proc: fill 'lib' field in /proc/<pid>/statm
in [Kernel]
|
Prev: score: add asm/asm-offsets.h wrapper
Next: Performance regression in scsi sequential throughput (iozone) due to "e084b - page-allocator: preserve PFN ordering when __GFP_COLD is set"
From: Hugh Dickins on 7 Dec 2009 06:30 On Mon, 7 Dec 2009, Amerigo Wang wrote: > Currently, the 'lib' field of /proc/<pid>/statm is > always 0, we should fill it with the right value, > the same with 'VmLib'. The right value (if you're looking for consistency with Linux 2.4) is the number of currently resident "library" pages: and we don't know that number - we can't even define what a library is. We could add some code to make it show the same bogus number as we show somewhere else, but it has said 0 ever since 2.5.37: so I don't think it's worth a line of code myself, but bow to others. > > Signed-off-by: WANG Cong <amwang(a)redhat.com> > Cc: Eric W. Biederman <ebiederm(a)xmission.com> > Cc: Alexey Dobriyan <adobriyan(a)gmail.com> > Cc: Al Viro <viro(a)zeniv.linux.org.uk> > Cc: Mel Gorman <mel(a)csn.ul.ie> > Cc: stable(a)kernel.org What does alarm me is that you think this is fit for -stable! Hugh -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
From: Cong Wang on 7 Dec 2009 06:50
Hugh Dickins wrote: > On Mon, 7 Dec 2009, Amerigo Wang wrote: > >> Currently, the 'lib' field of /proc/<pid>/statm is >> always 0, we should fill it with the right value, >> the same with 'VmLib'. > > The right value (if you're looking for consistency with Linux 2.4) > is the number of currently resident "library" pages: and we don't > know that number - we can't even define what a library is. Hmm, the current algorithm is just kicking out text size of itself from ->exec_vm, it really makes some sense, but not always. > > We could add some code to make it show the same bogus number as > we show somewhere else, but it has said 0 ever since 2.5.37: so > I don't think it's worth a line of code myself, but bow to others. If you mean 'VmLib' in /proc/<pid>/status, this is the same with it. > >> Signed-off-by: WANG Cong <amwang(a)redhat.com> >> Cc: Eric W. Biederman <ebiederm(a)xmission.com> >> Cc: Alexey Dobriyan <adobriyan(a)gmail.com> >> Cc: Al Viro <viro(a)zeniv.linux.org.uk> >> Cc: Mel Gorman <mel(a)csn.ul.ie> >> Cc: stable(a)kernel.org > > What does alarm me is that you think this is fit for -stable! > Oh, sorry, I thought missing this field is a mistake... Dropped. Thanks. -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo(a)vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/ |