From mboxrd@z Thu Jan 1 00:00:00 1970 From: Boris Derzhavets Subject: Re: Re: 2.6.37-rc1 mainline domU - BUG: unable to handle kernel paging request Date: Fri, 19 Nov 2010 06:32:26 -0800 (PST) Message-ID: <79696.74864.qm@web56104.mail.re3.yahoo.com> References: <129542.3339.qm@web56104.mail.re3.yahoo.com> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============1487317567==" Return-path: In-Reply-To: <129542.3339.qm@web56104.mail.re3.yahoo.com> List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xensource.com Errors-To: xen-devel-bounces@lists.xensource.com To: Konrad Rzeszutek Wilk , Bruce Edge , Jeremy Fitzhardinge Cc: xen-devel@lists.xensource.com List-Id: xen-devel@lists.xenproject.org --===============1487317567== Content-Type: multipart/alternative; boundary="0-492930146-1290177146=:74864" --0-492930146-1290177146=:74864 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable I've also noticed , that if i change file say under /mnt/nfs/fedora 1. ls runs fine 2. `ls -l` - generates page fault But doesn't crash DomU, regardless of stack trace printed in particular ter= minal session. I can close crashed terminal and open second one. # cd /mnt/nfs/fedora # ls -l will succeed until i make some new changes to file descriptors,eg. edit som= e file. Then crashes second terminal session and third should be opened to be able work with file descriptors ( `ls -l`). When number of page faults reaches some critical value ( more then 5 , but = in general unpredictable) DomU crashes. So , i cannot any more open new termin= al session. This=A0 is stable and obvious regression in 2.6.37-rc2 vs 2.6.36 a= s PV DomU kernel. Boris --- On Thu, 11/18/10, Boris Derzhavets wrote: From: Boris Derzhavets Subject: Re: [Xen-devel] Re: 2.6.37-rc1 mainline domU - BUG: unable to hand= le kernel paging request To: "Bruce Edge" Cc: "Jeremy Fitzhardinge" , xen-devel@lists.xensource.com,= "Konrad Rzeszutek Wilk" Date: Thursday, November 18, 2010, 12:05 PM Bruce, You should be able to apply patches to mainline 2.6.37-rc2 cleanly. This pa= tches are taken out from MY's kernel-2.6.37-rc2.git0.fc15.src.rpm. I already applied them on Ubuntu 10.10 to uncompressed mainline rc2. I have also to notice, that # mount IP-Dom0:/home/user1 /mnt/nfs # cd /mnt/nfs # ls -l crashes DomU immediately in text mode. In graphics mode it doesn't necessary happen every time. DomU might survive this "hack" and crashed one hour latter by another reason. Boris. --- On Thu, 11/18/10, Bruce Edge wrote: From: Bruce Edge Subject: Re: [Xen-devel] Re: 2.6.37-rc1 mainline domU - BUG: unable to hand= le kernel paging=0A request To: "Boris Derzhavets" Cc: "Konrad Rzeszutek Wilk" , "Jeremy Fitzhardinge"= , xen-devel@lists.xensource.com Date: Thursday, November 18, 2010, 11:40 AM On Thu, Nov 18, 2010 at 2:34 AM, Boris Derzhavets w= rote: =0ACould you apply two attached patches on top of 2.6.37-rc2 and see whether it gives some improvement or no ( with active NFS client at DomU) =0A Boris=A0=20 Hi Boris, Are you using the mainline kernel or a pvops branch with these patches? Maybe I'm doing something wrong, but they don't apply cleanly with 2.6.37-r= c2: =0A %> patch --dry-run <../patches.2.6.37/xen.next-2.6.37.patch=20 patching file pgtable.h Hunk #1 FAILED at 399. 1 out of 1 hunk FAILED -- saving rejects to file pgtable.h.rej patching file pgtable.c =0AHunk #1 FAILED at 15. 1 out of 1 hunk FAILED -- saving rejects to file pgtable.c.rej patching file ttm_bo_vm.c Hunk #1 FAILED at 273. Hunk #2 FAILED at 288. 2 out of 2 hunks FAILED -- saving rejects to file ttm_bo_vm.c.rej =0A...... =A0%> patch --dry-run <../patches.2.6.37/xen.pcifront.fixes.patch patching file enlighten.c Hunk #1 FAILED at 1090. Hunk #2 FAILED at 1202. 2 out of 2 hunks FAILED -- saving rejects to file enlighten.c.rej =0Apatching file setup.c Hunk #1 FAILED at 337. Hunk #2 FAILED at 356. 2 out of 2 hunks FAILED -- saving rejects to file setup.c.rej Same result for linux-2.6.37-xen-next branch. -Bruce =0A --- On Wed, 11/17/10, Bruce Edge wrote: =0A From: Bruce Edge =0A Subject: Re: [Xen-devel] Re: 2.6.37-rc1 mainline domU - BUG: unable to hand= le kernel paging request To: "Boris Derzhavets" =0ACc: "Konrad Rzeszutek Wilk" , "Jeremy Fitzhardin= ge" , xen-devel@lists.xensource.com =0ADate: Wednesday, November 17, 2010, 4:28 PM On Tue, Nov 16, 2010 at 1:49 PM, Boris Derzhavets w= rote: =0A=0AYes, here we are [=A0 186.975228] ------------[ cut here ]------------ [=A0 186.975245] kernel BUG at mm/mmap.c:2399! =0A=0A[=A0 186.975254] invalid opcode: 0000 [#1] SMP=20 [=A0 186.975269] last sysfs file: /sys/devices/system/cpu/cpu1/cache/index2= /shared_cpu_map [=A0 186.975284] CPU 0=20 [=A0 186.975290] Modules linked in: nfs fscache deflate zlib_deflate ctr ca= mellia cast5 rmd160 crypto_null ccm serpent blowfish twofish_generic twofis= h_x86_64 twofish_common ecb xcbc cbc sha256_generic sha512_generic des_gene= ric cryptd aes_x86_64 aes_generic ah6 ah4 esp6 esp4 xfrm4_mode_beet xfrm4_t= unnel tunnel4 xfrm4_mode_tunnel xfrm4_mode_transport xfrm6_mode_transport x= frm6_mode_ro xfrm6_mode_beet xfrm6_mode_tunnel ipcomp ipcomp6 xfrm_ipcomp x= frm6_tunnel tunnel6 af_key nfsd lockd nfs_acl auth_rpcgss exportfs sunrpc i= pv6 uinput xen_netfront=0A microcode xen_blkfront [last unloaded: scsi_wait= _scan] [=A0 186.975507]=20 [=A0 186.975515] Pid: 1562, comm: ls Not tainted 2.6.37-0.1.rc1.git8.xendom= 0.fc14.x86_64 #1 / [=A0 186.975529] RIP: e030:[]=A0 [] exi= t_mmap+0x10c/0x119 =0A=0A[=A0 186.975550] RSP: e02b:ffff8800781bde18=A0 EFLAGS: 00010202 [=A0 186.975560] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 000000000= 0000000 [=A0 186.975573] RDX: 00000000914a9149 RSI: 0000000000000001 RDI: ffffea000= 00c0280 =0A=0A[=A0 186.975585] RBP: ffff8800781bde48 R08: ffffea00000c0280 R09: 000= 0000000000001 [=A0 186.975598] R10: ffffffff8100750f R11: ffffea0000967778 R12: ffff88007= 6c68b00 [=A0 186.975610] R13: ffff88007f83f1e0 R14: ffff880076c68b68 R15: 000000000= 0000001 =0A=0A[=A0 186.975625] FS:=A0 00007f8e471d97c0(0000) GS:ffff88007f831000(00= 00) knlGS:0000000000000000 [=A0 186.975639] CS:=A0 e033 DS: 0000 ES:=0A 0000 CR0: 000000008005003b [=A0 186.975650] CR2: 00007f8e464a9940 CR3: 0000000001a03000 CR4: 000000000= 0002660 [=A0 186.975663] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 000000000= 0000000 [=A0 186.976012] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 000000000= 0000400 =0A=0A[=A0 186.976012] Process ls (pid: 1562, threadinfo ffff8800781bc000, = task ffff8800788223e0) [=A0 186.976012] Stack: [=A0 186.976012]=A0 000000000000006b ffff88007f83f1e0 ffff8800781bde38 ffff= 880076c68b00 [=A0 186.976012]=A0 ffff880076c68c40 ffff8800788229d0 ffff8800781bde68 ffff= ffff810505fc =0A=0A[=A0 186.976012]=A0 ffff8800788223e0 ffff880076c68b00 ffff8800781bdeb= 8 ffffffff81056747 [=A0 186.976012] Call Trace: [=A0 186.976012]=A0 [] mmput+0x65/0xd8 [=A0 186.976012]=A0 [] exit_mm+0x13e/0x14b =0A=0A[=A0 186.976012]=A0 []=0A do_exit+0x222/0x7c6 [=A0 186.976012]=A0 [] ? xen_restore_fl_direct_end+0x0/0x= 1 [=A0 186.976012]=A0 [] ? arch_local_irq_restore+0xb/0xd [=A0 186.976012]=A0 [] ? lockdep_sys_exit_thunk+0x35/0x67 =0A=0A[=A0 186.976012]=A0 [] do_group_exit+0x88/0xb6 [=A0 186.976012]=A0 [] sys_exit_group+0x17/0x1b [=A0 186.976012]=A0 [] system_call_fastpath+0x16/0x1b [=A0 186.976012] Code: 8d 7d 18 e8 c3 8a 00 00 41 c7 45 08 00 00 00 00 48 8= 9 df e8 0d e9 ff ff 48 85 c0 48 89 c3 75 f0 49 83 bc 24 98 01 00 00 00 74 0= 2 <0f> 0b 48 83 c4 18 5b 41 5c 41 5d c9 c3 55 48 89 e5 41 54 53 48=20 =0A=0A[=A0 186.976012] RIP=A0 [] exit_mmap+0x10c/0x119 [=A0 186.976012]=A0 RSP [=A0 186.976012] ---[ end trace c0f4eff4054a67e4=0A ]--- [=A0 186.976012] Fixing recursive fault but reboot is needed! Message from syslogd@fedora14 at Nov 17 00:47:40 ... =A0kernel:[=A0 186.975228] ------------[ cut here ]------------ Message from syslogd@fedora14 at Nov 17 00:47:40 ... =0A=0A=A0kernel:[=A0 186.975254] invalid opcode: 0000 [#1] SMP=20 Message from syslogd@fedora14 at Nov 17 00:47:40 ... =A0kernel:[=A0 186.975269] last sysfs file: /sys/devices/system/cpu/cpu1/ca= che/index2/shared_cpu_map =0A=0AMessage from syslogd@fedora14 at Nov 17 00:47:40 ... =A0kernel:[=A0 186.976012] Stack: Message from syslogd@fedora14 at Nov 17 00:47:40 ... =A0kernel:[=A0 186.976012] Call Trace: Message from syslogd@fedora14 at Nov 17 00:47:40 ... =0A=0A=A0kernel:[=A0 186.976012] Code: 8d 7d 18 e8 c3 8a 00 00 41 c7 45 08 = 00 00 00 00 48 89 df e8 0d e9 ff ff 48 85 c0 48 89 c3 75 f0 49 83 bc 24 98 = 01 00 00 00 74 02 <0f> 0b 48 83 c4=0A 18 5b 41 5c 41 5d c9 c3 55 48 89 e5 4= 1 54 53 48=20 --- On Tue, 11/16/10, Konrad Rzeszutek Wilk wrote: =0A=0A From: Konrad Rzeszutek Wilk =0A=0ASubject: Re: [Xen-devel] Re: 2.6.37-rc1 mainline domU - BUG: unable t= o handle kernel paging request To: "Boris Derzhavets" =0A=0ACc: "Jeremy Fitzhardinge" , xen-devel@lists.xensourc= e.com, "Bruce Edge" =0A=0ADate: Tuesday, November 16, 2010, 4:15 PM On Tue, Nov 16, 2010 at 12:43:28PM -0800, Boris Derzhavets wrote: > > Huh. I .. what? I am confused. I thought we established that the issue =0A=0A> > was not related to Xen PCI front? You also seem to uncomment the > > upstream.core.patches and the xen.pvhvm.patch -=0A why? >=20 > I cannot uncomment upstream.core.patches and the xen.pvhvm.patch > it gives failed HUNKs Uhh.. I am even more confused. >=20 > > Ok, they are.. v2.6.37-rc2 which came out today has the fixes =0A=0A>=20 > I am pretty sure rc2 doesn't contain everything from xen.next-2.6.37.patc= h, > gntdev's stuff for sure. I've built 2.6.37-rc2 kernel rpms and loaded=20 > kernel-2.6.27-rc2.git0.xendom0.x86_64 under Xen 4.0.1.=20 =0A=0A> Device /dev/xen/gntdev has not been created. I understand that it's > unrelated to DomU ( related to Dom0) , but once again with rc2 in DomU i = cannot > get 3.2 GB copied over to DomU from NFS share at Dom0. =0A=0A So what I think you are saying is that you keep on getting the bug in DomU? Is the stack-trace the same as in rc1? =0A=0A =20 I haven't had much time to look into the broken/working version issues here= , but I did confirm a couple of points: 1) The 2.6.37-rc2 has the same problem still 2) This problem goes away of one is not using NFS. =0A=0A Not staggeringly helpful I know, but it's one small data point. -Bruce =0A =0A=0A =20 =0A =0A=0A =20 -----Inline Attachment Follows----- _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel =0A=0A=0A --0-492930146-1290177146=:74864 Content-Type: text/html; charset=iso-8859-1 Content-Transfer-Encoding: quoted-printable
I've also noticed , that if i change file say= under /mnt/nfs/fedora

1. ls runs fine
2. `ls -l` - generates pag= e fault

But doesn't crash DomU, regardless of stack trace printed in= particular terminal session.

I can close crashed terminal and open = second one.
# cd /mnt/nfs/fedora
# ls -l
will succeed until i make= some new changes to file descriptors,eg. edit some file.
Then crashes s= econd terminal session and third should be opened to be able
work with f= ile descriptors ( `ls -l`).
When number of page faults reaches some crit= ical value ( more then 5 , but in
general unpredictable) DomU crashes. S= o , i cannot any more open new terminal
session. This  is stable an= d obvious regression in 2.6.37-rc2 vs 2.6.36 as PV DomU
kernel.

B= oris

--- On Thu, 11/18/10, Boris Derzhavets <bderzhavets@yahoo.com> wrote:
Cc: "Jeremy= Fitzhardinge" <jeremy@goop.org>, xen-devel@lists.xensource.com, "Kon= rad Rzeszutek Wilk" <konrad.wilk@oracle.com>
Date: Thursday, Novem= ber 18, 2010, 12:05 PM

Bruce,
You should be able to apply patches to mainline 2= .6.37-rc2 cleanly. This patches are
taken out from MY's kernel-2.6.37-rc= 2.git0.fc15.src.rpm.
I already applied them on Ubuntu 10.10 to uncompres= sed mainline rc2.

I have also to notice, that
# mount IP-Dom0:/home/user1 /mnt/nfs
# cd /mnt/nfs
# ls -l
crashes DomU i= mmediately in text mode.
In graphics mode it doesn't necessary happen ev= ery time.
DomU might survive this "hack" and crashed one
hour latter = by another reason.

Boris.

--- On Thu, 11/18/10, Bruce Edge= <bruce.edge@gmail.com> wrote:

From: Bruce Edge <bruce.edge@gmail.com>
Subject: Re: [Xen-dev= el] Re: 2.6.37-rc1 mainline domU - BUG: unable to handle kernel paging=0A r= equest
To: "Boris Derzhavets" <bderzhavets@yahoo.com>
Cc: "Konr= ad Rzeszutek Wilk" <konrad.wilk@oracle.com>, "Jeremy Fitzhardinge" &l= t;jeremy@goop.org>, xen-devel@lists.xensource.com
Date: Thursday, Nov= ember 18, 2010, 11:40 AM


<= br>
On Thu, Nov 18, 2010= at 2:34 AM, Boris Derzhavets <bde= rzhavets@yahoo.com> wrote:
=0A
Could you apply two attached patches on top of 2.6.37-rc2 a= nd see
whether it gives some improvement or no ( with active NFS client = at DomU)
=0A
Boris 
<= div>
Hi Boris,

Are you using the mainline kernel or a pvops branc= h with these patches?
Maybe I'm doing something wrong, but they don't ap= ply cleanly with 2.6.37-rc2:
=0A
%> patch --dry-run <../patches= .2.6.37/xen.next-2.6.37.patch

patching file pgtable.h
Hunk #1 FA= ILED at 399.
1 out of 1 hunk FAILED -- saving rejects to file pgtable.h.= rej
patching file pgtable.c
=0AHunk #1 FAILED at 15.
1 out of 1 hu= nk FAILED -- saving rejects to file pgtable.c.rej
patching file ttm_bo_v= m.c
Hunk #1 FAILED at 273.
Hunk #2 FAILED at 288.
2 out of 2 hunks= FAILED -- saving rejects to file ttm_bo_vm.c.rej
=0A......

 = ;%> patch --dry-run <../patches.2.6.37/xen.pcifront.fixes.patch
patching file enlighten.c
Hunk #1 FAILED at 1090.
Hunk #2 FAILED at= 1202.
2 out of 2 hunks FAILED -- saving rejects to file enlighten.c.rej=
=0Apatching file setup.c
Hunk #1 FAILED at 337.
Hunk #2 FAILED at= 356.
2 out of 2 hunks FAILED -- saving rejects to file setup.c.rej
<= br>Same result for linux-2.6.37-xen-next branch.

-Bruce

=0A= =

--- On Wed, 11/17/10= , Bruce Edge <bruce.edge@gmail.com> wrote:
=0A

From: Bruce Edge <bruce.edge@gmail.com>
=0A
Subject: Re: [Xen-devel] Re: 2.6.37-rc1 mainline domU - BUG: u= nable to handle kernel paging request
To: "Boris Derzhavets" <bderzhavets@yahoo.com>
=0A
Cc: "Konrad Rzeszut= ek Wilk" <konrad.wilk@oracle.com>, "Jeremy Fi= tzhardinge" <jeremy@goop.org>, xen-devel@lists.xensource.com
=0ADate: Wednesday, November 17,= 2010, 4:28 PM

On Tue, Nov 16, 2010 at 1:49 PM, Boris Derzhavets <bderzhavets@yahoo.com> wrote:=0A
=0A
Yes, here we are

[  186.975228] ------------[ cut here ]--= ----------
[  186.975245] kernel BUG at mm/mmap.c:2399!
=0A=0A[&= nbsp; 186.975254] invalid opcode: 0000 [#1] SMP
[  186.975269] las= t sysfs file: /sys/devices/system/cpu/cpu1/cache/index2/shared_cpu_map
[=   186.975284] CPU 0
[  186.975290] Modules linked in: nfs fsc= ache deflate zlib_deflate ctr camellia cast5 rmd160 crypto_null ccm serpent= blowfish twofish_generic twofish_x86_64 twofish_common ecb xcbc cbc sha256= _generic sha512_generic des_generic cryptd aes_x86_64 aes_generic ah6 ah4 e= sp6 esp4 xfrm4_mode_beet xfrm4_tunnel tunnel4 xfrm4_mode_tunnel xfrm4_mode_= transport xfrm6_mode_transport xfrm6_mode_ro xfrm6_mode_beet xfrm6_mode_tun= nel ipcomp ipcomp6 xfrm_ipcomp xfrm6_tunnel tunnel6 af_key nfsd lockd nfs_a= cl auth_rpcgss exportfs sunrpc ipv6 uinput xen_netfront=0A microcode xen_bl= kfront [last unloaded: scsi_wait_scan]
[  186.975507]
[  1= 86.975515] Pid: 1562, comm: ls Not tainted 2.6.37-0.1.rc1.git8.xendom0.fc14= .x86_64 #1 /
[  186.975529] RIP: e030:[<ffffffff8110ada1>]&nb= sp; [<ffffffff8110ada1>] exit_mmap+0x10c/0x119
=0A=0A[  186.9= 75550] RSP: e02b:ffff8800781bde18  EFLAGS: 00010202
[  186.975= 560] RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
[=   186.975573] RDX: 00000000914a9149 RSI: 0000000000000001 RDI: ffffea0= 0000c0280
=0A=0A[  186.975585] RBP: ffff8800781bde48 R08: ffffea000= 00c0280 R09: 0000000000000001
[  186.975598] R10: ffffffff8100750f = R11: ffffea0000967778 R12: ffff880076c68b00
[  186.975610] R13: fff= f88007f83f1e0 R14: ffff880076c68b68 R15: 0000000000000001
=0A=0A[  = 186.975625] FS:  00007f8e471d97c0(0000) GS:ffff88007f831000(0000) knlG= S:0000000000000000
[  186.975639] CS:  e033 DS: 0000 ES:=0A 00= 00 CR0: 000000008005003b
[  186.975650] CR2: 00007f8e464a9940 CR3: = 0000000001a03000 CR4: 0000000000002660
[  186.975663] DR0: 00000000= 00000000 DR1: 0000000000000000 DR2: 0000000000000000
[  186.976012]= DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
=0A= =0A[  186.976012] Process ls (pid: 1562, threadinfo ffff8800781bc000, = task ffff8800788223e0)
[  186.976012] Stack:
[  186.976012]=   000000000000006b ffff88007f83f1e0 ffff8800781bde38 ffff880076c68b00<= br>[  186.976012]  ffff880076c68c40 ffff8800788229d0 ffff8800781b= de68 ffffffff810505fc
=0A=0A[  186.976012]  ffff8800788223e0 f= fff880076c68b00 ffff8800781bdeb8 ffffffff81056747
[  186.976012] Ca= ll Trace:
[  186.976012]  [<ffffffff810505fc>] mmput+0x6= 5/0xd8
[  186.976012]  [<ffffffff81056747>] exit_mm+0x13= e/0x14b
=0A=0A[  186.976012]  [<ffffffff81056976>]=0A do= _exit+0x222/0x7c6
[  186.976012]  [<ffffffff8100750f>] ?= xen_restore_fl_direct_end+0x0/0x1
[  186.976012]  [<ffffff= ff8107ea7c>] ? arch_local_irq_restore+0xb/0xd
[  186.976012]&nbs= p; [<ffffffff814b3949>] ? lockdep_sys_exit_thunk+0x35/0x67
=0A=0A[=   186.976012]  [<ffffffff810571b0>] do_group_exit+0x88/0xb6=
[  186.976012]  [<ffffffff810571f5>] sys_exit_group+0x1= 7/0x1b
[  186.976012]  [<ffffffff8100acf2>] system_call_= fastpath+0x16/0x1b
[  186.976012] Code: 8d 7d 18 e8 c3 8a 00 00 41 = c7 45 08 00 00 00 00 48 89 df e8 0d e9 ff ff 48 85 c0 48 89 c3 75 f0 49 83 = bc 24 98 01 00 00 00 74 02 <0f> 0b 48 83 c4 18 5b 41 5c 41 5d c9 c3 5= 5 48 89 e5 41 54 53 48
=0A=0A[  186.976012] RIP  [<fffffff= f8110ada1>] exit_mmap+0x10c/0x119
[  186.976012]  RSP <f= fff8800781bde18>
[  186.976012] ---[ end trace c0f4eff4054a67e4= =0A ]---
[  186.976012] Fixing recursive fault but reboot is needed= !

Message from syslogd@fedora14 at Nov 17 00:47:40 ...
 kern= el:[  186.975228] ------------[ cut here ]------------

Message = from syslogd@fedora14 at Nov 17 00:47:40 ...
=0A=0A kernel:[  = 186.975254] invalid opcode: 0000 [#1] SMP

Message from syslogd@fedo= ra14 at Nov 17 00:47:40 ...
 kernel:[  186.975269] last sysfs = file: /sys/devices/system/cpu/cpu1/cache/index2/shared_cpu_map

=0A= =0AMessage from syslogd@fedora14 at Nov 17 00:47:40 ...
 kernel:[&n= bsp; 186.976012] Stack:

Message from syslogd@fedora14 at Nov 17 00:4= 7:40 ...
 kernel:[  186.976012] Call Trace:

Message fro= m syslogd@fedora14 at Nov 17 00:47:40 ...
=0A=0A kernel:[  186= .976012] Code: 8d 7d 18 e8 c3 8a 00 00 41 c7 45 08 00 00 00 00 48 89 df e8 = 0d e9 ff ff 48 85 c0 48 89 c3 75 f0 49 83 bc 24 98 01 00 00 00 74 02 <0f= > 0b 48 83 c4=0A 18 5b 41 5c 41 5d c9 c3 55 48 89 e5 41 54 53 48

--- On Tue, 11/16/10, Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> wrote:
=0A=0A

From: Konrad Rzeszutek Wilk <ko= nrad.wilk@oracle.com>
=0A=0ASubject: Re: [Xen-devel] Re: 2.6.37-r= c1 mainline domU - BUG: unable to handle kernel paging request
To: "Bori= s Derzhavets" <bderzhavets@yahoo.com>
=0A=0AC= c: "Jeremy Fitzhardinge" <jeremy@goop.org>, xen-devel@lists.xensource.com, "Bruce Edge" <bruce.edge@gmail.com>
=0A=0ADate: Tuesday, November 16,= 2010, 4:15 PM

On Tue, Nov 16, 2010 = at 12:43:28PM -0800, Boris Derzhavets wrote:
> > Huh. I .. what? I= am confused. I thought we established that the issue
=0A=0A> > wa= s not related to Xen PCI front? You also seem to uncomment the
> >= upstream.core.patches and the xen.pvhvm.patch -=0A why?
>
> I= cannot uncomment upstream.core.patches and the xen.pvhvm.patch
> it = gives failed HUNKs

Uhh.. I am even more confused.
>
> &= gt; Ok, they are.. v2.6.37-rc2 which came out today has the fixes
=0A=0A= >
> I am pretty sure rc2 doesn't contain everything from xen.next= -2.6.37.patch,
> gntdev's stuff for sure. I've built 2.6.37-rc2 kerne= l rpms and loaded
> kernel-2.6.27-rc2.git0.xendom0.x86_64 under Xen = 4.0.1.
=0A=0A> Device /dev/xen/gntdev has not been created. I unders= tand that it's
> unrelated to DomU ( related to Dom0) , but once agai= n with rc2 in DomU i cannot
> get 3.2 GB copied over to DomU from NFS= share at Dom0.
=0A=0A
So what I think you are saying is that you kee= p on getting the bug in DomU?
Is the stack-trace the same as in rc1?
=

=0A=0A =

I haven't had much time to look into the broken/wo= rking version issues here, but I did confirm a couple of points:
1) The = 2.6.37-rc2 has the same problem still
2) This problem goes away of one i= s not using NFS.
=0A=0A
Not staggeringly helpful I know, but it's one= small data point.

-Bruce
=0A

=0A=0A

=0A

=0A=0A

-----Inline Atta= chment Follows-----

_______________________= ________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-dev= el

=0A=0A --0-492930146-1290177146=:74864-- --===============1487317567== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xensource.com http://lists.xensource.com/xen-devel --===============1487317567==--