public inbox for libc-hacker@sourceware.org
 help / color / mirror / Atom feed
* [PATCH] Document M_PERTURB
@ 2010-07-26 14:37 Andreas Schwab
  2010-07-26 20:01 ` Roland McGrath
  0 siblings, 1 reply; 5+ messages in thread
From: Andreas Schwab @ 2010-07-26 14:37 UTC (permalink / raw)
  To: libc-hacker

2010-07-26  Andreas Schwab  <schwab@redhat.com>

	* manual/memory.texi (Malloc Tunable Parameters): Document
	M_PERTURB.

diff --git a/manual/memory.texi b/manual/memory.texi
index 59ea1ee..6ae408d 100644
--- a/manual/memory.texi
+++ b/manual/memory.texi
@@ -702,6 +702,11 @@ be allocated via @code{mmap}.
 @item M_MMAP_MAX
 The maximum number of chunks to allocate with @code{mmap}.  Setting this
 to zero disables all use of @code{mmap}.
+@item M_PERTURB
+If non-zero, memory blocks are filled with a value depending on
+this parameter when they are allocated (except when allocated by
+@code{calloc}) and freed.  This can be used to debug the use of
+uninitialized or freed heap memory.
 @end table
 
 @end deftypefun
-- 
1.7.1.1


-- 
Andreas Schwab, schwab@redhat.com
GPG Key fingerprint = D4E8 DBE3 3813 BB5D FA84  5EC7 45C6 250E 6F00 984E
"And now for something completely different."

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] Document M_PERTURB
  2010-07-26 14:37 [PATCH] Document M_PERTURB Andreas Schwab
@ 2010-07-26 20:01 ` Roland McGrath
  2010-07-26 20:17   ` Jakub Jelinek
  2010-07-27  9:13   ` Andreas Schwab
  0 siblings, 2 replies; 5+ messages in thread
From: Roland McGrath @ 2010-07-26 20:01 UTC (permalink / raw)
  To: Andreas Schwab; +Cc: libc-hacker

I would s/a value/values/ since there is one value for alloc and one value
for free, both based on the perturb byte.  I think it should also say
"... depending on the low-order byte of this value," so nobody can think
that passing in more than 8 bits has any effect.  With those tweaks, it's
fine to commit it.

Thanks,
Roland

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] Document M_PERTURB
  2010-07-26 20:01 ` Roland McGrath
@ 2010-07-26 20:17   ` Jakub Jelinek
  2010-07-27  9:13   ` Andreas Schwab
  1 sibling, 0 replies; 5+ messages in thread
From: Jakub Jelinek @ 2010-07-26 20:17 UTC (permalink / raw)
  To: Roland McGrath; +Cc: Andreas Schwab, libc-hacker

On Mon, Jul 26, 2010 at 01:01:11PM -0700, Roland McGrath wrote:
> I would s/a value/values/ since there is one value for alloc and one value
> for free, both based on the perturb byte.  I think it should also say
> "... depending on the low-order byte of this value," so nobody can think
> that passing in more than 8 bits has any effect.

It actually has.  MALLOC_PERTURB_=0 doesn't do anything, while
MALLOC_PERTURB_=256 does.

	Jakub

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] Document M_PERTURB
  2010-07-26 20:01 ` Roland McGrath
  2010-07-26 20:17   ` Jakub Jelinek
@ 2010-07-27  9:13   ` Andreas Schwab
  2010-07-27  9:32     ` Roland McGrath
  1 sibling, 1 reply; 5+ messages in thread
From: Andreas Schwab @ 2010-07-27  9:13 UTC (permalink / raw)
  To: Roland McGrath; +Cc: libc-hacker

Roland McGrath <roland@redhat.com> writes:

> I think it should also say "... depending on the low-order byte of
> this value," so nobody can think that passing in more than 8 bits has
> any effect.

Depends on how much of it is considered an implementation detail.

Andreas.

-- 
Andreas Schwab, schwab@redhat.com
GPG Key fingerprint = D4E8 DBE3 3813 BB5D FA84  5EC7 45C6 250E 6F00 984E
"And now for something completely different."

^ permalink raw reply	[flat|nested] 5+ messages in thread

* Re: [PATCH] Document M_PERTURB
  2010-07-27  9:13   ` Andreas Schwab
@ 2010-07-27  9:32     ` Roland McGrath
  0 siblings, 0 replies; 5+ messages in thread
From: Roland McGrath @ 2010-07-27  9:32 UTC (permalink / raw)
  To: Andreas Schwab; +Cc: libc-hacker

> Roland McGrath <roland@redhat.com> writes:
> 
> > I think it should also say "... depending on the low-order byte of
> > this value," so nobody can think that passing in more than 8 bits has
> > any effect.
> 
> Depends on how much of it is considered an implementation detail.

True.  Perhaps just say "some low-order bits of this parameter".

^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2010-07-27  9:32 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-07-26 14:37 [PATCH] Document M_PERTURB Andreas Schwab
2010-07-26 20:01 ` Roland McGrath
2010-07-26 20:17   ` Jakub Jelinek
2010-07-27  9:13   ` Andreas Schwab
2010-07-27  9:32     ` Roland McGrath

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).