linux-kernel - Re: Experiencing freezes with kernel 4.16.3 on a desktop with E5500 CPU (bisect included)

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]

Message-ID: <CA+ToGPHsw4zxfshMynkCQ24ZsrzRt3YnSKZJo+OZzjAu7rQE3g@mail.gmail.com>
Date:   Sun, 22 Apr 2018 15:24:12 -0300
From:   Diego Viola <diego.viola@...il.com>
To:     linux-kernel@...r.kernel.org, peterz@...radead.org,
        tglx@...utronix.de, len.brown@...el.com, rui.zhang@...el.com
Subject: Re: Experiencing freezes with kernel 4.16.3 on a desktop with E5500
 CPU (bisect included)

On Sun, Apr 22, 2018 at 2:47 PM, Diego Viola <diego.viola@...il.com> wrote:
> On Sat, Apr 21, 2018 at 11:19 PM, Diego Viola <diego.viola@...il.com> wrote:
>> Hi,
>>
>> I'm currently experiencing freezes after resuming from suspend when
>> using various applications on my desktop.
>>
>> For example, when using Firefox and switching between tabs the screen
>> would freeze, allowing me to only move the mouse pointer but focusing
>> the mouse pointer on a application won't have any effect.
>>
>> The keyboard would still work and the only way to make the issue go
>> away would be to switch to a virtual desktop and then switch back,
>> then I'd be able to use the mouse again until the same problem
>> re-occurs.
>>
>> This never happened before upgrading to 4.16.3 and 4.15.18 is unaffected.
>>
>> The problem only happens *after* resuming from suspend, in normal S0
>> state it never happens.
>>
>> My machine is a desktop and has a dual core E5500 CPU with 2GB of RAM,
>> I run Arch Linux and I use the i3wm.
>>
>> I don't think it's a problem with user space because I run exactly the
>> same Arch Linux setup on my T450 and the issue can't be reproduced on
>> that machine.
>>
>> I did a bisect and the cause of the bug seems to be this commit:
>>
>> aa83c45762a242acce9b35020363225a7b59d7c9 is the first bad commit
>>
>> Here is the bisect log:
>>
>> git bisect start
>> # bad: [0adb32858b0bddf4ada5f364a84ed60b196dbcda] Linux 4.16
>> git bisect bad 0adb32858b0bddf4ada5f364a84ed60b196dbcda
>> # good: [d8a5b80568a9cb66810e75b182018e9edb68e8ff] Linux 4.15
>> git bisect good d8a5b80568a9cb66810e75b182018e9edb68e8ff
>> # bad: [c14376de3a1befa70d9811ca2872d47367b48767] printk: Wake klogd
>> when passing console_lock owner
>> git bisect bad c14376de3a1befa70d9811ca2872d47367b48767
>> # bad: [a103950e0dd2058df5e8a8d4a915707bdcf205f0] Merge branch 'linus'
>> of git://git.kernel.org/pub/scm/linux/kernel/git/herbert/crypto-2.6
>> git bisect bad a103950e0dd2058df5e8a8d4a915707bdcf205f0
>> # bad: [d8b91dde38f4c43bd0bbbf17a90f735b16aaff2c] Merge branch
>> 'perf-core-for-linus' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
>> git bisect bad d8b91dde38f4c43bd0bbbf17a90f735b16aaff2c
>> # good: [9697e9da84299d0d715d515dd2cc48f1eceb277d] Merge tag
>> 'edac_for_4.16' of git://git.kernel.org/pub/scm/linux/kernel/git/bp/bp
>> git bisect good 9697e9da84299d0d715d515dd2cc48f1eceb277d
>> # good: [d0bd31dc5c0b46b9c778112900cf8f910ac26e1b] Merge tag
>> 'xtensa-20180129' of git://github.com/jcmvbkbc/linux-xtensa
>> git bisect good d0bd31dc5c0b46b9c778112900cf8f910ac26e1b
>> # good: [0337cf74ccf2a43437bff2e23b278e4f2dc4c6e2] perf util:
>> Introduce architecture specific errno/name mapping
>> git bisect good 0337cf74ccf2a43437bff2e23b278e4f2dc4c6e2
>> # bad: [6304672b7f0a5c010002e63a075160856dc4f88d] Merge branch
>> 'x86-pti-for-linus' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
>> git bisect bad 6304672b7f0a5c010002e63a075160856dc4f88d
>> # good: [a46d3f9b1c9888a244ed1ce8da0eca98c3f378e2] Merge branch
>> 'timers-core-for-linus' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
>> git bisect good a46d3f9b1c9888a244ed1ce8da0eca98c3f378e2
>> # good: [669c0f762ed19bd9ec128ebc97ae8641b6e1a4a3] Merge branch
>> 'x86-platform-for-linus' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
>> git bisect good 669c0f762ed19bd9ec128ebc97ae8641b6e1a4a3
>> # good: [7a32fc51ca938e67974cbb9db31e1a43f98345a9] x86/nospec: Fix
>> header guards names
>> git bisect good 7a32fc51ca938e67974cbb9db31e1a43f98345a9
>> # bad: [942633523cde99c5bfa0329996dd839ca66a87aa] Merge branch
>> 'x86-mm-for-linus' of
>> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
>> git bisect bad 942633523cde99c5bfa0329996dd839ca66a87aa
>> # bad: [aa83c45762a242acce9b35020363225a7b59d7c9] x86/tsc: Introduce
>> early tsc clocksource
>> git bisect bad aa83c45762a242acce9b35020363225a7b59d7c9
>> # good: [6d671e1b85c63e7a337ba76c1a154c091545cff8] x86/time:
>> Unconditionally register legacy timer interrupt
>> git bisect good 6d671e1b85c63e7a337ba76c1a154c091545cff8
>> # first bad commit: [aa83c45762a242acce9b35020363225a7b59d7c9]
>> x86/tsc: Introduce early tsc clocksource
>>
>> Current git HEAD is also affected.
>>
>> I already tried HEAD with that commit reverted and the problem can't
>> be reproduced anymore.
>>
>> Please let me know if I can include more info, thanks in advance.
>>
>> Diego
>
> When booting with clocksource=hpet I cannot reproduce the problem.

I can reproduce the freeze with clocksource=tsc as well.