Re: bcache: Fix a writeback performance regression

2013-08-29 Thread Stefan Priebe
sorry seems i see something similiar: http://pastebin.com/raw.php?i=ZqgLf9gp Stefan Am 28.08.2013 22:15, schrieb Stefan Priebe: sorry but that's completely wrong. please use branch bcache-for-3.10 http://evilpiepirate.org/git/linux-bcache.git/log/?h=bcache-for-3.10 Stefan Am 28.08.2013 22:1

Re: bcache: Fix a writeback performance regression

2013-08-28 Thread Stefan Priebe
sorry but that's completely wrong. please use branch bcache-for-3.10 http://evilpiepirate.org/git/linux-bcache.git/log/?h=bcache-for-3.10 Stefan Am 28.08.2013 22:12, schrieb kernel neophyte: On Wed, Aug 28, 2013 at 12:20 PM, Stefan Priebe wrote: Am 28.08.2013 20:47, schrieb kernel neophyte:

Re: bcache: Fix a writeback performance regression

2013-08-28 Thread kernel neophyte
On Wed, Aug 28, 2013 at 12:20 PM, Stefan Priebe wrote: > Am 28.08.2013 20:47, schrieb kernel neophyte: > >> On Wed, Aug 28, 2013 at 11:38 AM, Stefan Priebe - Profihost AG >> wrote: >>> >>> I don't had one for a few days. Which kernel so you use? >> >> >> 3.10 kernel with all of kent's stable patc

Re: bcache: Fix a writeback performance regression

2013-08-28 Thread Stefan Priebe
Am 28.08.2013 20:47, schrieb kernel neophyte: On Wed, Aug 28, 2013 at 11:38 AM, Stefan Priebe - Profihost AG wrote: I don't had one for a few days. Which kernel so you use? 3.10 kernel with all of kent's stable patches and perf patches.. Which exact 3.10 version? Which patches on top? and

Re: bcache: Fix a writeback performance regression

2013-08-28 Thread kernel neophyte
On Wed, Aug 28, 2013 at 11:38 AM, Stefan Priebe - Profihost AG wrote: > I don't had one for a few days. Which kernel so you use? 3.10 kernel with all of kent's stable patches and perf patches.. and all defaults except for: echo writeback > /sys/block/bcache0/bcache/cache_mode echo 0 > /sys/bloc

Re: bcache: Fix a writeback performance regression

2013-08-28 Thread Stefan Priebe - Profihost AG
I don't had one for a few days. Which kernel so you use? Stefan This mail was sent with my iPhone. Am 28.08.2013 um 20:12 schrieb kernel neophyte : > On Tue, Aug 27, 2013 at 11:05 PM, Kent Overstreet wrote: >> On Tue, Aug 27, 2013 at 08:37:54PM -0700, kernel neophyte wrote: >>> Hi Kent, >>> >

Re: bcache: Fix a writeback performance regression

2013-08-28 Thread kernel neophyte
On Tue, Aug 27, 2013 at 11:05 PM, Kent Overstreet wrote: > On Tue, Aug 27, 2013 at 08:37:54PM -0700, kernel neophyte wrote: >> Hi Kent, >> >> I am still seeing deadlock: > > Seeing blk_throtl_bio (what asshole misspells words in their function > names, so when you go to grep for them you grep for

Re: bcache: Fix a writeback performance regression

2013-08-27 Thread Kent Overstreet
On Tue, Aug 27, 2013 at 08:37:54PM -0700, kernel neophyte wrote: > Hi Kent, > > I am still seeing deadlock: Seeing blk_throtl_bio (what asshole misspells words in their function names, so when you go to grep for them you grep for the wrong thing?) in the backtrace - is that what it takes to hit i

Re: bcache: Fix a writeback performance regression

2013-08-27 Thread kernel neophyte
Hi Kent, I am still seeing deadlock: [ 930.169234] bcache: bch_cached_dev_attach() Caching sda3 as bcache0 on set 06496904-10ca-489a-ae75-68c6a07d3db1 [ 2522.956188] INFO: task bcache_writebac:2058 blocked for more than 120 seconds. [ 2522.956199] "echo 0 > /proc/sys/kernel/hung_task_timeout_sec

Re: bcache: Fix a writeback performance regression

2013-08-26 Thread Stefan Priebe
Hi Kent, a new one under 3.10: 2013-08-26 21:05:30 INFO: task ceph-osd:8939 blocked for more than 120 seconds. 2013-08-26 21:05:30 "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message. 2013-08-26 21:05:30 ceph-osd D 8160f760 0 8939 1 0x 2013-08-

Re: bcache: Fix a writeback performance regression

2013-08-22 Thread Kent Overstreet
On Thu, Aug 22, 2013 at 09:32:52AM +0200, Stefan Priebe - Profihost AG wrote: > great! > > Everything seems to work fine now! Except read_dirty always going to > negative values after a reboot. That one's fixed in 3.11 :) -- To unsubscribe from this list: send the line "unsubscribe linux-kernel"

Re: bcache: Fix a writeback performance regression

2013-08-22 Thread Stefan Priebe - Profihost AG
great! Everything seems to work fine now! Except read_dirty always going to negative values after a reboot. Stefan Am 22.08.2013 08:02, schrieb Kent Overstreet: > On Thu, Aug 22, 2013 at 07:59:04AM +0200, Stefan Priebe wrote: >> >>> schedule_timeout() is not the same as >>> schedule_timeout_inte

Re: bcache: Fix a writeback performance regression

2013-08-21 Thread Kent Overstreet
On Thu, Aug 22, 2013 at 07:59:04AM +0200, Stefan Priebe wrote: > > >schedule_timeout() is not the same as > >schedule_timeout_interruptible(). > > just search and replace? So i can try on my own. The one in read_dirty(), line ~330 -- To unsubscribe from this list: send the line "unsubscribe linu

Re: bcache: Fix a writeback performance regression

2013-08-21 Thread Stefan Priebe
>schedule_timeout() is not the same as >schedule_timeout_interruptible(). just search and replace? So i can try on my own. Stefan Am 22.08.2013 07:43, schrieb Kent Overstreet: On Thu, Aug 22, 2013 at 07:27:12AM +0200, Stefan Priebe wrote: today i had this one: Heh, I finally tracked it dow

Re: bcache: Fix a writeback performance regression

2013-08-21 Thread Kent Overstreet
On Thu, Aug 22, 2013 at 07:27:12AM +0200, Stefan Priebe wrote: > today i had this one: Heh, I finally tracked it down earlier today. Turned out to be a rather embarassing bug... schedule_timeout() is not the same as schedule_timeout_interruptible(). I'll get the fix out in the morning, for the mo

Re: bcache: Fix a writeback performance regression

2013-08-21 Thread Stefan Priebe
today i had this one: 2013-08-22 06:28:55 [] schedule+0x24/0x70 2013-08-22 06:28:55 [] io_schedule+0x87/0xd0 2013-08-22 06:28:55 [] sleep_on_page+0x9/0x10 2013-08-22 06:28:55 [] __wait_on_bit+0x57/0x80 2013-08-22 06:28:55 [] ? find_get_pages_tag+0xcc/0x180 2013-08-22 06:28:55

Re: bcache: Fix a writeback performance regression

2013-08-21 Thread Stefan Priebe
Am 22.08.2013 01:47, schrieb Kent Overstreet: On Tue, Aug 20, 2013 at 10:07:45AM +0200, Stefan Priebe - Profihost AG wrote: Am 20.08.2013 10:01, schrieb Stefan Priebe - Profihost AG: Am 20.08.2013 00:27, schrieb Kent Overstreet: On Mon, Aug 19, 2013 at 12:09:24AM +0200, Stefan Priebe wrote:

Re: bcache: Fix a writeback performance regression

2013-08-21 Thread Kent Overstreet
On Tue, Aug 20, 2013 at 10:07:45AM +0200, Stefan Priebe - Profihost AG wrote: > Am 20.08.2013 10:01, schrieb Stefan Priebe - Profihost AG: > > Am 20.08.2013 00:27, schrieb Kent Overstreet: > >> On Mon, Aug 19, 2013 at 12:09:24AM +0200, Stefan Priebe wrote: > >>> > >>> Vanilla 3.10.7 + bcache: Fix a

Re: bcache: Fix a writeback performance regression

2013-08-21 Thread Kent Overstreet
On Tue, Aug 20, 2013 at 10:07:45AM +0200, Stefan Priebe - Profihost AG wrote: > Am 20.08.2013 10:01, schrieb Stefan Priebe - Profihost AG: > > Am 20.08.2013 00:27, schrieb Kent Overstreet: > >> On Mon, Aug 19, 2013 at 12:09:24AM +0200, Stefan Priebe wrote: > >>> > >>> Vanilla 3.10.7 + bcache: Fix a

Re: bcache: Fix a writeback performance regression

2013-08-20 Thread Stefan Priebe - Profihost AG
Am 20.08.2013 10:01, schrieb Stefan Priebe - Profihost AG: > Am 20.08.2013 00:27, schrieb Kent Overstreet: >> On Mon, Aug 19, 2013 at 12:09:24AM +0200, Stefan Priebe wrote: >>> >>> Vanilla 3.10.7 + bcache: Fix a writeback performance regression >>> >>> http://pastebin.com/raw.php?i=LXZk4cMH >> >> W

Re: bcache: Fix a writeback performance regression

2013-08-20 Thread Stefan Priebe - Profihost AG
Am 20.08.2013 00:27, schrieb Kent Overstreet: > On Mon, Aug 19, 2013 at 12:09:24AM +0200, Stefan Priebe wrote: >> >> Vanilla 3.10.7 + bcache: Fix a writeback performance regression >> >> http://pastebin.com/raw.php?i=LXZk4cMH > > Whoops, at first I thought this was the same bug as one I'd already

Re: bcache: Fix a writeback performance regression

2013-08-19 Thread Kent Overstreet
On Mon, Aug 19, 2013 at 12:09:24AM +0200, Stefan Priebe wrote: > > Vanilla 3.10.7 + bcache: Fix a writeback performance regression > > http://pastebin.com/raw.php?i=LXZk4cMH Whoops, at first I thought this was the same bug as one I'd already been chasing down that had been a harmless bug - turns

Re: bcache: Fix a writeback performance regression

2013-08-18 Thread Stefan Priebe
Vanilla 3.10.7 + bcache: Fix a writeback performance regression http://pastebin.com/raw.php?i=LXZk4cMH Stefan Am 16.08.2013 12:11, schrieb Stefan Priebe - Profihost AG: Hi, bcache: Fix a writeback performance regression this one results in 3.10 into hung tasks in bcache_writeback read_dirty