From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from psmtp.com (na3sys010amx177.postini.com [74.125.245.177]) by kanga.kvack.org (Postfix) with SMTP id 2ED0D6B0044 for ; Thu, 8 Nov 2012 01:39:24 -0500 (EST) Received: by mail-ea0-f169.google.com with SMTP id k11so1124805eaa.14 for ; Wed, 07 Nov 2012 22:39:22 -0800 (PST) MIME-Version: 1.0 In-Reply-To: <509B533B.7090907@redhat.com> References: <1352193295-26815-1-git-send-email-mgorman@suse.de> <509A2970.9000408@redhat.com> <20121107152558.GZ8218@suse.de> <509B533B.7090907@redhat.com> Date: Thu, 8 Nov 2012 14:39:22 +0800 Message-ID: Subject: Re: [RFC PATCH 00/19] Foundation for automatic NUMA balancing From: =?GB2312?B?0e7W8Q==?= Content-Type: multipart/alternative; boundary=047d7b621ee2703fd204cdf618c2 Sender: owner-linux-mm@kvack.org List-ID: To: Zhouping Liu Cc: Mel Gorman , Peter Zijlstra , Andrea Arcangeli , Ingo Molnar , Rik van Riel , Johannes Weiner , Hugh Dickins , Thomas Gleixner , Linus Torvalds , Andrew Morton , Linux-MM , LKML , CAI Qian --047d7b621ee2703fd204cdf618c2 Content-Type: text/plain; charset=GB2312 Content-Transfer-Encoding: quoted-printable Hi all: I got a problem=A3=BA 1. on intel cpu xeon E5000 family which support xapic =A3=ACone N= IC irq can share on the CPUs basic on smp_affinity. 2. but on intel cpu xeon E5-2600 family which support x2apic, one NIC irq only on CPU0 whatever i set the smp_affinfiy like as "aa"; "55"; "ff". My OS is CentOS 6.2 x32 =A3=ACi test 4 cpus=A1=A3 the result is w= hich only support apic can share one irq to all cpus=A3=ACwhich support x2apic only m= ake the irq to one cpu=A1=A3 want help me richard 2012/11/8 Zhouping Liu > On 11/07/2012 11:25 PM, Mel Gorman wrote: > >> On Wed, Nov 07, 2012 at 05:27:12PM +0800, Zhouping Liu wrote: >> >>> Hello Mel, >>> >>> my 2 nodes machine hit a panic fault after applied the patch >>> set(based on kernel-3.7.0-rc4), please review it: >>> >>> >>> >> Early initialisation problem by the looks of things. Try this please >> > > Tested the patch, and the issue is gone. > > >> ---8<--- >> mm: numa: Check that preferred_node_policy is initialised >> >> Zhouping Liu reported the following >> >> [ 0.000000] ------------[ cut here ]------------ >> [ 0.000000] kernel BUG at mm/mempolicy.c:1785! >> [ 0.000000] invalid opcode: 0000 [#1] SMP >> [ 0.000000] Modules linked in: >> [ 0.000000] CPU 0 >> .... >> [ 0.000000] Call Trace: >> [ 0.000000] [] alloc_pages_current+0xa6/0x170 >> [ 0.000000] [] __get_free_pages+0x14/0x50 >> [ 0.000000] [] kmem_cache_init+0x53/0x2d2 >> [ 0.000000] [] start_kernel+0x1e0/0x3c7 >> >> Problem is that early in boot preferred_nod_policy and SLUB >> initialisation trips up. Check it is initialised. >> >> Signed-off-by: Mel Gorman >> > > Tested-by: Zhouping Liu > > Thanks, > Zhouping > > --- >> mm/mempolicy.c | 4 ++++ >> 1 file changed, 4 insertions(+) >> >> diff --git a/mm/mempolicy.c b/mm/mempolicy.c >> index 11d4b6b..8cfa6dc 100644 >> --- a/mm/mempolicy.c >> +++ b/mm/mempolicy.c >> @@ -129,6 +129,10 @@ static struct mempolicy *get_task_policy(struct >> task_struct *p) >> node =3D numa_node_id(); >> if (node !=3D -1) >> pol =3D &preferred_node_policy[node]; >> + >> + /* preferred_node_policy is not initialised early in boo= t >> */ >> + if (!pol->mode) >> + pol =3D NULL; >> } >> return pol; >> >> -- >> To unsubscribe, send a message with 'unsubscribe linux-mm' in >> the body to majordomo@kvack.org. For more info on Linux MM, >> see: http://www.linux-mm.org/ . >> Don't email: email@kvack.org >> > > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" i= n > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/**majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/ > --047d7b621ee2703fd204cdf618c2 Content-Type: text/html; charset=GB2312 Content-Transfer-Encoding: quoted-printable Hi all:<= div style=3D"font-family:arial,sans-serif;font-size:14px">    &nb= sp;     I got a problem=A3=BA 
       = ;   1. on intel cpu xeon E5000 family which support xapic =A3=ACone NI= C irq  can share on the CPUs basic on smp_affinity. 
    &n= bsp;     2. but on intel cpu xeon E5-2600 family which support x2= apic, one NIC irq only on CPU0 whatever  i set the smp_affinfiy like a= s "aa"; "55"; "ff".
    &n= bsp;    My OS is CentOS 6.2  x32 =A3=ACi test 4 cpus=A1=A3 t= he result is which only support apic can share one irq to all cpus=A3=ACwhi= ch support x2apic only make the irq to one cpu=A1=A3 


want help me

 =                     &nbs= p;                     &n= bsp;                richard


2012/11/8 Zhoup= ing Liu <zliu@redhat.com>
On 11/07/2012 11:25 PM, Mel Gorman wrote:
On Wed, Nov 07, 2012 at 05:27:12PM +0800, Zhouping Liu wrote:
Hello Mel,

my 2 nodes machine hit a panic fault after applied the patch
set(based on kernel-3.7.0-rc4), please review it:

<SNIP>
Early initialisation problem by the looks of things. Try this please

Tested the patch, and the issue is gone.


---8<---
mm: numa: Check that preferred_node_policy is initialised

Zhouping Liu reported the following

[ 0.000000] ------------[ cut here ]------------
[ 0.000000] kernel BUG at mm/mempolicy.c:1785!
[ 0.000000] invalid opcode: 0000 [#1] SMP
[ 0.000000] Modules linked in:
[ 0.000000] CPU 0
....
[    0.000000] Call Trace:
[    0.000000] [<ffffffff81176966>] alloc_pages_current+0xa= 6/0x170
[    0.000000] [<ffffffff81137a44>] __get_free_pages+0x14/0= x50
[    0.000000] [<ffffffff819efd9b>] kmem_cache_init+0x53/0x= 2d2
[    0.000000] [<ffffffff819caa53>] start_kernel+0x1e0/0x3c= 7

Problem is that early in boot preferred_nod_policy and SLUB
initialisation trips up. Check it is initialised.

Signed-off-by: Mel Gorman <mgorman@suse.de>

Tested-by: Zhouping Liu <zliu@redhat.com>

Thanks,
Zhouping

---
  mm/mempolicy.c |    4 ++++
  1 file changed, 4 insertions(+)

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 11d4b6b..8cfa6dc 100644
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -129,6 +129,10 @@ static struct mempolicy *get_task_policy(struct task_s= truct *p)
                node =3D numa_node_= id();
                if (node !=3D -1)                      = ;   pol =3D &preferred_node_policy[node];
+
+               /* preferred_node_policy= is not initialised early in boot */
+               if (!pol->mode)
+                     &nb= sp; pol =3D NULL;
        }
        return pol;

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majord= omo@kvack.org.  For more info on Linux MM,
see: http://www.linu= x-mm.org/ .
Don't email: <a href=3Dmailto:"dont@kvack.org"> email@kvack.org </a>

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

--047d7b621ee2703fd204cdf618c2-- -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@kvack.org. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: email@kvack.org