drm/amd/display: Support external tunneling feature

[Why & How]
The original code only supports the tunneling for embedded one.
To support external tunneling feature, it needs to check
Tunneling_Support bit register.

Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Reviewed-by: Jun Lei <jun.lei@amd.com>
Signed-off-by: Cruise Hung <Cruise.Hung@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: init local variable to fix format errors

[WHY & HOW]
Uninitialized local variables will cause format checker complain
about them.

Reviewed-by: Charlene Liu <charlene.liu@amd.com>
Signed-off-by: Yihan Zhu <Yihan.Zhu@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Extend dc_plane_get_status with flags

[WHY]
dc_plane_get_status may be used for reading other plane properties
in the future.

[HOW]
Provide API for choosing plane properties to read.

Reviewed-by: Charlene Liu <charlene.liu@amd.com>
Reviewed-by: Aric Cyr <aric.cyr@amd.com>
Reviewed-by: Swapnil Patel <swapnil.patel@amd.com>
Signed-off-by: Tomasz Siemek <Tomasz.Siemek@amd.com>
Signed-off-by: Tom Chung <chiahsuan.chung@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix use-after-unlock in eviction fence destroy

The eviction fence destroy path incorrectly calls dma_fence_put() on
evf_mgr->ev_fence after releasing the ev_fence_lock. This introduces a
potential use-after-unlock or race because another thread concurrently
modifies evf_mgr->ev_fence.

Fix this by grabbing a local reference to evf_mgr->ev_fence under the
lock and using that for dma_fence_put() after waiting.

Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Sunil Khatri <sunil.khatri@amd.com>
Signed-off-by: Arvind Yadav <Arvind.Yadav@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Allow NPS2-CPX combination for VFs

CPX partition mode is compatible with NPS2 on aquavanjaram VFs.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/mmsch: Add MMSCH v5_0 support for sriov

These structures are basically ported from MMSCH v4_0
The structures are the same as v4_0 except for the
init header

Signed-off-by: fanhuang <FangSheng.Huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Use compatible NPS mode info

Compatible NPS modes for a partition mode are exposed through xcp_config
interface. To determine if a compute partition mode is valid, check if
the current NPS mode is part of compatible NPS modes.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: Move SMUv13.0.12 function declarations

Move them to SMUv13.0.6 header file as they are used only in SMU
v13.0.6.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add pldm version reporting

Add pldm version reporting through sysfs node

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: Support chain runlists of XNACK+/XNACK-

If the MEC firmware supports chaining runlists of XNACK+/XNACK-
processes, set SQ_CONFIG1 chicken bit and SET_RESOURCES bit 28.

When the MEC/HWS supports it, KFD checks the XNACK+/XNACK- processes mix
happens or not. If it does, enter over-subscription.

Signed-off-by: Amber Lin <Amber.Lin@amd.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/radeon/cik: Clean up doorbells

Free doorbells in the error paths of cik_init and in cik_fini.

Build tested only.

Suggested-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add debugfs for spirom IFWI dump

Expose the debugfs file node for user space to dump the IFWI image
on spirom.

For one transaction between PSP and host, it will read out the
images on both active and inactive partitions so a buffer with two
times the size of maximum IFWI image (currently 16MByte) is needed.

v2: move the vbios gfl macros to the common header and rename the
    bo triplet struct to spirom_bo for this specific usage (Hawking)

v3: return directly the result of last command execution (Lijo)

Signed-off-by: Shiwu Zhang <shiwu.zhang@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix userq resource double freed

As the userq resource was already freed at the drm_release
early phase, it should avoid freeing userq resource again
at the later kms postclose callback.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix circular locking in userq creation

A circular locking dependency was detected between the global
`adev->userq_mutex` and per-file `userq_mgr->userq_mutex` when
creating user queues. The issue occurs because:

1. `amdgpu_userq_suspend()` and `amdgpu_userq_resume` take `adev->userq_mutex` first, then
   `userq_mgr->userq_mutex`
2. While `amdgpu_userq_create()` takes them in reverse order

This patch resolves the issue by:
1. Moving the `adev->userq_mutex` lock earlier in `amdgpu_userq_create()`
   to cover the `amdgpu_userq_ensure_ev_fence()` call
2. Releasing it after we're done with both queue creation and the
   scheduling halt check

v2: remove unused adev->userq_mutex lock (Prike)

Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Reviewed-by: Prike Liang <Prike.Liang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: read back register after written for VCN v4.0.5

On VCN v4.0.5 there is a race condition where the WPTR is not
updated after starting from idle when doorbell is used. Adding
register read-back after written at function end is to ensure
all register writes are done before they can be used.

Closes: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12528
Signed-off-by: David (Ming Qiang) Wu <David.Wu3@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Tested-by: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Ruijing Dong <ruijing.dong@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Revert "drm/amd/display: Hardware cursor changes color when switched to software cursor"

This reverts commit 272e6aab14bbf98d7a06b2b1cd6308a02d4a10a1.

Applying degamma curve to the cursor by default breaks Linux userspace
expectation.

On Linux, AMD display manager enables cursor degamma ROM just for
implict sRGB on HW versions where degamma is split into two blocks:
degamma ROM for pre-defined TFs and `gamma correction` for user/custom
curves, and degamma ROM settings doesn't apply to cursor plane.

Link: https://gitlab.freedesktop.org/drm/amd/-/issues/1513
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/2803
Reported-by: Michel Dänzer <michel.daenzer@mailbox.org>
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4144
Signed-off-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/userq: Fix DEBUG_LOCKS_WARN_ON(lock->magic != lock)

Fix DEBUG_LOCKS_WARN_ON(lock->magic != lock) warning logs.

Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix userq ttm_bo_pin and ttm_bo_unpin lockdep warnings

The ttm_bo_pin and ttm_bo_unpin warnings are resolved by moving the
doorbell bo reserve up before pin/unpin.

WARNING: CPU: 11 PID: 1818 at drivers/gpu/drm/ttm/ttm_bo.c:592 ttm_bo_pin+0x1f6/0x270 [ttm]
[  +0.000277] CPU: 11 UID: 1000 PID: 1818 Comm: Xwayland Tainted: G        W          6.12.0+ #15
[  +0.000006] Tainted: [W]=WARN
[  +0.000004] Hardware name: ASUS System Product Name/TUF GAMING B650-PLUS, BIOS 3072 12/20/2024
[  +0.000004] RIP: 0010:ttm_bo_pin+0x1f6/0x270 [ttm]
[  +0.000005] RSP: 0018:ffff88846ca879d0 EFLAGS: 00010246
[  +0.000007] RAX: 0000000000000000 RBX: ffff88810b7ca848 RCX: 0000000000000000
[  +0.000004] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[  +0.000005] RBP: ffff88846ca879e8 R08: 0000000000000000 R09: 0000000000000000
[  +0.000004] R10: 0000000000000000 R11: 0000000000000000 R12: ffff88810b7ca848
[  +0.000004] R13: ffff88846c666250 R14: 1ffff1108d950f44 R15: ffff88846ca87aa0
[  +0.000005] FS:  00007c45ff436d00(0000) GS:ffff888409580000(0000) knlGS:0000000000000000
[  +0.000004] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  +0.000005] CR2: 00005b0c142a60e0 CR3: 000000012ce5a000 CR4: 0000000000f50ef0
[  +0.000004] PKRU: 55555554
[  +0.000004] Call Trace:
[  +0.000004]  <TASK>
[  +0.000005]  ? show_regs+0x6c/0x80
[  +0.000007]  ? __warn+0xd2/0x2d0
[  +0.000007]  ? ttm_bo_pin+0x1f6/0x270 [ttm]
[  +0.000031]  ? report_bug+0x282/0x2f0
[  +0.000012]  ? handle_bug+0x6e/0xc0
[  +0.000007]  ? exc_invalid_op+0x18/0x50
[  +0.000007]  ? asm_exc_invalid_op+0x1b/0x20
[  +0.000017]  ? ttm_bo_pin+0x1f6/0x270 [ttm]
[  +0.000014]  amdgpu_bo_pin+0x365/0x9d0 [amdgpu]
[  +0.000191]  ? __pfx_amdgpu_bo_pin+0x10/0x10 [amdgpu]
[  +0.000185]  ? drm_gem_object_lookup+0x81/0xc0
[  +0.000008]  ? kasan_save_alloc_info+0x37/0x60
[  +0.000007]  ? __kasan_kmalloc+0xc3/0xd0
[  +0.000013]  amdgpu_userqueue_get_doorbell_index+0xee/0x5f0 [amdgpu]
[  +0.000209]  amdgpu_userq_ioctl+0x6b4/0xd40 [amdgpu]
[  +0.000193]  ? __pfx_amdgpu_userq_ioctl+0x10/0x10 [amdgpu]
[  +0.000211]  ? lock_acquire+0x7c/0xc0
[  +0.000006]  ? drm_dev_enter+0x51/0x190
[  +0.000015]  drm_ioctl_kernel+0x18b/0x330
[  +0.000007]  ? __pfx_amdgpu_userq_ioctl+0x10/0x10 [amdgpu]
[  +0.000190]  ? __pfx_drm_ioctl_kernel+0x10/0x10
[  +0.000005]  ? lock_acquire+0x7c/0xc0
[  +0.000009]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? __kasan_check_write+0x14/0x30
[  +0.000005]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000011]  drm_ioctl+0x589/0xd00
[  +0.000005]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000006]  ? __pfx_amdgpu_userq_ioctl+0x10/0x10 [amdgpu]
[  +0.000194]  ? __pfx_drm_ioctl+0x10/0x10
[  +0.000006]  ? __pm_runtime_resume+0x80/0x110
[  +0.000021]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? trace_hardirqs_on+0x53/0x60
[  +0.000005]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? _raw_spin_unlock_irqrestore+0x51/0x80
[  +0.000013]  amdgpu_drm_ioctl+0xd2/0x1c0 [amdgpu]
[  +0.000185]  __x64_sys_ioctl+0x13a/0x1c0
[  +0.000010]  x64_sys_call+0x11ad/0x25f0
[  +0.000007]  do_syscall_64+0x91/0x180
[  +0.000007]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? irqentry_exit+0x77/0xb0
[  +0.000005]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? exc_page_fault+0x93/0x150
[  +0.000009]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  +0.000005] RIP: 0033:0x7c45ff924ded
[  +0.000005] RSP: 002b:00007ffff7167810 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[  +0.000008] RAX: ffffffffffffffda RBX: 00000000c0486456 RCX: 00007c45ff924ded
[  +0.000004] RDX: 00007ffff7167870 RSI: 00000000c0486456 RDI: 000000000000000b
[  +0.000004] RBP: 00007ffff7167860 R08: ffff800100000000 R09: 0000000000010000
[  +0.000005] R10: 00007ffff7167950 R11: 0000000000000246 R12: 00005b0c2a51bc48
[  +0.000004] R13: 000000000000000b R14: 0000000000000000 R15: 00007ffff7167950
[  +0.000022]  </TASK>
[  +0.000004] irq event stamp: 80693
[  +0.000004] hardirqs last  enabled at (80699): [<ffffffff86a693a9>] __up_console_sem+0x79/0xa0
[  +0.000005] hardirqs last disabled at (80704): [<ffffffff86a6938e>] __up_console_sem+0x5e/0xa0
[  +0.000005] softirqs last  enabled at (80390): [<ffffffff8687377e>] __irq_exit_rcu+0x17e/0x1d0
[  +0.000005] softirqs last disabled at (80385): [<ffffffff8687377e>] __irq_exit_rcu+0x17e/0x1d0
[  +0.000006] ---[ end trace 0000000000000000 ]---
------------------------------------------------------------------------------------------------------

[  +0.000006] WARNING: CPU: 10 PID: 1818 at drivers/gpu/drm/ttm/ttm_bo.c:611 ttm_bo_unpin+0x21f/0x2c0 [ttm]
[  +0.000280] CPU: 10 UID: 1000 PID: 1818 Comm: Xwayland Tainted: G        W          6.12.0+ #15
[  +0.000006] Tainted: [W]=WARN
[  +0.000004] Hardware name: ASUS System Product Name/TUF GAMING B650-PLUS, BIOS 3072 12/20/2024
[  +0.000004] RIP: 0010:ttm_bo_unpin+0x21f/0x2c0 [ttm]
[  +0.000005] RSP: 0018:ffff88846ca87888 EFLAGS: 00010246
[  +0.000007] RAX: 0000000000000000 RBX: ffff88810b7ca848 RCX: 0000000000000000
[  +0.000005] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[  +0.000004] RBP: ffff88846ca878a0 R08: 0000000000000000 R09: 0000000000000000
[  +0.000004] R10: 0000000000000000 R11: 0000000000000000 R12: ffff888164e90050
[  +0.000005] R13: ffff88846c666200 R14: 0000000000000001 R15: ffff888168402d28
[  +0.000004] FS:  00007c45ff436d00(0000) GS:ffff888409500000(0000) knlGS:0000000000000000
[  +0.000005] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  +0.000004] CR2: 00007c45f7373b20 CR3: 000000012ce5a000 CR4: 0000000000f50ef0
[  +0.000005] PKRU: 55555554
[  +0.000004] Call Trace:
[  +0.000004]  <TASK>
[  +0.000005]  ? show_regs+0x6c/0x80
[  +0.000008]  ? __warn+0xd2/0x2d0
[  +0.000007]  ? ttm_bo_unpin+0x21f/0x2c0 [ttm]
[  +0.000012]  ? report_bug+0x282/0x2f0
[  +0.000013]  ? handle_bug+0x6e/0xc0
[  +0.000006]  ? exc_invalid_op+0x18/0x50
[  +0.000008]  ? asm_exc_invalid_op+0x1b/0x20
[  +0.000017]  ? ttm_bo_unpin+0x21f/0x2c0 [ttm]
[  +0.000011]  ? ttm_bo_unpin+0x217/0x2c0 [ttm]
[  +0.000011]  amdgpu_bo_unpin+0x45/0x250 [amdgpu]
[  +0.000216]  amdgpu_userq_ioctl+0x2c3/0xd40 [amdgpu]
[  +0.000226]  ? drm_dev_exit+0x2d/0x60
[  +0.000010]  ? __pfx_amdgpu_userq_ioctl+0x10/0x10 [amdgpu]
[  +0.000201]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? lock_acquire+0x7c/0xc0
[  +0.000006]  ? drm_dev_enter+0x51/0x190
[  +0.000015]  drm_ioctl_kernel+0x18b/0x330
[  +0.000007]  ? __pfx_amdgpu_userq_ioctl+0x10/0x10 [amdgpu]
[  +0.000188]  ? __pfx_drm_ioctl_kernel+0x10/0x10
[  +0.000006]  ? lock_acquire+0x7c/0xc0
[  +0.000008]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? __kasan_check_write+0x14/0x30
[  +0.000006]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000010]  drm_ioctl+0x589/0xd00
[  +0.000005]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000006]  ? __pfx_amdgpu_userq_ioctl+0x10/0x10 [amdgpu]
[  +0.000211]  ? __pfx_drm_ioctl+0x10/0x10
[  +0.000006]  ? __pm_runtime_resume+0x80/0x110
[  +0.000020]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000006]  ? trace_hardirqs_on+0x53/0x60
[  +0.000005]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? _raw_spin_unlock_irqrestore+0x51/0x80
[  +0.000013]  amdgpu_drm_ioctl+0xd2/0x1c0 [amdgpu]
[  +0.000186]  __x64_sys_ioctl+0x13a/0x1c0
[  +0.000010]  x64_sys_call+0x11ad/0x25f0
[  +0.000007]  do_syscall_64+0x91/0x180
[  +0.000007]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? do_syscall_64+0x9d/0x180
[  +0.000007]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000010]  ? __pfx___rseq_handle_notify_resume+0x10/0x10
[  +0.000005]  ? __pfx_blkcg_maybe_throttle_current+0x10/0x10
[  +0.000013]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000009]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000008]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? syscall_exit_to_user_mode+0x95/0x260
[  +0.000008]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? do_syscall_64+0x9d/0x180
[  +0.000007]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? do_syscall_64+0x9d/0x180
[  +0.000011]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000010]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000009]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000008]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? irqentry_exit_to_user_mode+0x8b/0x260
[  +0.000007]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000006]  ? irqentry_exit+0x77/0xb0
[  +0.000004]  ? srso_alias_return_thunk+0x5/0xfbef5
[  +0.000005]  ? exc_page_fault+0x93/0x150
[  +0.000010]  entry_SYSCALL_64_after_hwframe+0x76/0x7e
[  +0.000005] RIP: 0033:0x7c45ff924ded
[  +0.000005] RSP: 002b:00007ffff7168790 EFLAGS: 00000246 ORIG_RAX: 0000000000000010
[  +0.000008] RAX: ffffffffffffffda RBX: 00000000c0486456 RCX: 00007c45ff924ded
[  +0.000005] RDX: 00007ffff71687f0 RSI: 00000000c0486456 RDI: 000000000000000b
[  +0.000004] RBP: 00007ffff71687e0 R08: 00005b0c2a49b010 R09: 0000000000000007
[  +0.000004] R10: 00005b0c2a4d7140 R11: 0000000000000246 R12: 000000000000000b
[  +0.000004] R13: 00007c45ff19e5cc R14: 00005b0c2a51c538 R15: 00005b0c2a51bbd8
[  +0.000022]  </TASK>
[  +0.000005] irq event stamp: 87419
[  +0.000004] hardirqs last  enabled at (87425): [<ffffffff86a693a9>] __up_console_sem+0x79/0xa0
[  +0.000005] hardirqs last disabled at (87430): [<ffffffff86a6938e>] __up_console_sem+0x5e/0xa0
[  +0.000005] softirqs last  enabled at (87058): [<ffffffff8687377e>] __irq_exit_rcu+0x17e/0x1d0
[  +0.000006] softirqs last disabled at (87053): [<ffffffff8687377e>] __irq_exit_rcu+0x17e/0x1d0
[  +0.000005] ---[ end trace 0000000000000000 ]---

Signed-off-by: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/userq: Fix lock contention in userq fence

Fix lockdep warnings.

[  +0.000637] ================================
[  +0.000004] WARNING: inconsistent lock state
[  +0.000004] 6.12.0+ #18 Tainted: G        W  OE
[  +0.000004] --------------------------------
[  +0.000004] inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage.
[  +0.000004] Xwayland/1952 [HC0[0]:SC0[0]:HE1:SE1] takes:
[  +0.000005] ffff8884636f4740 (&fence_drv->fence_list_lock){?...}-{2:2}, at: amdgpu_userq_fence_driver_destroy+0xb8/0x540 [amdgpu]
[  +0.000208] {IN-HARDIRQ-W} state was registered at:
[  +0.000004]   lock_acquire.part.0+0x116/0x360
[  +0.000005]   lock_acquire+0x7c/0xc0
[  +0.000005]   _raw_spin_lock+0x2f/0x60
[  +0.000005]   amdgpu_userq_fence_driver_process+0x75/0x400 [amdgpu]
[  +0.000185]   gfx_v12_0_eop_irq+0x29f/0x420 [amdgpu]
[  +0.000210]   amdgpu_irq_dispatch+0x2a4/0x7b0 [amdgpu]
[  +0.000191]   amdgpu_ih_process+0x1e1/0x3d0 [amdgpu]
[  +0.000185]   amdgpu_irq_handler+0x28/0xc0 [amdgpu]
[  +0.000183]   __handle_irq_event_percpu+0x1bb/0x590
[  +0.000005]   handle_irq_event+0xab/0x1d0
[  +0.000005]   handle_edge_irq+0x1fd/0xc10
[  +0.000005]   __common_interrupt+0x83/0x190
[  +0.000004]   common_interrupt+0xb1/0xe0
[  +0.000005]   asm_common_interrupt+0x27/0x40
[  +0.000004]   cpuidle_enter_state+0x2ba/0x530
[  +0.000005]   cpuidle_enter+0x4f/0xb0
[  +0.000006]   call_cpuidle+0x46/0xd0
[  +0.000005]   do_idle+0x367/0x430
[  +0.000004]   cpu_startup_entry+0x58/0x70
[  +0.000005]   start_secondary+0x224/0x2b0
[  +0.000005]   common_startup_64+0x13e/0x141
[  +0.000005] irq event stamp: 88271
[  +0.000004] hardirqs last  enabled at (88271): [<ffffffffad9ca7a1>] _raw_spin_unlock_irqrestore+0x51/0x80
[  +0.000005] hardirqs last disabled at (88270): [<ffffffffad9ca424>] _raw_spin_lock_irqsave+0x74/0x80
[  +0.000005] softirqs last  enabled at (87858): [<ffffffffaa67377e>] __irq_exit_rcu+0x17e/0x1d0
[  +0.000005] softirqs last disabled at (87849): [<ffffffffaa67377e>] __irq_exit_rcu+0x17e/0x1d0
[  +0.000005]
              other info that might help us debug this:
[  +0.000004]  Possible unsafe locking scenario:

[  +0.000003]        CPU0
[  +0.000004]        ----
[  +0.000003]   lock(&fence_drv->fence_list_lock);

v2:
  Drop fence_list_flags and use xa_lock_irqsave() flags parameter (Christian)
  Fix merge conflicts.

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>

drm/amd/display: Avoid flooding unnecessary info messages

It's expected that we'll encounter temporary exceptions
during aux transactions. Adjust logging from drm_info to
drm_dbg_dp to prevent flooding with unnecessary log messages.

Fixes: 3637e457eb00 ("drm/amd/display: Fix wrong handling for AUX_DEFER case")
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Wayne Lin <Wayne.Lin@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250513032026.838036-1-Wayne.Lin@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Documentation/gpu: Disambiguate SPI term

Documentation/userspace-api/media/glossary.rst:170: WARNING: duplicate term description of SPI, other instance in gpu/amdgpu/amdgpu-glossary

That's because SPI of amdgpu (Shader Processor Input) shares the same
global glossary term as SPI of media subsystem (which is Serial
Peripheral Interface Bus). Disambiguate the former from the latter to
fix the warning.

Note that adding context qualifiers in the term is strictly necessary
in order to make Sphinx happy.

Fixes: dd3d035a7838 ("Documentation/gpu: Add new entries to amdgpu glossary")
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Closes: https://lore.kernel.org/linux-next/20250509185845.60bf5e7b@canb.auug.org.au/
Signed-off-by: Bagas Sanjaya <bagasdotme@gmail.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add GFX 9.5.0 support for per-queue/pipe reset

This patch enables per-queue and per-pipe reset functionality for
GFX IP v9.5.0 when using MEC firmware version 21 (0x15) or later.

This change:
1. Refactors the pipe reset support check in gfx_v9_4_3_pipe_reset_support()
   to use the compute_supported_reset flags instead of hardcoding
   version checks.
2. Adds support for GFX9.5.0 (IP 9.5.0) with MEC firmware version >= 21
   to enable per-queue and per-pipe reset capabilities.

v2: Replaced mec version check with !!(adev->gfx.compute_supported_reset & AMDGPU_RESET_TYPE_PER_PIPE) (Lijo)

Signed-off-by: Asad Kamal <asad.kamal@amd.com>
Signed-off-by: Jesse Zhang <jesse.zhang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: set vram type for GC 9.5.0

Set vram type so we can take different actions according to the type.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: set flip bits for RAS bad pages

Make the code more general, user doesn't need to pay attention to the
detail of flip bits setting.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display/dc/irq: Remove duplications of hpd_ack function from IRQ

The major of dcn and dce irqs share a copy-pasted collection
of copy-pasted function, which is: hpd_ack.

This patch removes the multiple copy-pasted by moving them to
the irq_service.c and make the irq_service's
calls the functions implemented by the irq_service.c
instead.

The hpd_ack function is replaced by hpd0_ack and hpd1_ack, the
required constants are also added.

The changes were not tested on actual hardware. I am only able
to verify that the changes keep the code compileable and do my
best to look repeatedly if I am not actually changing any code.

Signed-off-by: Sebastian Aguilera Novoa <saguileran@ime.usp.br>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix null check of pipe_ctx->plane_state for update_dchubp_dpp

Similar to commit 6a057072ddd1 ("drm/amd/display: Fix null check for
pipe_ctx->plane_state in dcn20_program_pipe") that addresses a null
pointer dereference on dcn20_update_dchubp_dpp. This is the same
function hooked for update_dchubp_dpp in dcn401, with the same issue.
Fix possible null pointer deference on dcn401_program_pipe too.

Fixes: 63ab80d9ac0a ("drm/amd/display: DML2.1 Post-Si Cleanup")
Signed-off-by: Melissa Wen <mwen@igalia.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Modify the count method of defer error

The number of newly added de counts and the number of
newly added error addresses remain consistent

Signed-off-by: Ce Sun <cesun102@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: drop warning in event_interrupt_isr_v1*()

Commit ded8b3c36f17 ("drm/amdgpu: properly handle GC vs MM in amdgpu_vmid_mgr_init()")
enables all 16 vmids for MMHUB on GC 10 and newer for KGD since
there are no KFD resources using MMHUB.  With this change, KFD
starts seeing MMHUB vmids in it's range with no pasid set.  As
such there is no need to warn, we can just ignore those interrupts.

Fixes: aded8b3c36f1 ("drm/amdgpu: properly handle GC vs MM in amdgpu_vmid_mgr_init()")
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Log RAS errors during load

During driver load, RAS event manager may not be initialized. This will
cause any ATHUB event during driver load to be skipped in dmesg log. Log
the error in dmesg log for easier diagnosis.

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix user queue deadlock by reordering mutex locking

This resolves a deadlock between user queue management and GPU reset
paths by enforcing consistent lock ordering.

The deadlock occurred when:

1. Process exit path (amdgpu_userq_mgr_fini) would:
   - Take uqm->userq_mutex
   - Then try to take adev->userq_mutex for list operations

2. GPU reset path (amdgpu_userq_pre_reset) would:
   - Take adev->userq_mutex first (for list traversal)
   - Then take uqm->userq_mutex

The solution establishes a strict top-down locking order:
1. Always take adev->userq_mutex before any uqm->userq_mutex
2. Maintain this order consistently across all code paths

Changes made:
- Reordered locking in amdgpu_userq_mgr_fini() to take device lock first
- Kept existing proper order in amdgpu_userq_pre_reset()
- Simplified the fini flow by removing redundant operations

This prevents circular dependencies while maintaining thread safety
during both normal operation and GPU reset scenarios.

Fixes: 4ce60dbada96 ("drm/amdgpu: store userq_managers in a list in adev")
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Arvind Yadav <Arvind.Yadav@amd.com>
Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix the kernel panic caused by RAS records exceed threshold

kernel panic caused by RAS records exceeding the threshold when load driver
specifying RMA(bad_page_threshold=128)

1.Fix the warnings caused by disabling the interrupt source
before it was enabled
2.Fix kernel panic when xcp sysfs is not initialized,null pointer
appears during fini
3.Fix the memory leak caused by the device's early exit due to rma

The first reason:
[ 2744.246650] ------------[ cut here ]------------
[ 2744.246651] WARNING: CPU: 0 PID: 289 at /tmp/amd.BkfTLqYV/amd/amdgpu/amdgpu_irq.c:635 amdgpu_irq_put.cold+0x42/0x6e [amdgpu]
[ 2744.247108] Modules linked in: amdgpu(OE+) amddrm_ttm_helper(OE) amdttm(OE) amdxcp(OE) amddrm_buddy(OE) amddrm_exec(OE) amd_sched(OE) amdkcl(OE) xt_conntrack nft_chain_nat xt_MASQUERADE nf_nat nf_conntrack_netlink nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 xfrm_user xfrm_algo nft_counter xt_addrtype nft_compat nf_tables nfnetlink br_netfilter bridge stp llc overlay binfmt_misc intel_rapl_msr intel_rapl_common i10nm_edac nfit x86_pkg_temp_thermal intel_powerclamp coretemp ipmi_ssif kvm_intel nls_iso8859_1 kvm rapl isst_if_mbox_pci isst_if_mmio pmt_telemetry pmt_crashlog isst_if_common pmt_class mei_me mei acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_power_meter acpi_pad mac_hid sch_fq_codel dm_multipath scsi_dh_rdac scsi_dh_emc scsi_dh_alua msr ramoops reed_solomon pstore_blk pstore_zone efi_pstore ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath
[ 2744.247167]  linear mlx5_ib ib_uverbs ib_core ast i2c_algo_bit drm_vram_helper drm_ttm_helper ttm drm_kms_helper crct10dif_pclmul syscopyarea crc32_pclmul ghash_clmulni_intel mlx5_core sysfillrect sysimgblt aesni_intel mlxfw fb_sys_fops psample cec crypto_simd cryptd rc_core i2c_i801 nvme xhci_pci tls intel_pmt drm pci_hyperv_intf nvme_core i2c_smbus i2c_ismt xhci_pci_renesas wmi pinctrl_emmitsburg
[ 2744.247194] CPU: 0 PID: 289 Comm: kworker/0:1 Tainted: G           OE     5.15.0-70-generic #77-Ubuntu
[ 2744.247197] Hardware name: Microsoft C278A/C278A, BIOS C2789.5.BS.1C23.AG.2 11/21/2024
[ 2744.247198] Workqueue: events work_for_cpu_fn
[ 2744.247206] RIP: 0010:amdgpu_irq_put.cold+0x42/0x6e [amdgpu]
[ 2744.247634] Code: 79 7f ff 44 89 ee 48 c7 c7 4d 5a 42 c2 89 55 d4 e8 90 09 bc bf 8b 55 d4 4c 89 e6 4c 89 ff e8 3c 76 7f ff 8b 55 d4 84 c0 75 07 <0f> 0b e9 95 79 7f ff 49 03 5c 24 08 f0 ff 0b 75 13 4c 89 e6 4c 89
[ 2744.247636] RSP: 0018:ffa0000019e27cb0 EFLAGS: 00010246
[ 2744.247639] RAX: 0000000000000000 RBX: 0000000000000000 RCX: ff11000150fa87c0
[ 2744.247641] RDX: 0000000000000000 RSI: ffffffffc2222430 RDI: ff1100019f200000
[ 2744.247642] RBP: ffa0000019e27ce0 R08: 0000000000000003 R09: ffffffffffe41a08
[ 2744.247643] R10: 0000000000ffff0a R11: 0000000000000001 R12: ff1100019f22ce60
[ 2744.247644] R13: 0000000000000000 R14: 00000000ffffffea R15: ff1100019f200000
[ 2744.247645] FS:  0000000000000000(0000) GS:ff11007e7e400000(0000) knlGS:0000000000000000
[ 2744.247647] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 2744.247649] CR2: 00007f3d2002819c CR3: 0000000006810003 CR4: 0000000000771ef0
[ 2744.247650] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 2744.247651] DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400
[ 2744.247652] PKRU: 55555554
[ 2744.247653] Call Trace:
[ 2744.247654]  <TASK>
[ 2744.247656]  sdma_v4_4_2_hw_fini+0x7a/0xc0 [amdgpu]
[ 2744.247997]  ? vcn_v4_0_3_hw_fini+0x5f/0xa0 [amdgpu]
[ 2744.248336]  amdgpu_ip_block_hw_fini+0x31/0x61 [amdgpu]
[ 2744.248776]  amdgpu_device_fini_hw+0x3bb/0x47b [amdgpu]
[ 2744.249197]  ? blocking_notifier_chain_unregister+0x56/0xb0
[ 2744.249202]  amdgpu_driver_unload_kms+0x51/0x60 [amdgpu]
[ 2744.249482]  amdgpu_driver_load_kms.cold+0x18/0x2e [amdgpu]
[ 2744.249913]  amdgpu_pci_probe+0x23e/0x590 [amdgpu]
[ 2744.250187]  local_pci_probe+0x48/0x90
[ 2744.250191]  work_for_cpu_fn+0x17/0x30
[ 2744.250196]  process_one_work+0x228/0x3d0
[ 2744.250198]  worker_thread+0x223/0x420
[ 2744.250200]  ? process_one_work+0x3d0/0x3d0
[ 2744.250201]  kthread+0x127/0x150
[ 2744.250204]  ? set_kthread_struct+0x50/0x50
[ 2744.250207]  ret_from_fork+0x1f/0x30
[ 2744.250212]  </TASK>
[ 2744.250213] ---[ end trace 488c997a88508bc3 ]---

The second reason:
[ 5139.303446] Memory manager not clean during takedown.
[ 5139.303509] WARNING: CPU: 145 PID: 117699 at drivers/gpu/drm/drm_mm.c:998 drm_mm_takedown+0x27/0x30 [drm]
[ 5139.303542] Modules linked in: amdgpu(OE+) amddrm_ttm_helper(OE) amdttm(OE) amdxcp(OE) amddrm_buddy(OE) amddrm_exec(OE) amd_sched(OE) amdkcl(OE) xt_conntrack nft_chain_nat xt_MASQUERADE nf_nat nf_conntrack_netlink nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 xfrm_user xfrm_algo nft_counter xt_addrtype nft_compat nf_tables nfnetlink br_netfilter bridge stp llc overlay intel_rapl_msr intel_rapl_common i10nm_edac nfit x86_pkg_temp_thermal intel_powerclamp coretemp ipmi_ssif kvm_intel binfmt_misc kvm nls_iso8859_1 rapl isst_if_mbox_pci pmt_telemetry pmt_crashlog isst_if_mmio pmt_class isst_if_common mei_me mei acpi_ipmi ipmi_si ipmi_devintf ipmi_msghandler acpi_pad acpi_power_meter mac_hid sch_fq_codel dm_multipath scsi_dh_rdac scsi_dh_emc scsi_dh_alua msr ramoops reed_solomon pstore_blk pstore_zone efi_pstore ip_tables x_tables autofs4 btrfs blake2b_generic zstd_compress raid10 raid456 async_raid6_recov async_memcpy async_pq async_xor async_tx xor raid6_pq libcrc32c raid1 raid0 multipath
[ 5139.303572]  linear mlx5_ib ib_uverbs ib_core crct10dif_pclmul ast crc32_pclmul i2c_algo_bit ghash_clmulni_intel aesni_intel crypto_simd drm_vram_helper cryptd drm_ttm_helper mlx5_core ttm drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops cec rc_core mlxfw psample intel_pmt nvme xhci_pci drm tls i2c_i801 pci_hyperv_intf nvme_core i2c_smbus i2c_ismt xhci_pci_renesas wmi pinctrl_emmitsburg [last unloaded: amdkcl]
[ 5139.303588] CPU: 145 PID: 117699 Comm: modprobe Tainted: G     U     OE     5.15.0-70-generic #77-Ubuntu
[ 5139.303590] Hardware name: Microsoft C278A/C278A, BIOS C2789.5.BS.1C23.AG.2 11/21/2024
[ 5139.303591] RIP: 0010:drm_mm_takedown+0x27/0x30 [drm]
[ 5139.303605] Code: cc 66 90 0f 1f 44 00 00 48 8b 47 38 48 83 c7 38 48 39 f8 75 05 c3 cc cc cc cc 55 48 c7 c7 18 d0 10 c0 48 89 e5 e8 5a bc c3 c1 <0f> 0b 5d c3 cc cc cc cc 90 0f 1f 44 00 00 55 b9 15 00 00 00 48 89
[ 5139.303607] RSP: 0018:ffa00000325c3940 EFLAGS: 00010286
[ 5139.303608] RAX: 0000000000000000 RBX: ff1100012f5cecb0 RCX: 0000000000000027
[ 5139.303609] RDX: ff11007e7fa60588 RSI: 0000000000000001 RDI: ff11007e7fa60580
[ 5139.303610] RBP: ffa00000325c3940 R08: 0000000000000003 R09: fffffffff00c2b78
[ 5139.303610] R10: 000000000000002b R11: 0000000000000001 R12: ff1100012f5cec00
[ 5139.303611] R13: ff1100012138f068 R14: 0000000000000000 R15: ff1100012f5cec90
[ 5139.303611] FS:  00007f42ffca0000(0000) GS:ff11007e7fa40000(0000) knlGS:0000000000000000
[ 5139.303612] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 5139.303613] CR2: 00007f23d945ab68 CR3: 00000001212ce005 CR4: 0000000000771ee0
[ 5139.303614] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 5139.303615] DR3: 0000000000000000 DR6: 00000000fffe07f0 DR7: 0000000000000400
[ 5139.303615] PKRU: 55555554
[ 5139.303616] Call Trace:
[ 5139.303617]  <TASK>
[ 5139.303619]  amdttm_range_man_fini_nocheck+0xfe/0x1c0 [amdttm]
[ 5139.303625]  amdgpu_ttm_fini+0x2ed/0x390 [amdgpu]
[ 5139.303800]  amdgpu_bo_fini+0x27/0xc0 [amdgpu]
[ 5139.303959]  gmc_v9_0_sw_fini+0x63/0x90 [amdgpu]
[ 5139.304144]  amdgpu_device_fini_sw+0x125/0x6a0 [amdgpu]
[ 5139.304302]  amdgpu_driver_release_kms+0x16/0x30 [amdgpu]
[ 5139.304455]  devm_drm_dev_init_release+0x4a/0x80 [drm]
[ 5139.304472]  devm_action_release+0x12/0x20
[ 5139.304476]  release_nodes+0x3d/0xb0
[ 5139.304478]  devres_release_all+0x9b/0xd0
[ 5139.304480]  really_probe+0x11d/0x420
[ 5139.304483]  __driver_probe_device+0x119/0x190
[ 5139.304485]  driver_probe_device+0x23/0xc0
[ 5139.304487]  __driver_attach+0xf7/0x1f0
[ 5139.304489]  ? __device_attach_driver+0x140/0x140
[ 5139.304491]  bus_for_each_dev+0x7c/0xd0
[ 5139.304493]  driver_attach+0x1e/0x30
[ 5139.304494]  bus_add_driver+0x148/0x220
[ 5139.304496]  driver_register+0x95/0x100
[ 5139.304498]  __pci_register_driver+0x68/0x70
[ 5139.304500]  amdgpu_init+0xbc/0x1000 [amdgpu]
[ 5139.304655]  ? 0xffffffffc0b8f000
[ 5139.304657]  do_one_initcall+0x46/0x1e0
[ 5139.304659]  ? kmem_cache_alloc_trace+0x19e/0x2e0
[ 5139.304663]  do_init_module+0x52/0x260
[ 5139.304665]  load_module+0xb2b/0xbc0
[ 5139.304667]  __do_sys_finit_module+0xbf/0x120
[ 5139.304669]  __x64_sys_finit_module+0x18/0x20
[ 5139.304670]  do_syscall_64+0x59/0xc0
[ 5139.304673]  ? exit_to_user_mode_prepare+0x37/0xb0
[ 5139.304676]  ? syscall_exit_to_user_mode+0x27/0x50
[ 5139.304678]  ? __x64_sys_mmap+0x33/0x50
[ 5139.304680]  ? do_syscall_64+0x69/0xc0
[ 5139.304681]  entry_SYSCALL_64_after_hwframe+0x61/0xcb
[ 5139.304684] RIP: 0033:0x7f42ffdbf88d
[ 5139.304686] Code: 5b 41 5c c3 66 0f 1f 84 00 00 00 00 00 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 73 b5 0f 00 f7 d8 64 89 01 48
[ 5139.304687] RSP: 002b:00007ffcb7427158 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
[ 5139.304688] RAX: ffffffffffffffda RBX: 000055ce8b8f3150 RCX: 00007f42ffdbf88d
[ 5139.304689] RDX: 0000000000000000 RSI: 000055ce8b8f9a70 RDI: 000000000000000a
[ 5139.304690] RBP: 0000000000040000 R08: 0000000000000000 R09: 0000000000000011
[ 5139.304690] R10: 000000000000000a R11: 0000000000000246 R12: 000055ce8b8f9a70
[ 5139.304691] R13: 000055ce8b8f2ec0 R14: 000055ce8b8f2ab0 R15: 000055ce8b8f9aa0
[ 5139.304692]  </TASK>
[ 5139.304693] ---[ end trace 8536b052f7883003 ]---

Signed-off-by: Ce Sun <cesun102@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgu: get RAS retire flip bits for new type of HBM

Get RAS retire flip bits for HBM with different types in various NPS modes.
Also set flip row bit and MCA R13 bit in PA in different NPS modes.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: implement get_retire_flip_bits for UMC v12

The RAS bad page retire flip bits can be set per vram type,
vram vendor and NPS mode.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add get_retire_flip_bits for UMC

Add the general interface to get flip bits for RAS bad page retirement.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: add vcn v5_0_0 ip headers

Add vcn v5_0_0 register offset and shift masks
header files
Only include the registers required for MMSCH
initialization

Signed-off-by: fanhuang <FangSheng.Huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: adjust high bits for RAS retired page

Per UMC address conversion algorithm, the high row bits of UMC MCA
address are changed when they're converted into normalized address
on specific ASICs.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd: add definition for new memory type

Support new version of HBM.

Signed-off-by: Tao Zhou <tao.zhou1@amd.com>
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Refine RAS bad page records counting and parsing in eeprom V3

there is only MCA records in V3, no need to care about PA records.
recalculate the value of ras_num_bad_pages when parsing failed and
go on with the left records instead of quit.

Signed-off-by: ganglxie <ganglxie@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Promote DC to 3.2.333

Summary

* Refactor DMI quirks
* Fix link-off issue triggered by quick unplug/replug
* Fix race condition when submitting DMUB commands
* Correct reply value when AUX Write incomplete
* Backup / restore plane config only on update

Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Add early 8b/10b channel equalization test pattern sequence

[WHY]
Early EQ pattern sequence is required for some LTTPR + old dongle
combinations.

[HOW]
If DP_EARLY_8B10B_TPS2 chip cap is set, this new sequence programs phy
to output TPS2 before initiating link training and writes TPS1 to
LTTPR training pattern register as instructed by vendor.

Add function to get embedded LTTPR target address offset.

Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Michael Strauss <michael.strauss@amd.com>
Signed-off-by: TungYu Lu <tungyu.lu@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Program triplebuffer on all pipes

[WHY]
Triplebuffer should be programmed on all pipes.
Some code assumed it only needed to be called on top
pipe, but as the HWSS function does not account
for that, it must be called on every pipe.

[HOW]
Remove condition to not program triplebuffer
on non-top/next pipe. Call the function
unconditionally on all pipes.

Reviewed-by: Dillon Varone <dillon.varone@amd.com>
Reviewed-by: Aric Cyr <aric.cyr@amd.com>
Signed-off-by: Sung Lee <Sung.Lee@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: [FW Promotion] Release 0.1.10.0

Refactoring some IPS and panel replay structs

Acked-by: Tom Chung <chiahsuan.chung@amd.com>
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: disable EASF narrow filter sharpening

[Why & How]
Default should be 1 to disable EASF narrow filter sharpening.

Reviewed-by: Alvin Lee <alvin.lee2@amd.com>
Signed-off-by: Samson Tam <Samson.Tam@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Return the exact value for debugging

[Why]
It's unnecessary to set operation_result as invalid reply when
p_notify->result != AUX_RET_SUCCESS.

[How]
Set operation_result as p_notify->result to better understand
the reason for the error

Reviewed-by: Ray Wu <ray.wu@amd.com>
Signed-off-by: Wayne Lin <Wayne.Lin@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Restructure DMI quirks

[Why]
DMI quirks are relatively big code that makes amdgpu_dm 200 lines
larger.

[How]
Move DMI quirks into a dedicated source file and make all quirks
variables for `struct amdgpu_display_manager`.

Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: check stream id dml21 wrapper to get plane_id

[Why & How]
Fix a false positive warning which occurs due to lack of correct checks
when querying plane_id in DML21. This fixes the warning when performing a
mode1 reset (cat /sys/kernel/debug/dri/1/amdgpu_gpu_recover):

[   35.751250] WARNING: CPU: 11 PID: 326 at /tmp/amd.PHpyAl7v/amd/amdgpu/../display/dc/dml2/dml2_dc_resource_mgmt.c:91 dml2_map_dc_pipes+0x243d/0x3f40 [amdgpu]
[   35.751434] Modules linked in: amdgpu(OE) amddrm_ttm_helper(OE) amdttm(OE) amddrm_buddy(OE) amdxcp(OE) amddrm_exec(OE) amd_sched(OE) amdkcl(OE) drm_suballoc_helper drm_ttm_helper ttm drm_display_helper cec rc_core i2c_algo_bit rfcomm qrtr cmac algif_hash algif_skcipher af_alg bnep amd_atl intel_rapl_msr intel_rapl_common snd_hda_codec_hdmi snd_hda_intel edac_mce_amd snd_intel_dspcfg snd_intel_sdw_acpi snd_hda_codec kvm_amd snd_hda_core snd_hwdep snd_pcm kvm snd_seq_midi snd_seq_midi_event snd_rawmidi crct10dif_pclmul polyval_clmulni polyval_generic btusb ghash_clmulni_intel sha256_ssse3 btrtl sha1_ssse3 snd_seq btintel aesni_intel btbcm btmtk snd_seq_device crypto_simd sunrpc cryptd bluetooth snd_timer ccp binfmt_misc rapl snd i2c_piix4 wmi_bmof gigabyte_wmi k10temp i2c_smbus soundcore gpio_amdpt mac_hid sch_fq_codel msr parport_pc ppdev lp parport efi_pstore nfnetlink dmi_sysfs ip_tables x_tables autofs4 hid_generic usbhid hid crc32_pclmul igc ahci xhci_pci libahci xhci_pci_renesas video wmi
[   35.751501] CPU: 11 UID: 0 PID: 326 Comm: kworker/u64:9 Tainted: G           OE      6.11.0-21-generic #21~24.04.1-Ubuntu
[   35.751504] Tainted: [O]=OOT_MODULE, [E]=UNSIGNED_MODULE
[   35.751505] Hardware name: Gigabyte Technology Co., Ltd. X670E AORUS PRO X/X670E AORUS PRO X, BIOS F30 05/22/2024
[   35.751506] Workqueue: amdgpu-reset-dev amdgpu_debugfs_reset_work [amdgpu]
[   35.751638] RIP: 0010:dml2_map_dc_pipes+0x243d/0x3f40 [amdgpu]
[   35.751794] Code: 6d 0c 00 00 8b 84 24 88 00 00 00 41 3b 44 9c 20 0f 84 fc 07 00 00 48 83 c3 01 48 83 fb 06 75 b3 4c 8b 64 24 68 4c 8b 6c 24 40 <0f> 0b b8 06 00 00 00 49 8b 94 24 a0 49 00 00 89 c3 83 f8 07 0f 87
[   35.751796] RSP: 0018:ffffbfa3805d7680 EFLAGS: 00010246
[   35.751798] RAX: 0000000000010000 RBX: 0000000000000006 RCX: 0000000000000000
[   35.751799] RDX: 0000000000000000 RSI: 0000000000000005 RDI: 0000000000000000
[   35.751800] RBP: ffffbfa3805d78f0 R08: 0000000000000000 R09: 0000000000000000
[   35.751801] R10: 0000000000000000 R11: 0000000000000000 R12: ffffbfa383249000
[   35.751802] R13: ffffa0e68f280000 R14: ffffbfa383249658 R15: 0000000000000000
[   35.751803] FS:  0000000000000000(0000) GS:ffffa0edbe580000(0000) knlGS:0000000000000000
[   35.751804] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   35.751805] CR2: 00005d847ef96c58 CR3: 000000041de3e000 CR4: 0000000000f50ef0
[   35.751806] PKRU: 55555554
[   35.751807] Call Trace:
[   35.751810]  <TASK>
[   35.751816]  ? show_regs+0x6c/0x80
[   35.751820]  ? __warn+0x88/0x140
[   35.751822]  ? dml2_map_dc_pipes+0x243d/0x3f40 [amdgpu]
[   35.751964]  ? report_bug+0x182/0x1b0
[   35.751969]  ? handle_bug+0x6e/0xb0
[   35.751972]  ? exc_invalid_op+0x18/0x80
[   35.751974]  ? asm_exc_invalid_op+0x1b/0x20
[   35.751978]  ? dml2_map_dc_pipes+0x243d/0x3f40 [amdgpu]
[   35.752117]  ? math_pow+0x48/0xa0 [amdgpu]
[   35.752256]  ? srso_alias_return_thunk+0x5/0xfbef5
[   35.752260]  ? math_pow+0x48/0xa0 [amdgpu]
[   35.752400]  ? srso_alias_return_thunk+0x5/0xfbef5
[   35.752403]  ? math_pow+0x11/0xa0 [amdgpu]
[   35.752524]  ? srso_alias_return_thunk+0x5/0xfbef5
[   35.752526]  ? core_dcn4_mode_programming+0xe4d/0x20d0 [amdgpu]
[   35.752663]  ? srso_alias_return_thunk+0x5/0xfbef5
[   35.752669]  dml21_validate+0x3d4/0x980 [amdgpu]

Reviewed-by: Austin Zheng <austin.zheng@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: fix link_set_dpms_off multi-display MST corner case

[Why & How]
When MST config is unplugged/replugged too quickly, it can potentially
result in a scenario where previous DC state has not been reset before
the HPD link detection sequence begins. In this case, driver will
disable the streams/link prior to re-enabling the link for link
training.

There is a bug in the current logic that does not account for the fact
that current_state can be released and cleared prior to swapping to a
new state (resulting in the pipe_ctx stream pointers to be cleared) in
between disabling streams.

To resolve this, cache the original streams prior to committing any
stream updates.

Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: George Shen <george.shen@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Defer BW-optimization-blocked DRR adjustments

[Why & How]
Instead of dropping DRR updates, defer them. This fixes issues where
monitor continues to see incorrect refresh rate after VRR was turned off
by userspace.

Fixes: 32953485c558 ("drm/amd/display: Do not update DRR while BW optimizations pending")
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3546
Reviewed-by: Sun peng Li <sunpeng.li@amd.com>
Signed-off-by: John Olender <john.olender@gmail.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Revert: "drm/amd/display: Enable urgent latency adjustment on DCN35"

This reverts commit 756c85e4d0dd ("drm/amd/display: Enable urgent latency adjustment on DCN35")

Reason for revert: Negative power impact.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Gabe Teeger <Gabe.Teeger@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Fix race in dmub_srv_wait_for_pending

[WHY]
If commands are being submitted to DMCUB while concurrently waiting for
pending commands to complete, rptr and wptr may never match again, and
reported command count will not update.

[HOW]
Modify dmub_srv_wait_for_pending to constantly check wptr and rptr
match, and update inbox status whenever a message is sent to avoid the
race and determine message completion or idle as quickly as possible.

Reviewed-by: Nicholas Kazlauskas <nicholas.kazlauskas@amd.com>
Signed-off-by: Dillon Varone <dillon.varone@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Correct the reply value when AUX write incomplete

[Why]
Now forcing aux->transfer to return 0 when incomplete AUX write is
inappropriate. It should return bytes have been transferred.

[How]
aux->transfer is asked not to change original msg except reply field of
drm_dp_aux_msg structure. Copy the msg->buffer when it's write request,
and overwrite the first byte when sink reply 1 byte indicating partially
written byte number. Then we can return the correct value without
changing the original msg.

Fixes: 3637e457eb00 ("drm/amd/display: Fix wrong handling for AUX_DEFER case")
Cc: Mario Limonciello <mario.limonciello@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Ray Wu <ray.wu@amd.com>
Signed-off-by: Wayne Lin <Wayne.Lin@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Backup and restore plane configuration only on update

[WHY&HOW]
When backing up and restoring plane states for minimal transition
cases, only configuration should be backed up and restored. Information
only relevant to the object/allocation (like refcount) should be
excluded. Also move this interface to dc_plane.h.

Reviewed-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Signed-off-by: Dillon Varone <Dillon.Varone@amd.com>
Signed-off-by: Ray Wu <ray.wu@amd.com>
Tested-by: Daniel Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix incorrect MALL size for GFX1151

On GFX1151, the reported MALL cache size reflects only
half of its actual size; this adjustment corrects the discrepancy.

Signed-off-by: Tim Huang <tim.huang@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Yifan Zhang <yifan1.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: Remove remainder of mode2_reset_is_support

The previous patch removed smu_mode2_reset_is_support()
which was the only function to call through the mode2_reset_is_support()
method pointer.

Remove the remaining functions that were assigned to it
and the pointer itself.

See discussion at:
https://lore.kernel.org/all/DM4PR12MB5165D85BD85BC8FC8BF7A3B48E88A@DM4PR12MB5165.namprd12.prod.outlook.com/

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm: Remove unused smu_mode2_reset_is_support

smu_mode2_reset_is_support() was added in 2020 by
commit 5c03e5843e6b ("drm/amdgpu:add smu mode1/2 support for aldebaran")
but has remained unused.

See discussion at:
https://lore.kernel.org/all/DM4PR12MB5165D85BD85BC8FC8BF7A3B48E88A@DM4PR12MB5165.namprd12.prod.outlook.com/

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm/smu13: Remove unused smu_v13_0_init_display_count

smu_v13_0_init_display_count() was added in 2020 by
commit c05d1c401572 ("drm/amd/swsmu: add aldebaran smu13 ip support (v3)")
but is unused.

See discussion on:
https://lore.kernel.org/all/DM4PR12MB5165D85BD85BC8FC8BF7A3B48E88A@DM4PR12MB5165.namprd12.prod.outlook.com/
that it really isn't neede.

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix amdgpu_userq_wait_ioctl() warn missing error code 'r'

To resolve the warning regarding the missing error code 'r' in
amdgpu_userq_wait_ioctl(), assign the value 'r = -EINVAL'.

Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/r/202505080458.rnV8YfiY-lkp@intel.com/
Cc: Alex Deucher <alexander.deucher@amd.com>
Cc: Christian König <christian.koenig@amd.com>
Cc: Arunpravin Paneer Selvam <Arunpravin.PaneerSelvam@amd.com>
Reviewed-by: Sunil Khatri <sunil.khatri@amd.com>
Signed-off-by: Arvind Yadav <Arvind.Yadav@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Fix NULL dereference in amdgpu_userq_restore_worker

Switch cancel_delayed_work() to cancel_delayed_work_sync() to ensure
the delayed work has finished executing before proceeding with
resource cleanup. This prevents a potential use-after-free or
NULL dereference if the resume_work is still running during finalization.

BUG: kernel NULL pointer dereference, address: 0000000000000140
[  +0.000050] #PF: supervisor read access in kernel mode
[  +0.000019] #PF: error_code(0x0000) - not-present page
[  +0.000021] PGD 0 P4D 0
[  +0.000015] Oops: Oops: 0000 [#1] PREEMPT SMP NOPTI
[  +0.000021] CPU: 17 UID: 0 PID: 196299 Comm: kworker/17:0 Tainted: G     U             6.14.0-org-staging #1
[  +0.000032] Tainted: [U]=USER
[  +0.000015] Hardware name: Gigabyte Technology Co., Ltd. X570 AORUS ELITE/X570 AORUS ELITE, BIOS F39 03/22/2024
[  +0.000029] Workqueue: events amdgpu_userq_restore_worker [amdgpu]
[  +0.000426] RIP: 0010:drm_exec_lock_obj+0x32/0x210 [drm_exec]
[  +0.000025] Code: e5 41 57 41 56 41 55 49 89 f5 41 54 49 89 fc 48 83 ec 08 4c 8b 77 30 4d 85 f6 0f 85 c0 00 00 00 4c 8d 7f 08 48 39 77 38 74 54 <49> 8b bd f8 00 00 00 4c 89 fe 41 f6 04 24 01 75 3c e8 08 50 bc e0
[  +0.000046] RSP: 0018:ffffab1b04da3ce8 EFLAGS: 00010297
[  +0.000020] RAX: 0000000000000001 RBX: ffff930cc60e4bc0 RCX: 0000000000000000
[  +0.000025] RDX: 0000000000000004 RSI: 0000000000000048 RDI: ffffab1b04da3d88
[  +0.000028] RBP: ffffab1b04da3d10 R08: ffff930cc60e4000 R09: 0000000000000000
[  +0.000022] R10: ffffab1b04da3d18 R11: 0000000000000001 R12: ffffab1b04da3d88
[  +0.000023] R13: 0000000000000048 R14: 0000000000000000 R15: ffffab1b04da3d90
[  +0.000023] FS:  0000000000000000(0000) GS:ffff9313dea80000(0000) knlGS:0000000000000000
[  +0.000024] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  +0.000021] CR2: 0000000000000140 CR3: 000000018351a000 CR4: 0000000000350ef0
[  +0.000025] Call Trace:
[  +0.000018]  <TASK>
[  +0.000015]  ? show_regs+0x69/0x80
[  +0.000022]  ? __die+0x25/0x70
[  +0.000019]  ? page_fault_oops+0x15d/0x510
[  +0.000024]  ? do_user_addr_fault+0x312/0x690
[  +0.000024]  ? sched_clock_cpu+0x10/0x1a0
[  +0.000028]  ? exc_page_fault+0x78/0x1b0
[  +0.000025]  ? asm_exc_page_fault+0x27/0x30
[  +0.000024]  ? drm_exec_lock_obj+0x32/0x210 [drm_exec]
[  +0.000024]  drm_exec_prepare_obj+0x21/0x60 [drm_exec]
[  +0.000021]  amdgpu_vm_lock_pd+0x22/0x30 [amdgpu]
[  +0.000266]  amdgpu_userq_validate_bos+0x6c/0x320 [amdgpu]
[  +0.000333]  amdgpu_userq_restore_worker+0x4a/0x120 [amdgpu]
[  +0.000316]  process_one_work+0x189/0x3c0
[  +0.000021]  worker_thread+0x2a4/0x3b0
[  +0.000022]  kthread+0x109/0x220
[  +0.000018]  ? __pfx_worker_thread+0x10/0x10
[  +0.000779]  ? _raw_spin_unlock_irq+0x1f/0x40
[  +0.000560]  ? __pfx_kthread+0x10/0x10
[  +0.000543]  ret_from_fork+0x3c/0x60
[  +0.000507]  ? __pfx_kthread+0x10/0x10
[  +0.000515]  ret_from_fork_asm+0x1a/0x30
[  +0.000515]  </TASK>

v2: Replace cancel_delayed_work() to cancel_delayed_work_sync()
    in amdgpu_userq_destroy() and amdgpu_userq_evict().

Cc: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Sunil Khatri <sunil.khatri@amd.com>
Signed-off-by: Arvind Yadav <arvind.yadav@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: csa unmap use uninterruptible lock

After process exit to unmap csa and free GPU vm, if signal is accepted
and then waiting to take vm lock is interrupted and return, it causes
memory leaking and below warning backtrace.

Change to use uninterruptible wait lock fix the issue.

WARNING: CPU: 69 PID: 167800 at amd/amdgpu/amdgpu_kms.c:1525
 amdgpu_driver_postclose_kms+0x294/0x2a0 [amdgpu]
 Call Trace:
  <TASK>
  drm_file_free.part.0+0x1da/0x230 [drm]
  drm_close_helper.isra.0+0x65/0x70 [drm]
  drm_release+0x6a/0x120 [drm]
  amdgpu_drm_release+0x51/0x60 [amdgpu]
  __fput+0x9f/0x280
  ____fput+0xe/0x20
  task_work_run+0x67/0xa0
  do_exit+0x217/0x3c0
  do_group_exit+0x3b/0xb0
  get_signal+0x14a/0x8d0
  arch_do_signal_or_restart+0xde/0x100
  exit_to_user_mode_loop+0xc1/0x1a0
  exit_to_user_mode_prepare+0xf4/0x100
  syscall_exit_to_user_mode+0x17/0x40
  do_syscall_64+0x69/0xc0

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>

drm/amd/display: use drm_dbg_driver() in amdgpu_dm.c

Replace all use of DRM_DEBUG_DRIVER in amdgpu_dm.c with
drm_dbg_driver(). The latter prints the instance of the drm device
associated with the error which would helpful in debugging scenarios
involving multiple GPUs

Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Harry Wentland <harry.wentland@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Merge tag 'amd-drm-next-6.16-2025-05-09' of https://gitlab.freedesktop.org/agd5f/linux into drm-next

amd-drm-next-6.16-2025-05-09:

amdgpu:
- IPS fixes
- DSC cleanup
- DC Scaling updates
- DC FP fixes
- Fused I2C-over-AUX updates
- SubVP fixes
- Freesync fix
- DMUB AUX fixes
- VCN fix
- Hibernation fixes
- HDP fixes
- DCN 2.1 fixes
- DPIA fixes
- DMUB updates
- Use drm_file_err in amdgpu
- Enforce isolation updates
- Use new dma_fence helpers
- USERQ fixes
- Documentation updates
- Misc code cleanups
- SR-IOV updates
- RAS updates
- PSP 12 cleanups

amdkfd:
- Update error messages for SDMA
- Userptr updates

drm:
- Add drm_file_err function

dma-buf:
- Add a helper to sort and deduplicate dma_fence arrays

From: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250509230951.3871914-1-alexander.deucher@amd.com
Signed-off-by: Dave Airlie <airlied@redhat.com>

Merge tag 'drm-misc-next-2025-05-08' of https://gitlab.freedesktop.org/drm/misc/kernel into drm-next

drm-misc-next for v6.16-rc1:

Cross-subsystem Changes:
- Change vsprintf %p4cn to %p4chR, remove %p4cn.

Core Changes:
- Documentation updates (fb rendering, actual_brightness)

Driver Changes:
- Small fixes to appletbdrm, panthor, st7571-i2c, rockchip, renesas,
  panic handler, gpusvm, vkms, panel timings.
- Add AUO B140QAN08.H, BOE NE140WUM-N6S, CSW MNE007QS3-8, BOE TD4320 panels.
- Convert rk3066_hdmi to bridge driver.
- Improve HPD on anx7625.
- Speed up loading tegra firmware, and other small fixes to tegra & host1x.

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Link: https://lore.kernel.org/r/5428be12-fc08-4e28-8c5f-85d73b8a7e04@linux.intel.com

Merge tag 'drm-intel-next-2025-05-08' of https://gitlab.freedesktop.org/drm/i915/kernel into drm-next

Non-display related:
- Fix undefined reference to `intel_pxp_gsccs_is_ready_for_sessions'

Display related:
- More work towards display separation (Jani)
- Stop writing VRR_CTL_IGN_MAX_SHIFT for MTL onwards (Jouni)
- DSC checks for 3 engines (Ankit)
- Add link rate and lane count to i915_display_info (Khaled)
- PSR fixes and workaround for underrun on idle (Jouni)
- LOBF enablement and ALMP fixes (Animesh)
- Clean up VGA plane handling (Ville)
- Use an intel_connector pointer everywhere (Imre)
- Fix warning for coffeelake on SunrisePoint PCH (Jiajia)
- Rework/Correction on minimum hblank calculation (Arun)
- Dmesg clean up (Jani)
- Add a couple of simple display workarounds (Ankit, Vinod)
- Refactor HDCP GSC (Jani)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
Link: https://lore.kernel.org/r/aByyL3bEufPu79OM@intel.com

Merge tag 'drm-xe-next-2025-05-08' of https://gitlab.freedesktop.org/drm/xe/kernel into drm-next

UAPI Changes:
- Expose PCIe link downgrade attributes (Raag)

Cross-subsystem Changes:

Core Changes:
- gpusvm has_dma_mapping fix (Dafna)

Driver Changes:
- Forcewake hold fix (Tejas)
- Fix guc_info debugfs for VFs (Daniele)
- Fix devcoredump chunk alignment calculation (Arnd)
- Don't print timedout job message on killed exec queues (Matt Brost)
- Don't flush the GSC worker from the reset path (Daniele)
- Use copy_from_user() instead of __copy_from_user() (Harish)
- Only flush SVM garbage collector if CONFIG_DRM_XE_GPUSVM (Shuicheng)
- Fix forcewake vs runtime pm ref release ordering (Shuicheng)
- Move xe_device_sysfs_init() to xe_device_probe() (Raag)
- Append PCIe Gen5 limitations to xe_firmware document (Raag)

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Thomas Hellstrom <thomas.hellstrom@linux.intel.com>
Link: https://lore.kernel.org/r/aBzUwbzCzz7Qo7fA@fedora

Merge tag 'drm-intel-gt-next-2025-05-08-1' of https://gitlab.freedesktop.org/drm/i915/kernel into drm-next

Driver Changes:

- Fix SLPC wait boosting reference counting to avoid getting stuck on non-boost
  frequency on power saving profile on DG1/DG2 (Vinay)
- Add 20ms delay to engine reset for robustness on HSW (Nitin)

- Use proper sleeping functions for timeouts shorter than 20ms (Andi)
- Fix fence not released on early probe errors for HuC (Janusz)

- Remove const from struct i915_wa list allocation (Kees)
- Apply SPDX license format where missing and use single-line format (Andi)
- Whitespace fixes (Dan, Andi)
- Selftest improvements (Mikolaj, Badal, Sk,

Signed-off-by: Dave Airlie <airlied@redhat.com>
From: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Link: https://lore.kernel.org/r/aBxNYp0IviE23zy-@jlahtine-mobl

Reapply: drm/amdgpu: Use generic hdp flush function

Except HDP v5.2 all use a common logic for HDP flush. Use a generic
function. HDP v5.2 forces NO_KIQ logic, revisit it later.

Reapply after fixing up an HDP regression.

v2: merge the fix (Alex)

Signed-off-by: Lijo Lazar <lijo.lazar@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com> (v1)
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/hdp7: use memcfg register to post the write for HDP flush

Reading back the remapped HDP flush register seems to cause
problems on some platforms. All we need is a read, so read back
the memcfg register.

Fixes: 689275140cb8 ("drm/amdgpu/hdp7.0: do a posting read when flushing HDP")
Reported-by: Alexey Klimov <alexey.klimov@linaro.org>
Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/hdp6: use memcfg register to post the write for HDP flush

Reading back the remapped HDP flush register seems to cause
problems on some platforms. All we need is a read, so read back
the memcfg register.

Fixes: abe1cbaec6cf ("drm/amdgpu/hdp6.0: do a posting read when flushing HDP")
Reported-by: Alexey Klimov <alexey.klimov@linaro.org>
Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: cleanup sriov function for psp v12

PSP v12 won't have SRIOV function.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/hdp5.2: use memcfg register to post the write for HDP flush

Reading back the remapped HDP flush register seems to cause
problems on some platforms. All we need is a read, so read back
the memcfg register.

Fixes: f756dbac1ce1 ("drm/amdgpu/hdp5.2: do a posting read when flushing HDP")
Reported-by: Alexey Klimov <alexey.klimov@linaro.org>
Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/hdp5: use memcfg register to post the write for HDP flush

Reading back the remapped HDP flush register seems to cause
problems on some platforms. All we need is a read, so read back
the memcfg register.

Fixes: cf424020e040 ("drm/amdgpu/hdp5.0: do a posting read when flushing HDP")
Reported-by: Alexey Klimov <alexey.klimov@linaro.org>
Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: remove re-route ih in psp v12

APU doesn't have second IH ring, so re-routing action here is a no-op.
It will take a lot of time to wait timeout from PSP during the
initialization. So remove the function in psp v12.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd: Add per-ring reset for vcn v5.0.0 use

If there is a problem requiring a reset of the VCN engine, it is better to
reset the VCN engine rather than the entire GPU.

Add a reset callback for the ring which will stop and start VCN if an
issue happens.

Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Link: https://lore.kernel.org/r/20250506204948.12048-4-mario.limonciello@amd.com
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd: Add per-ring reset for vcn v4.0.0 use

If there is a problem requiring a reset of the VCN engine, it is better to
reset the VCN engine rather than the entire GPU.

Add a reset callback for the ring which will stop and start VCN if an
issue happens.

Link: https://lore.kernel.org/r/20250506204948.12048-3-mario.limonciello@amd.com
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd: Add per-ring reset for vcn v4.0.5 use

There is a problem occurring on VCN 4.0.5 where in some situations a job
is timing out.  This triggers a job timeout which then causes a GPU
reset for recovery.  That has exposed a number of issues with GPU reset
that have since been fixed. But also a GPU reset isn't actually needed
for this circumstance. Just restarting the ring is enough.

Add a reset callback for the ring which will stop and start VCN if the
issue happens.

Link: https://gitlab.freedesktop.org/mesa/mesa/-/issues/12528
Link: https://gitlab.freedesktop.org/drm/amd/-/issues/3909
Link: https://lore.kernel.org/r/20250506204948.12048-2-mario.limonciello@amd.com
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/hdp4: use memcfg register to post the write for HDP flush

Reading back the remapped HDP flush register seems to cause
problems on some platforms. All we need is a read, so read back
the memcfg register.

Fixes: c9b8dcabb52a ("drm/amdgpu/hdp4.0: do a posting read when flushing HDP")
Reported-by: Alexey Klimov <alexey.klimov@linaro.org>
Link: https://lists.freedesktop.org/archives/amd-gfx/2025-April/123150.html
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4119
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3908
Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Revert "drm/amdgpu: Use generic hdp flush function"

This reverts commit 18a878fd8aef0ec21648a3782f55a79790cd4073.

Revert this temporarily to make it easier to fix a regression
in the HDP handling.

Reviewed-by: Felix Kuehling <felix.kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm/smu13: Remove unused smu_v3 functions

smu_v13_0_display_clock_voltage_request() and
smu_v13_0_set_min_deep_sleep_dcefclk() were added in 2020 by
commit c05d1c401572 ("drm/amd/swsmu: add aldebaran smu13 ip support (v3)")
but have remained unused.

Remove them.

smu_v13_0_display_clock_voltage_request() was the only user
of smu_v13_0_set_hard_freq_limited_range().  Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm/smu11: Remove unused smu_v11_0_get_dpm_level_range

The last use of smu_v11_0_get_dpm_level_range() was removed in 2020 by
commit 46a301e14e8a ("drm/amd/powerplay: drop unnecessary Navi1x specific
APIs")

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/pm/smu7: Remove unused smu7_copy_bytes_from_smc

smu7_copy_bytes_from_smc() was added in 2016 by
commit 1ff55f465103 ("drm/amd/powerplay: implement smu7_smumgr for asics
with smu ip version 7.")

but never used.

Remove it.

Signed-off-by: Dr. David Alan Gilbert <linux@treblig.org>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix the indentation

fix the indentation
drivers/gpu/drm/amd/amdgpu/gfx_v11_0.c:6992 gfx_v11_ip_dump

compiler: gcc-11 (Debian 11.3.0-12) 11.3.0

Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/r/202505071619.7sHTLpNg-lkp@intel.com/
Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
Reviewed-by: Arvind Yadav <Arvind.Yadav@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: remove mdelay in psp v12

Since secure firmware is more stable than bring up phase, I believe we
don't need such mdelays any more before wait PSP response on PSP v12.

Signed-off-by: Huang Rui <ray.huang@amd.com>
Reviewed-by: Trigger Huang <Trigger.Huang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

amd/amdkfd: Trigger segfault for early userptr unmmapping

If applications unmap the memory before destroying the userptr, it needs
trigger a segfault to notify user space to correct the free sequence in
VM debug mode.

v2: Send gpu access fault to user space
v3: Report gpu address to user space, remove unnecessary params
v4: update pr_err into one line, remove userptr log info

Signed-off-by: Shane Xiao <shane.xiao@amd.com>
Acked-by: Christian König <christian.koenig@amd.com>
Reviewed-by: Felix Kuehling <Felix.Kuehling@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add debug bit for userptr usage

In VM debug mode, it is desirable to notify the application
to correct the freeing sequence by unmapping the memory before
destroying the userptr in the old userptr path. Add a bitmask
to decide whether to send gpu vm fault to the applition.

Signed-off-by: Shane Xiao <shane.xiao@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: unreserve the gem BO before returning from attach error

It requires unlocking the reserved gem BO before returning from
attaching the eviction fence error.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: promote the implicit sync to the dependent read fences

The driver doesn't want to implicitly sync on the DMA_RESV_USAGE_BOOKKEEP
usage fences, and the BOOKEEP fences should be synced explicitly. So, as
the VM implicit syncing only need to return and sync the dependent read
fences.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/psp: mark securedisplay TA as optional

This is an optional TA which is only available on
certain embedded systems.  Mark it as optional to avoid
user confusion.  This mirrors what we already do for
other optional TAs.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4181
Reviewed-by: Hawking Zhang <Hawking.Zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix pm notifier handling

Set the s3/s0ix and s4 flags in the pm notifier so that we can skip
the resource evictions properly in pm prepare based on whether
we are suspending or hibernating.  Drop the eviction as processes
are not frozen at this time, we we can end up getting stuck trying
to evict VRAM while applications continue to submit work which
causes the buffers to get pulled back into VRAM.

v2: Move suspend flags out of pm notifier (Mario)

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/4178
Fixes: 2965e6355dcd ("drm/amd: Add Suspend/Hibernate notification callback support")
Cc: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Implement unrecoverable error message handling for VFs

This notification may arrive in VF mailbox while polling for response from
another event.

This patches covers the following scenarios:

- If VF is already in RMA state, then do not attempt to contact the host.
  Host will ignore the VF after sending the notification.

- If the notification is detected during polling, then set the RMA status,
  and return error to caller.

- If the notification arrives by interrupt, then set the RMA status and
  queue a reset.  This reset will fail and VF will stop runtime services.

Reviewed-by: Shravan Kumar Gande <Shravankumar.Gande@amd.com>
Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com>
Signed-off-by: Ellen Pan <yunru.pan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add unrecoverable error message definitions for VFs

Host may stop runtime services after reaching a bad page threshold.

This notification will indicate to the VF that it no longer has
access to the GPU.

Reviewed-by: Shravan Kumar Gande <Shravankumar.Gande@amd.com>
Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com>
Signed-off-by: Ellen Pan <yunru.pan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Revert "drm/amd: Stop evicting resources on APUs in suspend"

This reverts commit 3a9626c816db901def438dc2513622e281186d39.

This breaks S4 because we end up setting the s3/s0ix flags
even when we are entering s4 since prepare is used by both
flows.  The causes both the S3/s0ix and s4 flags to be set
which breaks several checks in the driver which assume they
are mutually exclusive.

Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3634
Cc: Mario Limonciello <mario.limonciello@amd.com>
Reviewed-by: Mario Limonciello <mario.limonciello@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu/vcn: using separate VCN1_AON_SOC offset

VCN1_AON_SOC_ADDRESS_3_0 offset varies on different
VCN generations, the issue in vcn4.0.5 is caused by
a different VCN1_AON_SOC_ADDRESS_3_0 offset.

This patch does the following:

    1. use the same offset for other VCN generations.
    2. use the vcn4.0.5 special offset
    3. update vcn_4_0 and vcn_5_0

Acked-by: Saleemkhan Jamadar <saleemkhan.jamadar@amd.com>
Reviewed-by: Leo Liu <leo.liu@amd.com>
Signed-off-by: Ruijing Dong <ruijing.dong@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: fix the eviction fence dereference

The dma_resv_add_fence() already refers to the added fence.
So when attaching the evciton fence to the gem bo, it needn't
refer to it anymore.

Signed-off-by: Prike Liang <Prike.Liang@amd.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Implement Runtime Bad Page query for VFs

Host will send a notification when new bad pages are available.

Uopn guest request, the first 256 bad page addresses
will be placed into the PF2VF region.
Guest should pause the PF2VF worker thread while
the copy is in progress.

Reviewed-by: Shravan Kumar Gande <Shravankumar.Gande@amd.com>
Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com>
Signed-off-by: Ellen Pan <yunru.pan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add Runtime Bad Page message definitions for VFs

Currently VFs rely on poison consumption interrupt from HW
to kick off the bad page retirement process. Part of this process
includes a VF reset.

This patch adds the following:

1) Host Bad Pages notification message.
2) Guest request bad pages message.

When combined, VFs are able to reserve the pages early, and potentially
avoid future poison consumption that will disrupt user services
from consequent FLR.

Reviewed-by: Shravan Kumar Gande <Shravankumar.Gande@amd.com>
Signed-off-by: Victor Skvortsov <victor.skvortsov@amd.com>
Signed-off-by: Ellen Pan <yunru.pan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

Documentation/gpu: Add new entries to amdgpu glossary

Add some additional entries.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdgpu: Add documentation to some parts of the AMDGPU ring and wb

Add some random documentation associated with the ring buffer
manipulations and writeback.

Signed-off-by: Rodrigo Siqueira <siqueira@igalia.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amdkfd: change error to warning message for SDMA queues creation

SDMA doesn't support oversubsciption, it is the user matter to create
queues over HW limit, but not supposed to be a KFD error.

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
Reviewed-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Don't check for NULL divisor in fixpt code

[Why]
We check for a NULL divisor but don't act on it.
This check does nothing other than throw a warning.
It does confuse static checkers though:
See https://lkml.org/lkml/2025/4/26/371

[How]
Drop the ASSERTs in both DC and SPL variants.

Fixes: 4562236b3bc0 ("drm/amd/dc: Add dc display driver (v2)")
Fixes: 6efc0ab3b05d ("drm/amd/display: add back quality EASF and ISHARP and dc dependency changes")
Signed-off-by: Harry Wentland <harry.wentland@amd.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Leo Li <sunpeng.li@amd.com>
Cc: Alex Deucher <alexander.deucher@amd.com>
Acked-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>

drm/amd/display: Use true/false for boolean variables in DML2 core files

Replace 0 and 1 with false and true for boolean variables in
dml2_core_dcn4_calcs.c and dml2_core_utils.c to align with the Linux
kernel coding style guidelines, which recommend using C99 bool type
with true/false values.

Signed-off-by: Ivan Shamliev <ivan.shamliev.dev@abv.bg>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>