Re: S2ram + btrfs with recent kernel => soft lockup warning and hard hang

2017-02-13 Thread Bastien ROUCARIES
On Sun, Feb 12, 2017 at 4:45 PM, Bastien ROUCARIES wrote: > Hi, > > Since 3.8 s2ram on btrfs system lead to hang. Sorry for the typo means >4.8 > > Bastien

Re: S2ram + btrfs with recent kernel => soft lockup warning and hard hang

2017-02-13 Thread Bastien ROUCARIES
Hi, On Mon, Feb 13, 2017 at 11:43 AM, Filipe Manana wrote: > > 3.8 is very old, and so is 3.19 (which is what you're running as shown > in the github issue). Many fixes landed after 3.19. Just try a recent > kernel. > Sorry I made a typo I mean 4.8. I have just tried 4.9

S2ram + btrfs with recent kernel => soft lockup warning and hard hang

2017-02-12 Thread Bastien ROUCARIES
Hi, Since 3.8 s2ram on btrfs system lead to hang. Could not save a trace will try to get a few picture but trace are similar to : https://github.com/docker/docker/issues/15654 and https://patchwork.kernel.org/patch/8352811/ It is 100% reproducible. mount table /dev/mapper/portable2015--bastien-

Re: [3.10.7] Reproductible soft lockup, recover but freeze even the mouse

2013-09-09 Thread Bastien ROUCARIES
On Mon, Sep 9, 2013 at 11:48 AM, NeilBrown wrote: > On Mon, 9 Sep 2013 10:58:12 +0200 Bastien ROUCARIES > wrote: > >> Hi, >> >> I experiment a soft lockup. during 20 s I have no mouse, could not >> switch back to console and so on. >> >> I achieve

[3.10.7] Reproductible soft lockup, recover but freeze even the mouse

2013-09-09 Thread Bastien ROUCARIES
Hi, I experiment a soft lockup. during 20 s I have no mouse, could not switch back to console and so on. I achieve to get a backtrace by echo 15 > /proc/sys/kernel/hung_task_timeout_secs Seems file system related (hdd led is on). My filesystems are ext4 over raid0 so add some cc How can I debu

Re: [PATCH 0/4] fat: fix ESTALE errors

2012-08-21 Thread Bastien ROUCARIES
On Tue, Aug 21, 2012 at 8:41 AM, OGAWA Hirofumi wrote: > Namjae Jeon writes: > >> And.. Hi Ogawa. >> I checked other filesystem about unlink - inode issue. but I found >> Ext4 have same issue. >> Although other filesysm is having this issue, Can we think It could be >> only FAT issue ? > > (I ass

Re: [PATCH] unifdef: set a secure umask before calling mkstemp()

2012-08-20 Thread Bastien ROUCARIES
On Sat, Aug 18, 2012 at 1:43 AM, Tony Finch wrote: > Jesper Juhl wrote: > >> In newer glibc's (versions > 2.06) reasonably secure permissions of >> 0600 are used when creating a temporary file with mkstemp(). But for >> older glibc's (versions <= 2.06) 0666 is used which is not secure. > > Thanks