doing gpu reset/recover Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Tested-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-28drm/amdgpu: move WB_FREE to correct placeMonk Liu1-5/+7
WB_FREE should be put after all engines's hw_fini done, otherwise the invalid wptr/rptr_addr would still be used by engines which trigger abnormal bugs. This fixes couple DMAR reading error in host side for SRIOV after guest kmd is unloaded. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-28drm/amdgpu: fix&cleanups for wb_clearMonk Liu1-2/+3
fix: should do right shift on wb before clearing cleanups: 1,should memset all wb buffer 2,set max wb number to 128 (total 4KB) is big enough Signed-off-by: Monk Liu <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: Unify the dm resume calls into oneMikita Lipski1-9/+0
amdgpu_dm_display_resume is now called from dm_resume to unify DAL resume call into a single function call There is no more need to separately call 2 resume functions for DM. Initially they were separated to resume display state after cursor is pinned. But because there is no longer any corruption with the cursor - the calls can be merged into one function hook. Signed-off-by: Mikita Lipski <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Reviewed-by: Andrey Grodzovsky <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: Add place holder for soc15 asic init on emulationShaoyun Liu1-14/+4
Add common smu_soc_asic_init function to emulate the sillicon post sequence Signed-off-by: Shaoyun Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: Double the timeout count on emulation modeShaoyun Liu1-0/+2
Signed-off-by: Shaoyun Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: Fix none-powerplay issue when load driver on emulation modeShaoyun Liu1-0/+3
On emulation mode , driver will be loaded with powerplay disabled Signed-off-by: Shaoyun Liu <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: Basic emulation supportShaoyun Liu1-3/+23
Add amdgpu_emu_mode module parameter to control the emulation mode Avoid vbios operation on emulation since there is no vbios post duirng emulation, use the common hw_init to simulate the post Signed-off-by: Shaoyun Liu <[email protected]> Acked-By: Alex Deucher <[email protected]> Acked-By: Christian Konig <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: Enable ip block bit mask print out info by defaultShaoyun Liu1-1/+1
Signed-off-by: Shaoyun Liu <[email protected]> Reviewed-by: Christian Konig <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: remove DC special casing for KB/MLAlex Deucher1-3/+2
It seems to be working now. Bug: https://bugs.freedesktop.org/show_bug.cgi?id=102372 Reviewed-by: Mike Lothian <[email protected]> Reviewed-by: Harry Wentland <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: drop the drm irq pre/post/un install callbacksAlex Deucher1-0/+3
The preinstall callback didn't do anything because not all of the IPs were initialized when it was called. Move the postinstall setup into sequence in the driver. The uninstall callback disabled all interrupt source, but it got called too late in the driver sequence and caused problems with IPs who already freed the relevant data structures. Move the call into the right place in the driver sequence. Acked-by: Christian König <[email protected]> Acked-by: Harry Wentland <[email protected]> Tested-By: Mikita Lipski <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: move struct gart_funcs into amdgpu_gmc.hChristian König1-1/+1
And rename it to struct gmc_funcs. Signed-off-by: Christian König <[email protected]> Reviewed-by: Samuel Li <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-02-19drm/amdgpu: move struct amdgpu_mc into amdgpu_gmc.hChristian König1-5/+5
And rename it to amdgpu_gmc as well. Signed-off-by: Christian König <[email protected]> Reviewed-by: Samuel Li <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-01-23drm/amdgpu: Avoid leaking PM domain on driver unbind (v2)Alex Deucher1-2/+0
We only support vga_switcheroo and runtime pm on PX/HG systems so forcing runpm to 1 doesn't do anything useful anyway. Only call vga_switcheroo_init_domain_pm_ops() for PX/HG so that the cleanup path is correct as well. This mirrors what radeon does as well. v2: rework the patch originally sent by Lukas (Alex) Acked-by: Lukas Wunner <[email protected]> Reported-by: Lukas Wunner <[email protected]> Signed-off-by: Alex Deucher <[email protected]> Signed-off-by: Lukas Wunner <[email protected]> (v1) Cc: [email protected]
2018-01-23drm/amdgpu: Reenable manual GPU reset from sysfsAndrey Grodzovsky1-1/+1
Otherwise it keeps rejecting the reset. Signed-off-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2018-01-10drm/amdgpu: fix 64bit BAR detectionChristian König1-1/+1
Windows added by the BIOS are not marked as 64bit because they are usually not changeable anyway. This fixes large BAR support on my new Ryzen build system. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: rename amdgpu_get_pcie_infoAlex Deucher1-1/+1
add device to the name for consistency. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: move amdgpu_need_backup to amdgpu_object.cAlex Deucher1-8/+0
It's the only place it's used. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: rename amdgpu_gpu_recoverAlex Deucher1-2/+3
add device to the name for consistency. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: move dummy page functions to amdgpu_gart.cAlex Deucher1-45/+0
It's the only place they are used. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: rename amdgpu_need_postAlex Deucher1-4/+4
add device to the name for consistency. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: rename ip block helper functionsAlex Deucher1-24/+26
add device to the name for consistency. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: move fw_reserve functions to amdgpu_ttm.cAlex Deucher1-95/+0
It's the only place they are used. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: rename amdgpu_*_location functionsAlex Deucher1-4/+6
add device to the name for consistency. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: move amdgpu_doorbell_get_kfd_info to amdgpu_amdkfd.cAlex Deucher1-32/+1
It's the only place it's used. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: rename amdgpu_pci_config_resetAlex Deucher1-1/+1
add device for consistency with other functions in this file. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: rename amdgpu_program_register_sequenceAlex Deucher1-4/+4
add device for consistency with other functions in this file. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: rename amdgpu_wb_* functionsAlex Deucher1-4/+4
add device for consistency. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: move debugfs functions to their own fileAlex Deucher1-768/+1
amdgpu_device.c was getting pretty cluttered. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: rename amdgpu_suspend to amdgpu_device_ip_suspendAlex Deucher1-4/+4
for consistency with the other functions in that file. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: use consistent naming for static funcs in amdgpu_device.cAlex Deucher1-86/+90
Prefix the functions with device or device_ip for functions which deal with ip blocks for consistency. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-18drm/amdgpu: move atom functions from amdgpu_device.cAlex Deucher1-231/+0
and move them to amdgpu_atombios.c for consistency. Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-15drm/amdgpu: Simplify amdgpu_lockup_timeout usage.Andrey Grodzovsky1-1/+6
With introduction of amdgpu_gpu_recovery we don't need any more to rely on amdgpu_lockup_timeout == 0 for disabling GPU reset. Signed-off-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-15drm/amdgpu: Add gpu_recovery parameterAndrey Grodzovsky1-1/+8
Add new parameter to control GPU recovery procedure. v2: Add auto logic where reset is disabled for bare metal and enabled for SR-IOV. Allow forced reset from debugfs. Signed-off-by: Andrey Grodzovsky <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-13drm/amdgpu: drop scratch regs save and restore from GPU reset handlingAlex Deucher1-2/+0
The expectation is that the base driver doesn't mess with these. Some components interact with these directly so let the components handle these directly. Reviewed-by: Harry Wentland <[email protected]> Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-13drm/amdgpu: drop scratch regs save and restore from S3/S4 handlingAlex Deucher1-2/+0
The expectation is that the base driver doesn't mess with these. Some components interact with these directly so let the components handle these directly. Reviewed-by: Harry Wentland <[email protected]> Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-12drm/amdgpu: no need to evict VRAM in device_finiMonk Liu1-2/+1
this VRAM evict is not needed and also cost 2seconds to finish because the IRQ is software side disabled before it. Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-12drm/amdgpu: add amdgpu_evict_vram debugfs fileChristian König1-1/+12
Torture test for MM and VM support, can be used to evict all VRAM while the system is under load. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-12drm/amdgpu: cleanup debugfs handling a bitChristian König1-35/+11
Remove the superflous .debugfs_init callback and register all files in amdgpu_device.c in just one function. Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-07drm: move amd_gpu_scheduler into common locationLucas Stach1-2/+2
This moves and renames the AMDGPU scheduler to a common location in DRM in order to facilitate re-use by other drivers. This is mostly a straight forward rename with no code changes. One notable exception is the function to_drm_sched_fence(), which is no longer a inline header function to avoid the need to export the drm_sched_fence_ops_scheduled and drm_sched_fence_ops_finished structures. Reviewed-by: Chunming Zhou <[email protected]> Tested-by: Dieter Nützel <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Lucas Stach <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu: allow specifying vm_block_size for multi level PDs v2Christian König1-13/+1
This patch allows specifying the vm_block_size even when multi level page directories are active. v2: fix signed/unsigned compare warning Signed-off-by: Christian König <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu: move validation of the VM size into the VM codeChristian König1-15/+1
This moves validation of the VM size parameter into amdgpu_vm_adjust_size(). Signed-off-by: Christian König <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu: allow non pot VM size valuesChristian König1-6/+0
The VM size actually doesn't need to be a power of two. Signed-off-by: Christian König <[email protected]> Reviewed-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/ttm: use an operation context for ttm_bo_mem_space v2Christian König1-2/+3
Instead of specifying interruptible and no_wait_gpu manually. v2: rebase Signed-off-by: Christian König <[email protected]> Reviewed-by: Michel Dänzer <[email protected]> Reviewed-by: Chunming Zhou <[email protected]> Tested-by: Dieter Nützel <[email protected]> Tested-by: Michel Dänzer <[email protected]> Acked-by: Felix Kuehling <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu: align GTT start to 4GB v2Christian König1-1/+4
For VCE to work properly the start of the GTT space must be aligned to a 4GB boundary. v2: add comment why we do this Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu: remove VRAM size reduction v2Christian König1-29/+1
Remove some outdated comments and all code which tries to reduce the VRAM size mapped into the MC. This is superfluous and misleading since we never actually program the size. v2: handle gmc_v6_0.c as well Signed-off-by: Christian König <[email protected]> Reviewed-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu: require a root bus window above 4GB for BAR resizeChristian König1-0/+18
Don't even try to resize the BAR when there is no window above 4GB. Signed-off-by: Christian König <[email protected]> Acked-by: Alex Deucher <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu:show error message if fail on event4Monk Liu1-1/+2
Signed-off-by: Monk Liu <[email protected]> Acked-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
2017-12-06drm/amdgpu:free CSA in unified placeMonk Liu1-0/+1
instead of doing it in each GFX ip's sw_fini Signed-off-by: Monk Liu <[email protected]> Reviewed-by: Christian König <[email protected]> Signed-off-by: Alex Deucher <[email protected]>
Powered by cgit, Gruvboxed by Blaster4385.