This is the mail archive of the glibc-bugs@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug libc/11357] SIGBUS in getenv().


------- Additional Comments From pluto at agmk dot net  2010-03-08 13:35 -------
(In reply to comment #1)
> It hurts, don't do it then.

i know it hurts :) but i prefer to catch such things on local
x86 workstation than on remote sparc system or arm netbooks.

> E.g. GCC assumes that on x86_64 and i?86 it can
> dereference an unaligned pointer (except for vector insns),
> albeit sometimes slower than an aligned pointer.

gcc assumptions about (un)aligned access on x86 are in fact
a problem but afaics it currently uses quite optimal and safe
mov{b,w,l,q} combinations. is it worth to consider this as
an enhancement on gcc mailinglist?

> Thus there can be millions of other places where you'd get a SIGBUS.

i'm not interesting in 'other' broken software.
i would like to have only a working C library for my tuned process
with customized eflags. glibc works on sparc-sigbus-linux :) so why
it can't work on x86-sigbus-linux? afaics glibc already uses such
policy switches like e.g. _STRING_ARCH_unaligned macro. can i safely
#undef this macro (and maybe others) on x86?

-- 


http://sourceware.org/bugzilla/show_bug.cgi?id=11357

------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]