Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

My question still stands: Why are none of these patches submitted to mainline or backported to stable kernels? Thanks. On 2015-09-16 2:08 PM, Denis Bychkov wrote: Hi Kent, Vojtech, list Your fix works perfectly, I can finally remove my patch that disables the partial_stripes_expensive

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

My question still stands: Why are none of these patches submitted to mainline or backported to stable kernels? Thanks. On 2015-09-16 2:08 PM, Denis Bychkov wrote: Hi Kent, Vojtech, list Your fix works perfectly, I can finally remove my patch that disables the partial_stripes_expensive

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Hi Kent, After running a day with this new version of your patch, I did not notice any problems, so I assume, it works. I'll keep an eye on it and report back if find anything bad. I believe, you finally fixed that long lasting bug with writeback threads spinning CPU. To Vojtech Pavlik: thank you

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Hi Kent, After running a day with this new version of your patch, I did not notice any problems, so I assume, it works. I'll keep an eye on it and report back if find anything bad. I believe, you finally fixed that long lasting bug with writeback threads spinning CPU. To Vojtech Pavlik: thank you

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Yes, sure, I'll try it. And yes, this was exactly my setup - several backing devices sharing the same cache set, all in write back mode. When I decided to try your patch yesterday, I created a few more and attached them to the same cache set. Did not want to experiment with the live ones - so just

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

On Thu, Sep 17, 2015 at 08:40:54AM -0800, Kent Overstreet wrote: > On Thu, Sep 17, 2015 at 11:30:17AM -0400, Denis Bychkov wrote: > > Well, it turns out my celebration was a bit premature. > > > > PLEASE, DO NOT APPLY THE PATCH POSTED BY KENT (not the one Vojtech > > posted) ON A PRODUCTION

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

On Thu, Sep 17, 2015 at 11:30:17AM -0400, Denis Bychkov wrote: > Well, it turns out my celebration was a bit premature. > > PLEASE, DO NOT APPLY THE PATCH POSTED BY KENT (not the one Vojtech > posted) ON A PRODUCTION SYSTEM, IT CAUSES DATA CORRUPTION. > > The interesting thing is that it somehow

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Well, it turns out my celebration was a bit premature. PLEASE, DO NOT APPLY THE PATCH POSTED BY KENT (not the one Vojtech posted) ON A PRODUCTION SYSTEM, IT CAUSES DATA CORRUPTION. The interesting thing is that it somehow damaged the partition that was not supposed to receive any writes (the

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Well, it turns out my celebration was a bit premature. PLEASE, DO NOT APPLY THE PATCH POSTED BY KENT (not the one Vojtech posted) ON A PRODUCTION SYSTEM, IT CAUSES DATA CORRUPTION. The interesting thing is that it somehow damaged the partition that was not supposed to receive any writes (the

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

On Thu, Sep 17, 2015 at 11:30:17AM -0400, Denis Bychkov wrote: > Well, it turns out my celebration was a bit premature. > > PLEASE, DO NOT APPLY THE PATCH POSTED BY KENT (not the one Vojtech > posted) ON A PRODUCTION SYSTEM, IT CAUSES DATA CORRUPTION. > > The interesting thing is that it somehow

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

On Thu, Sep 17, 2015 at 08:40:54AM -0800, Kent Overstreet wrote: > On Thu, Sep 17, 2015 at 11:30:17AM -0400, Denis Bychkov wrote: > > Well, it turns out my celebration was a bit premature. > > > > PLEASE, DO NOT APPLY THE PATCH POSTED BY KENT (not the one Vojtech > > posted) ON A PRODUCTION

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Yes, sure, I'll try it. And yes, this was exactly my setup - several backing devices sharing the same cache set, all in write back mode. When I decided to try your patch yesterday, I created a few more and attached them to the same cache set. Did not want to experiment with the live ones - so just

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Hi Kent, Vojtech, list Your fix works perfectly, I can finally remove my patch that disables the partial_stripes_expensive branch and unleash the full bcache performance on my RAID-6 array. I was aware of this problem for some time now, but never got to learning the bcache codebase enough to find

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

On Sat, Sep 05, 2015 at 01:10:12PM +0200, Vojtech Pavlik wrote: > Fix writeback_thread never finishing writing back all dirty data in bcache > when > partial_stripes_expensive is set, and spinning, consuming 100% of CPU instead. > > Signed-off-by: Vojtech Pavlik > --- > > This is a fix for the

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

On Sat, Sep 05, 2015 at 01:10:12PM +0200, Vojtech Pavlik wrote: > Fix writeback_thread never finishing writing back all dirty data in bcache > when > partial_stripes_expensive is set, and spinning, consuming 100% of CPU instead. > > Signed-off-by: Vojtech Pavlik > --- > >

Re: [PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Hi Kent, Vojtech, list Your fix works perfectly, I can finally remove my patch that disables the partial_stripes_expensive branch and unleash the full bcache performance on my RAID-6 array. I was aware of this problem for some time now, but never got to learning the bcache codebase enough to find

[PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Fix writeback_thread never finishing writing back all dirty data in bcache when partial_stripes_expensive is set, and spinning, consuming 100% of CPU instead. Signed-off-by: Vojtech Pavlik --- This is a fix for the current upstream bcache, not the devel branch. If partial_stripes_expensive is

[PATCH] bcache: Fix writeback_thread never writing back incomplete stripes.

Fix writeback_thread never finishing writing back all dirty data in bcache when partial_stripes_expensive is set, and spinning, consuming 100% of CPU instead. Signed-off-by: Vojtech Pavlik --- This is a fix for the current upstream bcache, not the devel branch. If