Re: Thread model (was: Ext2 superblock fault)

2008-03-19 Thread olafBuddenhagen
Hi, On Wed, Mar 19, 2008 at 10:45:15AM +, Samuel Thibault wrote: > [EMAIL PROTECTED], le Tue 18 Mar 2008 11:02:43 +0100, a écrit : > > I don't know how the syncing works, so I can't really tell what the > > problem is. If there are blocking points before the superblock read, > > we need to ch

Re: Thread model (was: Ext2 superblock fault)

2008-03-19 Thread Samuel Thibault
Michal Suchanek, le Wed 19 Mar 2008 16:55:48 +0100, a écrit : > On 19/03/2008, Samuel Thibault <[EMAIL PROTECTED]> wrote: > > Yes, that's what I meant actually: the diskfs_sync_everything() function > > is able to trigger a lot of thread creations. > > > > A way to have things work correctly woul

Re: Thread model (was: Ext2 superblock fault)

2008-03-19 Thread Michal Suchanek
On 19/03/2008, Samuel Thibault <[EMAIL PROTECTED]> wrote: > > Yes, that's what I meant actually: the diskfs_sync_everything() function > is able to trigger a lot of thread creations. > > A way to have things work correctly would be by marking threads with a > "level", i.e. diskfs_sync_everythin

Re: Thread model (was: Ext2 superblock fault)

2008-03-19 Thread Samuel Thibault
[EMAIL PROTECTED], le Tue 18 Mar 2008 11:02:43 +0100, a écrit : > On Mon, Mar 17, 2008 at 10:41:01AM +, Samuel Thibault wrote: > > [EMAIL PROTECTED], le Sun 16 Mar 2008 08:52:56 +0100, a écrit : > > > > What makes me wonder is, how can it happen in the first place that > > > so many requests a

Re: Thread model (was: Ext2 superblock fault)

2008-03-18 Thread olafBuddenhagen
Hi, On Mon, Mar 17, 2008 at 10:41:01AM +, Samuel Thibault wrote: > [EMAIL PROTECTED], le Sun 16 Mar 2008 08:52:56 +0100, a écrit : > > What makes me wonder is, how can it happen in the first place that > > so many requests are generated before the superblock is requested > > during handling o

Re: Thread model (was: Ext2 superblock fault)

2008-03-17 Thread Samuel Thibault
Thomas Bushnell BSG, le Mon 17 Mar 2008 15:09:12 -0400, a écrit : > > On Sun, 2008-03-16 at 08:52 +0100, [EMAIL PROTECTED] wrote: > > Hi, > > > > On Tue, Mar 11, 2008 at 11:19:32AM +, Samuel Thibault wrote: > > > [EMAIL PROTECTED], le Tue 11 Mar 2008 04:53:45 +0100, a écrit : > > > > > > [I]

Re: Thread model (was: Ext2 superblock fault)

2008-03-17 Thread Thomas Bushnell BSG
On Sun, 2008-03-16 at 08:52 +0100, [EMAIL PROTECTED] wrote: > Hi, > > On Tue, Mar 11, 2008 at 11:19:32AM +, Samuel Thibault wrote: > > [EMAIL PROTECTED], le Tue 11 Mar 2008 04:53:45 +0100, a écrit : > > > > [I] suggested a more adaptive approach: Keep track of the existing > > > threads, and

Re: Thread model (was: Ext2 superblock fault)

2008-03-17 Thread Samuel Thibault
[EMAIL PROTECTED], le Sun 16 Mar 2008 08:52:56 +0100, a écrit : > On Tue, Mar 11, 2008 at 11:19:32AM +, Samuel Thibault wrote: > > [EMAIL PROTECTED], le Tue 11 Mar 2008 04:53:45 +0100, a écrit : > > > > [I] suggested a more adaptive approach: Keep track of the existing > > > threads, and if no

Re: Thread model (was: Ext2 superblock fault)

2008-03-17 Thread olafBuddenhagen
Hi, On Tue, Mar 11, 2008 at 11:19:32AM +, Samuel Thibault wrote: > [EMAIL PROTECTED], le Tue 11 Mar 2008 04:53:45 +0100, a écrit : > > [I] suggested a more adaptive approach: Keep track of the existing > > threads, and if none of them makes progress in a certain amount of > > time (say 100 ms

Re: Thread model (was: Ext2 superblock fault)

2008-03-11 Thread Samuel Thibault
Hello, [EMAIL PROTECTED], le Tue 11 Mar 2008 04:53:45 +0100, a écrit : > [I] suggested a more adaptive approach: Keep track of the existing > threads, and if none of them makes progress in a certain amount of > time (say 100 ms), allow creating some more threads. But that was > never implemented.

Thread model (was: Ext2 superblock fault)

2008-03-11 Thread olafBuddenhagen
Hi, On Sun, Mar 09, 2008 at 10:17:05PM -0400, Thomas Bushnell BSG wrote: > On Mon, 2008-03-10 at 01:19 +, Samuel Thibault wrote: > > This thread is syncing everything, i.e. asking a lot of writes, > > which triggers the creation of a lot of threads. Unfortunately the > > superblock was paged

Re: Ext2 superblock fault

2008-03-10 Thread Samuel Thibault
Thomas Bushnell BSG, le Sun 09 Mar 2008 22:17:05 -0400, a écrit : > On Mon, 2008-03-10 at 01:19 +, Samuel Thibault wrote: > > This thread is syncing everything, i.e. asking a lot of writes, which > > triggers the creation of a lot of threads. Unfortunately the superblock > > was paged out, so

Re: Ext2 superblock fault

2008-03-09 Thread Thomas Bushnell BSG
On Mon, 2008-03-10 at 01:19 +, Samuel Thibault wrote: > > This thread is syncing everything, i.e. asking a lot of writes, which > triggers the creation of a lot of threads. Unfortunately the superblock > was paged out, so they all block on reading it. Unfortunately, since in > Debian there

Re: Ext2 superblock fault

2008-03-09 Thread Samuel Thibault
While having a beer with Gianluca, I thought about a possible case: > Thread 7 (thread 588.7): > #5 0x08052472 in write_all_disknodes () > at /build/mbanck/hurd-20071119/build-tree/hurd/ext2fs/inode.c:592 > #6 0x08054a90 in diskfs_sync_everything (wait=0) This thread is syncing everything,

Ext2 superblock fault

2008-03-01 Thread Samuel Thibault
Hello, Sometimes ext2fs just hangs, and gdb-ing it shows a huge lot of threads almost all blocked at hurd/ext2fs/getblk.c:236, i.e. on a dereference of sblock (unfortunately I forgot to ask for a backtrace). I was wondering: what happens if because of memory pressure the superblock gets swapped o