Beta binary

Thiago Macieira thiago at macieira.org
Thu Nov 28 12:48:23 UTC 2013


On quinta-feira, 28 de novembro de 2013 11:29:07, Linus Torvalds wrote:
> but I get a *lot* or warnings from valgrind deep in Qt. Most of them
> seem to be about 4-byte accesses that go past the end of a string,
> which is almost certainly the usual "somebody just optimized a string
> function to work a word at a time" issue. I think Thiago at some point
> said that SuSE had better annotations for some of these things.

Can you share what those were? I know strlen, strcmp, the mem versions are 
optimised in glibc to use word-at-a-time (or more, using SSE2) if the 
processor can do it. Also, GCC often does exactly that for the built-ins it 
inserts into the code.

Unfortunately, valgrind is at odds with the compiler. Any aligned load 
containing at least one valid byte cannot segfault, so it's valid to do that 
provided you don't act on any bytes past the end of your buffer.

-- 
Thiago Macieira - thiago (AT) macieira.info - thiago (AT) kde.org
   Software Architect - Intel Open Source Technology Center
      PGP/GPG: 0x6EF45358; fingerprint:
      E067 918B B660 DBD1 105C  966C 33F5 F005 6EF4 5358
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 190 bytes
Desc: This is a digitally signed message part.
URL: <http://lists.hohndel.org/pipermail/subsurface/attachments/20131128/19872e0e/attachment.sig>


More information about the subsurface mailing list