Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-11 Thread Wolfgang Walter
Am Donnerstag, 7. Februar 2019, 11:27:50 schrieb Jan Kara: > On Fri 01-02-19 09:19:04, Dave Chinner wrote: > > Maybe for memcgs, but that's exactly the oppose of what we want to > > do for global caches (e.g. filesystem metadata caches). We need to > > make sure that a single, heavily pressured

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-08 Thread Roman Gushchin
On Fri, Feb 08, 2019 at 02:49:44PM -0800, Andrew Morton wrote: > On Fri, 8 Feb 2019 13:50:49 +0100 Jan Kara wrote: > > > > > Has anyone done significant testing with Rik's maybe-fix? > > > > > > I will give it a spin with bonnie++ today. We'll see what comes out. > > > > OK, I did a bonnie++

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-08 Thread Andrew Morton
On Fri, 8 Feb 2019 13:50:49 +0100 Jan Kara wrote: > > > Has anyone done significant testing with Rik's maybe-fix? > > > > I will give it a spin with bonnie++ today. We'll see what comes out. > > OK, I did a bonnie++ run with Rik's patch (on top of 4.20 to rule out other > differences). This

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-08 Thread Dave Chinner
On Thu, Feb 07, 2019 at 09:37:27PM -0800, Andrew Morton wrote: > On Thu, 7 Feb 2019 11:27:50 +0100 Jan Kara wrote: > > > On Fri 01-02-19 09:19:04, Dave Chinner wrote: > > > Maybe for memcgs, but that's exactly the oppose of what we want to > > > do for global caches (e.g. filesystem metadata

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-08 Thread Jan Kara
On Fri 08-02-19 10:55:07, Jan Kara wrote: > On Thu 07-02-19 21:37:27, Andrew Morton wrote: > > On Thu, 7 Feb 2019 11:27:50 +0100 Jan Kara wrote: > > > > > On Fri 01-02-19 09:19:04, Dave Chinner wrote: > > > > Maybe for memcgs, but that's exactly the oppose of what we want to > > > > do for

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-08 Thread Jan Kara
On Thu 07-02-19 21:37:27, Andrew Morton wrote: > On Thu, 7 Feb 2019 11:27:50 +0100 Jan Kara wrote: > > > On Fri 01-02-19 09:19:04, Dave Chinner wrote: > > > Maybe for memcgs, but that's exactly the oppose of what we want to > > > do for global caches (e.g. filesystem metadata caches). We need to

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-07 Thread Andrew Morton
On Thu, 7 Feb 2019 11:27:50 +0100 Jan Kara wrote: > On Fri 01-02-19 09:19:04, Dave Chinner wrote: > > Maybe for memcgs, but that's exactly the oppose of what we want to > > do for global caches (e.g. filesystem metadata caches). We need to > > make sure that a single, heavily pressured cache

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-07 Thread Jan Kara
On Fri 01-02-19 09:19:04, Dave Chinner wrote: > Maybe for memcgs, but that's exactly the oppose of what we want to > do for global caches (e.g. filesystem metadata caches). We need to > make sure that a single, heavily pressured cache doesn't evict small > caches that lower pressure but are

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-04 Thread Dave Chinner
On Fri, Feb 01, 2019 at 09:19:04AM +1100, Dave Chinner wrote: > > So, assuming all this, can we, please, first check if Rik's patch is > > addressing > > the regression? > > Nope, it's broken and buggy, and reintroduces problems with racing > deferred counts that were fixed years ago when I

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-02-01 Thread Dave Chinner
On Thu, Jan 31, 2019 at 03:48:11PM +, Chris Mason wrote: > On 30 Jan 2019, at 20:34, Dave Chinner wrote: > > > On Wed, Jan 30, 2019 at 12:21:07PM +, Chris Mason wrote: > >> > >> > >> On 29 Jan 2019, at 23:17, Dave Chinner wrote: > >> > >>> From: Dave Chinner > >>> > >>> This reverts

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-01-31 Thread Dave Chinner
On Thu, Jan 31, 2019 at 06:57:10PM +, Roman Gushchin wrote: > On Thu, Jan 31, 2019 at 10:10:11AM +0100, Michal Hocko wrote: > > On Thu 31-01-19 12:34:03, Dave Chinner wrote: > > > On Wed, Jan 30, 2019 at 12:21:07PM +, Chris Mason wrote: > > > > > > > > > > > > On 29 Jan 2019, at 23:17,

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-01-31 Thread Roman Gushchin
On Thu, Jan 31, 2019 at 10:10:11AM +0100, Michal Hocko wrote: > On Thu 31-01-19 12:34:03, Dave Chinner wrote: > > On Wed, Jan 30, 2019 at 12:21:07PM +, Chris Mason wrote: > > > > > > > > > On 29 Jan 2019, at 23:17, Dave Chinner wrote: > > > > > > > From: Dave Chinner > > > > > > > > This

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-01-31 Thread Chris Mason
On 30 Jan 2019, at 20:34, Dave Chinner wrote: > On Wed, Jan 30, 2019 at 12:21:07PM +, Chris Mason wrote: >> >> >> On 29 Jan 2019, at 23:17, Dave Chinner wrote: >> >>> From: Dave Chinner >>> >>> This reverts commit a76cf1a474d7dbcd9336b5f5afb0162baa142cf0. >>> >>> This change causes serious

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-01-31 Thread Michal Hocko
On Thu 31-01-19 12:34:03, Dave Chinner wrote: > On Wed, Jan 30, 2019 at 12:21:07PM +, Chris Mason wrote: > > > > > > On 29 Jan 2019, at 23:17, Dave Chinner wrote: > > > > > From: Dave Chinner > > > > > > This reverts commit a76cf1a474d7dbcd9336b5f5afb0162baa142cf0. > > > > > > This change

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-01-30 Thread Dave Chinner
On Wed, Jan 30, 2019 at 12:21:07PM +, Chris Mason wrote: > > > On 29 Jan 2019, at 23:17, Dave Chinner wrote: > > > From: Dave Chinner > > > > This reverts commit a76cf1a474d7dbcd9336b5f5afb0162baa142cf0. > > > > This change causes serious changes to page cache and inode cache > > behaviour

Re: [PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-01-30 Thread Chris Mason
On 29 Jan 2019, at 23:17, Dave Chinner wrote: > From: Dave Chinner > > This reverts commit a76cf1a474d7dbcd9336b5f5afb0162baa142cf0. > > This change causes serious changes to page cache and inode cache > behaviour and balance, resulting in major performance regressions > when combining

[PATCH 1/2] Revert "mm: don't reclaim inodes with many attached pages"

2019-01-29 Thread Dave Chinner
From: Dave Chinner This reverts commit a76cf1a474d7dbcd9336b5f5afb0162baa142cf0. This change causes serious changes to page cache and inode cache behaviour and balance, resulting in major performance regressions when combining worklaods such as large file copies and kernel compiles.