* size of swapped-out part of the process [not found] <ab4346860811170452t2cbcb451u407edca19c9e0ba6@mail.gmail.com> @ 2008-11-17 12:53 ` Victoria Muntean 2008-11-17 13:33 ` J.R. Mauro 2008-11-17 14:02 ` Hugh Dickins 0 siblings, 2 replies; 12+ messages in thread From: Victoria Muntean @ 2008-11-17 12:53 UTC (permalink / raw) To: Kernel Linux How can I calculate size of swapped-out part of the process from /proc/PID/*, even if approximately ? For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global swap-in-use==0. Hence (VmSize - VmRSS) is far from being swapped-out part of the process. What is ? Thanks Viki ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 12:53 ` size of swapped-out part of the process Victoria Muntean @ 2008-11-17 13:33 ` J.R. Mauro 2008-11-17 13:47 ` Victoria Muntean 2008-11-17 14:02 ` Hugh Dickins 1 sibling, 1 reply; 12+ messages in thread From: J.R. Mauro @ 2008-11-17 13:33 UTC (permalink / raw) To: Victoria Muntean; +Cc: Kernel Linux On Mon, Nov 17, 2008 at 7:53 AM, Victoria Muntean <vikimun@gmail.com> wrote: > How can I calculate size of swapped-out part of the process from > /proc/PID/*, even if approximately ? > For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global > swap-in-use==0. > Hence (VmSize - VmRSS) is far from being swapped-out part of the > process. What is ? Doesn't top's SWAP field give you this? > > Thanks > Viki > -- > To unsubscribe from this list: send the line "unsubscribe linux-kernel" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html > Please read the FAQ at http://www.tux.org/lkml/ > ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 13:33 ` J.R. Mauro @ 2008-11-17 13:47 ` Victoria Muntean 2008-11-17 13:55 ` J.R. Mauro 0 siblings, 1 reply; 12+ messages in thread From: Victoria Muntean @ 2008-11-17 13:47 UTC (permalink / raw) To: J.R. Mauro, Kernel Linux On Mon, Nov 17, 2008 at 3:33 PM, J.R. Mauro <jrm8005@gmail.com> wrote: > On Mon, Nov 17, 2008 at 7:53 AM, Victoria Muntean <vikimun@gmail.com> wrote: >> How can I calculate size of swapped-out part of the process from >> /proc/PID/*, even if approximately ? >> For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global >> swap-in-use==0. >> Hence (VmSize - VmRSS) is far from being swapped-out part of the >> process. What is ? > > Doesn't top's SWAP field give you this? No. SWAP is global. I need per-process. Per-process participation in swap. ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 13:47 ` Victoria Muntean @ 2008-11-17 13:55 ` J.R. Mauro 2008-11-17 14:08 ` Victoria Muntean 0 siblings, 1 reply; 12+ messages in thread From: J.R. Mauro @ 2008-11-17 13:55 UTC (permalink / raw) To: Victoria Muntean; +Cc: Kernel Linux On Mon, Nov 17, 2008 at 8:47 AM, Victoria Muntean <vikimun@gmail.com> wrote: > On Mon, Nov 17, 2008 at 3:33 PM, J.R. Mauro <jrm8005@gmail.com> wrote: >> On Mon, Nov 17, 2008 at 7:53 AM, Victoria Muntean <vikimun@gmail.com> wrote: >>> How can I calculate size of swapped-out part of the process from >>> /proc/PID/*, even if approximately ? >>> For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global >>> swap-in-use==0. >>> Hence (VmSize - VmRSS) is far from being swapped-out part of the >>> process. What is ? >> >> Doesn't top's SWAP field give you this? > > No. SWAP is global. I need per-process. Per-process participation in swap. > Are you sure? I'm not talking about the Swap: BigNumber in the header. If you hit f you can add a SWAP column which is per process. Or if it isn't per-process, it's horribly misleading. ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 13:55 ` J.R. Mauro @ 2008-11-17 14:08 ` Victoria Muntean 2008-11-17 14:14 ` J.R. Mauro 0 siblings, 1 reply; 12+ messages in thread From: Victoria Muntean @ 2008-11-17 14:08 UTC (permalink / raw) To: J.R. Mauro; +Cc: Kernel Linux On Mon, Nov 17, 2008 at 3:55 PM, J.R. Mauro <jrm8005@gmail.com> wrote: > On Mon, Nov 17, 2008 at 8:47 AM, Victoria Muntean <vikimun@gmail.com> wrote: >> On Mon, Nov 17, 2008 at 3:33 PM, J.R. Mauro <jrm8005@gmail.com> wrote: >>> On Mon, Nov 17, 2008 at 7:53 AM, Victoria Muntean <vikimun@gmail.com> wrote: >>>> How can I calculate size of swapped-out part of the process from >>>> /proc/PID/*, even if approximately ? >>>> For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global >>>> swap-in-use==0. >>>> Hence (VmSize - VmRSS) is far from being swapped-out part of the >>>> process. What is ? >>> >>> Doesn't top's SWAP field give you this? >> >> No. SWAP is global. I need per-process. Per-process participation in swap. >> > > Are you sure? I'm not talking about the Swap: BigNumber in the header. > If you hit f you can add a SWAP column which is per process. Or if it > isn't per-process, it's horribly misleading. 'top' naively shows (VmSize - VmRSS) in the SWAP column, which is not helpful. ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 14:08 ` Victoria Muntean @ 2008-11-17 14:14 ` J.R. Mauro 0 siblings, 0 replies; 12+ messages in thread From: J.R. Mauro @ 2008-11-17 14:14 UTC (permalink / raw) To: Victoria Muntean; +Cc: Kernel Linux On Mon, Nov 17, 2008 at 9:08 AM, Victoria Muntean <vikimun@gmail.com> wrote: > On Mon, Nov 17, 2008 at 3:55 PM, J.R. Mauro <jrm8005@gmail.com> wrote: >> On Mon, Nov 17, 2008 at 8:47 AM, Victoria Muntean <vikimun@gmail.com> wrote: >>> On Mon, Nov 17, 2008 at 3:33 PM, J.R. Mauro <jrm8005@gmail.com> wrote: >>>> On Mon, Nov 17, 2008 at 7:53 AM, Victoria Muntean <vikimun@gmail.com> wrote: >>>>> How can I calculate size of swapped-out part of the process from >>>>> /proc/PID/*, even if approximately ? >>>>> For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global >>>>> swap-in-use==0. >>>>> Hence (VmSize - VmRSS) is far from being swapped-out part of the >>>>> process. What is ? >>>> >>>> Doesn't top's SWAP field give you this? >>> >>> No. SWAP is global. I need per-process. Per-process participation in swap. >>> >> >> Are you sure? I'm not talking about the Swap: BigNumber in the header. >> If you hit f you can add a SWAP column which is per process. Or if it >> isn't per-process, it's horribly misleading. > > 'top' naively shows (VmSize - VmRSS) in the SWAP column, which is not helpful. > Ah, I see. The smap solution Hugh mentioned is what you're after, I guess. ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 12:53 ` size of swapped-out part of the process Victoria Muntean 2008-11-17 13:33 ` J.R. Mauro @ 2008-11-17 14:02 ` Hugh Dickins 2008-11-17 14:09 ` Victoria Muntean 2008-11-17 19:42 ` Victoria Muntean 1 sibling, 2 replies; 12+ messages in thread From: Hugh Dickins @ 2008-11-17 14:02 UTC (permalink / raw) To: Victoria Muntean; +Cc: Kernel Linux On Mon, 17 Nov 2008, Victoria Muntean wrote: > How can I calculate size of swapped-out part of the process from > /proc/PID/*, even if approximately ? > For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global > swap-in-use==0. > Hence (VmSize - VmRSS) is far from being swapped-out part of the > process. What is ? I'm sorry, we don't record those numbers per process or per mm, so you won't find them in /proc/PID/status or /proc/PID/statm. And we'd probably resist accepting a patch to add them, so as not to enlarge key data structures to hold them. There's also an ambiguity about what "swapped-out" means: does it include shared file pages which were once mapped into the process, but have since been freed under memory pressure? But it looks like you're interested in swapped-out to mean written on swap. In that case, if you have CONFIG_PROC_PAGE_MONITOR=y, you can read /proc/PID/smaps and add up all the "Swap:" lines. But don't keep reading it in a tight loop: that would tend to detract from the performance of your system. Hugh ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 14:02 ` Hugh Dickins @ 2008-11-17 14:09 ` Victoria Muntean 2008-11-17 19:42 ` Victoria Muntean 1 sibling, 0 replies; 12+ messages in thread From: Victoria Muntean @ 2008-11-17 14:09 UTC (permalink / raw) To: Hugh Dickins; +Cc: Kernel Linux On Mon, Nov 17, 2008 at 4:02 PM, Hugh Dickins <hugh@veritas.com> wrote: > On Mon, 17 Nov 2008, Victoria Muntean wrote: >> How can I calculate size of swapped-out part of the process from >> /proc/PID/*, even if approximately ? >> For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global >> swap-in-use==0. >> Hence (VmSize - VmRSS) is far from being swapped-out part of the >> process. What is ? > > I'm sorry, we don't record those numbers per process or per mm, > so you won't find them in /proc/PID/status or /proc/PID/statm. > > And we'd probably resist accepting a patch to add them, > so as not to enlarge key data structures to hold them. > > There's also an ambiguity about what "swapped-out" means: does > it include shared file pages which were once mapped into the > process, but have since been freed under memory pressure? > > But it looks like you're interested in swapped-out to mean written > on swap. In that case, if you have CONFIG_PROC_PAGE_MONITOR=y, > you can read /proc/PID/smaps and add up all the "Swap:" lines. ok thanks > But don't keep reading it in a tight loop: that would > tend to detract from the performance of your system. ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 14:02 ` Hugh Dickins 2008-11-17 14:09 ` Victoria Muntean @ 2008-11-17 19:42 ` Victoria Muntean 2008-11-17 20:31 ` Bryan Donlan ` (2 more replies) 1 sibling, 3 replies; 12+ messages in thread From: Victoria Muntean @ 2008-11-17 19:42 UTC (permalink / raw) To: Hugh Dickins; +Cc: Kernel Linux On 11/17/08, Hugh Dickins <hugh@veritas.com> wrote: > On Mon, 17 Nov 2008, Victoria Muntean wrote: >> How can I calculate size of swapped-out part of the process from >> /proc/PID/*, even if approximately ? >> For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global >> swap-in-use==0. >> Hence (VmSize - VmRSS) is far from being swapped-out part of the >> process. What is ? > > I'm sorry, we don't record those numbers per process or per mm, > so you won't find them in /proc/PID/status or /proc/PID/statm. > > And we'd probably resist accepting a patch to add them, > so as not to enlarge key data structures to hold them. > > There's also an ambiguity about what "swapped-out" means: does > it include shared file pages which were once mapped into the > process, but have since been freed under memory pressure? > > But it looks like you're interested in swapped-out to mean written > on swap. In that case, if you have CONFIG_PROC_PAGE_MONITOR=y, > you can read /proc/PID/smaps and add up all the "Swap:" lines. Can you possibly explain what types of pages create "VSZ>>RSZ" effect in total absence of swapping and plenty of free memory ? ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 19:42 ` Victoria Muntean @ 2008-11-17 20:31 ` Bryan Donlan 2008-11-17 21:14 ` Andreas Schwab 2008-11-18 1:33 ` David Schwartz 2 siblings, 0 replies; 12+ messages in thread From: Bryan Donlan @ 2008-11-17 20:31 UTC (permalink / raw) To: Victoria Muntean; +Cc: Hugh Dickins, Kernel Linux On Mon, Nov 17, 2008 at 2:42 PM, Victoria Muntean <vikimun@gmail.com> wrote: > On 11/17/08, Hugh Dickins <hugh@veritas.com> wrote: >> On Mon, 17 Nov 2008, Victoria Muntean wrote: >>> How can I calculate size of swapped-out part of the process from >>> /proc/PID/*, even if approximately ? >>> For example, I have VmSize=1216192 kB, VmRSS = 628788 kB, but global >>> swap-in-use==0. >>> Hence (VmSize - VmRSS) is far from being swapped-out part of the >>> process. What is ? >> >> I'm sorry, we don't record those numbers per process or per mm, >> so you won't find them in /proc/PID/status or /proc/PID/statm. >> >> And we'd probably resist accepting a patch to add them, >> so as not to enlarge key data structures to hold them. >> >> There's also an ambiguity about what "swapped-out" means: does >> it include shared file pages which were once mapped into the >> process, but have since been freed under memory pressure? >> >> But it looks like you're interested in swapped-out to mean written >> on swap. In that case, if you have CONFIG_PROC_PAGE_MONITOR=y, >> you can read /proc/PID/smaps and add up all the "Swap:" lines. > > Can you possibly explain what types of pages create "VSZ>>RSZ" > effect in total absence of swapping and plenty of free memory ? mmap(2) a file, then don't touch some pages in it. This happens all the time with shared libraries - for example, if you link in imagemagick, but never use JPEG encoding/decoding, the pages of memory corresponding to libjpeg will never be touched, and so will be accounted in VSZ but not RSZ. If there's memory pressure (even if you have plenty of cache, this can occur just from IO activity), long-unaccessed pages from these mmaps may be evicted after being read as well. Also, some programs have a habit of allocating much more memory than they actually use - the pages they don't access are not mapped, and so are not accounted in RSZ either. If they never write to them, they won't be present in swap either. ^ permalink raw reply [flat|nested] 12+ messages in thread
* Re: size of swapped-out part of the process 2008-11-17 19:42 ` Victoria Muntean 2008-11-17 20:31 ` Bryan Donlan @ 2008-11-17 21:14 ` Andreas Schwab 2008-11-18 1:33 ` David Schwartz 2 siblings, 0 replies; 12+ messages in thread From: Andreas Schwab @ 2008-11-17 21:14 UTC (permalink / raw) To: Victoria Muntean; +Cc: Hugh Dickins, Kernel Linux "Victoria Muntean" <vikimun@gmail.com> writes: > Can you possibly explain what types of pages create "VSZ>>RSZ" > effect in total absence of swapping and plenty of free memory ? Any mapping that hasn't been touched yet. A page that is never read nor written will never become resident. Andreas. -- Andreas Schwab, SuSE Labs, schwab@suse.de SuSE Linux Products GmbH, Maxfeldstraße 5, 90409 Nürnberg, Germany PGP key fingerprint = 58CA 54C7 6D53 942B 1756 01D3 44D5 214B 8276 4ED5 "And now for something completely different." ^ permalink raw reply [flat|nested] 12+ messages in thread
* RE: size of swapped-out part of the process 2008-11-17 19:42 ` Victoria Muntean 2008-11-17 20:31 ` Bryan Donlan 2008-11-17 21:14 ` Andreas Schwab @ 2008-11-18 1:33 ` David Schwartz 2 siblings, 0 replies; 12+ messages in thread From: David Schwartz @ 2008-11-18 1:33 UTC (permalink / raw) To: Hugh Dickins; +Cc: Kernel Linux > On 11/17/08, Hugh Dickins <hugh@veritas.com> wrote: > Can you possibly explain what types of pages create "VSZ>>RSZ" > effect in total absence of swapping and plenty of free memory ? Rarely-used functions like 'malloc' and 'mmap' enlarge the virtual memory size but (often) have no effect on the resident set size. DS ^ permalink raw reply [flat|nested] 12+ messages in thread
end of thread, other threads:[~2008-11-18 1:33 UTC | newest]
Thread overview: 12+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
[not found] <ab4346860811170452t2cbcb451u407edca19c9e0ba6@mail.gmail.com>
2008-11-17 12:53 ` size of swapped-out part of the process Victoria Muntean
2008-11-17 13:33 ` J.R. Mauro
2008-11-17 13:47 ` Victoria Muntean
2008-11-17 13:55 ` J.R. Mauro
2008-11-17 14:08 ` Victoria Muntean
2008-11-17 14:14 ` J.R. Mauro
2008-11-17 14:02 ` Hugh Dickins
2008-11-17 14:09 ` Victoria Muntean
2008-11-17 19:42 ` Victoria Muntean
2008-11-17 20:31 ` Bryan Donlan
2008-11-17 21:14 ` Andreas Schwab
2008-11-18 1:33 ` David Schwartz
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.