Re: [PATCH] PSS(proportional set size) accounting in smaps
On Tue, Aug 14, 2007 at 01:19:40AM -0500, Matt Mackall wrote: > On Tue, Aug 14, 2007 at 09:33:50AM +0800, Fengguang Wu wrote: > > The "proportional set size" (PSS) of a process is the count of pages it has > > in > > memory, where each page is divided by the number of processes sharing it. > > So if > > a process has 1000 pages all to itself, and 1000 shared with one other > > process, > > its PSS will be 1500. > >- lwn.net: "ELC: How much memory are applications really > > using?" > > > > The PSS proposed by Matt Mackall is a very nice metic for measuring an > > process's > > memory footprint. So collect and export it via /proc//smaps. > > > > Matt Mackall's pagemap/kpagemap and John Berthels's exmap can also do the > > job, > > providing pretty much details. But for PSS, let's do it in a simple way. > > Yes, if people actually want to use this particular metric a lot (and > I obviously personally think it makes a lot of sense), then it should > be done in kernel like this. Thank you for the acknowledge, Matt. > > Cc: Matt Mackall <[EMAIL PROTECTED]> > > Cc: John Berthels <[EMAIL PROTECTED]> > > Signed-off-by: Fengguang Wu <[EMAIL PROTECTED]> > > --- > > fs/proc/task_mmu.c | 13 ++--- > > 1 file changed, 10 insertions(+), 3 deletions(-) > > > > --- linux-2.6.23-rc2-mm2.orig/fs/proc/task_mmu.c > > +++ linux-2.6.23-rc2-mm2/fs/proc/task_mmu.c > > @@ -319,6 +319,7 @@ const struct file_operations proc_maps_o > > struct mem_size_stats > > { > > unsigned long resident; > > + u64 pss; /* proportional set size: my share of rss */ > > 64 bits? Yes, to accommodate the extra 12 bits for error shifting. > > unsigned long shared_clean; > > unsigned long shared_dirty; > > unsigned long private_clean; > > @@ -341,6 +342,7 @@ static int smaps_pte_range(pmd_t *pmd, u > > pte_t *pte, ptent; > > spinlock_t *ptl; > > struct page *page; > > + int mapcount; > > > > pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); > > for (; addr != end; pte++, addr += PAGE_SIZE) { > > @@ -357,16 +359,19 @@ static int smaps_pte_range(pmd_t *pmd, u > > /* Accumulate the size in pages that have been accessed. */ > > if (pte_young(ptent) || PageReferenced(page)) > > mss->referenced += PAGE_SIZE; > > - if (page_mapcount(page) >= 2) { > > + mapcount = page_mapcount(page); > > + if (mapcount >= 2) { > > if (pte_dirty(ptent)) > > mss->shared_dirty += PAGE_SIZE; > > else > > mss->shared_clean += PAGE_SIZE; > > + mss->pss += (PAGE_SIZE << 12) / mapcount; > > Hmm, what's that shift for? Oh, you're doing fixed-point math. > > 64-bit divisions are quite expensive on some platforms. The compiler > might be able to do something smarter with common constants like: > >if (mapcount == 1) > mss->pss += PAGE_SIZE; >else if (mapcount == 2) > mss->pss += PAGE_SIZE / 2; >else if (mapcount == 3) > mss->pss += PAGE_SIZE / 3; >else if (mapcount == 4) > mss->pss += PAGE_SIZE / 4; >else > mss->pss += PAGE_SIZE / mapcount; > > ..but I don't know. I suspect we'll at least want to special-case > mapcount == 1 though. Don't worry, the PAGE_SIZE being divided is unsigned long. So there's no 64bit division on 32bit CPU :) And we do avoid the division for the common case of mapcount == 1. > > + sarg.mss.resident >> 10, > > + (unsigned long)(mss->pss >> 22), > > And then you're throwing away 22 bits of precision. 10 bits wasn't > enough? Hmmm.. Looks like the worst case is sharing a 4k page 2049 > ways, where we'll be off by .999 bytes per 4k page for nearly 50% > error. Your extra 12 bits drops this to .2% error, so I suppose it's > worth it. > > But it probably needs a comment. OK, I introduced PSS_ERROR_BITS=12, and some comments for it. Note that the output unit of 1KB could be the most significant source of errors :) > > - sarg.mss.referenced >> 10); > > + sarg.mss.referenced>> 10); > > Unrelated change. Ok, removed it. Thank you, Fengguang === PSS(proportional set size) accounting in smaps The "proportional set size" (PSS) of a process is the count of pages it has in memory, where each page is divided by the number of processes sharing it. So if a process has 1000 pages all to itself, and 1000 shared with one other process, its PSS will be 1500. - lwn.net: "ELC: How much memory are applications really using?" The PSS proposed by Matt Mackall is a very nice metic for measuring an process's memory footprint. So collect and export it via /proc//smaps. Matt Mackall's pagemap/kpagemap and John Berthels's exmap can also do the job. They are comprehensive tools. But for PSS, let's do it in the simple way. Cc: Matt Mackall <[EMAIL PROTECTED]> Cc:
Re: [PATCH] PSS(proportional set size) accounting in smaps
On Tue, Aug 14, 2007 at 10:57:31AM +0530, Balbir Singh wrote: > I keep forgetting to check that you are on the cc. My email client > loves dropping you from the to/cc list. hehe, sorry for my crappy smtp server ;) > Fengguang Wu wrote: > > The "proportional set size" (PSS) of a process is the count of pages it has > > in > > memory, where each page is divided by the number of processes sharing it. > > So if > > a process has 1000 pages all to itself, and 1000 shared with one other > > process, > > its PSS will be 1500. > >- lwn.net: "ELC: How much memory are applications really > > using?" > > > > The PSS proposed by Matt Mackall is a very nice metic for measuring an > > process's > > memory footprint. So collect and export it via /proc//smaps. > > > > Matt Mackall's pagemap/kpagemap and John Berthels's exmap can also do the > > job, > > providing pretty much details. But for PSS, let's do it in a simple way. > > > > Cc: Matt Mackall <[EMAIL PROTECTED]> > > Cc: John Berthels <[EMAIL PROTECTED]> > > Signed-off-by: Fengguang Wu <[EMAIL PROTECTED]> > > I like the idea of moving towards PSS. I had sent some patches back in > December > last year > > http://marc.info/?l=linux-mm&m=116738715329816&w=4 Thank you. That's a pretty comprehensive work. > > - if (page_mapcount(page) >= 2) { > > + mapcount = page_mapcount(page); > > + if (mapcount >= 2) { > > This accounting is of-course racy. Mapcount can change any moment. Sure it is: I never expect to provide accurate numbers. The mapcount here is to prevent divide-by-zero errors. > If we are reasonably sure that mapping will not change at the time > of page_rmap_x() operations, we could handle shared accounting > at those points and implement accurate shared accounting. That would be desirable, if only we can keep the cost low ;) - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH] PSS(proportional set size) accounting in smaps
On Tue, Aug 14, 2007 at 09:33:50AM +0800, Fengguang Wu wrote: > The "proportional set size" (PSS) of a process is the count of pages it has in > memory, where each page is divided by the number of processes sharing it. So > if > a process has 1000 pages all to itself, and 1000 shared with one other > process, > its PSS will be 1500. >- lwn.net: "ELC: How much memory are applications really > using?" > > The PSS proposed by Matt Mackall is a very nice metic for measuring an > process's > memory footprint. So collect and export it via /proc//smaps. > > Matt Mackall's pagemap/kpagemap and John Berthels's exmap can also do the job, > providing pretty much details. But for PSS, let's do it in a simple way. Yes, if people actually want to use this particular metric a lot (and I obviously personally think it makes a lot of sense), then it should be done in kernel like this. > Cc: Matt Mackall <[EMAIL PROTECTED]> > Cc: John Berthels <[EMAIL PROTECTED]> > Signed-off-by: Fengguang Wu <[EMAIL PROTECTED]> > --- > fs/proc/task_mmu.c | 13 ++--- > 1 file changed, 10 insertions(+), 3 deletions(-) > > --- linux-2.6.23-rc2-mm2.orig/fs/proc/task_mmu.c > +++ linux-2.6.23-rc2-mm2/fs/proc/task_mmu.c > @@ -319,6 +319,7 @@ const struct file_operations proc_maps_o > struct mem_size_stats > { > unsigned long resident; > + u64 pss; /* proportional set size: my share of rss */ 64 bits? > unsigned long shared_clean; > unsigned long shared_dirty; > unsigned long private_clean; > @@ -341,6 +342,7 @@ static int smaps_pte_range(pmd_t *pmd, u > pte_t *pte, ptent; > spinlock_t *ptl; > struct page *page; > + int mapcount; > > pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); > for (; addr != end; pte++, addr += PAGE_SIZE) { > @@ -357,16 +359,19 @@ static int smaps_pte_range(pmd_t *pmd, u > /* Accumulate the size in pages that have been accessed. */ > if (pte_young(ptent) || PageReferenced(page)) > mss->referenced += PAGE_SIZE; > - if (page_mapcount(page) >= 2) { > + mapcount = page_mapcount(page); > + if (mapcount >= 2) { > if (pte_dirty(ptent)) > mss->shared_dirty += PAGE_SIZE; > else > mss->shared_clean += PAGE_SIZE; > + mss->pss += (PAGE_SIZE << 12) / mapcount; Hmm, what's that shift for? Oh, you're doing fixed-point math. 64-bit divisions are quite expensive on some platforms. The compiler might be able to do something smarter with common constants like: if (mapcount == 1) mss->pss += PAGE_SIZE; else if (mapcount == 2) mss->pss += PAGE_SIZE / 2; else if (mapcount == 3) mss->pss += PAGE_SIZE / 3; else if (mapcount == 4) mss->pss += PAGE_SIZE / 4; else mss->pss += PAGE_SIZE / mapcount; ..but I don't know. I suspect we'll at least want to special-case mapcount == 1 though. > +sarg.mss.resident >> 10, > +(unsigned long)(mss->pss >> 22), And then you're throwing away 22 bits of precision. 10 bits wasn't enough? Hmmm.. Looks like the worst case is sharing a 4k page 2049 ways, where we'll be off by .999 bytes per 4k page for nearly 50% error. Your extra 12 bits drops this to .2% error, so I suppose it's worth it. But it probably needs a comment. > -sarg.mss.referenced >> 10); > +sarg.mss.referenced>> 10); Unrelated change. -- Mathematics is the supreme nostalgia of our time. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
[Fwd: Re: [PATCH] PSS(proportional set size) accounting in smaps]
I keep forgetting to check that you are on the cc. My email client loves dropping you from the to/cc list. Original Message Subject: Re: [PATCH] PSS(proportional set size) accounting in smaps Date: Tue, 14 Aug 2007 10:56:12 +0530 From: Balbir Singh <[EMAIL PROTECTED]> Reply-To: [EMAIL PROTECTED] Organization: IBM To: Andrew Morton <[EMAIL PROTECTED]>, Matt Mackall <[EMAIL PROTECTED]>, John Berthels <[EMAIL PROTECTED]>, linux-kernel References: <[EMAIL PROTECTED]> Fengguang Wu wrote: > The "proportional set size" (PSS) of a process is the count of pages it has in > memory, where each page is divided by the number of processes sharing it. So > if > a process has 1000 pages all to itself, and 1000 shared with one other > process, > its PSS will be 1500. >- lwn.net: "ELC: How much memory are applications really > using?" > > The PSS proposed by Matt Mackall is a very nice metic for measuring an > process's > memory footprint. So collect and export it via /proc//smaps. > > Matt Mackall's pagemap/kpagemap and John Berthels's exmap can also do the job, > providing pretty much details. But for PSS, let's do it in a simple way. > > Cc: Matt Mackall <[EMAIL PROTECTED]> > Cc: John Berthels <[EMAIL PROTECTED]> > Signed-off-by: Fengguang Wu <[EMAIL PROTECTED]> I like the idea of moving towards PSS. I had sent some patches back in December last year http://marc.info/?l=linux-mm&m=116738715329816&w=4 > --- > fs/proc/task_mmu.c | 13 ++--- > 1 file changed, 10 insertions(+), 3 deletions(-) > > --- linux-2.6.23-rc2-mm2.orig/fs/proc/task_mmu.c > +++ linux-2.6.23-rc2-mm2/fs/proc/task_mmu.c > @@ -319,6 +319,7 @@ const struct file_operations proc_maps_o > struct mem_size_stats > { > unsigned long resident; > + u64 pss; /* proportional set size: my share of rss */ > unsigned long shared_clean; > unsigned long shared_dirty; > unsigned long private_clean; > @@ -341,6 +342,7 @@ static int smaps_pte_range(pmd_t *pmd, u > pte_t *pte, ptent; > spinlock_t *ptl; > struct page *page; > + int mapcount; > > pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); > for (; addr != end; pte++, addr += PAGE_SIZE) { > @@ -357,16 +359,19 @@ static int smaps_pte_range(pmd_t *pmd, u > /* Accumulate the size in pages that have been accessed. */ > if (pte_young(ptent) || PageReferenced(page)) > mss->referenced += PAGE_SIZE; > - if (page_mapcount(page) >= 2) { > + mapcount = page_mapcount(page); > + if (mapcount >= 2) { This accounting is of-course racy. Mapcount can change any moment. > if (pte_dirty(ptent)) > mss->shared_dirty += PAGE_SIZE; > else > mss->shared_clean += PAGE_SIZE; > + mss->pss += (PAGE_SIZE << 12) / mapcount; > } else { > if (pte_dirty(ptent)) > mss->private_dirty += PAGE_SIZE; > else > mss->private_clean += PAGE_SIZE; > + mss->pss += (PAGE_SIZE << 12); > } > } > pte_unmap_unlock(pte - 1, ptl); > @@ -395,18 +400,20 @@ static int show_smap(struct seq_file *m, > seq_printf(m, > "Size: %8lu kB\n" > "Rss:%8lu kB\n" > +"Pss:%8lu kB\n" > "Shared_Clean: %8lu kB\n" > "Shared_Dirty: %8lu kB\n" > "Private_Clean: %8lu kB\n" > "Private_Dirty: %8lu kB\n" > "Referenced: %8lu kB\n", > (vma->vm_end - vma->vm_start) >> 10, > -sarg.mss.resident >> 10, > +sarg.mss.resident >> 10, > +(unsigned long)(mss->pss >> 22), > sarg.mss.shared_clean >> 10, > sarg.mss.shared_dirty >> 10, > sarg.mss.private_clean >> 10, > sarg.mss.private_dirty >> 10, > -sarg.mss.referenced >> 10); > +sarg.mss.referenced>> 10); > > return ret; > } > If we are reasonably sure that mapping will not change at the time of page_rmap_x() operations, we could handle shared accounting at those points and implement accurate shared accounting. -- Warm Regards, Balbir Singh Linux Technology Center IBM, ISTL - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Re: [PATCH] PSS(proportional set size) accounting in smaps
Fengguang Wu wrote: > The "proportional set size" (PSS) of a process is the count of pages it has in > memory, where each page is divided by the number of processes sharing it. So > if > a process has 1000 pages all to itself, and 1000 shared with one other > process, > its PSS will be 1500. >- lwn.net: "ELC: How much memory are applications really > using?" > > The PSS proposed by Matt Mackall is a very nice metic for measuring an > process's > memory footprint. So collect and export it via /proc//smaps. > > Matt Mackall's pagemap/kpagemap and John Berthels's exmap can also do the job, > providing pretty much details. But for PSS, let's do it in a simple way. > > Cc: Matt Mackall <[EMAIL PROTECTED]> > Cc: John Berthels <[EMAIL PROTECTED]> > Signed-off-by: Fengguang Wu <[EMAIL PROTECTED]> I like the idea of moving towards PSS. I had sent some patches back in December last year http://marc.info/?l=linux-mm&m=116738715329816&w=4 > --- > fs/proc/task_mmu.c | 13 ++--- > 1 file changed, 10 insertions(+), 3 deletions(-) > > --- linux-2.6.23-rc2-mm2.orig/fs/proc/task_mmu.c > +++ linux-2.6.23-rc2-mm2/fs/proc/task_mmu.c > @@ -319,6 +319,7 @@ const struct file_operations proc_maps_o > struct mem_size_stats > { > unsigned long resident; > + u64 pss; /* proportional set size: my share of rss */ > unsigned long shared_clean; > unsigned long shared_dirty; > unsigned long private_clean; > @@ -341,6 +342,7 @@ static int smaps_pte_range(pmd_t *pmd, u > pte_t *pte, ptent; > spinlock_t *ptl; > struct page *page; > + int mapcount; > > pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); > for (; addr != end; pte++, addr += PAGE_SIZE) { > @@ -357,16 +359,19 @@ static int smaps_pte_range(pmd_t *pmd, u > /* Accumulate the size in pages that have been accessed. */ > if (pte_young(ptent) || PageReferenced(page)) > mss->referenced += PAGE_SIZE; > - if (page_mapcount(page) >= 2) { > + mapcount = page_mapcount(page); > + if (mapcount >= 2) { This accounting is of-course racy. Mapcount can change any moment. > if (pte_dirty(ptent)) > mss->shared_dirty += PAGE_SIZE; > else > mss->shared_clean += PAGE_SIZE; > + mss->pss += (PAGE_SIZE << 12) / mapcount; > } else { > if (pte_dirty(ptent)) > mss->private_dirty += PAGE_SIZE; > else > mss->private_clean += PAGE_SIZE; > + mss->pss += (PAGE_SIZE << 12); > } > } > pte_unmap_unlock(pte - 1, ptl); > @@ -395,18 +400,20 @@ static int show_smap(struct seq_file *m, > seq_printf(m, > "Size: %8lu kB\n" > "Rss:%8lu kB\n" > +"Pss:%8lu kB\n" > "Shared_Clean: %8lu kB\n" > "Shared_Dirty: %8lu kB\n" > "Private_Clean: %8lu kB\n" > "Private_Dirty: %8lu kB\n" > "Referenced: %8lu kB\n", > (vma->vm_end - vma->vm_start) >> 10, > -sarg.mss.resident >> 10, > +sarg.mss.resident >> 10, > +(unsigned long)(mss->pss >> 22), > sarg.mss.shared_clean >> 10, > sarg.mss.shared_dirty >> 10, > sarg.mss.private_clean >> 10, > sarg.mss.private_dirty >> 10, > -sarg.mss.referenced >> 10); > +sarg.mss.referenced>> 10); > > return ret; > } > If we are reasonably sure that mapping will not change at the time of page_rmap_x() operations, we could handle shared accounting at those points and implement accurate shared accounting. -- Warm Regards, Balbir Singh Linux Technology Center IBM, ISTL - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
[PATCH] PSS(proportional set size) accounting in smaps
The "proportional set size" (PSS) of a process is the count of pages it has in memory, where each page is divided by the number of processes sharing it. So if a process has 1000 pages all to itself, and 1000 shared with one other process, its PSS will be 1500. - lwn.net: "ELC: How much memory are applications really using?" The PSS proposed by Matt Mackall is a very nice metic for measuring an process's memory footprint. So collect and export it via /proc//smaps. Matt Mackall's pagemap/kpagemap and John Berthels's exmap can also do the job, providing pretty much details. But for PSS, let's do it in a simple way. Cc: Matt Mackall <[EMAIL PROTECTED]> Cc: John Berthels <[EMAIL PROTECTED]> Signed-off-by: Fengguang Wu <[EMAIL PROTECTED]> --- fs/proc/task_mmu.c | 13 ++--- 1 file changed, 10 insertions(+), 3 deletions(-) --- linux-2.6.23-rc2-mm2.orig/fs/proc/task_mmu.c +++ linux-2.6.23-rc2-mm2/fs/proc/task_mmu.c @@ -319,6 +319,7 @@ const struct file_operations proc_maps_o struct mem_size_stats { unsigned long resident; + u64 pss; /* proportional set size: my share of rss */ unsigned long shared_clean; unsigned long shared_dirty; unsigned long private_clean; @@ -341,6 +342,7 @@ static int smaps_pte_range(pmd_t *pmd, u pte_t *pte, ptent; spinlock_t *ptl; struct page *page; + int mapcount; pte = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl); for (; addr != end; pte++, addr += PAGE_SIZE) { @@ -357,16 +359,19 @@ static int smaps_pte_range(pmd_t *pmd, u /* Accumulate the size in pages that have been accessed. */ if (pte_young(ptent) || PageReferenced(page)) mss->referenced += PAGE_SIZE; - if (page_mapcount(page) >= 2) { + mapcount = page_mapcount(page); + if (mapcount >= 2) { if (pte_dirty(ptent)) mss->shared_dirty += PAGE_SIZE; else mss->shared_clean += PAGE_SIZE; + mss->pss += (PAGE_SIZE << 12) / mapcount; } else { if (pte_dirty(ptent)) mss->private_dirty += PAGE_SIZE; else mss->private_clean += PAGE_SIZE; + mss->pss += (PAGE_SIZE << 12); } } pte_unmap_unlock(pte - 1, ptl); @@ -395,18 +400,20 @@ static int show_smap(struct seq_file *m, seq_printf(m, "Size: %8lu kB\n" "Rss:%8lu kB\n" + "Pss:%8lu kB\n" "Shared_Clean: %8lu kB\n" "Shared_Dirty: %8lu kB\n" "Private_Clean: %8lu kB\n" "Private_Dirty: %8lu kB\n" "Referenced: %8lu kB\n", (vma->vm_end - vma->vm_start) >> 10, - sarg.mss.resident >> 10, + sarg.mss.resident >> 10, + (unsigned long)(mss->pss >> 22), sarg.mss.shared_clean >> 10, sarg.mss.shared_dirty >> 10, sarg.mss.private_clean >> 10, sarg.mss.private_dirty >> 10, - sarg.mss.referenced >> 10); + sarg.mss.referenced>> 10); return ret; } - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [EMAIL PROTECTED] More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/