lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
Open Source and information security mailing list archives
| ||
|
Date: Tue, 27 Nov 2012 17:31:39 -0500 From: Steven Rostedt <rostedt@...dmis.org> To: akpm@...ux-foundation.org Cc: mm-commits@...r.kernel.org, robert.jarzmik@...e.fr, david@...morbit.com, fweisbec@...il.com, hughd@...gle.com, mingo@...hat.com, LKML <linux-kernel@...r.kernel.org> Subject: Re: + mm-trace-filemap-add-and-del.patch added to -mm tree On Tue, 2012-11-27 at 14:15 -0800, akpm@...ux-foundation.org wrote: > The patch titled > Subject: mm: trace filemap add and del > has been added to the -mm tree. Its filename is > mm-trace-filemap-add-and-del.patch > > Before you just go and hit "reply", please: > a) Consider who else should be cc'ed > b) Prefer to cc a suitable mailing list as well > c) Ideally: find the original patch on the mailing list and do a > reply-to-all to that, adding suitable additional cc's > > *** Remember to use Documentation/SubmitChecklist when testing your code *** > > The -mm tree is included into linux-next and is updated > there every 3-4 working days > > ------------------------------------------------------ > From: Robert Jarzmik <robert.jarzmik@...e.fr> > Subject: mm: trace filemap add and del > Sorry I missed the first post of this. > Use the events API to trace filemap loading and unloading of file pieces > into the page cache. > > This patch aims at tracing the eviction reload cycle of executable and > shared libraries pages in a memory constrained environment. > > The typical usage is to spot a specific device and inode (for example > /lib/libc.so) to see the eviction cycles, and find out if frequently used > code is rather spread across many pages (bad) or coallesced (good). > > Signed-off-by: Robert Jarzmik <robert.jarzmik@...e.fr> > Cc: Dave Chinner <david@...morbit.com> > Cc: Hugh Dickins <hughd@...gle.com> > Cc: Steven Rostedt <rostedt@...dmis.org> > Cc: Frederic Weisbecker <fweisbec@...il.com> > Cc: Ingo Molnar <mingo@...hat.com> > Signed-off-by: Andrew Morton <akpm@...ux-foundation.org> > --- > > include/trace/events/filemap.h | 79 +++++++++++++++++++++++++++++++ > mm/filemap.c | 5 + > 2 files changed, 84 insertions(+) > > diff -puN /dev/null include/trace/events/filemap.h > --- /dev/null > > +++ a/include/trace/events/filemap.h > @@ -0,0 +1,79 @@ > +#undef TRACE_SYSTEM > +#define TRACE_SYSTEM filemap > + > +#if !defined(_TRACE_FILEMAP_H) || defined(TRACE_HEADER_MULTI_READ) > +#define _TRACE_FILEMAP_H > + > +#include <linux/types.h> > +#include <linux/tracepoint.h> > +#include <linux/mm.h> > +#include <linux/memcontrol.h> > +#include <linux/device.h> > +#include <linux/kdev_t.h> > + > +TRACE_EVENT(mm_filemap_delete_from_page_cache, > + > + TP_PROTO(struct page *page), > + > + TP_ARGS(page), > + > + TP_STRUCT__entry( > + __field(struct page *, page) > + __field(unsigned long, i_ino) > + __field(unsigned long, index) > + __field(dev_t, s_dev) > + ), > + > + TP_fast_assign( > + __entry->page = page; > + __entry->i_ino = page->mapping->host->i_ino; > + __entry->index = page->index; > + if (page->mapping->host->i_sb) > + __entry->s_dev = page->mapping->host->i_sb->s_dev; > + else > + __entry->s_dev = page->mapping->host->i_rdev; > + ), > + > + TP_printk("dev %d:%d ino %lx page=%p pfn=%lu ofs=%lu", > + MAJOR(__entry->s_dev), MINOR(__entry->s_dev), > + __entry->i_ino, > + __entry->page, > + page_to_pfn(__entry->page), > + __entry->index << PAGE_SHIFT) > +); > + > +TRACE_EVENT(mm_filemap_add_to_page_cache, > + > + TP_PROTO(struct page *page), > + > + TP_ARGS(page), > + > + TP_STRUCT__entry( > + __field(struct page *, page) > + __field(unsigned long, i_ino) > + __field(unsigned long, index) > + __field(dev_t, s_dev) > + ), > + > + TP_fast_assign( > + __entry->page = page; > + __entry->i_ino = page->mapping->host->i_ino; > + __entry->index = page->index; > + if (page->mapping->host->i_sb) > + __entry->s_dev = page->mapping->host->i_sb->s_dev; > + else > + __entry->s_dev = page->mapping->host->i_rdev; > + ), > + > + TP_printk("dev %d:%d ino %lx page=%p pfn=%lu ofs=%lu", > + MAJOR(__entry->s_dev), MINOR(__entry->s_dev), > + __entry->i_ino, > + __entry->page, > + page_to_pfn(__entry->page), > + __entry->index << PAGE_SHIFT) > +); The above two events are identical. Please use DECLARE_EVENT_CLASS() and DEFINE_EVENT() for duplicates. Each TRACE_EVENT() can add around 5k of text and data. Where as using DECLARE_EVENT_CLASS() and DEFINE_EVENT() will add 5k for the first installment (as TRACE_EVENT() really is just the two anyway) and then just around 200 bytes for every DEFINE_EVENT() after that. It also makes this cleaner (the events are quite complex). Thus do the following: DECLARE_EVENT_CLASS(mm_filemap_page_cache_template, TP_PROTO(struct page *page), TP_ARGS(page), TP_STRUCT__entry( __field(struct page *, page) __field(unsigned long, i_ino) __field(unsigned long, index) __field(dev_t, s_dev) ), TP_fast_assign( __entry->page = page; __entry->i_ino = page->mapping->host->i_ino; __entry->index = page->index; if (page->mapping->host->i_sb) __entry->s_dev = page->mapping->host->i_sb->s_dev; else __entry->s_dev = page->mapping->host->i_rdev; ), TP_printk("dev %d:%d ino %lx page=%p pfn=%lu ofs=%lu", MAJOR(__entry->s_dev), MINOR(__entry->s_dev), __entry->i_ino, __entry->page, page_to_pfn(__entry->page), __entry->index << PAGE_SHIFT) ); DEFINE_EVENT(mm_filemap_page_cache_template, mm_filemap_delete_from_page_cache, TP_PROTO(struct page *page), TP_ARGS(page)); DEFINE_EVENT(mm_filemap_page_cache_template, mm_filemap_add_to_page_cache, TP_PROTO(struct page *page), TP_ARGS(page)); -- Steve > + > +#endif /* _TRACE_FILEMAP_H */ > + > +/* This part must be outside protection */ > +#include <trace/define_trace.h> > diff -puN mm/filemap.c~mm-trace-filemap-add-and-del mm/filemap.c > --- a/mm/filemap.c~mm-trace-filemap-add-and-del > +++ a/mm/filemap.c > @@ -35,6 +35,9 @@ > #include <linux/cleancache.h> > #include "internal.h" > > +#define CREATE_TRACE_POINTS > +#include <trace/events/filemap.h> > + > /* > * FIXME: remove all knowledge of the buffer layer from the core VM > */ > @@ -113,6 +116,7 @@ void __delete_from_page_cache(struct pag > { > struct address_space *mapping = page->mapping; > > + trace_mm_filemap_delete_from_page_cache(page); > /* > * if we're uptodate, flush out into the cleancache, otherwise > * invalidate any existing cleancache entries. We can't leave > @@ -463,6 +467,7 @@ int add_to_page_cache_locked(struct page > if (likely(!error)) { > mapping->nrpages++; > __inc_zone_page_state(page, NR_FILE_PAGES); > + trace_mm_filemap_add_to_page_cache(page); > spin_unlock_irq(&mapping->tree_lock); > } else { > page->mapping = NULL; > _ > > Patches currently in -mm which might be from robert.jarzmik@...e.fr are > > linux-next.patch > mm-trace-filemap-add-and-del.patch -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@...r.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists