This is the mail archive of the
glibc-bugs@sourceware.org
mailing list for the glibc project.
[Bug libc/4937] RFE: micro-efficiency in symbol lookup for GNU_HASH chains
- From: "jreiser at BitWagon dot com" <sourceware-bugzilla at sourceware dot org>
- To: glibc-bugs at sources dot redhat dot com
- Date: 19 Aug 2007 17:00:07 -0000
- Subject: [Bug libc/4937] RFE: micro-efficiency in symbol lookup for GNU_HASH chains
- References: <20070819144917.4937.jreiser@BitWagon.com>
- Reply-to: sourceware-bugzilla at sourceware dot org
------- Additional Comments From jreiser at BitWagon dot com 2007-08-19 17:00 -------
The second re-writing is incorrect.
if (*hasharr == (new_hash | (1u & *hasharr)))
fails when (1u == (1u & new_hash)) and (0 == (1u & *hasharr)). This happens
when the hash of a non-last symbol has the low bit set. The proposed re-writing
will skip over the chain member even though the original code does not, and the
detailed comparison of the THEN might find the symbol.
--
http://sourceware.org/bugzilla/show_bug.cgi?id=4937
------- You are receiving this mail because: -------
You are on the CC list for the bug, or are watching someone who is.