* [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
@ 2023-03-19 2:48 Philip Müller
2023-03-19 3:01 ` Philip Müller
0 siblings, 1 reply; 10+ messages in thread
From: Philip Müller @ 2023-03-19 2:48 UTC (permalink / raw)
To: John Harrison, Greg Kroah-Hartman
Cc: Chris Wilson, Joonas Lahtinen, Jani Nikula, Rodrigo Vivi,
Tvrtko Ursulin, intel-gfx, stable
Hi there,
similar to the last patch breaking i915 in stable kernels now also this
patch creates unbootable systems on Intel GFX:
"kernel: [drm:drm_mode_config_cleanup.cold [drm]] *ERROR* connector DP-3
leaked!"
mrt 18 16:01:23 imac kernel: WARN_ON(({ void *__mptr = (void *)(state);
do { extern void __compiletime_assert_262(void)
__attribute__((error("pointer type mismatch in container_of()"))); if
(!(!(!__builtin_types_compatible_p(typeof(*(state)), typeof(((>
mrt 18 16:01:23 imac kernel: WARNING: CPU: 0 PID: 153 at
drivers/gpu/drm/i915/intel_atomic_plane.c:105
intel_plane_destroy_state+0x37/0x50 [i915]
mrt 18 16:01:23 imac kernel: Modules linked in: i915(+) kvmgt vfio_mdev
sd_mod mdev vfio_iommu_type1 vfio kvm irqbypass intel_gtt i2c_algo_bit
ahci drm_kms_helper libahci syscopyarea sdhci_pci libata xhci_pci cqhci
sysfillrect sysimgblt fb_sys_fops xhc>
mrt 18 16:01:23 imac kernel: CPU: 0 PID: 153 Comm: (udev-worker)
Tainted: G W 4.19.278-1-MANJARO #1
mrt 18 16:01:23 imac kernel: Hardware name: Apple Inc.
iMac16,2/Mac-FFE5EF870D7BA81A, BIOS 428.60.3.0.0 10/27/2021
mrt 18 16:01:23 imac kernel: RIP:
0010:intel_plane_destroy_state+0x37/0x50 [i915]
mrt 18 16:01:23 imac kernel: Code: 90 00 00 00 00 48 89 f3 75 0d 48 89
de 48 89 ef 5b 5d e9 ac 6c e3 ff 48 c7 c6 28 38 7b c0 48 c7 c7 11 f1 79
c0 e8 4e d3 57 e3 <0f> 0b 48 89 de 48 89 ef 5b 5d e9 8a 6c e3 ff 66 2e
0f 1f 84 00 00
mrt 18 16:01:23 imac kernel: RSP: 0018:ffffb0d0011bbb38 EFLAGS: 00010282
mrt 18 16:01:23 imac kernel: RAX: 0000000000000000 RBX: ffffa0ad21eba600
RCX: 0000000000000000
mrt 18 16:01:23 imac kernel: RDX: 00000000000001d7 RSI: ffffffffa4ceb497
RDI: 0000000000000246
mrt 18 16:01:23 imac kernel: RBP: ffffa0ad21490400 R08: 0000000000000002
R09: ffffffffa4ceb2bf
mrt 18 16:01:23 imac kernel: R10: 0000000000000001 R11: 0000000000000000
R12: ffffa0ad21b38000
mrt 18 16:01:23 imac kernel: R13: ffffa0ad21b3c9e8 R14: 00000000fffffffa
R15: ffffa0ad24e05000
mrt 18 16:01:23 imac kernel: FS: 00007f86e003b080(0000)
GS:ffffa0ad26a00000(0000) knlGS:0000000000000000
mrt 18 16:01:23 imac kernel: CS: 0010 DS: 0000 ES: 0000 CR0:
0000000080050033
mrt 18 16:01:23 imac kernel: CR2: 00007f8026f80000 CR3: 000000026193a002
CR4: 00000000003606f0
mrt 18 16:01:23 imac kernel: DR0: 0000000000000000 DR1: 0000000000000000
DR2: 0000000000000000
mrt 18 16:01:23 imac kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0
DR7: 0000000000000400
mrt 18 16:01:23 imac kernel: Call Trace:
mrt 18 16:01:23 imac kernel: drm_plane_cleanup+0xad/0xf0 [drm]
mrt 18 16:01:23 imac kernel: intel_plane_destroy+0xe/0x20 [i915]
mrt 18 16:01:23 imac kernel: drm_mode_config_cleanup+0x104/0x280 [drm]
mrt 18 16:01:23 imac kernel: intel_modeset_cleanup+0xbc/0x130 [i915]
mrt 18 16:01:23 imac kernel: i915_driver_load+0xbad/0xdc0 [i915]
mrt 18 16:01:23 imac kernel: ? acpi_dev_found+0x5f/0x70
mrt 18 16:01:23 imac kernel: local_pci_probe+0x3e/0x80
mrt 18 16:01:23 imac kernel: pci_device_probe+0x102/0x1a0
mrt 18 16:01:23 imac kernel: really_probe+0x245/0x3c0
mrt 18 16:01:23 imac kernel: driver_probe_device+0xb3/0xf0
mrt 18 16:01:23 imac kernel: __driver_attach+0x104/0x110
mrt 18 16:01:23 imac kernel: ? driver_probe_device+0xf0/0xf0
mrt 18 16:01:23 imac kernel: bus_for_each_dev+0x71/0xb0
mrt 18 16:01:23 imac kernel: bus_add_driver+0x16a/0x220
mrt 18 16:01:23 imac kernel: driver_register+0x89/0xd0
mrt 18 16:01:23 imac kernel: ? 0xffffffffc0871000
mrt 18 16:01:23 imac kernel: do_one_initcall+0x46/0x1d0
mrt 18 16:01:23 imac kernel: ? kmem_cache_alloc_trace+0x2f/0x1b0
mrt 18 16:01:23 imac kernel: do_init_module+0x4c/0x260
mrt 18 16:01:23 imac kernel: __se_sys_finit_module+0xb1/0x110
mrt 18 16:01:23 imac kernel: do_syscall_64+0x4e/0x110
mrt 18 16:01:23 imac kernel: entry_SYSCALL_64_after_hwframe+0x5c/0xc1
mrt 18 16:01:23 imac kernel: RIP: 0033:0x7f86e0b660dd
mrt 18 16:01:23 imac kernel: Code: 5d 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 63 7c 0d 00 f7
d8 64 89 01 48
mrt 18 16:01:23 imac kernel: RSP: 002b:00007fff34b17fc8 EFLAGS: 00000246
ORIG_RAX: 0000000000000139
mrt 18 16:01:23 imac kernel: RAX: ffffffffffffffda RBX: 00005566daeded40
RCX: 00007f86e0b660dd
mrt 18 16:01:23 imac kernel: RDX: 0000000000000000 RSI: 00007f86e0cb9343
RDI: 000000000000001b
mrt 18 16:01:23 imac kernel: RBP: 00007f86e0cb9343 R08: 0000000000000000
R09: 0000000000000000
mrt 18 16:01:23 imac kernel: R10: 000000000000001b R11: 0000000000000246
R12: 0000000000020000
mrt 18 16:01:23 imac kernel: R13: 00005566daece520 R14: 00005566daeded40
R15: 00005566daee28f0
mrt 18 16:01:23 imac kernel: ---[ end trace 07e565c61a1cf4c3 ]---
We have again a full house of broken stable kernels, as this patch got
applied to the following releases: 4.14.310, 4.19.278, 5.4.237,
5.10.175, 5.15.103, 6.1.20, 6.2.7
It wouild be good to check which patch-set is needed to get the issue
tried to been fixed here solved and re-apply them in a working set to
stable kernel series next time, instead of let them trickerling down the
stream one-by-one and constantly break i915 driver due this.
A revert of the named patch fixed the "regression".
--
Best, Philip
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
2023-03-19 2:48 [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels Philip Müller
@ 2023-03-19 3:01 ` Philip Müller
2023-03-20 13:28 ` Greg Kroah-Hartman
0 siblings, 1 reply; 10+ messages in thread
From: Philip Müller @ 2023-03-19 3:01 UTC (permalink / raw)
To: John Harrison, Greg Kroah-Hartman
Cc: Chris Wilson, Joonas Lahtinen, Jani Nikula, Rodrigo Vivi,
Tvrtko Ursulin, intel-gfx, stable
Have to correct the affected kernels to these: 4.14.310, 4.19.278,
5.4.237, 5.10.175
On 19.03.23 09:48, Philip Müller wrote:
> Hi there,
>
> similar to the last patch breaking i915 in stable kernels now also this
> patch creates unbootable systems on Intel GFX:
>
> "kernel: [drm:drm_mode_config_cleanup.cold [drm]] *ERROR* connector DP-3
> leaked!"
>
> mrt 18 16:01:23 imac kernel: WARN_ON(({ void *__mptr = (void *)(state);
> do { extern void __compiletime_assert_262(void)
> __attribute__((error("pointer type mismatch in container_of()"))); if
> (!(!(!__builtin_types_compatible_p(typeof(*(state)), typeof(((>
> mrt 18 16:01:23 imac kernel: WARNING: CPU: 0 PID: 153 at
> drivers/gpu/drm/i915/intel_atomic_plane.c:105
> intel_plane_destroy_state+0x37/0x50 [i915]
> mrt 18 16:01:23 imac kernel: Modules linked in: i915(+) kvmgt vfio_mdev
> sd_mod mdev vfio_iommu_type1 vfio kvm irqbypass intel_gtt i2c_algo_bit
> ahci drm_kms_helper libahci syscopyarea sdhci_pci libata xhci_pci cqhci
> sysfillrect sysimgblt fb_sys_fops xhc>
> mrt 18 16:01:23 imac kernel: CPU: 0 PID: 153 Comm: (udev-worker)
> Tainted: G W 4.19.278-1-MANJARO #1
> mrt 18 16:01:23 imac kernel: Hardware name: Apple Inc.
> iMac16,2/Mac-FFE5EF870D7BA81A, BIOS 428.60.3.0.0 10/27/2021
> mrt 18 16:01:23 imac kernel: RIP:
> 0010:intel_plane_destroy_state+0x37/0x50 [i915]
> mrt 18 16:01:23 imac kernel: Code: 90 00 00 00 00 48 89 f3 75 0d 48 89
> de 48 89 ef 5b 5d e9 ac 6c e3 ff 48 c7 c6 28 38 7b c0 48 c7 c7 11 f1 79
> c0 e8 4e d3 57 e3 <0f> 0b 48 89 de 48 89 ef 5b 5d e9 8a 6c e3 ff 66 2e
> 0f 1f 84 00 00
> mrt 18 16:01:23 imac kernel: RSP: 0018:ffffb0d0011bbb38 EFLAGS: 00010282
> mrt 18 16:01:23 imac kernel: RAX: 0000000000000000 RBX: ffffa0ad21eba600
> RCX: 0000000000000000
> mrt 18 16:01:23 imac kernel: RDX: 00000000000001d7 RSI: ffffffffa4ceb497
> RDI: 0000000000000246
> mrt 18 16:01:23 imac kernel: RBP: ffffa0ad21490400 R08: 0000000000000002
> R09: ffffffffa4ceb2bf
> mrt 18 16:01:23 imac kernel: R10: 0000000000000001 R11: 0000000000000000
> R12: ffffa0ad21b38000
> mrt 18 16:01:23 imac kernel: R13: ffffa0ad21b3c9e8 R14: 00000000fffffffa
> R15: ffffa0ad24e05000
> mrt 18 16:01:23 imac kernel: FS: 00007f86e003b080(0000)
> GS:ffffa0ad26a00000(0000) knlGS:0000000000000000
> mrt 18 16:01:23 imac kernel: CS: 0010 DS: 0000 ES: 0000 CR0:
> 0000000080050033
> mrt 18 16:01:23 imac kernel: CR2: 00007f8026f80000 CR3: 000000026193a002
> CR4: 00000000003606f0
> mrt 18 16:01:23 imac kernel: DR0: 0000000000000000 DR1: 0000000000000000
> DR2: 0000000000000000
> mrt 18 16:01:23 imac kernel: DR3: 0000000000000000 DR6: 00000000fffe0ff0
> DR7: 0000000000000400
> mrt 18 16:01:23 imac kernel: Call Trace:
> mrt 18 16:01:23 imac kernel: drm_plane_cleanup+0xad/0xf0 [drm]
> mrt 18 16:01:23 imac kernel: intel_plane_destroy+0xe/0x20 [i915]
> mrt 18 16:01:23 imac kernel: drm_mode_config_cleanup+0x104/0x280 [drm]
> mrt 18 16:01:23 imac kernel: intel_modeset_cleanup+0xbc/0x130 [i915]
> mrt 18 16:01:23 imac kernel: i915_driver_load+0xbad/0xdc0 [i915]
> mrt 18 16:01:23 imac kernel: ? acpi_dev_found+0x5f/0x70
> mrt 18 16:01:23 imac kernel: local_pci_probe+0x3e/0x80
> mrt 18 16:01:23 imac kernel: pci_device_probe+0x102/0x1a0
> mrt 18 16:01:23 imac kernel: really_probe+0x245/0x3c0
> mrt 18 16:01:23 imac kernel: driver_probe_device+0xb3/0xf0
> mrt 18 16:01:23 imac kernel: __driver_attach+0x104/0x110
> mrt 18 16:01:23 imac kernel: ? driver_probe_device+0xf0/0xf0
> mrt 18 16:01:23 imac kernel: bus_for_each_dev+0x71/0xb0
> mrt 18 16:01:23 imac kernel: bus_add_driver+0x16a/0x220
> mrt 18 16:01:23 imac kernel: driver_register+0x89/0xd0
> mrt 18 16:01:23 imac kernel: ? 0xffffffffc0871000
> mrt 18 16:01:23 imac kernel: do_one_initcall+0x46/0x1d0
> mrt 18 16:01:23 imac kernel: ? kmem_cache_alloc_trace+0x2f/0x1b0
> mrt 18 16:01:23 imac kernel: do_init_module+0x4c/0x260
> mrt 18 16:01:23 imac kernel: __se_sys_finit_module+0xb1/0x110
> mrt 18 16:01:23 imac kernel: do_syscall_64+0x4e/0x110
> mrt 18 16:01:23 imac kernel: entry_SYSCALL_64_after_hwframe+0x5c/0xc1
> mrt 18 16:01:23 imac kernel: RIP: 0033:0x7f86e0b660dd
> mrt 18 16:01:23 imac kernel: Code: 5d 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 63 7c 0d 00 f7
> d8 64 89 01 48
> mrt 18 16:01:23 imac kernel: RSP: 002b:00007fff34b17fc8 EFLAGS: 00000246
> ORIG_RAX: 0000000000000139
> mrt 18 16:01:23 imac kernel: RAX: ffffffffffffffda RBX: 00005566daeded40
> RCX: 00007f86e0b660dd
> mrt 18 16:01:23 imac kernel: RDX: 0000000000000000 RSI: 00007f86e0cb9343
> RDI: 000000000000001b
> mrt 18 16:01:23 imac kernel: RBP: 00007f86e0cb9343 R08: 0000000000000000
> R09: 0000000000000000
> mrt 18 16:01:23 imac kernel: R10: 000000000000001b R11: 0000000000000246
> R12: 0000000000020000
> mrt 18 16:01:23 imac kernel: R13: 00005566daece520 R14: 00005566daeded40
> R15: 00005566daee28f0
> mrt 18 16:01:23 imac kernel: ---[ end trace 07e565c61a1cf4c3 ]---
>
> We have again a full house of broken stable kernels, as this patch got
> applied to the following releases: 4.14.310, 4.19.278, 5.4.237,
> 5.10.175, 5.15.103, 6.1.20, 6.2.7
>
> It wouild be good to check which patch-set is needed to get the issue
> tried to been fixed here solved and re-apply them in a working set to
> stable kernel series next time, instead of let them trickerling down the
> stream one-by-one and constantly break i915 driver due this.
>
> A revert of the named patch fixed the "regression".
>
>
--
Best, Philip
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Intel-gfx] [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
2023-03-19 3:01 ` Philip Müller
@ 2023-03-20 13:28 ` Greg Kroah-Hartman
0 siblings, 0 replies; 10+ messages in thread
From: Greg Kroah-Hartman @ 2023-03-20 13:28 UTC (permalink / raw)
To: Philip Müller; +Cc: intel-gfx, Chris Wilson, stable, Rodrigo Vivi
On Sun, Mar 19, 2023 at 10:01:01AM +0700, Philip Müller wrote:
> Have to correct the affected kernels to these: 4.14.310, 4.19.278, 5.4.237,
> 5.10.175
Please don't top-post :(
Anyway, should be fixed in the next round of releases in a few days, if
not, please let us know.
greg k-h
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
@ 2023-03-20 13:28 ` Greg Kroah-Hartman
0 siblings, 0 replies; 10+ messages in thread
From: Greg Kroah-Hartman @ 2023-03-20 13:28 UTC (permalink / raw)
To: Philip Müller
Cc: John Harrison, Chris Wilson, Joonas Lahtinen, Jani Nikula,
Rodrigo Vivi, Tvrtko Ursulin, intel-gfx, stable
On Sun, Mar 19, 2023 at 10:01:01AM +0700, Philip Müller wrote:
> Have to correct the affected kernels to these: 4.14.310, 4.19.278, 5.4.237,
> 5.10.175
Please don't top-post :(
Anyway, should be fixed in the next round of releases in a few days, if
not, please let us know.
greg k-h
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
2023-03-20 13:28 ` Greg Kroah-Hartman
@ 2023-03-21 0:58 ` Philip Müller
-1 siblings, 0 replies; 10+ messages in thread
From: Philip Müller @ 2023-03-21 0:58 UTC (permalink / raw)
To: Greg Kroah-Hartman
Cc: John Harrison, Chris Wilson, Joonas Lahtinen, Jani Nikula,
Rodrigo Vivi, Tvrtko Ursulin, intel-gfx, stable
On 20.03.23 20:28, Greg Kroah-Hartman wrote:
> On Sun, Mar 19, 2023 at 10:01:01AM +0700, Philip Müller wrote:
>> Have to correct the affected kernels to these: 4.14.310, 4.19.278, 5.4.237,
>> 5.10.175
>
> Please don't top-post :(
>
> Anyway, should be fixed in the next round of releases in a few days, if
> not, please let us know.
>
> greg k-h
Hi Greg,
seems the RCs work out. 4.19.279-rc1 and 5.10.176-rc1 were tested by a
user who had the issue on i915. In 5.15 series the drm patch got
reverted. I only see "drm/i915: Don't use stolen memory for ring buffers
with LLC" there as "drm/i915: Don't use BAR mappings for ring buffers
with LLC" was reverted with 5.15.101.
--
Best, Philip
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Intel-gfx] [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
@ 2023-03-21 0:58 ` Philip Müller
0 siblings, 0 replies; 10+ messages in thread
From: Philip Müller @ 2023-03-21 0:58 UTC (permalink / raw)
To: Greg Kroah-Hartman; +Cc: intel-gfx, Chris Wilson, stable, Rodrigo Vivi
On 20.03.23 20:28, Greg Kroah-Hartman wrote:
> On Sun, Mar 19, 2023 at 10:01:01AM +0700, Philip Müller wrote:
>> Have to correct the affected kernels to these: 4.14.310, 4.19.278, 5.4.237,
>> 5.10.175
>
> Please don't top-post :(
>
> Anyway, should be fixed in the next round of releases in a few days, if
> not, please let us know.
>
> greg k-h
Hi Greg,
seems the RCs work out. 4.19.279-rc1 and 5.10.176-rc1 were tested by a
user who had the issue on i915. In 5.15 series the drm patch got
reverted. I only see "drm/i915: Don't use stolen memory for ring buffers
with LLC" there as "drm/i915: Don't use BAR mappings for ring buffers
with LLC" was reverted with 5.15.101.
--
Best, Philip
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Intel-gfx] [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
2023-03-21 0:58 ` [Intel-gfx] " Philip Müller
@ 2023-03-21 7:12 ` Greg Kroah-Hartman
-1 siblings, 0 replies; 10+ messages in thread
From: Greg Kroah-Hartman @ 2023-03-21 7:12 UTC (permalink / raw)
To: Philip Müller; +Cc: intel-gfx, Chris Wilson, stable, Rodrigo Vivi
On Tue, Mar 21, 2023 at 07:58:44AM +0700, Philip Müller wrote:
> On 20.03.23 20:28, Greg Kroah-Hartman wrote:
> > On Sun, Mar 19, 2023 at 10:01:01AM +0700, Philip Müller wrote:
> > > Have to correct the affected kernels to these: 4.14.310, 4.19.278, 5.4.237,
> > > 5.10.175
> >
> > Please don't top-post :(
> >
> > Anyway, should be fixed in the next round of releases in a few days, if
> > not, please let us know.
> >
> > greg k-h
>
> Hi Greg,
>
> seems the RCs work out. 4.19.279-rc1 and 5.10.176-rc1 were tested by a user
> who had the issue on i915. In 5.15 series the drm patch got reverted. I only
> see "drm/i915: Don't use stolen memory for ring buffers with LLC" there as
> "drm/i915: Don't use BAR mappings for ring buffers with LLC" was reverted
> with 5.15.101.
So is 5.15.y ok or not? Sorry, I could not parse your response here.
thanks,
greg k-h
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
@ 2023-03-21 7:12 ` Greg Kroah-Hartman
0 siblings, 0 replies; 10+ messages in thread
From: Greg Kroah-Hartman @ 2023-03-21 7:12 UTC (permalink / raw)
To: Philip Müller
Cc: John Harrison, Chris Wilson, Joonas Lahtinen, Jani Nikula,
Rodrigo Vivi, Tvrtko Ursulin, intel-gfx, stable
On Tue, Mar 21, 2023 at 07:58:44AM +0700, Philip Müller wrote:
> On 20.03.23 20:28, Greg Kroah-Hartman wrote:
> > On Sun, Mar 19, 2023 at 10:01:01AM +0700, Philip Müller wrote:
> > > Have to correct the affected kernels to these: 4.14.310, 4.19.278, 5.4.237,
> > > 5.10.175
> >
> > Please don't top-post :(
> >
> > Anyway, should be fixed in the next round of releases in a few days, if
> > not, please let us know.
> >
> > greg k-h
>
> Hi Greg,
>
> seems the RCs work out. 4.19.279-rc1 and 5.10.176-rc1 were tested by a user
> who had the issue on i915. In 5.15 series the drm patch got reverted. I only
> see "drm/i915: Don't use stolen memory for ring buffers with LLC" there as
> "drm/i915: Don't use BAR mappings for ring buffers with LLC" was reverted
> with 5.15.101.
So is 5.15.y ok or not? Sorry, I could not parse your response here.
thanks,
greg k-h
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
2023-03-21 7:12 ` Greg Kroah-Hartman
@ 2023-03-21 12:16 ` Philip Müller
-1 siblings, 0 replies; 10+ messages in thread
From: Philip Müller @ 2023-03-21 12:16 UTC (permalink / raw)
To: Greg Kroah-Hartman
Cc: John Harrison, Chris Wilson, Joonas Lahtinen, Jani Nikula,
Rodrigo Vivi, Tvrtko Ursulin, intel-gfx, stable
On 21.03.23 14:12, Greg Kroah-Hartman wrote:
> On Tue, Mar 21, 2023 at 07:58:44AM +0700, Philip Müller wrote:
>> On 20.03.23 20:28, Greg Kroah-Hartman wrote:
>>> On Sun, Mar 19, 2023 at 10:01:01AM +0700, Philip Müller wrote:
>>>> Have to correct the affected kernels to these: 4.14.310, 4.19.278, 5.4.237,
>>>> 5.10.175
>>>
>>> Please don't top-post :(
>>>
>>> Anyway, should be fixed in the next round of releases in a few days, if
>>> not, please let us know.
>>>
>>> greg k-h
>>
>> Hi Greg,
>>
>> seems the RCs work out. 4.19.279-rc1 and 5.10.176-rc1 were tested by a user
>> who had the issue on i915. In 5.15 series the drm patch got reverted. I only
>> see "drm/i915: Don't use stolen memory for ring buffers with LLC" there as
>> "drm/i915: Don't use BAR mappings for ring buffers with LLC" was reverted
>> with 5.15.101.
>
> So is 5.15.y ok or not? Sorry, I could not parse your response here.
>
> thanks,
>
> greg k-h
Hi Greg,
yes, 5.15.y seems to work. You could even re-add the reverted patch from
5.15.101 as we also have tested with it applied on top of 5.15.104-rc1.
--
Best, Philip
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [Intel-gfx] [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels
@ 2023-03-21 12:16 ` Philip Müller
0 siblings, 0 replies; 10+ messages in thread
From: Philip Müller @ 2023-03-21 12:16 UTC (permalink / raw)
To: Greg Kroah-Hartman; +Cc: intel-gfx, Chris Wilson, stable, Rodrigo Vivi
On 21.03.23 14:12, Greg Kroah-Hartman wrote:
> On Tue, Mar 21, 2023 at 07:58:44AM +0700, Philip Müller wrote:
>> On 20.03.23 20:28, Greg Kroah-Hartman wrote:
>>> On Sun, Mar 19, 2023 at 10:01:01AM +0700, Philip Müller wrote:
>>>> Have to correct the affected kernels to these: 4.14.310, 4.19.278, 5.4.237,
>>>> 5.10.175
>>>
>>> Please don't top-post :(
>>>
>>> Anyway, should be fixed in the next round of releases in a few days, if
>>> not, please let us know.
>>>
>>> greg k-h
>>
>> Hi Greg,
>>
>> seems the RCs work out. 4.19.279-rc1 and 5.10.176-rc1 were tested by a user
>> who had the issue on i915. In 5.15 series the drm patch got reverted. I only
>> see "drm/i915: Don't use stolen memory for ring buffers with LLC" there as
>> "drm/i915: Don't use BAR mappings for ring buffers with LLC" was reverted
>> with 5.15.101.
>
> So is 5.15.y ok or not? Sorry, I could not parse your response here.
>
> thanks,
>
> greg k-h
Hi Greg,
yes, 5.15.y seems to work. You could even re-add the reverted patch from
5.15.101 as we also have tested with it applied on top of 5.15.104-rc1.
--
Best, Philip
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2023-03-22 20:08 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-03-19 2:48 [Regression] drm/i915: Don't use BAR mappings for ring buffers with LLC alone creates issues in stable kernels Philip Müller
2023-03-19 3:01 ` Philip Müller
2023-03-20 13:28 ` [Intel-gfx] " Greg Kroah-Hartman
2023-03-20 13:28 ` Greg Kroah-Hartman
2023-03-21 0:58 ` Philip Müller
2023-03-21 0:58 ` [Intel-gfx] " Philip Müller
2023-03-21 7:12 ` Greg Kroah-Hartman
2023-03-21 7:12 ` Greg Kroah-Hartman
2023-03-21 12:16 ` Philip Müller
2023-03-21 12:16 ` [Intel-gfx] " Philip Müller
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.