Once drm_framebuffer_init has returned 0, the framebuffer is hooked up
to the reference counting machinery and can no longer be destroyed with
a simple kfree. Therefore, it must be called last.
If drm_framebuffer_init returns 0 but its caller then returns non-0,
there will likely be memory corruption fireworks down the road.
The following lead me to this fix:
[ 12.891228] kernel BUG at lib/list_debug.c:25!
[...]
[ 12.891263] RIP: 0010:__list_add_valid+0x4b/0x70
[...]
[ 12.891324] Call Trace:
[ 12.891330] drm_framebuffer_init+0xb5/0x100 [drm]
[ 12.891378] amdgpu_display_gem_fb_verify_and_init+0x47/0x120 [amdgpu]
[ 12.891592] ? amdgpu_display_user_framebuffer_create+0x10d/0x1f0 [amdgpu]
[ 12.891794] amdgpu_display_user_framebuffer_create+0x126/0x1f0 [amdgpu]
[ 12.891995] drm_internal_framebuffer_create+0x378/0x3f0 [drm]
[ 12.892036] ? drm_internal_framebuffer_create+0x3f0/0x3f0 [drm]
[ 12.892075] drm_mode_addfb2+0x34/0xd0 [drm]
[ 12.892115] ? drm_internal_framebuffer_create+0x3f0/0x3f0 [drm]
[ 12.892153] drm_ioctl_kernel+0xe2/0x150 [drm]
[ 12.892193] drm_ioctl+0x3da/0x460 [drm]
[ 12.892232] ? drm_internal_framebuffer_create+0x3f0/0x3f0 [drm]
[ 12.892274] amdgpu_drm_ioctl+0x43/0x80 [amdgpu]
[ 12.892475] __se_sys_ioctl+0x72/0xc0
[ 12.892483] do_syscall_64+0x33/0x40
[ 12.892491] entry_SYSCALL_64_after_hwframe+0x44/0xae
Fixes: f258907fdd "drm/amdgpu: Verify bo size can fit framebuffer size on init."
Signed-off-by: Michel Dänzer <mdaenzer@redhat.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
It will cause error when alloc memory larger than 128KB in
amdgpu_bo_create->kzalloc. So it needs to switch kzalloc to kvzalloc.
Call Trace:
alloc_pages_current+0x6a/0xe0
kmalloc_order+0x32/0xb0
kmalloc_order_trace+0x1e/0x80
__kmalloc+0x249/0x2d0
amdgpu_bo_create+0x102/0x500 [amdgpu]
? xas_create+0x264/0x3e0
amdgpu_bo_create_vm+0x32/0x60 [amdgpu]
amdgpu_vm_pt_create+0xf5/0x260 [amdgpu]
amdgpu_vm_init+0x1fd/0x4d0 [amdgpu]
Signed-off-by: Changfeng <Changfeng.Zhu@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
When some tools performing psp mailbox attack, the readback value
of register can be a random value which may break psp.
[How]
Use a psp wptr cache machanism to aovid the change made by attack.
v2: unify change and add detailed reason
Signed-off-by: Victor Zhao <Victor.Zhao@amd.com>
Signed-off-by: Jingwen Chen <Jingwen.Chen2@amd.com>
Reviewed-by: Monk Liu <monk.liu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
On QUERY2 IOCTL don't query counts of correctable
and uncorrectable errors, since when RAS is
enabled and supported on Vega20 server boards,
this takes insurmountably long time, in O(n^3),
which slows the system down to the point of it
being unusable when we have GUI up.
Fixes: ae363a212b ("drm/amdgpu: Add a new flag to AMDGPU_CTX_OP_QUERY_STATE2")
Cc: Alexander Deucher <Alexander.Deucher@amd.com>
Cc: stable@vger.kernel.org
Signed-off-by: Luben Tuikov <luben.tuikov@amd.com>
Reviewed-by: Alexander Deucher <Alexander.Deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
1.correct KFD SDMA RLC queue register offset error.
(all sdma rlc register offset is base on SDMA0.RLC0_RLC0_RB_CNTL)
2.HQD_N_REGS (19+6+7+12)
12: the 2 more resgisters than navi1x (SDMAx_RLCy_MIDCMD_DATA{9,10})
the patch also can be fixed NULL pointer issue when read
/sys/kernel/debug/kfd/hqds on sienna_cichlid chip.
Signed-off-by: Kevin Wang <kevin1.wang@amd.com>
Reviewed-by: Likun Gao <Likun.Gao@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
This unfortunately comes up in regular intervals and breaks
GPU reset for the engine in question.
The sched.ready flag controls if an engine can't get working
during hw_init, but should never be set to false during hw_fini.
v2: squash in unused variable fix (Alex)
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
the gem object rfb->base.obj[0] is get according to num_planes
in amdgpufb_create, but is not put according to num_planes
[How]
put rfb->base.obj[0] in amdgpu_fbdev_destroy according to num_planes
Signed-off-by: Jingwen Chen <Jingwen.Chen2@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
When PAGE_SIZE is larger than AMDGPU_PAGE_SIZE, the number of GPU TLB
entries which need to update in amdgpu_map_buffer() should be multiplied
by AMDGPU_GPU_PAGES_IN_CPU_PAGE (PAGE_SIZE / AMDGPU_PAGE_SIZE).
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Yi Li <liyi@loongson.cn>
Signed-off-by: Huacai Chen <chenhuacai@loongson.cn>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Update the method of disabling VCN IP for specific SKU for navi1x ASIC,
it will judge whether should add the related IP at the function of
amdgpu_device_ip_block_add().
Signed-off-by: Likun Gao <Likun.Gao@amd.com>
Reviewed-by: Guchun Chen <guchun.chen@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
amd-drm-fixes-5.13-2021-05-05:
amdgpu:
- MPO hang workaround
- Fix for concurrent VM flushes on vega/navi
- dcefclk is not adjustable on navi1x and newer
- MST HPD debugfs fix
- Suspend/resumes fixes
- Register VGA clients late in case driver fails to load
- Fix GEM leak in user framebuffer create
- Add support for polaris12 with 32 bit memory interface
- Fix duplicate cursor issue when using overlay
- Fix corruption with tiled surfaces on VCN3
- Add BO size and stride check to fix BO size verification
radeon:
- Fix off-by-one in power state parsing
- Fix possible memory leak in power state parsing
Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Alex Deucher <alexander.deucher@amd.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20210506033929.3875-1-alexander.deucher@amd.com
The builtin size check isn't really the right thing for AMD
modifiers due to a couple of reasons:
1) In the format structs we don't do set any of the tilesize / blocks
etc. to avoid having format arrays per modifier/GPU
2) The pitch on the main plane is pixel_pitch * bytes_per_pixel even
for tiled ...
3) The pitch for the DCC planes is really the pixel pitch of the main
surface that would be covered by it ...
Note that we only handle GFX9+ case but we do this after converting
the implicit modifier to an explicit modifier, so on GFX9+ all
framebuffers should be checked here.
There is a TODO about DCC alignment, but it isn't worse than before
and I'd need to dig a bunch into the specifics. Getting this out in
a reasonable timeframe to make sure it gets the appropriate testing
seemed more important.
Finally as I've found that debugging addfb2 failures is a pita I was
generous adding explicit error messages to every failure case.
Fixes: f258907fdd ("drm/amdgpu: Verify bo size can fit framebuffer size on init.")
Tested-by: Simon Ser <contact@emersion.fr>
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Otherwise tiling modes that require the values form this field
(In particular _*_X) would be corrupted upon video decode.
Copied from the VCN v2 code.
Fixes: 99541f392b ("drm/amdgpu: add mc resume DPG mode for VCN3.0")
Reviewed-and-Tested by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Bas Nieuwenhuizen <bas@basnieuwenhuizen.nl>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
The runtime resume PM op disregards the return value from
amdgpu_device_resume(), masking errors for failed resumes at the PM
layer.
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Pavan Kumar Ramayanam <pavan.ramayanam@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Sriov gets suspend of IP block <dce_virtual> failed as return
value was not initialized.
v2: return 0 directly to align original code semantic before this
was broken out into a separate helper function instead of setting
initial values
Signed-off-by: Victor Zhao <Victor.Zhao@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
Starting with Vega the hardware supports concurrent flushes
of VMID which can be used to implement per process VMID
allocation.
But concurrent flushes are mutual exclusive with back to
back VMID allocations, fix this to avoid a VMID used in
two ways at the same time.
v2: don't set ring to NULL
Signed-off-by: Christian König <christian.koenig@amd.com>
Reviewed-by: James Zhu <James.Zhu@amd.com>
Tested-by: James Zhu <James.Zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Cc: stable@vger.kernel.org
The buffer of SA bo will be used by many cases. So it's better
to invalidate the cache of indirect buffer allocated by SA before
commit the IB.
Signed-off-by: Jinzhou Su <Jinzhou.Su@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fix the following issues with SDMA RAS error reporting:
1. Read the EDC_COUNTER2 register also to fetch error counts
for all sub-blocks in SDMA.
2. SDMA RAS on Aldebaran suports single-bit uncorrectable errors
only. So, report error count in UE count instead of CE count.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-By: John Clements <John.Clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Reset the RAS error count and error status registers after
reading to prevent over reporting error counts on Aldebaran.
Signed-off-by: Mukul Joshi <mukul.joshi@amd.com>
Reviewed-By: John Clements <John.Clements@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
