amd-staging-drm-next breaks suspend

Thu Jan 20 00:48:12 UTC 2022

Bisected the error and found the first bad commit to be
d015e9861e55928a78137a2c95897bc50637fc47 is the first bad commit
commit d015e9861e55928a78137a2c95897bc50637fc47
Author: Jonathan Kim <jonathan.kim at amd.com>
Date:   Thu Dec 9 16:48:56 2021 -0500

    drm/amdgpu: improve debug VRAM access performance using sdma

    For better performance during VRAM access for debugged processes,
do
    read/write copies over SDMA.

    In order to fulfill post mortem debugging on a broken device,
fallback to
    stable MMIO access when gpu recovery is disabled or when job
submission
    time outs are set to max.  Failed SDMA access should automatically
fall
    back to MMIO access.

    Use a pre-allocated GTT bounce buffer pre-mapped into GART to avoid
    page-table updates and TLB flushes on access.

    Signed-off-by: Jonathan Kim <jonathan.kim at amd.com>
    Reviewed-by: Felix Kuehling <felix.kuehling at amd.com>

 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.c | 78
+++++++++++++++++++++++++++++++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_ttm.h |  4 ++
 2 files changed, 82 insertions(+)


Am Donnerstag, dem 20.01.2022 um 00:22 +0100 schrieb Bert Karwatzki:
> Reverting commit 72f686438de13f121c52f58d7445570a33dfdc61 does not
> change the errors:
> [    1.310550] ------------[ cut here ]------------
> [    1.310551] trying to bind memory to uninitialized GART !
> [    1.310556] WARNING: CPU: 9 PID: 252 at
> drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c:254
> amdgpu_gart_bind+0x2e/0x40
> [amdgpu]
> [    1.310659] Modules linked in: amdgpu(+) gpu_sched i2c_algo_bit
> drm_ttm_helper hid_sensor_hub ttm hid_generic nvme drm_kms_helper
> nvme_core cec xhci_pci t10_pi r8169 rc_core crc32_pclmul crc_t10dif
> i2c_hid_acpi realtek xhci_hcd psmouse crc32c_intel crct10dif_generic
> i2c_hid amd_sfh mdio_devres crct10dif_pclmul drm i2c_piix4 usbcore
> libphy crct10dif_common wmi button battery video fjes(-) hid
> [    1.310672] CPU: 9 PID: 252 Comm: systemd-udevd Not tainted
> 5.13.0+
> #4
> [    1.310673] Hardware name: Micro-Star International Co., Ltd.
> Alpha
> 15 B5EEK/MS-158L, BIOS E158LAMS.107 11/10/2021
> [    1.310674] RIP: 0010:amdgpu_gart_bind+0x2e/0x40 [amdgpu]
> [    1.310762] Code: 00 80 bf 34 25 00 00 00 74 14 4c 8b 8f 20 25 00
> 00
> 4d 85 c9 74 05 e9 01 ff ff ff 31 c0 c3 48 c7 c7 68 36 dd c0 e8 86 db
> 19
> e8 <0f> 0b b8 ea ff ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 44
> 00
> [    1.310763] RSP: 0018:ffffb19d00c33920 EFLAGS: 00010282
> [    1.310764] RAX: 0000000000000000 RBX: 0000000000000067 RCX:
> ffffffffa9abb208
> [    1.310765] RDX: 0000000000000000 RSI: 00000000ffffefff RDI:
> ffffffffa9a63200
> [    1.310766] RBP: ffff985ce2a796c0 R08: 0000000000000000 R09:
> ffffb19d00c33748
> [    1.310766] R10: ffffb19d00c33740 R11: ffffffffa9ad3248 R12:
> 0000000000000000
> [    1.310766] R13: ffff985cd45a0000 R14: ffff985cd45a0000 R15:
> 0000000000000000
> [    1.310767] FS:  00007f69fabdc8c0(0000) GS:ffff985f9e640000(0000)
> knlGS:0000000000000000
> [    1.310768] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [    1.310768] CR2: 00007f69fabc5dca CR3: 00000001139ec000 CR4:
> 0000000000750ee0
> [    1.310769] PKRU: 55555554
> [    1.310770] Call Trace:
> [    1.310772]  amdgpu_ttm_gart_bind+0x79/0xc0 [amdgpu]
> [    1.310858]  amdgpu_ttm_alloc_gart+0x146/0x1a0 [amdgpu]
> [    1.310942]  amdgpu_bo_create_reserved.part.0+0xf8/0x1b0 [amdgpu]
> [    1.311025]  ? amdgpu_ttm_debugfs_init+0x110/0x110 [amdgpu]
> [    1.311145]  amdgpu_bo_create_kernel+0x3b/0xa0 [amdgpu]
> [    1.311229]  amdgpu_ttm_init.cold+0x165/0x17f [amdgpu]
> [    1.311349]  gmc_v10_0_sw_init+0x2dc/0x430 [amdgpu]
> [    1.311455]  amdgpu_device_init.cold+0x1544/0x1b54 [amdgpu]
> [    1.311570]  ? acpi_ns_get_node+0x4f/0x5a
> [    1.311574]  ? acpi_get_handle+0x8e/0xb7
> [    1.311576]  amdgpu_driver_load_kms+0x67/0x320 [amdgpu]
> [    1.311664]  amdgpu_pci_probe+0x1bc/0x290 [amdgpu]
> [    1.311750]  local_pci_probe+0x42/0x80
> [    1.311753]  ? __cond_resched+0x16/0x40
> [    1.311755]  pci_device_probe+0xfd/0x1b0
> [    1.311756]  really_probe+0xf2/0x460
> [    1.311759]  driver_probe_device+0xe8/0x160
> [    1.311760]  device_driver_attach+0xa1/0xb0
> [    1.311761]  __driver_attach+0x8f/0x150
> [    1.311763]  ? device_driver_attach+0xb0/0xb0
> [    1.311764]  ? device_driver_attach+0xb0/0xb0
> [    1.311765]  bus_for_each_dev+0x78/0xc0
> [    1.311766]  bus_add_driver+0x12b/0x1e0
> [    1.311768]  driver_register+0x8f/0xe0
> [    1.311769]  ? 0xffffffffc1828000
> [    1.311770]  do_one_initcall+0x44/0x1d0
> [    1.311772]  ? kmem_cache_alloc_trace+0x103/0x240
> [    1.311775]  do_init_module+0x5c/0x270
> [    1.311777]  __do_sys_finit_module+0xb1/0x110
> [    1.311779]  do_syscall_64+0x40/0xb0
> [    1.311781]  entry_SYSCALL_64_after_hwframe+0x44/0xae
> [    1.311783] RIP: 0033:0x7f69fb094679
> [    1.311785] Code: 48 8d 3d 9a a1 0c 00 0f 05 eb a5 66 0f 1f 44 00
> 00
> 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 c7 57 0c 00 f7 d8 64 89 01
> 48
> [    1.311786] RSP: 002b:00007ffce4131708 EFLAGS: 00000246 ORIG_RAX:
> 0000000000000139
> [    1.311788] RAX: ffffffffffffffda RBX: 000055d71350a3a0 RCX:
> 00007f69fb094679
> [    1.311788] RDX: 0000000000000000 RSI: 00007f69fb234eed RDI:
> 0000000000000013
> [    1.311789] RBP: 0000000000020000 R08: 0000000000000000 R09:
> 000055d7134f3930
> [    1.311789] R10: 0000000000000013 R11: 0000000000000246 R12:
> 00007f69fb234eed
> [    1.311790] R13: 0000000000000000 R14: 000055d7134da0f0 R15:
> 000055d71350a3a0
> [    1.311791] ---[ end trace ff47998e3140e95d ]---
> [    1.311793] [drm:amdgpu_ttm_gart_bind [amdgpu]] *ERROR* failed to
> bind 1 pages at 0x00400000
> [    1.312100] amdgpu 0000:03:00.0: amdgpu: 00000000989bdfac bind
> failed
>
> and using https://patchwork.freedesktop.org/patch/469907/
> gives a this message:
>
> [    1.311502] ------------[ cut here ]------------
> [    1.311502] WARNING: CPU: 9 PID: 221 at
> drivers/gpu/drm/amd/amdgpu/amdgpu_gart.c:244
> amdgpu_gart_bind+0x16/0x20
> [amdgpu]
> [    1.311602] Modules linked in: amdgpu(+) gpu_sched i2c_algo_bit
> drm_ttm_helper hid_sensor_hub ttm hid_generic nvme xhci_pci
> drm_kms_helper nvme_core t10_pi xhci_hcd crc_t10dif r8169 cec
> crct10dif_generic i2c_hid_acpi amd_sfh rc_core crct10dif_pclmul
> realtek
> i2c_hid crc32_pclmul mdio_devres psmouse usbcore crc32c_intel drm
> libphy i2c_piix4 crct10dif_common button wmi battery video fjes(-)
> hid
> [    1.311614] CPU: 9 PID: 221 Comm: systemd-udevd Not tainted
> 5.13.0+
> #6
> [    1.311616] Hardware name: Micro-Star International Co., Ltd.
> Alpha
> 15 B5EEK/MS-158L, BIOS E158LAMS.107 11/10/2021
> [    1.311617] RIP: 0010:amdgpu_gart_bind+0x16/0x20 [amdgpu]
> [    1.311701] Code: 39 df 74 aa eb dc e8 19 a2 f6 f0 66 0f 1f 84 00
> 00
> 00 00 00 0f 1f 44 00 00 4c 8b 8f 20 25 00 00 4d 85 c9 74 05 e9 3a ff
> ff
> ff <0f> 0b c3 0f 1f 80 00 00 00 00 0f 1f 44 00 00 55 48 89 fd 53 0f
> ae
> [    1.311702] RSP: 0018:ffffb5be80b17948 EFLAGS: 00010246
> [    1.311703] RAX: 0000000000000022 RBX: ffff98670583e858 RCX:
> ffff98677c8a8738
> [    1.311704] RDX: 0000000000000001 RSI: 0000000000400000 RDI:
> ffff986720ce0000
> [    1.311704] RBP: ffff986705841a08 R08: 0000000000000067 R09:
> 0000000000000000
> [    1.311705] R10: ffff986705841a08 R11: 0000000000000400 R12:
> 0000000000000000
> [    1.311705] R13: ffff98670dc50e40 R14: ffff986720ce0000 R15:
> 0000000000000000
> [    1.311706] FS:  00007ff4ee0968c0(0000) GS:ffff9869de840000(0000)
> knlGS:0000000000000000
> [    1.311707] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [    1.311707] CR2: 00007ff4ee07fdca CR3: 000000017b436000 CR4:
> 0000000000750ee0
> [    1.311708] PKRU: 55555554
> [    1.311708] Call Trace:
> [    1.311710]  amdgpu_ttm_alloc_gart+0x147/0x190 [amdgpu]
> [    1.311793]  amdgpu_bo_create_reserved.part.0+0xf8/0x1b0 [amdgpu]
> [    1.311873]  ? amdgpu_ttm_debugfs_init+0x110/0x110 [amdgpu]
> [    1.311952]  amdgpu_bo_create_kernel+0x3b/0xa0 [amdgpu]
> [    1.312031]  amdgpu_ttm_init.cold+0x165/0x17f [amdgpu]
> [    1.312181]  gmc_v10_0_sw_init+0x2dc/0x430 [amdgpu]
> [    1.312275]  amdgpu_device_init.cold+0x1544/0x1b54 [amdgpu]
> [    1.312385]  ? acpi_ns_get_node+0x4f/0x5a
> [    1.312388]  ? acpi_get_handle+0x8e/0xb7
> [    1.312390]  amdgpu_driver_load_kms+0x67/0x320 [amdgpu]
> [    1.312479]  amdgpu_pci_probe+0x1bc/0x290 [amdgpu]
> [    1.312573]  local_pci_probe+0x42/0x80
> [    1.312578]  ? __cond_resched+0x16/0x40
> [    1.312581]  pci_device_probe+0xfd/0x1b0
> [    1.312583]  really_probe+0xf2/0x460
> [    1.312587]  driver_probe_device+0xe8/0x160
> [    1.312589]  device_driver_attach+0xa1/0xb0
> [    1.312591]  __driver_attach+0x8f/0x150
> [    1.312592]  ? device_driver_attach+0xb0/0xb0
> [    1.312593]  ? device_driver_attach+0xb0/0xb0
> [    1.312594]  bus_for_each_dev+0x78/0xc0
> [    1.312595]  bus_add_driver+0x12b/0x1e0
> [    1.312597]  driver_register+0x8f/0xe0
> [    1.312598]  ? 0xffffffffc1696000
> [    1.312599]  do_one_initcall+0x44/0x1d0
> [    1.312602]  ? kmem_cache_alloc_trace+0x103/0x240
> [    1.312604]  do_init_module+0x5c/0x270
> [    1.312606]  __do_sys_finit_module+0xb1/0x110
> [    1.312608]  do_syscall_64+0x40/0xb0
> [    1.312610]  entry_SYSCALL_64_after_hwframe+0x44/0xae
> [    1.312614] RIP: 0033:0x7ff4ee54e679
> [    1.312616] Code: 48 8d 3d 9a a1 0c 00 0f 05 eb a5 66 0f 1f 44 00
> 00
> 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 c7 57 0c 00 f7 d8 64 89 01
> 48
> [    1.312617] RSP: 002b:00007fff9c9d5bc8 EFLAGS: 00000246 ORIG_RAX:
> 0000000000000139
> [    1.312618] RAX: ffffffffffffffda RBX: 000055aca9cf9230 RCX:
> 00007ff4ee54e679
> [    1.312619] RDX: 0000000000000000 RSI: 00007ff4ee6eeeed RDI:
> 0000000000000013
> [    1.312619] RBP: 0000000000020000 R08: 0000000000000000 R09:
> 000055aca9cdf480
> [    1.312620] R10: 0000000000000013 R11: 0000000000000246 R12:
> 00007ff4ee6eeeed
> [    1.312620] R13: 0000000000000000 R14: 000055aca9cf9ab0 R15:
> 000055aca9cf9230
> [    1.312622] ---[ end trace c4fc99f16e2a1eb7 ]---
>
> In both cases there's a similar error message for the oth GPU in the
> system (the integrated one at 08:00.0)
>
> Am Mittwoch, dem 19.01.2022 um 23:40 +0100 schrieb Das, Nirmoy:
> >
> > On 1/19/2022 10:59 PM, Limonciello, Mario wrote:
> > > [Public]
> > >
> > > > -----Original Message-----
> > > > From: Bert Karwatzki <spasswolf at web.de>
> > > > Sent: Wednesday, January 19, 2022 15:52
> > > > To: amd-gfx at lists.freedesktop.org
> > > > Cc: Limonciello, Mario <Mario.Limonciello at amd.com>; Kazlauskas,
> > > > Nicholas
> > > > <Nicholas.Kazlauskas at amd.com>; Zhuo, Qingqing (Lillian)
> > > > <Qingqing.Zhuo at amd.com>; Scott Bruce <smbruce at gmail.com>; Alex
> > > > Deucher
> > > > <alexdeucher at gmail.com>; Chris Hixon
> > > > <linux-kernel-bugs at hixontech.com>
> > > > Subject: amd-staging-drm-next breaks suspend
> > > >
> > > > I just tested drm-staging-drm-next with HEAD
> > > > f1b2924ee6929cb431440e6f961f06eb65d52beb:
> > > > Going into suspend leads to a hang again:
> > > > This is probably caused by
> > > > [ 1.310551] trying to bind memory to uninitialized GART !
> > > > and/or
> > > > [ 3.976438] trying to bind memory to uninitialized GART !
> >
> >
> > Could you please also try
> > https://patchwork.freedesktop.org/patch/469907/ ?
> >
> >
> > Regards,
> >
> > Nirmoy
> >
> > > >