mirrors/linux
2
0
Fork 0
mirror of https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git synced 2026-08-06 17:22:55 -04:00
Code Issues Packages Projects Releases Wiki Activity
1,429,592 Commits 1 Branch 941 Tags
d78ddeb8938a366aabfabf60255c1a94de8d8ea1
Commit Graph

122554 Commits

Author SHA1 Message Date
Michal Wajdeczko
cd11babcd0 drm/xe/pf: Update LMTT to use tile-based messages 
Since now we have tile-based SR-IOV printk macros, there is no
need to manually prepare the LMTT specific warning message (that
is now upgraded to proper error level message) nor to use generic
debug message without tile/LMTT identification.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Link: https://lore.kernel.org/r/20251005133641.2651-4-michal.wajdeczko@intel.com
 2025-10-06 19:39:25 +02:00
Michal Wajdeczko
c66e4b6cae drm/xe: Add tile-based SRIOV printk macros 
We already have device and GT level SR-IOV specific macros, but
unlike native case, we don't have yet tile-based ones.

Add macros to match native use case and also update GT-based
macros to rely on those new tile-based SR-IOV macros. This will
slightly rearrange the output of the GT logs and instead:

  [...] Tile0: GT0: PF: pushed VF1 config with 2 KLVs...

we might see:

  [...] PF: Tile0: GT0: pushed VF1 config with 2 KLVs...

but that's even better.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Link: https://lore.kernel.org/r/20251005133641.2651-3-michal.wajdeczko@intel.com
 2025-10-06 19:39:23 +02:00
Michal Wajdeczko
c95f180207 drm/xe: Update SRIOV printk macros 
Recently we introduced xe-based printk macros, use them instead
of plain drm-based ones.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Link: https://lore.kernel.org/r/20251005133641.2651-2-michal.wajdeczko@intel.com
 2025-10-06 19:39:22 +02:00
Michal Wajdeczko
9a54b5127f drm/xe/pf: Make the late-initialization really late 
While the late PF per-GT initialization is done quite late in the
single GT initialization flow, in case of multi-GT platforms, it
may still be done before other GT early initialization. That leads
to some issues during unwind, when there are cross-GT dependencies,
like resource cleanup that is shared by both GTs, but the other GT
may already be sanitized or disabled.

The following errors could be observed when trying to unload the PF
driver with some LMEM/VRAM already provisioned for few VFs:

 [ ] xe 0000:03:00.0: DEVRES REL ffff88814708f240 fini_config (16 bytes)
 [ ] xe 0000:03:00.0: [drm:lmtt_write_pte [xe]] PF: LMTT: WRITE level=2 index=1 pte=0x0
 [ ] xe 0000:03:00.0: [drm:lmtt_invalidate_hw [xe]] PF: LMTT: num_fences=2 err=-19
 [ ] xe 0000:03:00.0: [drm:lmtt_pt_free [xe]] PF: LMTT: level=0 addr=53a470000
 [ ] xe 0000:03:00.0: [drm:lmtt_pt_free [xe]] PF: LMTT: level=1 addr=53a4b0000
 [ ] xe 0000:03:00.0: [drm:lmtt_invalidate_hw [xe]] PF: LMTT: num_fences=2 err=-19
 [ ] xe 0000:03:00.0: [drm] PF: LMTT0 invalidation failed (-ENODEV)
 [ ] xe 0000:03:00.0: [drm:lmtt_write_pte [xe]] PF: LMTT: WRITE level=2 index=2 pte=0x0
 [ ] xe 0000:03:00.0: [drm:lmtt_invalidate_hw [xe]] PF: LMTT: num_fences=2 err=-19
 [ ] xe 0000:03:00.0: [drm:lmtt_pt_free [xe]] PF: LMTT: level=0 addr=539b70000
 [ ] xe 0000:03:00.0: [drm:lmtt_pt_free [xe]] PF: LMTT: level=1 addr=539bf0000
 [ ] xe 0000:03:00.0: [drm:lmtt_invalidate_hw [xe]] PF: LMTT: num_fences=2 err=-19
 [ ] xe 0000:03:00.0: [drm] PF: LMTT0 invalidation failed (-ENODEV)

Move all PF per-GT late initialization to the already defined late
SR-IOV initialization function to allow proper order of the cleanup
actions.

While around, format all PF function stubs as one-liners, like many
other stubs are defined in the Xe driver.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Link: https://lore.kernel.org/r/20251004162008.1782-1-michal.wajdeczko@intel.com
 2025-10-06 19:30:17 +02:00
Michal Wajdeczko
71f1939e0d drm/xe/xe_late_bind_fw: Fix and simplify parsing user input 
Code was wrongly passing sizeof(uval) as the number base to use,
and unlike other debugfs entries that represent bool data, it
wasn't using the dedicated function to parse user input as bool.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Badal Nilawar <badal.nilawar@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Link: https://lore.kernel.org/r/20251002192736.203186-1-michal.wajdeczko@intel.com
 2025-10-06 19:24:15 +02:00
Michal Wajdeczko
869580c415 drm/xe: Don't force DRM_XE_DEBUG_MEMIRQ for SR-IOV debug 
For pure SR-IOV debugging there is no need to select already
separated config for the debugging of the memory based interrupts,
as the latter is also very noisy on its own. Change config order
and use a weak reverse dependency instead.

Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Reviewed-by: Jonathan Cavitt <jonathan.cavitt@intel.com>
Reviewed-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Link: https://lore.kernel.org/r/20251002171308.203127-1-michal.wajdeczko@intel.com
 2025-10-06 19:11:30 +02:00
Shuicheng Lin
a908de69ce drm/xe: Fix copyright and function naming in xe_ttm_vram_mgr 
- Correct copyright year from "2002" to "2022".
- Rename ttm_vram_mgr_fini() to xe_ttm_vram_mgr_fini() to avoid
  confusion with generic TTM helpers.

No functional changes intended.

Signed-off-by: Shuicheng Lin <shuicheng.lin@intel.com>
Reviewed-by: Nitin Gote <nitin.r.gote@intel.com>
Link: https://lore.kernel.org/r/20251004000425.2489291-2-shuicheng.lin@intel.com
Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
 2025-10-06 12:46:28 -04:00
Ian Forbes
228c5d44df drm/vmwgfx: Fix copy-paste typo in validation 
'entry' should be 'val' which is the loop iterator.

Fixes: 9e931f2e09 ("drm/vmwgfx: Refactor resource validation hashtable to use linux/hashtable implementation.")
Signed-off-by: Ian Forbes <ian.forbes@broadcom.com>
Reviewed-by: Zack Rusin <zack.rusin@broadcom.com>
Signed-off-by: Zack Rusin <zack.rusin@broadcom.com>
Link: https://lore.kernel.org/r/20250926195427.1405237-2-ian.forbes@broadcom.com
 2025-10-06 11:57:59 -04:00
Ian Forbes
dfe1323ab3 drm/vmwgfx: Fix Use-after-free in validation 
Nodes stored in the validation duplicates hashtable come from an arena
allocator that is cleared at the end of vmw_execbuf_process. All nodes
are expected to be cleared in vmw_validation_drop_ht but this node escaped
because its resource was destroyed prematurely.

Fixes: 64ad2abfe9 ("drm/vmwgfx: Adapt validation code for reference-free lookups")
Reported-by: Kuzey Arda Bulut <kuzeyardabulut@gmail.com>
Signed-off-by: Ian Forbes <ian.forbes@broadcom.com>
Reviewed-by: Zack Rusin <zack.rusin@broadcom.com>
Signed-off-by: Zack Rusin <zack.rusin@broadcom.com>
Link: https://lore.kernel.org/r/20250926195427.1405237-1-ian.forbes@broadcom.com
 2025-10-06 11:57:38 -04:00
Zack Rusin
5ac2c02790 drm/vmwgfx: Fix a null-ptr access in the cursor snooper 
Check that the resource which is converted to a surface exists before
trying to use the cursor snooper on it.

vmw_cmd_res_check allows explicit invalid (SVGA3D_INVALID_ID) identifiers
because some svga commands accept SVGA3D_INVALID_ID to mean "no surface",
unfortunately functions that accept the actual surfaces as objects might
(and in case of the cursor snooper, do not) be able to handle null
objects. Make sure that we validate not only the identifier (via the
vmw_cmd_res_check) but also check that the actual resource exists before
trying to do something with it.

Fixes unchecked null-ptr reference in the snooping code.

Signed-off-by: Zack Rusin <zack.rusin@broadcom.com>
Fixes: c0951b797e ("drm/vmwgfx: Refactor resource management")
Reported-by: Kuzey Arda Bulut <kuzeyardabulut@gmail.com>
Cc: Broadcom internal kernel review list <bcm-kernel-feedback-list@broadcom.com>
Cc: dri-devel@lists.freedesktop.org
Reviewed-by: Ian Forbes <ian.forbes@broadcom.com>
Link: https://lore.kernel.org/r/20250917153655.1968583-1-zack.rusin@broadcom.com
 2025-10-06 11:57:17 -04:00
Janusz Krzysztofik
78b4d6463e drm/vgem-fence: Fix potential deadlock on release 
A timer that expires a vgem fence automatically in 10 seconds is now
released with timer_delete_sync() from fence->ops.release() called on last
dma_fence_put().  In some scenarios, it can run in IRQ context, which is
not safe unless TIMER_IRQSAFE is used.  One potentially risky scenario was
demonstrated in Intel DRM CI trybot, BAT run on machine bat-adlp-6, while
working on new IGT subtests syncobj_timeline@stress-* as user space
replacements of some problematic test cases of a dma-fence-chain selftest
[1].

[117.004338] ================================
[117.004340] WARNING: inconsistent lock state
[117.004342] 6.17.0-rc7-CI_DRM_17270-g7644974e648c+ #1 Tainted: G S   U
[117.004346] --------------------------------
[117.004347] inconsistent {HARDIRQ-ON-W} -> {IN-HARDIRQ-W} usage.
[117.004349] swapper/0/0 [HC1[1]:SC1[1]:HE0:SE0] takes:
[117.004352] ffff888138f86aa8 ((&fence->timer)){?.-.}-{0:0}, at: __timer_delete_sync+0x4b/0x190
[117.004361] {HARDIRQ-ON-W} state was registered at:
[117.004363]   lock_acquire+0xc4/0x2e0
[117.004366]   call_timer_fn+0x80/0x2a0
[117.004368]   __run_timers+0x231/0x310
[117.004370]   run_timer_softirq+0x76/0xe0
[117.004372]   handle_softirqs+0xd4/0x4d0
[117.004375]   __irq_exit_rcu+0x13f/0x160
[117.004377]   irq_exit_rcu+0xe/0x20
[117.004379]   sysvec_apic_timer_interrupt+0xa0/0xc0
[117.004382]   asm_sysvec_apic_timer_interrupt+0x1b/0x20
[117.004385]   cpuidle_enter_state+0x12b/0x8a0
[117.004388]   cpuidle_enter+0x2e/0x50
[117.004393]   call_cpuidle+0x22/0x60
[117.004395]   do_idle+0x1fd/0x260
[117.004398]   cpu_startup_entry+0x29/0x30
[117.004401]   start_secondary+0x12d/0x160
[117.004404]   common_startup_64+0x13e/0x141
[117.004407] irq event stamp: 2282669
[117.004409] hardirqs last  enabled at (2282668): [<ffffffff8289db71>] _raw_spin_unlock_irqrestore+0x51/0x80
[117.004414] hardirqs last disabled at (2282669): [<ffffffff82882021>] sysvec_irq_work+0x11/0xc0
[117.004419] softirqs last  enabled at (2254702): [<ffffffff8289fd00>] __do_softirq+0x10/0x18
[117.004423] softirqs last disabled at (2254725): [<ffffffff813d4ddf>] __irq_exit_rcu+0x13f/0x160
[117.004426]
other info that might help us debug this:
[117.004429]  Possible unsafe locking scenario:
[117.004432]        CPU0
[117.004433]        ----
[117.004434]   lock((&fence->timer));
[117.004436]   <Interrupt>
[117.004438]     lock((&fence->timer));
[117.004440]
 *** DEADLOCK ***
[117.004443] 1 lock held by swapper/0/0:
[117.004445]  #0: ffffc90000003d50 ((&fence->timer)){?.-.}-{0:0}, at: call_timer_fn+0x7a/0x2a0
[117.004450]
stack backtrace:
[117.004453] CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Tainted: G S   U              6.17.0-rc7-CI_DRM_17270-g7644974e648c+ #1 PREEMPT(voluntary)
[117.004455] Tainted: [S]=CPU_OUT_OF_SPEC, [U]=USER
[117.004455] Hardware name: Intel Corporation Alder Lake Client Platform/AlderLake-P DDR4 RVP, BIOS RPLPFWI1.R00.4035.A00.2301200723 01/20/2023
[117.004456] Call Trace:
[117.004456]  <IRQ>
[117.004457]  dump_stack_lvl+0x91/0xf0
[117.004460]  dump_stack+0x10/0x20
[117.004461]  print_usage_bug.part.0+0x260/0x360
[117.004463]  mark_lock+0x76e/0x9c0
[117.004465]  ? register_lock_class+0x48/0x4a0
[117.004467]  __lock_acquire+0xbc3/0x2860
[117.004469]  lock_acquire+0xc4/0x2e0
[117.004470]  ? __timer_delete_sync+0x4b/0x190
[117.004472]  ? __timer_delete_sync+0x4b/0x190
[117.004473]  __timer_delete_sync+0x68/0x190
[117.004474]  ? __timer_delete_sync+0x4b/0x190
[117.004475]  timer_delete_sync+0x10/0x20
[117.004476]  vgem_fence_release+0x19/0x30 [vgem]
[117.004478]  dma_fence_release+0xc1/0x3b0
[117.004480]  ? dma_fence_release+0xa1/0x3b0
[117.004481]  dma_fence_chain_release+0xe7/0x130
[117.004483]  dma_fence_release+0xc1/0x3b0
[117.004484]  ? _raw_spin_unlock_irqrestore+0x27/0x80
[117.004485]  dma_fence_chain_irq_work+0x59/0x80
[117.004487]  irq_work_single+0x75/0xa0
[117.004490]  irq_work_run_list+0x33/0x60
[117.004491]  irq_work_run+0x18/0x40
[117.004493]  __sysvec_irq_work+0x35/0x170
[117.004494]  sysvec_irq_work+0x47/0xc0
[117.004496]  asm_sysvec_irq_work+0x1b/0x20
[117.004497] RIP: 0010:_raw_spin_unlock_irqrestore+0x57/0x80
[117.004499] Code: 00 75 1c 65 ff 0d d9 34 68 01 74 20 5b 41 5c 5d 31 c0 31 d2 31 c9 31 f6 31 ff c3 cc cc cc cc e8 7f 9d d3 fe fb 0f 1f 44 00 00 <eb> d7 0f 1f 44 00 00 5b 41 5c 5d 31 c0 31 d2 31 c9 31 f6 31 ff c3
[117.004499] RSP: 0018:ffffc90000003cf0 EFLAGS: 00000246
[117.004500] RAX: 0000000000000000 RBX: ffff888155e94c40 RCX: 0000000000000000
[117.004501] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[117.004502] RBP: ffffc90000003d00 R08: 0000000000000000 R09: 0000000000000000
[117.004502] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000246
[117.004502] R13: 0000000000000001 R14: 0000000000000246 R15: ffff888155e94c80
[117.004506]  dma_fence_signal+0x49/0xb0
[117.004507]  ? __pfx_vgem_fence_timeout+0x10/0x10 [vgem]
[117.004508]  vgem_fence_timeout+0x12/0x20 [vgem]
[117.004509]  call_timer_fn+0xa1/0x2a0
[117.004512]  ? __pfx_vgem_fence_timeout+0x10/0x10 [vgem]
[117.004513]  __run_timers+0x231/0x310
[117.004514]  ? tmigr_handle_remote+0x2ac/0x560
[117.004517]  timer_expire_remote+0x46/0x70
[117.004518]  tmigr_handle_remote+0x433/0x560
[117.004520]  ? __run_timers+0x239/0x310
[117.004521]  ? run_timer_softirq+0x21/0xe0
[117.004522]  ? lock_release+0xce/0x2a0
[117.004524]  run_timer_softirq+0xcf/0xe0
[117.004525]  handle_softirqs+0xd4/0x4d0
[117.004526]  __irq_exit_rcu+0x13f/0x160
[117.004527]  irq_exit_rcu+0xe/0x20
[117.004528]  sysvec_apic_timer_interrupt+0xa0/0xc0
[117.004529]  </IRQ>
[117.004529]  <TASK>
[117.004529]  asm_sysvec_apic_timer_interrupt+0x1b/0x20
[117.004530] RIP: 0010:cpuidle_enter_state+0x12b/0x8a0
[117.004532] Code: 48 0f a3 05 97 ce 0e 01 0f 82 2e 03 00 00 31 ff e8 8a 41 bd fe 80 7d d0 00 0f 85 11 03 00 00 e8 8b 06 d5 fe fb 0f 1f 44 00 00 <45> 85 f6 0f 88 67 02 00 00 4d 63 ee 49 83 fd 0a 0f 83 34 06 00 00
[117.004532] RSP: 0018:ffffffff83403d88 EFLAGS: 00000246
[117.004533] RAX: 0000000000000000 RBX: ffff88888f046440 RCX: 0000000000000000
[117.004533] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[117.004534] RBP: ffffffff83403dd8 R08: 0000000000000000 R09: 0000000000000000
[117.004534] R10: 0000000000000000 R11: 0000000000000000 R12: ffffffff837cbe80
[117.004534] R13: 0000000000000004 R14: 0000000000000004 R15: 0000001ad1df466b
[117.004537]  ? cpuidle_enter_state+0x125/0x8a0
[117.004538]  ? sched_clock_noinstr+0x9/0x10
[117.004540]  cpuidle_enter+0x2e/0x50
[117.004542]  call_cpuidle+0x22/0x60
[117.004542]  do_idle+0x1fd/0x260
[117.004544]  cpu_startup_entry+0x29/0x30
[117.004546]  rest_init+0x104/0x200
[117.004548]  start_kernel+0x93d/0xbd0
[117.004550]  ? load_ucode_intel_bsp+0x2a/0x90
[117.004551]  ? sme_unmap_bootdata+0x14/0x80
[117.004554]  x86_64_start_reservations+0x18/0x30
[117.004555]  x86_64_start_kernel+0xfd/0x150
[117.004556]  ? soft_restart_cpu+0x14/0x14
[117.004558]  common_startup_64+0x13e/0x141
[117.004560]  </TASK>
[117.004565] ------------[ cut here ]------------
[117.004692] WARNING: CPU: 0 PID: 0 at kernel/time/timer.c:1610 __timer_delete_sync+0x126/0x190
[117.004697] Modules linked in: vgem snd_hda_codec_intelhdmi snd_hda_codec_hdmi i915 prime_numbers ttm drm_buddy drm_display_helper cec rc_core i2c_algo_bit hid_sensor_custom hid_sensor_hub hid_generic intel_ishtp_hid hid intel_uncore_frequency intel_uncore_frequency_common x86_pkg_temp_thermal intel_powerclamp cmdlinepart ee1004 r8153_ecm spi_nor coretemp cdc_ether mei_pxp mei_hdcp usbnet mtd intel_rapl_msr wmi_bmof kvm_intel snd_hda_intel snd_intel_dspcfg processor_thermal_device_pci kvm snd_hda_codec processor_thermal_device irqbypass processor_thermal_wt_hint polyval_clmulni platform_temperature_control snd_hda_core ghash_clmulni_intel processor_thermal_rfim spi_pxa2xx_platform snd_hwdep aesni_intel processor_thermal_rapl dw_dmac snd_pcm dw_dmac_core intel_rapl_common r8152 rapl mii intel_cstate spi_pxa2xx_core i2c_i801 processor_thermal_wt_req snd_timer i2c_mux mei_me intel_ish_ipc processor_thermal_power_floor e1000e snd i2c_smbus spi_intel_pci processor_thermal_mbox mei soundcore intel_ishtp thunderbolt idma64
[117.004733]  spi_intel int340x_thermal_zone igen6_edac binfmt_misc intel_skl_int3472_tps68470 intel_pmc_core tps68470_regulator video clk_tps68470 pmt_telemetry pmt_discovery nls_iso8859_1 pmt_class intel_pmc_ssram_telemetry intel_skl_int3472_discrete int3400_thermal intel_hid intel_skl_int3472_common acpi_thermal_rel intel_vsec wmi pinctrl_tigerlake acpi_tad sparse_keymap acpi_pad dm_multipath msr nvme_fabrics fuse efi_pstore nfnetlink autofs4
[117.004782] CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Tainted: G S   U              6.17.0-rc7-CI_DRM_17270-g7644974e648c+ #1 PREEMPT(voluntary)
[117.004787] Tainted: [S]=CPU_OUT_OF_SPEC, [U]=USER
[117.004789] Hardware name: Intel Corporation Alder Lake Client Platform/AlderLake-P DDR4 RVP, BIOS RPLPFWI1.R00.4035.A00.2301200723 01/20/2023
[117.004793] RIP: 0010:__timer_delete_sync+0x126/0x190
[117.004795] Code: 31 c0 45 31 c9 c3 cc cc cc cc 48 8b 75 d0 45 84 f6 74 63 49 c7 45 18 00 00 00 00 48 89 c7 e8 51 46 39 01 f3 90 e9 66 ff ff ff <0f> 0b e9 5f ff ff ff e8 ee e4 0c 00 49 8d 5d 28 45 31 c9 31 c9 4c
[117.004801] RSP: 0018:ffffc90000003a40 EFLAGS: 00010046
[117.004804] RAX: ffffffff815093fb RBX: ffff888138f86aa8 RCX: 0000000000000000
[117.004807] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[117.004809] RBP: ffffc90000003a70 R08: 0000000000000000 R09: 0000000000000000
[117.004812] R10: 0000000000000000 R11: 0000000000000000 R12: ffffffff815093fb
[117.004814] R13: ffff888138f86a80 R14: 0000000000000000 R15: 0000000000000000
[117.004817] FS:  0000000000000000(0000) GS:ffff88890b0f7000(0000) knlGS:0000000000000000
[117.004820] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[117.004823] CR2: 00005db8131eb7f0 CR3: 0000000003448000 CR4: 0000000000f52ef0
[117.004826] PKRU: 55555554
[117.004827] Call Trace:
[117.004829]  <IRQ>
[117.004831]  timer_delete_sync+0x10/0x20
[117.004833]  vgem_fence_release+0x19/0x30 [vgem]
[117.004836]  dma_fence_release+0xc1/0x3b0
[117.004838]  ? dma_fence_release+0xa1/0x3b0
[117.004841]  dma_fence_chain_release+0xe7/0x130
[117.004844]  dma_fence_release+0xc1/0x3b0
[117.004847]  ? _raw_spin_unlock_irqrestore+0x27/0x80
[117.004850]  dma_fence_chain_irq_work+0x59/0x80
[117.004853]  irq_work_single+0x75/0xa0
[117.004857]  irq_work_run_list+0x33/0x60
[117.004860]  irq_work_run+0x18/0x40
[117.004863]  __sysvec_irq_work+0x35/0x170
[117.004865]  sysvec_irq_work+0x47/0xc0
[117.004868]  asm_sysvec_irq_work+0x1b/0x20
[117.004871] RIP: 0010:_raw_spin_unlock_irqrestore+0x57/0x80
[117.004874] Code: 00 75 1c 65 ff 0d d9 34 68 01 74 20 5b 41 5c 5d 31 c0 31 d2 31 c9 31 f6 31 ff c3 cc cc cc cc e8 7f 9d d3 fe fb 0f 1f 44 00 00 <eb> d7 0f 1f 44 00 00 5b 41 5c 5d 31 c0 31 d2 31 c9 31 f6 31 ff c3
[117.004879] RSP: 0018:ffffc90000003cf0 EFLAGS: 00000246
[117.004882] RAX: 0000000000000000 RBX: ffff888155e94c40 RCX: 0000000000000000
[117.004884] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[117.004887] RBP: ffffc90000003d00 R08: 0000000000000000 R09: 0000000000000000
[117.004890] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000246
[117.004892] R13: 0000000000000001 R14: 0000000000000246 R15: ffff888155e94c80
[117.004897]  dma_fence_signal+0x49/0xb0
[117.004899]  ? __pfx_vgem_fence_timeout+0x10/0x10 [vgem]
[117.004902]  vgem_fence_timeout+0x12/0x20 [vgem]
[117.004904]  call_timer_fn+0xa1/0x2a0
[117.004908]  ? __pfx_vgem_fence_timeout+0x10/0x10 [vgem]
[117.004910]  __run_timers+0x231/0x310
[117.004913]  ? tmigr_handle_remote+0x2ac/0x560
[117.004917]  timer_expire_remote+0x46/0x70
[117.004919]  tmigr_handle_remote+0x433/0x560
[117.004923]  ? __run_timers+0x239/0x310
[117.004925]  ? run_timer_softirq+0x21/0xe0
[117.004928]  ? lock_release+0xce/0x2a0
[117.004931]  run_timer_softirq+0xcf/0xe0
[117.004933]  handle_softirqs+0xd4/0x4d0
[117.004936]  __irq_exit_rcu+0x13f/0x160
[117.004938]  irq_exit_rcu+0xe/0x20
[117.004940]  sysvec_apic_timer_interrupt+0xa0/0xc0
[117.004943]  </IRQ>
[117.004944]  <TASK>
[117.004946]  asm_sysvec_apic_timer_interrupt+0x1b/0x20
[117.004949] RIP: 0010:cpuidle_enter_state+0x12b/0x8a0
[117.004953] Code: 48 0f a3 05 97 ce 0e 01 0f 82 2e 03 00 00 31 ff e8 8a 41 bd fe 80 7d d0 00 0f 85 11 03 00 00 e8 8b 06 d5 fe fb 0f 1f 44 00 00 <45> 85 f6 0f 88 67 02 00 00 4d 63 ee 49 83 fd 0a 0f 83 34 06 00 00
[117.004961] RSP: 0018:ffffffff83403d88 EFLAGS: 00000246
[117.004963] RAX: 0000000000000000 RBX: ffff88888f046440 RCX: 0000000000000000
[117.004966] RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
[117.004968] RBP: ffffffff83403dd8 R08: 0000000000000000 R09: 0000000000000000
[117.004971] R10: 0000000000000000 R11: 0000000000000000 R12: ffffffff837cbe80
[117.004974] R13: 0000000000000004 R14: 0000000000000004 R15: 0000001ad1df466b
[117.004978]  ? cpuidle_enter_state+0x125/0x8a0
[117.004981]  ? sched_clock_noinstr+0x9/0x10
[117.004985]  cpuidle_enter+0x2e/0x50
[117.004989]  call_cpuidle+0x22/0x60
[117.004991]  do_idle+0x1fd/0x260
[117.005001]  cpu_startup_entry+0x29/0x30
[117.005004]  rest_init+0x104/0x200
[117.005008]  start_kernel+0x93d/0xbd0
[117.005011]  ? load_ucode_intel_bsp+0x2a/0x90
[117.005014]  ? sme_unmap_bootdata+0x14/0x80
[117.005017]  x86_64_start_reservations+0x18/0x30
[117.005020]  x86_64_start_kernel+0xfd/0x150
[117.005023]  ? soft_restart_cpu+0x14/0x14
[117.005026]  common_startup_64+0x13e/0x141
[117.005030]  </TASK>
[117.005032] irq event stamp: 2282669
[117.005034] hardirqs last  enabled at (2282668): [<ffffffff8289db71>] _raw_spin_unlock_irqrestore+0x51/0x80
[117.005038] hardirqs last disabled at (2282669): [<ffffffff82882021>] sysvec_irq_work+0x11/0xc0
[117.005043] softirqs last  enabled at (2254702): [<ffffffff8289fd00>] __do_softirq+0x10/0x18
[117.005047] softirqs last disabled at (2254725): [<ffffffff813d4ddf>] __irq_exit_rcu+0x13f/0x160
[117.005051] ---[ end trace 0000000000000000 ]---

Make the timer IRQ safe.

[1] https://patchwork.freedesktop.org/series/154987/#rev2

Fixes: 4077798484 ("drm/vgem: Attach sw fences to exported vGEM dma-buf (ioctl)")
Signed-off-by: Janusz Krzysztofik <janusz.krzysztofik@linux.intel.com>
Reviewed-by: Christian König <christian.koenig@amd.com>
Link: https://lore.kernel.org/r/20250926152628.2165080-2-janusz.krzysztofik@linux.intel.com
Signed-off-by: Maarten Lankhorst <dev@lankhorst.se>
 2025-10-06 15:05:17 +02:00
Maxime Ripard
40d2a29d86 drm/atomic: Convert drm_atomic_get_private_obj_state() to use new plane state 
The drm_atomic_get_private_obj_state() function tries to find if a
private_obj had already been allocated and was part of the given
drm_atomic_state. If one is found, it returns the existing state
pointer.

At the point in time where drm_atomic_get_private_obj_state() can be
called (ie, during atomic_check), the existing state is the new state
and we can thus replace the hand-crafted logic by a call to
drm_atomic_get_new_private_obj_state().

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-38-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:23 +02:00
Maxime Ripard
22c0dd1911 drm/ingenic: crtc: Switch to ingenic_drm_get_new_priv_state() 
The ingenic CRTC atomic_enable() implementation will indirectly call
drm_atomic_get_private_obj_state() through ingenic_drm_get_priv_state().

drm_atomic_get_private_obj_state() will either return the new state for
the object in the global state if it exists, or will allocate a new one
and add it to the global state.

atomic_enable() however isn't allowed to modify the global state. So
what the implementation should use is the
drm_atomic_get_new_private_obj_state() helper to get the new state for
the CRTC, without performing an extra allocation.

We still need to make sure the private state will be part of the global
state by the time atomic_enable runs, so we still need to call
ingenic_drm_get_priv_state() in atomic_check. We can then call
ingenic_drm_get_new_priv_state() in atomic_enable, which is a wrapper
around drm_atomic_get_new_private_obj_state().

Reported-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-37-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:22 +02:00
Maxime Ripard
d8ebc4162f drm/framebuffer: Switch to drm_atomic_get_new_crtc_state() 
The atomic_remove_fb() function uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

Despite its name, this function builds and commit a new
drm_atomic_state, and the call to drm_atomic_get_existing_crtc_state()
is part of the state building, thus happening before the states are
swapped.

As such, the existing state points to the new state, and we can use
drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-34-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:21 +02:00
Maxime Ripard
653ce3a3f6 drm/atomic: Switch to drm_atomic_get_new_crtc_state() 
The drm_atomic_connector_check() function uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-33-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:20 +02:00
Maxime Ripard
4bcd18bbf5 drm/vc4: Switch to drm_atomic_get_new_crtc_state() 
The vc4 atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Dave Stevenson <dave.stevenson@raspberrypi.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-32-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:20 +02:00
Maxime Ripard
218fa3e6a1 drm/vboxvideo: Switch to drm_atomic_get_new_crtc_state() 
The vboxvideo atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Hans de Goede <hansg@kernel.org>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-31-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:19 +02:00
Maxime Ripard
5d79c84930 drm/tilcdc: Switch to drm_atomic_get_new_crtc_state() 
The tilcdc atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Jyri Sarha <jyri.sarha@iki.fi>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-30-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:19 +02:00
Maxime Ripard
2ebd2179d3 drm/tegra: Switch to drm_atomic_get_new_crtc_state() 
The tegra atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-29-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:18 +02:00
Maxime Ripard
9bb3ceb961 drm/sun4i: Switch to drm_atomic_get_new_crtc_state() 
The sun4i atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-28-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:18 +02:00
Maxime Ripard
0a36e14bef drm/rockchip: Switch to drm_atomic_get_new_crtc_state() 
The rockchip atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-27-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:17 +02:00
Maxime Ripard
0a10e39982 drm/omap: Switch to drm_atomic_get_new_crtc_state() 
The omap atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-26-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:17 +02:00
Maxime Ripard
7be36c7b60 drm/msm/mdp5: Switch to drm_atomic_get_new_crtc_state() 
The msm atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-25-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:16 +02:00
Maxime Ripard
afea87890d drm/mediatek: Switch to drm_atomic_get_new_crtc_state() 
The mediatek atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-24-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:16 +02:00
Maxime Ripard
97481a9a00 drm/loongson: Switch to drm_atomic_get_new_crtc_state() 
The loongson atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-23-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:15 +02:00
Maxime Ripard
addc98e2f9 drm/logicvc: Switch to drm_atomic_get_new_crtc_state() 
The logicvc atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-22-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:15 +02:00
Maxime Ripard
afce57a71d drm/kmb: Switch to drm_atomic_get_new_crtc_state() 
The kmb atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-21-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:14 +02:00
Maxime Ripard
21c73cab4d drm/ingenic: Switch to drm_atomic_get_new_crtc_state() 
The ingenic atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Paul Cercueil <paul@crapouillou.net>
Tested-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-20-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:13 +02:00
Maxime Ripard
ba7c370cbe drm/imx-ipuv3: Switch to drm_atomic_get_new_crtc_state() 
The imx-ipuv3 atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-19-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:13 +02:00
Maxime Ripard
14ccb2861a drm/imx-dcss: Switch to drm_atomic_get_new_crtc_state() 
The imx-dcss atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Laurentiu Palcu <laurentiu.palcu@oss.nxp.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-18-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:12 +02:00
Maxime Ripard
7f3364784b drm/imx-dc: Switch to drm_atomic_get_new_crtc_state() 
The imx-dc atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Liu Ying <victor.liu@nxp.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-17-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:12 +02:00
Maxime Ripard
aa61da4324 drm/exynos: Switch to drm_atomic_get_new_crtc_state() 
The exynos atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-16-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:11 +02:00
Maxime Ripard
c0253fc4c8 drm/atmel-hlcdc: Switch to drm_atomic_get_new_crtc_state() 
The atmel-hlcdc atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Manikandan Muralidharan <manikandan.m@microchip.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-15-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:11 +02:00
Maxime Ripard
488b14a0ac drm/armada: Switch to drm_atomic_get_new_crtc_state() 
The armada atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-14-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:10 +02:00
Maxime Ripard
3d77587374 drm/armada: Drop always true condition in atomic_check 
The drm_atomic_state pointer passed to atomic_check is always valid, so
checking if it's NULL or not is pointless. Remove the check.

Suggested-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-13-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:10 +02:00
Maxime Ripard
41230da56d drm/arm/malidp: Switch to drm_atomic_get_new_crtc_state() 
The malidp atomic_check implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called as part of the global atomic_check, thus before the
states are swapped. The existing state thus points to the new state, and
we can use drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Liviu Dudau <liviu.dudau@arm.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-12-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:09 +02:00
Maxime Ripard
78f55a23d1 drm/ingenic: ipu: Switch to drm_atomic_get_new_crtc_state() 
The ingenic IPU atomic_set_property implementation uses the deprecated
drm_atomic_get_existing_crtc_state() helper.

This hook is called during the state building process, before
atomic_check, and thus before the states are swapped. The existing state
thus points to the new state, and we can use
drm_atomic_get_new_crtc_state() instead.

Reviewed-by: Paul Cercueil <paul@crapouillou.net>
Tested-by: Paul Cercueil <paul@crapouillou.net>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-11-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:09 +02:00
Maxime Ripard
b8563c9418 drm/atomic: Convert drm_atomic_get_crtc_state() to use new connector state 
The drm_atomic_get_crtc_state() function calls the deprecated
drm_atomic_get_existing_crtc_state() helper to get find if a crtc state
had already been allocated and was part of the given drm_atomic_state.

At the point in time where drm_atomic_get_crtc_state() can be
called (ie, during atomic_check), the existing state is the new state
and drm_atomic_get_existing_crtc_state() can thus be replaced by
drm_atomic_get_new_crtc_state().

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-10-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:08 +02:00
Maxime Ripard
c92e072dfa drm/tilcdc: crtc: Use drm_atomic_helper_check_crtc_primary_plane() 
In the tilcdc_crtc_atomic_check(), the tilcdc driver hand-crafts its own
implementation of drm_atomic_helper_check_crtc_primary_plane(). And it
does so by accessing the state pointer in drm_atomic_state->planes which
is deprecated.

Let's use the right helper here.

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Acked-by: Jyri Sarha <jyri.sarha@iki.fi>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-7-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:07 +02:00
Maxime Ripard
28af3ca1de drm/vkms: Convert vkms_crtc_atomic_check() to use new plane state 
The vkms_crtc_atomic_check() function calls the deprecated
drm_atomic_get_existing_plane_state() helper for its internal mode
checking logic.

During atomic_check, the existing state is the new state and
drm_atomic_get_existing_plane_state() can thus be replaced by
drm_atomic_get_new_plane_state().

Reviewed-by: Louis Chauvet <louis.chauvet@bootlin.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-6-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:06 +02:00
Maxime Ripard
618c234740 drm/atomic: Convert drm_atomic_get_plane_state() to use new plane state 
The drm_atomic_get_plane_state() function calls the deprecated
drm_atomic_get_existing_plane_state() helper to get find if a plane
state had already been allocated and was part of the given
drm_atomic_state.

At the point in time where drm_atomic_get_plane_state() can be called
(ie, during atomic_check), the existing state is the new state and
drm_atomic_get_existing_plane_state() can thus be replaced by
drm_atomic_get_new_plane_state().

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-5-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:59:05 +02:00
Maxime Ripard
fa02f9b2ff drm/atomic: Convert drm_atomic_get_connector_state() to use new connector state 
The drm_atomic_get_connector_state() function calls a hand-rolled
implementation of the deprecated
drm_atomic_get_existing_connector_state() helper to get find if a
connector state had already been allocated and was part of the given
drm_atomic_state.

At the point in time where drm_atomic_get_connector_state() can be
called (ie, during atomic_check), the existing state is the new state
and drm_atomic_get_existing_connector_state() can thus be replaced by
drm_atomic_get_new_connector_state().

Reviewed-by: Luca Ceresoli <luca.ceresoli@bootlin.com>
Tested-by: Luca Ceresoli <luca.ceresoli@bootlin.com> # on imx8mp
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com>
Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://lore.kernel.org/r/20250930-drm-no-more-existing-state-v5-1-eeb9e1287907@kernel.org
Signed-off-by: Maxime Ripard <mripard@kernel.org>
 2025-10-06 13:58:53 +02:00
Chia-I Wu
4e04683022 drm/panthor: add custom ASN_HASH support for mt8196 
Add panthor_soc_data to control custom ASN_HASH. Add compatible string
for "mediatek,mt8196-mali" and enable custom ASN_HASH for the soc.

Without custom ASN_HASH, FW fails to boot

  panthor 48000000.gpu: [drm] *ERROR* Unhandled Page fault in AS0 at VA 0x0000000000000000
  panthor 48000000.gpu: [drm] *ERROR* Failed to boot MCU (status=fatal)
  panthor 48000000.gpu: probe with driver panthor failed with error -110

With custom ASN_HASH, panthor probes fine and userspace boots to ui just
fine as well

  panthor 48000000.gpu: [drm] clock rate = 0
  panthor 48000000.gpu: EM: created perf domain
  panthor 48000000.gpu: [drm] Mali-G925-Immortalis id 0xd830 major 0x0 minor 0x1 status 0x5
  panthor 48000000.gpu: [drm] Features: L2:0x8130306 Tiler:0x809 Mem:0x301 MMU:0x2830 AS:0xff
  panthor 48000000.gpu: [drm] shader_present=0xee0077 l2_present=0x1 tiler_present=0x1
  panthor 48000000.gpu: [drm] Firmware protected mode entry not be supported, ignoring
  panthor 48000000.gpu: [drm] Firmware git sha: 27713280172c742d467a4b7d11180930094092ec
  panthor 48000000.gpu: [drm] CSF FW using interface v3.13.0, Features 0x10 Instrumentation features 0x71
  [drm] Initialized panthor 1.5.0 for 48000000.gpu on minor 1

Note that the clock and the regulator drivers are not upstreamed yet.
They might as well take a different form when upstreamed.

Signed-off-by: Chia-I Wu <olvaffe@gmail.com>
Reviewed-by: Boris Brezillon <boris.brezillon@collabora.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Steven Price <steven.price@arm.com>
Link: https://lore.kernel.org/r/20250913002155.1163908-3-olvaffe@gmail.com
 2025-10-06 10:50:59 +01:00
Piotr Piórkowski
8462d16d1b drm/xe: Combine userspace context check 
Both vm->xef and XE_LRC_CREATE_USER_CTX indicate in xe_lrc_init that
the context originates from userspace. However, XE_LRC_CREATE_USER_CTX
has a broader scope as it may be set even when no vm->xef is present.
The XE_BO_FLAG_PINNED_LATE_RESTORE flag can be extended to both cases,
so there is no point in handling the two cases separately.
Let's combine vm->xef and XE_LRC_CREATE_USER_CTX checks to detect
userspace context.

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Suggested-by: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251003162619.1984236-6-piotr.piorkowski@intel.com
 2025-10-06 08:33:52 +02:00
Piotr Piórkowski
b48140f446 drm/xe/pf: Force use user VRAM for LMEM provisioning 
The LMEM assigned to VFs should be allocated from the general-purpose
VRAM pool, not from the kernel-reserved region.
Let's force the use of general-purpose VRAM for BOs intended for VFs.

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251003162619.1984236-5-piotr.piorkowski@intel.com
 2025-10-06 08:33:51 +02:00
Piotr Piórkowski
3f6cd669d5 drm/xe: Force user context allocations in user VRAM 
In general, kernel structures should be allocated in the kernel-dedicated
VRAM region. However, userspace context data - while used by the kernel -
does not need to reside there.
Let's force the allocation of such data in the general-purpose VRAM region
accessible to userspace.

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251003162619.1984236-4-piotr.piorkowski@intel.com
 2025-10-06 08:33:49 +02:00
Piotr Piórkowski
9d290ab0b5 drm/xe: Introduce new BO flag XE_BO_FLAG_FORCE_USER_VRAM 
When using a separate VRAM region for kernel allocations,
some kernel structures, such as context userspace data,
should not reside in the VRAM region dedicated to the kernel.
The VRAM kernel region is intended only for allocations necessary
for driver operation. Allocations created via ioctl are long-lived
and not easily evictable. If this region runs out of space,
there may not be a fallback, which could cause failures.
To prevent this, add a new BO flag that explicitly forces the BO to be
allocated in the general-purpose VRAM region accessible to userspace,
avoiding the kernel-only VRAM region.

v2:
 - update commit message (Matthew)

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251003162619.1984236-3-piotr.piorkowski@intel.com
 2025-10-06 08:33:48 +02:00
Piotr Piórkowski
db7dde9904 drm/xe: Add initial support for separate kernel VRAM region on the tile 
So far, kernel and userspace allocations have shared the same VRAM region.
However, in some scenarios, it may be necessary to reserve a separate
VRAM area exclusively for kernel allocations.
Let's add preliminary support for such a configuration.

v2:
- replaced for_each_bo_flag_vram with the improved
  for_each_set_bo_vram_flag helper (Matthew)
- moved the VRAM flag iteration macro definition into xe_bo.c (Matthew)
- drop unused bo_flgas from bo_vram_flags_to_vram_placement (Matthew)
- use hweight32 helper in __xe_bo_fixed_placement for readability
  (Matthew)
v3: remove unnecessary VRAM fixup id

Signed-off-by: Piotr Piórkowski <piotr.piorkowski@intel.com>
Cc: Matthew Auld <matthew.auld@intel.com>
Reviewed-by: Matthew Auld <matthew.auld@intel.com>
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Link: https://lore.kernel.org/r/20251003162619.1984236-2-piotr.piorkowski@intel.com
 2025-10-06 08:33:46 +02:00
Ville Syrjälä
32fdf8f418 drm/i915: s/intel_get_linetime_us()/skl_wm_linetime_us()/ 
Rename intel_get_linetime_us() to skl_wm_linetime_us() to better
reflect that it's not meant to be used for anything apart from
the watermark calculations.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250919180838.10498-4-ville.syrjala@linux.intel.com
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
 2025-10-04 15:17:18 +03:00
Ville Syrjälä
09f21bee91 drm/i915: Deobfuscate wm linetime calculation 
intel_get_linetime_us() is a mess. Rewrite it in a straightforward
manner. Also the checks for the !active and pixel_rate==0 are
completely pointless here since we know that the plane is visible.

Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20250919180838.10498-3-ville.syrjala@linux.intel.com
Reviewed-by: Vinod Govindapillai <vinod.govindapillai@intel.com>
 2025-10-04 15:16:32 +03:00