linux-kernel - Re: [PATCH -v2 2/2] x86, acpi: Parse all SRAT cpu entries even have cpu num limitation

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, 22 Dec 2010 12:28:25 -0800
From:	Venkatesh Pallipadi <venki@...gle.com>
To:	David Rientjes <rientjes@...gle.com>
Cc:	Yinghai Lu <yinghai@...nel.org>, "H. Peter Anvin" <hpa@...or.com>,
	Ingo Molnar <mingo@...e.hu>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Thomas Gleixner <tglx@...utronix.de>,
	Wu Fengguang <fengguang.wu@...el.com>,
	Peter Zijlstra <peterz@...radead.org>,
	LKML <linux-kernel@...r.kernel.org>,
	Nikanth Karthikesan <knikanth@...e.de>,
	"Zheng, Shaohui" <shaohui.zheng@...el.com>,
	Eric Dumazet <eric.dumazet@...il.com>,
	Bjorn Helgaas <bjorn.helgaas@...com>,
	Nikhil Rao <ncrao@...gle.com>,
	Takuya Yoshikawa <yoshikawa.takuya@....ntt.co.jp>
Subject: Re: [PATCH -v2 2/2] x86, acpi: Parse all SRAT cpu entries even have
 cpu num limitation

On Tue, Dec 21, 2010 at 10:43 PM, David Rientjes <rientjes@...gle.com> wrote:
> On Mon, 20 Dec 2010, Venkatesh Pallipadi wrote:
>
>> git bisect seems to narrow this down to the change below.
>>
>> Thanks,
>> Venki
>>
>> $ git bisect visualize
>> commit 50f2d7f682f9c0ed58191d0982fe77888d59d162
>> Author: Nikanth Karthikesan <knikanth@...e.de>
>> Date:   Thu Sep 30 17:34:10 2010 +0530
>>
>>     x86, numa: Assign CPUs to nodes in round-robin manner on fake NUMA
>>
>>     commit d9c2d5ac6af87b4491bff107113aaf16f6c2b2d9 "x86, numa: Use near(er)
>>     online node instead of roundrobin for NUMA" changed NUMA initialization on
>>     Intel to choose the nearest online node or first node.  Fake NUMA would be
>>     better of with round-robin initialization, instead of the all CPUS on
>>     first node.  Change the choice of first node, back to round-robin.
>>
>>     For testing NUMA kernel behaviour without cpusets and NUMA aware
>>     applications, it would be better to have cpus in different nodes, rather
>>     than all in a single node.  With cpusets migration of tasks scenarios
>>     cannot not be tested.
>>
>>     I guess having it round-robin shouldn't affect the use cases for all cpus
>>     on the first node.
>>
>>     The code comments in arch/x86/mm/numa_64.c:759 indicate that this used to
>>     be the case, which was changed by commit d9c2d5ac6.  It changed from
>>     roundrobin to nearer or first node.  And I couldn't find any reason for
>>     this change in its changelog.
>>
>>     Signed-off-by: Nikanth Karthikesan <knikanth@...e.de>
>>     Cc: David Rientjes <rientjes@...gle.com>
>>     Signed-off-by: Andrew Morton <akpm@...ux-foundation.org>
>>
>
> Peter just merged my NUMA emulation fixes into the x86 tree, could you try
> applying Yinghai's series on top of x86/linux-2.6-tip.git#x86/numa and see
> if the problem persists?
>
> On a different topic: Yinghai, do you think you could base your series off
> of Tejun's x86_32/x86_64 NUMA unification series since it already
> duplicates some of the work?
>

Yes. #x86/numa kernel works fine both with and without Yinghai's series.
I am assuming those changes are lined up for .38. Is there any
specific fix that can make into .37 to fix this regression?

Thanks,
Venki
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/