Re: [PATCH v4 3/4] mm, shmem: Add shmem resident memory accounting

2015-10-04 Thread Hugh Dickins
On Fri, 2 Oct 2015, Vlastimil Babka wrote: > From: Jerome Marchand > > Currently looking at /proc//status or statm, there is no way to > distinguish shmem pages from pages mapped to a regular file (shmem > pages are mapped to /dev/zero), even though their implication in > actual memory use is

Re: [PATCH v4 3/4] mm, shmem: Add shmem resident memory accounting

2015-10-04 Thread Hugh Dickins
On Fri, 2 Oct 2015, Vlastimil Babka wrote: > From: Jerome Marchand > > Currently looking at /proc//status or statm, there is no way to > distinguish shmem pages from pages mapped to a regular file (shmem > pages are mapped to /dev/zero), even though their implication in >

Re: [PATCH v4 3/4] mm, shmem: Add shmem resident memory accounting

2015-10-02 Thread Andrew Morton
On Fri, 2 Oct 2015 15:35:50 +0200 Vlastimil Babka wrote: > From: Jerome Marchand Changelog is a bit weird. > Currently looking at /proc//status or statm, there is no way to > distinguish shmem pages from pages mapped to a regular file (shmem > pages are mapped to /dev/zero), even though

[PATCH v4 3/4] mm, shmem: Add shmem resident memory accounting

2015-10-02 Thread Vlastimil Babka
From: Jerome Marchand Currently looking at /proc//status or statm, there is no way to distinguish shmem pages from pages mapped to a regular file (shmem pages are mapped to /dev/zero), even though their implication in actual memory use is quite different. This patch adds MM_SHMEMPAGES counter to

[PATCH v4 3/4] mm, shmem: Add shmem resident memory accounting

2015-10-02 Thread Vlastimil Babka
From: Jerome Marchand Currently looking at /proc//status or statm, there is no way to distinguish shmem pages from pages mapped to a regular file (shmem pages are mapped to /dev/zero), even though their implication in actual memory use is quite different. This patch adds

Re: [PATCH v4 3/4] mm, shmem: Add shmem resident memory accounting

2015-10-02 Thread Andrew Morton
On Fri, 2 Oct 2015 15:35:50 +0200 Vlastimil Babka wrote: > From: Jerome Marchand Changelog is a bit weird. > Currently looking at /proc//status or statm, there is no way to > distinguish shmem pages from pages mapped to a regular file (shmem > pages are