Bug 3946 - Weak definitions aren't so weak
Summary: Weak definitions aren't so weak
Status: RESOLVED INVALID
Alias: None
Product: glibc
Classification: Unclassified
Component: libc (show other bugs)
Version: 2.3.6
: P2 normal
Target Milestone: ---
Assignee: Ulrich Drepper
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2007-01-30 23:26 UTC by Samuel Thibault
Modified: 2018-04-20 14:02 UTC (History)
1 user (show)

See Also:
Host:
Target:
Build:
Last reconfirmed:
fweimer: security-

Attachments
weak definitions testcase (466 bytes, application/octet-stream)
2007-01-30 23:27 UTC, Samuel Thibault 		Details
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Samuel Thibault 2007-01-30 23:26:46 UTC 
Attached is a testcase for weak references: two libraries define the safe symbol f, one as a weak definition, the other as a strong definition, and a program calls it. When run with ld-linux 2.3.6 and 2.5, the link order matters, while with ld-linux 2.2.5, the strong definition always wins.

I would have expected to always get the 2.2.5's "strong always wins" behavior (and that's what I can read when googling), is there a reason for 2.3.6 and above to behave differently? (I can't see what weak definitions can be used for in such case).
Comment 1 Samuel Thibault 2007-01-30 23:27:14 UTC 
Created attachment 1527 [details]
weak definitions testcase
Comment 2 Ulrich Drepper 2007-01-31 01:06:15 UTC 
Weak definition only and exclusively have a meaning for static archives.  Old
glibc versions mistakenly implemented it for dynamic linking, too.  This is now
not the default anymore.   Broken code an use LD_DYNAMIC_WEAK.