On Wed, Nov 05, 2008 at 12:28:53PM +0000, Paul LeoNerd Evans wrote: > On Tue, Nov 04, 2008 at 03:53:57PM -0500, Jan Harkes wrote: > > Yes, the libc version of makecontext is (was?) broken on amd64 and it > > was corrupting pointers passed as arguments. I think the high bits of > > the pointer were either cleared or filled with random data, not sure > > anymore. > > Actually, I've just read makecontext(3): > > When this context is later activated (using setcontext(2) or swapconâ > text()) the function func is called, and passed the series of integer > (int) arguments that follow argc; the caller must specify the number of > these arguments in argc. When this function returns, the successor > context is activated. If the successor context pointer is NULL, the > thread exits.
While Linux / glibc's manpage isn't too clear on the subject, the
following comment comes from NetBSD's version:
The standard does not clearly define the type of integer arguments passed
to func via makecontext(); portable applications should not rely on the
implementation detail that it may be possible to pass pointer arguments
to functions. This may be clarified in a future revision of the stan-
dard.
http://www.daemon-systems.org/man/makecontext.3.html
> That sounds to me like it won't guarantee to pass a pointer around
> anyway, only an int.
--
Paul "LeoNerd" Evans
[EMAIL PROTECTED]
ICQ# 4135350 | Registered Linux# 179460
http://www.leonerd.org.uk/
signature.asc
Description: Digital signature
