linux-kernel - Re: [RFC][PATCH] proc: export more page flags in /proc/kpageflags

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
Hash Suite for Android: free password hash cracker in your pocket
[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Date:	Tue, 14 Apr 2009 15:22:31 +0800
From:	Wu Fengguang <fengguang.wu@...el.com>
To:	KOSAKI Motohiro <kosaki.motohiro@...fujitsu.com>
Cc:	Andrew Morton <akpm@...ux-foundation.org>,
	Andi Kleen <andi@...stfloor.org>,
	LKML <linux-kernel@...r.kernel.org>,
	"linux-mm@...ck.org" <linux-mm@...ck.org>
Subject: Re: [RFC][PATCH] proc: export more page flags in /proc/kpageflags

On Tue, Apr 14, 2009 at 02:54:40PM +0800, KOSAKI Motohiro wrote:
> Hi
> 
> > On Tue, Apr 14, 2009 at 12:37:10PM +0800, KOSAKI Motohiro wrote:
> > > > Export the following page flags in /proc/kpageflags,
> > > > just in case they will be useful to someone:
> > > > 
> > > > - PG_swapcache
> > > > - PG_swapbacked
> > > > - PG_mappedtodisk
> > > > - PG_reserved
> > > > - PG_private
> > > > - PG_private_2
> > > > - PG_owner_priv_1
> > > > 
> > > > - PG_head
> > > > - PG_tail
> > > > - PG_compound
> > > > 
> > > > - PG_unevictable
> > > > - PG_mlocked
> > > > 
> > > > - PG_poison
> > > 
> > > Sorry, NAK this.
> > > We shouldn't expose internal flags. please choice useful flags only.
> > 
> > OK. So are there anyone interested in any of these flags? Thanks!
> > 
> > My rational to export most page flags is that hopefully they could
> > help debugging kernel at some random situations..
> 
> I think,
> 
> > > > - PG_mappedtodisk
> > > > - PG_reserved
> > > > - PG_private
> > > > - PG_private_2
> > > > - PG_owner_priv_1
> > > > 
> > > > - PG_head
> > > > - PG_tail

> > > > - PG_unevictable
> > > > - PG_mlocked
 
How about including PG_unevictable/PG_mlocked?
They shall be meaningful to administrators.

> this 9 flags shouldn't exported.
> I can't imazine administrator use what purpose those flags.

> > > > - PG_swapcache
> > > > - PG_swapbacked
> > > > - PG_poison
> > > > - PG_compound
>
> I can agree this 4 flags.
> However pagemap lack's hugepage considering.
> if PG_compound exporting, we need more work.

You mean to fold PG_head/PG_tail into PG_COMPOUND?
Yes, that's a good simplification for end users.

> > 
> > > > Also add the following two pseudo page flags:
> > > > 
> > > > - PG_MMAP:   whether the page is memory mapped
> 
> hm, I can agree it.
> 
> 
> > > > - PG_NOPAGE: whether the page is present
> 
> PM_NOT_PRESENT isn't enough?

That would not be usable if you are going to do a system wide scan.
PG_NOPAGE could help differentiate the 'no page' case from 'no flags'
case.

However PG_NOPAGE is more about the last resort. The system wide scan
can be made much more efficient if we know the exact memory layouts.

Thanks,
Fengguang

> > > > 
> > > > This increases the total number of exported page flags to 25.
> > > > 
> > > > Cc: Andi Kleen <andi@...stfloor.org>
> > > > Cc: Matt Mackall <mpm@...enic.com>
> > > > Cc: Alexey Dobriyan <adobriyan@...il.com>
> > > > Signed-off-by: Wu Fengguang <fengguang.wu@...el.com>
> > > > ---
> > > >  fs/proc/page.c |  112 +++++++++++++++++++++++++++++++++--------------
> > > >  1 file changed, 81 insertions(+), 31 deletions(-)
> > > > 
> > > > --- mm.orig/fs/proc/page.c
> > > > +++ mm/fs/proc/page.c
> > > > @@ -68,20 +68,86 @@ static const struct file_operations proc
> > > >  
> > > >  /* These macros are used to decouple internal flags from exported ones */
> > > >  
> > > > -#define KPF_LOCKED     0
> > > > -#define KPF_ERROR      1
> > > > -#define KPF_REFERENCED 2
> > > > -#define KPF_UPTODATE   3
> > > > -#define KPF_DIRTY      4
> > > > -#define KPF_LRU        5
> > > > -#define KPF_ACTIVE     6
> > > > -#define KPF_SLAB       7
> > > > -#define KPF_WRITEBACK  8
> > > > -#define KPF_RECLAIM    9
> > > > -#define KPF_BUDDY     10
> > > > +enum {
> > > > +	KPF_LOCKED,		/*  0 */
> > > > +	KPF_ERROR,		/*  1 */
> > > > +	KPF_REFERENCED,		/*  2 */
> > > > +	KPF_UPTODATE,		/*  3 */
> > > > +	KPF_DIRTY,		/*  4 */
> > > > +	KPF_LRU,		/*  5 */
> > > > +	KPF_ACTIVE,		/*  6 */
> > > > +	KPF_SLAB,		/*  7 */
> > > > +	KPF_WRITEBACK,		/*  8 */
> > > > +	KPF_RECLAIM,		/*  9 */
> > > > +	KPF_BUDDY,		/* 10 */
> > > > +	KPF_MMAP,		/* 11 */
> > > > +	KPF_SWAPCACHE,		/* 12 */
> > > > +	KPF_SWAPBACKED,		/* 13 */
> > > > +	KPF_MAPPEDTODISK,	/* 14 */
> > > > +	KPF_RESERVED,		/* 15 */
> > > > +	KPF_PRIVATE,		/* 16 */
> > > > +	KPF_PRIVATE2,		/* 17 */
> > > > +	KPF_OWNER_PRIVATE,	/* 18 */
> > > > +	KPF_COMPOUND_HEAD,	/* 19 */
> > > > +	KPF_COMPOUND_TAIL,	/* 20 */
> > > > +	KPF_UNEVICTABLE,	/* 21 */
> > > > +	KPF_MLOCKED,		/* 22 */
> > > > +	KPF_POISON,		/* 23 */
> > > > +	KPF_NOPAGE,		/* 24 */
> > > > +	KPF_NUM
> > > > +};
> > > 
> > > this is userland export value. then enum is wrong idea.
> > > explicit name-number relationship is better. it prevent unintetional
> > > ABI break.
> > 
> > Right, that's the reason I add the /* number */ comments.
> > Anyway, it would be better to use explicit #defines.
> > 
> > Thanks,
> > Fengguang
> > 
> > --
> > 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/
> 
> 
--
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/