root cause that S2A need to use deduct offset flag.
after setting this flag, vcn0 doorbell value works.
so return it as before
Signed-off-by: Jane Jian <Jane.Jian@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The basic problem here is that it's not allowed to page fault while
holding the reservation lock.
So it can happen that multiple processes try to validate an userptr
at the same time.
Work around that by putting the HMM range object into the mutex
protected bo list for now.
v2: make sure range is set to NULL in case of an error
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
CC: stable@vger.kernel.org
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
drm-misc-fixes for v6.1-rc6:
- Fix error handling in vc4_atomic_commit_tail()
- Set bpc for logictechno panels.
- Fix potential memory leak in drm_dev_init()
- Fix potential null-ptr-deref in drm_vblank_destroy_worker()
- Set lima's clkname corrrectly when regulator is missing.
- Small amdgpu fix to gang submission.
- Revert hiding unregistered connectors from userspace, as it breaks on DP-MST.
- Add workaround for DP++ dual mode adaptors that don't support
i2c subaddressing.
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/c7d02936-c550-199b-6cb7-cbf6cf104e4a@linux.intel.com
If we enable virtual display functionality on parts with
no display hardware we can end up trying to check for and
reserve the vbios FB area on devices where it doesn't exist.
Check if display hardware is actually present on the hardware
before trying to reserve the memory.
v2: move the check into common code
Acked-by: Evan Quan <evan.quan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
It is to resolve a regression, which fails to allocate
VRAM due to no free memory in application, the reason
is we add check of vram_pin_size for memory limit, and
application is pinning the memory for Peerdirect, KFD
should not count it in memory limit. So removing
vram_pin_size will resolve it.
Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When p->gang_size equals 0, amdgpu_cs_pass1() will return directly
without freeing chunk_array, which will cause a memory leak issue,
this patch fixes it.
Fixes: 4624459c84 ("drm/amdgpu: add gang submit frontend v6")
Reviewed-by: Luben Tuikov <luben.tuikov@amd.com>
Signed-off-by: Dong Chenchen <dongchenchen2@huawei.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Re-take the eviction lock immediately again after the allocation is
completed, to fix circular locking warning with drm_buddy allocator.
Move amdgpu_vm_eviction_lock/unlock/trylock to amdgpu_vm.h as they are
called from multiple files.
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Get below kernel WARNING backtrace when pressing ctrl-C to kill kfdtest
application.
If amdgpu_cs_parser_bos returns error after taking bo_list_mutex, as
caller amdgpu_cs_ioctl will not unlock bo_list_mutex, this generates the
kernel WARNING.
Add unlock bo_list_mutex after amdgpu_cs_parser_bos error handling to
cleanup bo_list userptr bo.
WARNING: kfdtest/2930 still has locks held!
1 lock held by kfdtest/2930:
(&list->bo_list_mutex){+.+.}-{3:3}, at: amdgpu_cs_ioctl+0xce5/0x1f10 [amdgpu]
stack backtrace:
dump_stack_lvl+0x44/0x57
get_signal+0x79f/0xd00
arch_do_signal_or_restart+0x36/0x7b0
exit_to_user_mode_prepare+0xfd/0x1b0
syscall_exit_to_user_mode+0x19/0x40
do_syscall_64+0x40/0x80
Signed-off-by: Philip Yang <Philip.Yang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The recent change brought a bug on SRIOV envrionment. It caused
unloading amdgpu failed on Guest VM. The reason is that the VF
FLR was requested while unloading amdgpu driver, but the VF FLR
of SRIOV sequence is wrong while removing PCI device.
For SRIOV, the guest driver should not trigger the whole XGMI hive
to do the reset. Host driver control how the device been reset.
Fixes: f5c7e77970 ("drm/amdgpu: Adjust removal control flow for smu v13_0_2")
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Shaoyun Liu <Shaoyun.Liu@amd.com>
Signed-off-by: Gavin Wan <Gavin.Wan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If a system does not have swap and memory is under 100% usage,
amdgpu will fail to evict resources. Currently the suspend
carries on proceeding to reset the GPU:
```
[drm] evicting device resources failed
[drm:amdgpu_device_ip_suspend_phase2 [amdgpu]] *ERROR* suspend of IP block <vcn_v3_0> failed -12
[drm] free PSP TMR buffer
[TTM] Failed allocating page table
[drm] evicting device resources failed
amdgpu 0000:03:00.0: amdgpu: MODE1 reset
amdgpu 0000:03:00.0: amdgpu: GPU mode1 reset
amdgpu 0000:03:00.0: amdgpu: GPU smu mode1 reset
```
At this point if the suspend actually succeeded I think that amdgpu
would have recovered because the GPU would have power cut off and
restored. However the kernel fails to continue the suspend from the
memory pressure and amdgpu fails to run the "resume" from the aborted
suspend.
```
ACPI: PM: Preparing to enter system sleep state S3
SLUB: Unable to allocate memory on node -1, gfp=0xdc0(GFP_KERNEL|__GFP_ZERO)
cache: Acpi-State, object size: 80, buffer size: 80, default order: 0, min order: 0
node 0: slabs: 22, objs: 1122, free: 0
ACPI Error: AE_NO_MEMORY, Could not update object reference count (20210730/utdelete-651)
[drm:psp_hw_start [amdgpu]] *ERROR* PSP load kdb failed!
[drm:psp_resume [amdgpu]] *ERROR* PSP resume failed
[drm:amdgpu_device_fw_loading [amdgpu]] *ERROR* resume of IP block <psp> failed -62
amdgpu 0000:03:00.0: amdgpu: amdgpu_device_ip_resume failed (-62).
PM: dpm_run_callback(): pci_pm_resume+0x0/0x100 returns -62
amdgpu 0000:03:00.0: PM: failed to resume async: error -62
```
To avoid this series of unfortunate events, fail amdgpu's suspend
when the memory eviction fails. This will let the system gracefully
recover and the user can try suspend again when the memory pressure
is relieved.
Reported-by: post@davidak.de
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2223
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Use mes.sched_version, mes.kiq_version for debugfs as
mes.ucode_fw_version does not contain correct versioning information.
Signed-off-by: Graham Sider <Graham.Sider@amd.com>
Reviewed-by: Jack Xiao <Jack.Xiao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This patch to fix the gdm3 start failure with virual display:
/usr/libexec/gdm-x-session[1711]: (II) AMDGPU(0): Setting screen physical size to 270 x 203
/usr/libexec/gdm-x-session[1711]: (EE) AMDGPU(0): Failed to make import prime FD as pixmap: 22
/usr/libexec/gdm-x-session[1711]: (EE) AMDGPU(0): failed to set mode: Invalid argument
/usr/libexec/gdm-x-session[1711]: (WW) AMDGPU(0): Failed to set mode on CRTC 0
/usr/libexec/gdm-x-session[1711]: (EE) AMDGPU(0): Failed to enable any CRTC
gnome-shell[1840]: Running GNOME Shell (using mutter 42.2) as a X11 window and compositing manager
/usr/libexec/gdm-x-session[1711]: (EE) AMDGPU(0): failed to set mode: Invalid argument
vkms doesn't have modifiers support, set fb_modifiers_not_supported to bring the gdm back.
Signed-off-by: Yifan Zhang <yifan1.zhang@amd.com>
Acked-by: Guchun Chen <guchun.chen@amd.com>
Reviewed-by: Tim Huang <Tim.Huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
In the S2idle suspend/resume phase the gfxoff is keeping functional so
some IP blocks will be likely to reinitialize at gfxoff entry and that
will result in failing to program GC registers.Therefore, let disallow
gfxoff until AMDGPU IPs reinitialized completely.
Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org # 5.15.x
IMU is a new firmware for GFX11.
There are four means by which firmware version can be queried
from the driver: device attributes, vf2pf, debugfs,
and the AMDGPU_INFO_FW_VERSION option in the amdgpu info ioctl.
Add IMU as an option for those four methods.
V2: Added debugfs
Reviewed-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: David Francis <David.Francis@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
If the number of pages from the userptr BO differs from the SG BO then the
allocated memory for the SG table doesn't get freed before returning
-EINVAL, which may lead to a memory leak in some error paths. Fix this by
checking the number of pages before allocating memory for the SG table.
Fixes: 264fb4d332 ("drm/amdgpu: Add multi-GPU DMA mapping helpers")
Signed-off-by: Rafael Mendonca <rafaelmendsr@gmail.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>
MMHUB 2.1.x versions don't have ATCL2. Remove accesses to ATCL2 registers.
Since they are non-existing registers, read access will cause a
'Completer Abort' and gets reported when AER is enabled with the below patch.
Tagging with the patch so that this is backported along with it.
v2: squash in uninitialized warning fix (Nathan Chancellor)
Fixes: 8795e182b0 ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()")
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
[why]
MES response time in sriov may be longer than default value
due to reset or init in other VF. A timeout value specific
to sriov is needed.
[how]
When in sriov, adjust the timeout value to calculated
worst case scenario.
Signed-off-by: Yiqing Yao <yiqing.yao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[WHY]
0, original pstate X
1, ctx_A_create -> ctx_A->stable_pstate = X
2, ctx_A_set_pstate (Y) -> current pstate is Y (PEAK or STANDARD)
3, ctx_B_create -> ctx_B->stable_pstate = Y
4, ctx_A_destroy -> restore pstate to X
5, ctx_B_destroy -> restore pstate to Y
Above sequence will cause final pstate is wrong (Y), should be original X.
[HOW]
When ctx_B create,
if ctx_A touched pstate setting
(not auto, stable_pstate_ctx != NULL),
set ctx_B->stable_pstate the same value as ctx_A saved,
if stable_pstate_ctx == NULL,
fetch current pstate to fill
ctx_B->stable_pstate.
Signed-off-by: Chengming Gui <Jack.Gui@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Commit 8795e182b0 ("PCI/portdrv: Don't disable AER reporting in get_port_device_capability()")
uncovered a bug in amdgpu that required a reordering of the driver
init sequence to avoid accessing a special register on the GPU
before it was properly set up leading to an PCI AER error. This
reordering uncovered a different hw programming ordering dependency
in some APUs where the SDMA doorbells need to be programmed before
the GFX doorbells. To fix this, move the SDMA doorbell programming
back into the soc15 common code, but use the actual doorbell range
values directly rather than the values stored in the ring structure
since those will not be initialized at this point.
This is a partial revert, but with the doorbell assignment
fixed so the proper doorbell index is set before it's used.
Fixes: e3163bc8ff ("drm/amdgpu: move nbio sdma_doorbell_range() into sdma code for vega")
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: skhan@linuxfoundation.org
Cc: stable@vger.kernel.org
A user reported a bug on CAPE VERDE system where uvd_v3_1
IP component failed to initialize as there is an issue with
BO move code from one memory to other.
In function amdgpu_mem_visible() called by amdgpu_bo_move(),
when there are no blocks to compare or if we have a single
block then break the loop.
Fixes: 312b4dc11d ("drm/amdgpu: Fix VRAM BO swap issue")
Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
If mes is not dequeued during fini, mes will be in an uncleaned state
during reload, then mes couldn't receive some commands which leads to
reload failure.
[How]
Perform MES dequeue via MMIO after all the unmap jobs are done by mes
and before kiq fini.
v2: Move the dequeue operation inside kiq_hw_fini.
Signed-off-by: YuBiao Wang <YuBiao.Wang@amd.com>
Reviewed-by: Jack Xiao <Jack.Xiao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
- refactor mode2 on v11.0.7 to align with aldebaran
- comment out using mode2 reset as default for now, will introduce
another controller to replace previous reset_level_mask
v2: squash in unused variable removal (Alex)
Signed-off-by: Victor Zhao <Victor.Zhao@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This reverts commit dac6b80818.
This commit reverted the AMDGPU_SKIP_MODE2_RESET as it conflicts with
the original design of reset handler. Will redesign it.
Fixes: dac6b80818 ("drm/amdgpu: let mode2 reset fallback to default when failure")
Signed-off-by: Victor Zhao <Victor.Zhao@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
This reverts commit 5bd8d53f6f.
This commit breaks the reset logic for aldebaran, revert it for now.
Will move the mask inside the reset handler.
Fixes: 5bd8d53f6f ("drm/amdgpu: add debugfs amdgpu_reset_level")
Signed-off-by: Victor Zhao <Victor.Zhao@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
For asic with VF MMIO access protection avoid using CPU for VM table updates.
CPU pagetable updates have issues with HDP flush as VF MMIO access protection
blocks write to mmBIF_BX_DEV0_EPF0_VF0_HDP_MEM_COHERENCY_FLUSH_CNTL register
during sriov runtime.
v3: introduce virtualization capability flag AMDGPU_VF_MMIO_ACCESS_PROTECT
which indicates that VF MMIO write access is not allowed in sriov runtime
Signed-off-by: Danijel Slivka <danijel.slivka@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
