linux-kernel - Re: [RESEND PATCH v4] drm: Don't free jobs in wait_event

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 25 Oct 2019 10:49:34 +0100
From:   Steven Price <steven.price@....com>
To:     Christian Gmeiner <christian.gmeiner@...il.com>
Cc:     David Airlie <airlied@...ux.ie>,
        Sharat Masetty <smasetty@...eaurora.org>,
        LKML <linux-kernel@...r.kernel.org>,
        DRI mailing list <dri-devel@...ts.freedesktop.org>,
        Alex Deucher <alexander.deucher@....com>,
        Sam Ravnborg <sam@...nborg.org>,
        Christian König <christian.koenig@....com>,
        Erico Nunes <nunes.erico@...il.com>
Subject: Re: [RESEND PATCH v4] drm: Don't free jobs in
 wait_event_interruptible()

On 25/10/2019 10:43, Christian Gmeiner wrote:
> Am Do., 24. Okt. 2019 um 18:25 Uhr schrieb Steven Price <steven.price@....com>:
>>
>> drm_sched_cleanup_jobs() attempts to free finished jobs, however because
>> it is called as the condition of wait_event_interruptible() it must not
>> sleep. Unfortuantly some free callbacks (notibly for Panfrost) do sleep.
>>
>> Instead let's rename drm_sched_cleanup_jobs() to
>> drm_sched_get_cleanup_job() and simply return a job for processing if
>> there is one. The caller can then call the free_job() callback outside
>> the wait_event_interruptible() where sleeping is possible before
>> re-checking and returning to sleep if necessary.
>>
>> Signed-off-by: Steven Price <steven.price@....com>
> 
> Tested-by: Christian Gmeiner <christian.gmeiner@...il.com>
> 
> Without this patch I get the following warning:

Thanks, if you've got an (easily) reproducible case, can you check which
commit this fixes. I *think*:

Fixes: 5918045c4ed4 ("drm/scheduler: rework job destruction")

But I haven't got a reliable way of reproducing this (with Panfrost).

Thanks,

Steve

> 
> [  242.935254] ------------[ cut here ]------------
> [  242.940044] WARNING: CPU: 2 PID: 109 at kernel/sched/core.c:6731
> __might_sleep+0x94/0xa8
> [  242.948242] do not call blocking ops when !TASK_RUNNING; state=1
> set at [<38751e36>] prepare_to_wait_event+0xa8/0x180
> [  242.958923] Modules linked in:
> [  242.962010] CPU: 2 PID: 109 Comm: 130000.gpu Not tainted 5.4.0-rc4 #10
> [  242.968551] Hardware name: Freescale i.MX6 Quad/DualLite (Device Tree)
> [  242.975112] [<c0113160>] (unwind_backtrace) from [<c010cf34>]
> (show_stack+0x10/0x14)
> [  242.982879] [<c010cf34>] (show_stack) from [<c0c065ec>]
> (dump_stack+0xd8/0x110)
> [  242.990213] [<c0c065ec>] (dump_stack) from [<c0128adc>] (__warn+0xc0/0x10c)
> [  242.997194] [<c0128adc>] (__warn) from [<c0128f10>]
> (warn_slowpath_fmt+0x8c/0xb8)
> [  243.004697] [<c0128f10>] (warn_slowpath_fmt) from [<c01598bc>]
> (__might_sleep+0x94/0xa8)
> [  243.012810] [<c01598bc>] (__might_sleep) from [<c0c246e4>]
> (__mutex_lock+0x38/0xa1c)
> [  243.020571] [<c0c246e4>] (__mutex_lock) from [<c0c250e4>]
> (mutex_lock_nested+0x1c/0x24)
> [  243.028600] [<c0c250e4>] (mutex_lock_nested) from [<c064f020>]
> (etnaviv_cmdbuf_free+0x40/0x8c)
> [  243.037233] [<c064f020>] (etnaviv_cmdbuf_free) from [<c06503a0>]
> (etnaviv_submit_put+0x38/0x1c8)
> [  243.046042] [<c06503a0>] (etnaviv_submit_put) from [<c064177c>]
> (drm_sched_cleanup_jobs+0xc8/0xec)
> [  243.055021] [<c064177c>] (drm_sched_cleanup_jobs) from [<c06419b4>]
> (drm_sched_main+0x214/0x298)
> [  243.063826] [<c06419b4>] (drm_sched_main) from [<c0152890>]
> (kthread+0x140/0x158)
> [  243.071329] [<c0152890>] (kthread) from [<c01010b4>]
> (ret_from_fork+0x14/0x20)
> [  243.078563] Exception stack(0xec691fb0 to 0xec691ff8)
> [  243.083630] 1fa0:                                     00000000
> 00000000 00000000 00000000
> [  243.091822] 1fc0: 00000000 00000000 00000000 00000000 00000000
> 00000000 00000000 00000000
> [  243.100013] 1fe0: 00000000 00000000 00000000 00000000 00000013 00000000
> [  243.106795] irq event stamp: 321
> [  243.110098] hardirqs last  enabled at (339): [<c0193854>]
> console_unlock+0x430/0x620
> [  243.117864] hardirqs last disabled at (346): [<c01934cc>]
> console_unlock+0xa8/0x620
> [  243.125592] softirqs last  enabled at (362): [<c01024e0>]
> __do_softirq+0x2c0/0x590
> [  243.133232] softirqs last disabled at (373): [<c0130ed0>]
> irq_exit+0x100/0x18c
> [  243.140517] ---[ end trace 8afcd79e9e2725b2 ]---
>