linux-kernel - Re: [PATCH 4.9 033/120] perf/x86/intel/uncore: Fix hardcoded socket 0 assumption in the Haswell init code

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:   Wed, 18 Jan 2017 18:20:11 +0100
From:   Greg Kroah-Hartman <gregkh@...uxfoundation.org>
To:     Prarit Bhargava <prarit@...hat.com>
Cc:     linux-kernel@...r.kernel.org, stable@...r.kernel.org,
        Alexander Shishkin <alexander.shishkin@...ux.intel.com>,
        Arnaldo Carvalho de Melo <acme@...hat.com>,
        Borislav Petkov <bp@...e.de>, "H. Peter Anvin" <hpa@...or.com>,
        Harish Chegondi <harish.chegondi@...el.com>,
        Jiri Olsa <jolsa@...hat.com>, Kan Liang <kan.liang@...el.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Stephane Eranian <eranian@...gle.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Vince Weaver <vincent.weaver@...ne.edu>,
        Ingo Molnar <mingo@...nel.org>
Subject: Re: [PATCH 4.9 033/120] perf/x86/intel/uncore: Fix hardcoded socket
 0 assumption in the Haswell init code

On Wed, Jan 18, 2017 at 11:55:58AM -0500, Prarit Bhargava wrote:
> 
> 
> On 01/18/2017 11:33 AM, Greg Kroah-Hartman wrote:
> > On Wed, Jan 18, 2017 at 09:38:07AM -0500, Prarit Bhargava wrote:
> >>
> >>
> >> On 01/18/2017 05:45 AM, Greg Kroah-Hartman wrote:
> >>> 4.9-stable review patch.  If anyone has any objections, please let me know.
> >>>
> >>
> >> Nack.
> >>
> >> The value of boot_cpu_data.logical_proc_id may be uninitialized and set to
> >> default -1 on systems that pick a random core as boot cpu.  This was
> >> inadvertently fixed by 9d85eb9119f4 ("x86/smpboot: Make logical package
> >> management more robust") which is in 4.10-rc1.
> >>
> >> Before 9d85eb9119f4:
> >>
> >> [    3.971539] hswep_uncore_cpu_init: cpu 5 pkg 0 boot_cpu_data.logical_proc_id
> >> 65535
> >> [    3.976504] hswep_uncore_cpu_init: cpu 5 pkg 0 cpu_data(0).logical_proc_id 0
> >>
> >> After 9d85eb9119f4:
> >>
> >> [    3.919112] hswep_uncore_cpu_init: cpu 5 pkg 0 boot_cpu_data.logical_proc_id 0
> >> [    3.923391] hswep_uncore_cpu_init: cpu 5 pkg 0 cpu_data(0).logical_proc_id 0
> >>
> >> This patch should not be applied to any stable branch.
> > 
> > So the fixes: line lies?  This isn't needed at all for 4.9?
> 
> No, the fixes: line does not lie.  This patch is not needed at all for 4.9.
> Other patches are required beyond this patch in order for 4.9 to remain stable.
> 
> This patch is 6d6daa20945f ("perf/x86/intel/uncore: Fix hardcoded socket 0
> assumption in the Haswell init code") which was tested on and applied to
> 4.10-rc3 IIRC.  This patch was applied to 4.10-rc4.
> 
> [prarit@...rit linux]$ git describe --contains 6d6daa20945f
> v4.10-rc4~9^2~5

Yes, but the patch says it is fixing a bug since 4.6-rc1.  The fixes
line says:
	    Fixes: cf6d445f6897 ("perf/x86/uncore: Track packages, not per CPU data")

$ git dc cf6d445f6897
v4.6-rc1~165^2~28

(dc is my git alias for "describe --contains" as I type it so often...)

> 4.9 is broken and requires additional patches beyond this patch.  Applying this
> patch to 4.9 stable without those additional fixes will result in kernel panics
> on some Haswell systems that boot on random cores.

So, does 4.9 on its own work properly on these systems?  If not, what
are the commits that are needed to fix it?

If 4.9 is fine as-is, great, we should drop this patch then, correct?
But then that fixes: line lied :(

still confused,

greg k-h