linux-stable.git/drivers/gpu/drm/amd/amdgpu, branch linux-5.13.y

drm/amdgpu: Fix a deadlock if previous GEM object allocation fails

2021-09-18T11:42:23+00:00

commit 703677d9345d87d7288ed8a2483ca424af7d4b3b upstream.

Fall through to handle the error instead of return.

Fixes: f8aab60422c37 ("drm/amdgpu: Initialise drm_gem_object_funcs for imported BOs")
Cc: stable@vger.kernel.org
Signed-off-by: xinhui pan 
Reviewed-by: Christian König 
Signed-off-by: Alex Deucher 
Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: Fix BUG_ON assert

2021-09-18T11:42:23+00:00

commit ea7acd7c5967542353430947f3faf699e70602e5 upstream.

With added CPU domain to placement you can have
now 3 placemnts at once.

CC: stable@kernel.org
Signed-off-by: Andrey Grodzovsky 
Reviewed-by: Christian König 
Link: https://patchwork.freedesktop.org/patch/msgid/20210622162339.761651-5-andrey.grodzovsky@amd.com
Signed-off-by: Greg Kroah-Hartman

gpu: drm: amd: amdgpu: amdgpu_i2c: fix possible uninitialized-variable access in amdgpu_i2c_router_select_ddc_port()

2021-09-18T11:42:11+00:00

[ Upstream commit a211260c34cfadc6068fece8c9e99e0fe1e2a2b6 ]

The variable val is declared without initialization, and its address is
passed to amdgpu_i2c_get_byte(). In this function, the value of val is
accessed in:
  DRM_DEBUG("i2c 0x%02x 0x%02x read failed\n",
       addr, *val);

Also, when amdgpu_i2c_get_byte() returns, val may remain uninitialized,
but it is accessed in:
  val &= ~amdgpu_connector->router.ddc_mux_control_pin;

To fix this possible uninitialized-variable access, initialize val to 0 in
amdgpu_i2c_router_select_ddc_port().

Reported-by: TOTE Robot 
Signed-off-by: Tuo Li 
Signed-off-by: Alex Deucher 
Signed-off-by: Sasha Levin

drm/amdgpu: Fix a printing message

2021-09-18T11:42:03+00:00

[ Upstream commit 95f71f12aa45d65b7f2ccab95569795edffd379a ]

The printing message "PSP loading VCN firmware" is mis-leading because
people might think driver is loading VCN firmware. Actually when this
message is printed, driver is just preparing some VCN ucode, not loading
VCN firmware yet. The actual VCN firmware loading will be in the PSP block
hw_init. Fix the printing message

Signed-off-by: Oak Zeng 
Reviewed-by: Christian Konig 
Signed-off-by: Alex Deucher 
Signed-off-by: Sasha Levin

drm/amdgpu: Fix amdgpu_ras_eeprom_init()

2021-09-18T11:41:58+00:00

[ Upstream commit dce4400e6516d18313d23de45b5be8a18980b00e ]

No need to account for the 2 bytes of EEPROM
address--this is now well abstracted away by
the fixes the the lower layers.

Cc: Andrey Grodzovsky 
Cc: Alexander Deucher 
Signed-off-by: Luben Tuikov 
Acked-by: Alexander Deucher 
Signed-off-by: Alex Deucher 
Signed-off-by: Sasha Levin

drm/amdgpu/acp: Make PM domain really work

2021-09-15T08:00:39+00:00

[ Upstream commit aff890288de2d818e4f83ec40c9315e2d735df07 ]

Devices created by mfd_add_hotplug_devices() don't really increase the
index of its name, so get_mfd_cell_dev() cannot find any device, hence a
NULL dev is passed to pm_genpd_add_device():
[   56.974926] (NULL device *): amdgpu: device acp_audio_dma.0.auto added to pm domain
[   56.974933] (NULL device *): amdgpu: Failed to add dev to genpd
[   56.974941] [drm:amdgpu_device_ip_init [amdgpu]] *ERROR* hw_init of IP block  failed -22
[   56.975810] amdgpu 0000:00:01.0: amdgpu: amdgpu_device_ip_init failed
[   56.975839] amdgpu 0000:00:01.0: amdgpu: Fatal error during GPU init
[   56.977136] ------------[ cut here ]------------
[   56.977143] kernel BUG at mm/slub.c:4206!
[   56.977158] invalid opcode: 0000 [#1] SMP NOPTI
[   56.977167] CPU: 1 PID: 1648 Comm: modprobe Not tainted 5.12.0-051200rc8-generic #202104182230
[   56.977175] Hardware name: To Be Filled By O.E.M. To Be Filled By O.E.M./FM2A68M-HD+, BIOS P5.20 02/13/2019
[   56.977180] RIP: 0010:kfree+0x3bf/0x410
[   56.977195] Code: 89 e7 48 d3 e2 f7 da e8 5f 0d 02 00 80 e7 02 75 3e 44 89 ee 4c 89 e7 e8 ef 5f fd ff e9 fa fe ff ff 49 8b 44 24 08 a8 01 75 b7 <0f> 0b 4c 8b 4d b0 48 8b 4d a8 48 89 da 4c 89 e6 41 b8 01 00 00 00
[   56.977202] RSP: 0018:ffffa48640ff79f0 EFLAGS: 00010246
[   56.977210] RAX: 0000000000000000 RBX: ffff9286127d5608 RCX: 0000000000000000
[   56.977215] RDX: 0000000000000000 RSI: ffffffffc099d0fb RDI: ffff9286127d5608
[   56.977220] RBP: ffffa48640ff7a48 R08: 0000000000000001 R09: 0000000000000001
[   56.977224] R10: 0000000000000000 R11: ffff9286087d8458 R12: fffff3ae0449f540
[   56.977229] R13: 0000000000000000 R14: dead000000000122 R15: dead000000000100
[   56.977234] FS:  00007f9de5929540(0000) GS:ffff928612e80000(0000) knlGS:0000000000000000
[   56.977240] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[   56.977245] CR2: 00007f697dd97160 CR3: 00000001110f0000 CR4: 00000000001506e0
[   56.977251] Call Trace:
[   56.977261]  amdgpu_dm_encoder_destroy+0x1b/0x30 [amdgpu]
[   56.978056]  drm_mode_config_cleanup+0x4f/0x2e0 [drm]
[   56.978147]  ? kfree+0x3dd/0x410
[   56.978157]  ? drm_managed_release+0xc8/0x100 [drm]
[   56.978232]  drm_mode_config_init_release+0xe/0x10 [drm]
[   56.978311]  drm_managed_release+0x9d/0x100 [drm]
[   56.978388]  devm_drm_dev_init_release+0x4d/0x70 [drm]
[   56.978450]  devm_action_release+0x15/0x20
[   56.978459]  release_nodes+0x77/0xc0
[   56.978469]  devres_release_all+0x3f/0x50
[   56.978477]  really_probe+0x245/0x460
[   56.978485]  driver_probe_device+0xe9/0x160
[   56.978492]  device_driver_attach+0xab/0xb0
[   56.978499]  __driver_attach+0x8f/0x150
[   56.978506]  ? device_driver_attach+0xb0/0xb0
[   56.978513]  bus_for_each_dev+0x7e/0xc0
[   56.978521]  driver_attach+0x1e/0x20
[   56.978528]  bus_add_driver+0x135/0x1f0
[   56.978534]  driver_register+0x91/0xf0
[   56.978540]  __pci_register_driver+0x54/0x60
[   56.978549]  amdgpu_init+0x77/0x1000 [amdgpu]
[   56.979246]  ? 0xffffffffc0dbc000
[   56.979254]  do_one_initcall+0x48/0x1d0
[   56.979265]  ? kmem_cache_alloc_trace+0x120/0x230
[   56.979274]  ? do_init_module+0x28/0x280
[   56.979282]  do_init_module+0x62/0x280
[   56.979288]  load_module+0x71c/0x7a0
[   56.979296]  __do_sys_finit_module+0xc2/0x120
[   56.979305]  __x64_sys_finit_module+0x1a/0x20
[   56.979311]  do_syscall_64+0x38/0x90
[   56.979319]  entry_SYSCALL_64_after_hwframe+0x44/0xae
[   56.979328] RIP: 0033:0x7f9de54f989d
[   56.979335] Code: 00 c3 66 2e 0f 1f 84 00 00 00 00 00 90 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 c3 f5 0c 00 f7 d8 64 89 01 48
[   56.979342] RSP: 002b:00007ffe3c395a28 EFLAGS: 00000246 ORIG_RAX: 0000000000000139
[   56.979350] RAX: ffffffffffffffda RBX: 0000560df3ef4330 RCX: 00007f9de54f989d
[   56.979355] RDX: 0000000000000000 RSI: 0000560df3a07358 RDI: 000000000000000f
[   56.979360] RBP: 0000000000040000 R08: 0000000000000000 R09: 0000000000000000
[   56.979365] R10: 000000000000000f R11: 0000000000000246 R12: 0000560df3a07358
[   56.979369] R13: 0000000000000000 R14: 0000560df3ef4460 R15: 0000560df3ef4330
[   56.979377] Modules linked in: amdgpu(+) iommu_v2 gpu_sched drm_ttm_helper ttm drm_kms_helper cec rc_core i2c_algo_bit fb_sys_fops syscopyarea sysfillrect sysimgblt nft_counter xt_tcpudp ipt_REJECT nf_reject_ipv4 xt_conntrack iptable_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 iptable_mangle iptable_raw iptable_security ip_set nf_tables libcrc32c nfnetlink ip6_tables iptable_filter bpfilter input_leds binfmt_misc edac_mce_amd kvm_amd ccp kvm snd_hda_codec_realtek snd_hda_codec_generic crct10dif_pclmul snd_hda_codec_hdmi ledtrig_audio ghash_clmulni_intel aesni_intel snd_hda_intel snd_intel_dspcfg snd_seq_midi crypto_simd snd_intel_sdw_acpi cryptd snd_hda_codec snd_seq_midi_event snd_rawmidi snd_hda_core snd_hwdep snd_seq fam15h_power k10temp snd_pcm snd_seq_device snd_timer snd mac_hid soundcore sch_fq_codel nct6775 hwmon_vid drm ip_tables x_tables autofs4 dm_mirror dm_region_hash dm_log hid_generic usbhid hid uas usb_storage r8169 crc32_pclmul realtek ahci xhci_pci i2c_piix4
[   56.979521]  xhci_pci_renesas libahci video
[   56.979541] ---[ end trace cb8f6a346f18da7b ]---

Instead of finding MFD hotplugged device by its name, simply iterate
over the child devices to avoid the issue.

Squash in unused variable removal (Alex)

BugLink: https://bugs.launchpad.net/bugs/1920674
Fixes: 25030321ba28 ("drm/amd: add pm domain for ACP IP sub blocks")
Signed-off-by: Kai-Heng Feng 
Signed-off-by: Alex Deucher 
Signed-off-by: Sasha Levin

drm/amdgpu: Fix build with missing pm_suspend_target_state module export

2021-09-03T08:22:49+00:00

commit c41a4e877a185241d8e83501453326fb98f67354 upstream.

Building a randconfig here triggered:

  ERROR: modpost: "pm_suspend_target_state" [drivers/gpu/drm/amd/amdgpu/amdgpu.ko] undefined!

because the module export of that symbol happens in
kernel/power/suspend.c which is enabled with CONFIG_SUSPEND.

The ifdef guards in amdgpu_acpi_is_s0ix_supported(), however, test for
CONFIG_PM_SLEEP which is defined like this:

  config PM_SLEEP
          def_bool y
          depends on SUSPEND || HIBERNATE_CALLBACKS

and that randconfig has:

  # CONFIG_SUSPEND is not set
  CONFIG_HIBERNATE_CALLBACKS=y

leading to the module export missing.

Change the ifdeffery to depend directly on CONFIG_SUSPEND.

Fixes: 5706cb3c910c ("drm/amdgpu: fix checking pmops when PM_SLEEP is not enabled")
Reviewed-by: Lijo Lazar 
Signed-off-by: Borislav Petkov 
Link: https://lkml.kernel.org/r/YSP6Lv53QV0cOAsd@zn.tnic
Signed-off-by: Alex Deucher 
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: use the preferred pin domain after the check

2021-09-03T08:22:49+00:00

commit 2a7b9a8437130fd328001f4edfac8eec98dfe298 upstream.

For some reason we run into an use case where a BO is already pinned
into GTT, but should be pinned into VRAM|GTT again.

Handle that case gracefully as well.

Reviewed-by: Shashank Sharma 
Reviewed-by: Alex Deucher 
Signed-off-by: Christian König 
Signed-off-by: Alex Deucher 
Cc: stable@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: Cancel delayed work when GFXOFF is disabled

2021-09-03T08:22:49+00:00

commit 32bc8f8373d2d6a681c96e4b25dca60d4d1c6016 upstream.

schedule_delayed_work does not push back the work if it was already
scheduled before, so amdgpu_device_delay_enable_gfx_off ran ~100 ms
after the first time GFXOFF was disabled and re-enabled, even if GFXOFF
was disabled and re-enabled again during those 100 ms.

This resulted in frame drops / stutter with the upcoming mutter 41
release on Navi 14, due to constantly enabling GFXOFF in the HW and
disabling it again (for getting the GPU clock counter).

To fix this, call cancel_delayed_work_sync when the disable count
transitions from 0 to 1, and only schedule the delayed work on the
reverse transition, not if the disable count was already 0. This makes
sure the delayed work doesn't run at unexpected times, and allows it to
be lock-free.

v2:
* Use cancel_delayed_work_sync & mutex_trylock instead of
  mod_delayed_work.
v3:
* Make amdgpu_device_delay_enable_gfx_off lock-free (Christian König)
v4:
* Fix race condition between amdgpu_gfx_off_ctrl incrementing
  adev->gfx.gfx_off_req_count and amdgpu_device_delay_enable_gfx_off
  checking for it to be 0 (Evan Quan)

Cc: stable@vger.kernel.org
Reviewed-by: Evan Quan 
Reviewed-by: Lijo Lazar  # v3
Acked-by: Christian König  # v3
Signed-off-by: Michel Dänzer 
Signed-off-by: Alex Deucher 
Signed-off-by: Greg Kroah-Hartman

drm/amdgpu: fix the doorbell missing when in CGPG issue for renoir.

2021-08-26T12:34:59+00:00

[ Upstream commit 1c0539a6fc8a4a4b77278e35d763073890de96b9 ]

If GC has entered CGPG, ringing doorbell > first page doesn't wakeup GC.
Enlarge CP_MEC_DOORBELL_RANGE_UPPER to workaround this issue.

Signed-off-by: Yifan Zhang 
Reviewed-by: Alex Deucher 
Signed-off-by: Alex Deucher 
Signed-off-by: Sasha Levin