--
What|Removed |Added
OtherBugsDependingO||6915
nThis||
http://sourceware.org/bugzilla/show_bug.cgi?id=6412
--
--- Additional Comments From hjl dot tools at gmail dot com 2008-10-12
01:48 ---
The Linux binutils has a patch for PR 6412. Please try this patch:
--- binutils/readelf.c.solaris 2008-10-11 18:47:31.0 -0700
+++ binutils/readelf.c 2008-10-11 18:46:18.0 -0700
@@ -4452,7
--- Additional Comments From hjl dot tools at gmail dot com 2008-10-11
23:47 ---
Arch may be changed by the .arch directive. However, when we
generate NOPs, we lost the .arch information. We need to know
the arch for NOP, which may not be necessarily the same as the
one specified by -mar
--- Additional Comments From schwab at suse dot de 2008-10-11 14:40 ---
In MRI mode, 1b is a number in base 2. MRI mode is only for compatibility with
existing MRI assembler code, don't use it.
--
What|Removed |Added
-
--- Additional Comments From swanson at ukfsn dot org 2008-10-11 14:28
---
Created an attachment (id=2995)
--> (http://sourceware.org/bugzilla/attachment.cgi?id=2995&action=view)
binutils-cvs-fix-i386-nop.patch
--
http://sourceware.org/bugzilla/show_bug.cgi?id=6957
--- You are
As recently discussed on the Linux kernel mailing list, in gas the semantics for
choosing the i386 NOP sequence are incorrect and will cause unexpected failures.
They should chosen depending on march where new instructions can be used and not
mtune where new instructions must not be used.
http://t
