This is the mail archive of the binutils@sourceware.org mailing list for the binutils 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] |
On Thursday, December 09, 2010 12:55:28 Mike Frysinger wrote: > i noticed that a bunch of linker tests are under an "isnative" check when > the test invokes an execution of the output. but these are often a bunch > of build/link tests followed by an exec test which means we should be able > to do the build/link tests fine for cross-targets. > > along these lines, i'm looking at the output of the ld-elfvers tests. the > first one (vers1) fails because the test expects the output: > Version definitions: > [1-4] 0x01 0x0c96425f vers1.so > [1-4] 0x00 0x0a7927b1 VERS_1.1 > [1-4] 0x00 0x0a7927b2 VERS_1.2 > VERS_1.1 > [1-4] 0x00 0x0a7922b0 VERS_2.0 > VERS_1.2 > > but in my case, i'm seeing VERS_1.2 given an index of 0x02: > 3 0x02 0x0a7927b2 VERS_1.2 > > i have only a slight understanding of GNU ELF versioning internals, but i > dont think this is a problem. am i off base here, or should i update the > test to match the index field with like 0x0[0-4] ? nm, the issue is due to the tests mixing C visible and linker visible symbols in the C code. i'll send a patch to address the issue. -mike
Attachment:
signature.asc
Description: This is a digitally signed message part.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |