Re: [Patch] Another fix for cycle detection

[Jeff Law <law at redhat dot com>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On 01/23/12 15:03, Roland McGrath wrote:
>> I've already seen it overflow 8 bit signed, seems silly to flip
>> to an unsigned which doesn't give us that much headroom.  But if
>> that's what you'd prefer, I'll do it.
> 
> If it has exceeded 127 in real-world cases, then going to 65535
> seems reasonable.  But any real-world case having exceeded the
> limit indicates that explicit overflow detection is warranted.
It certainly has exceeded 127 in real world cases :(  A quick survey
of /usr/bin/* on my F16 box shows dozens of executables that have
already exceeded 127 libraries.  Right now the largest is digikam at
182 DSO dependencies according to ldd.

So, pending a resolution on the prior patch, ISTM the way to go is 64k
+ overflow checking.



> 
>> In case it's not obvious, this patch is pointless without last
>> week's fix to cycle detection:
>> 
>> http://sources.redhat.com/ml/libc-alpha/2012-01/msg00072.html
> 
> That's a sufficiently more complex and subtle change that it merits
> more careful review than this one needed, and I haven't looked at
> it in detail.
Understood.

jeff

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJPHjFnAAoJEBRtltQi2kC78sIIAIg2AjuB2j38g73snS2U0uSX
JGeVacNpkLWOvn6fDPRRMMqO0U9c0vxq+0E91VQbdthIsIYFitoo4t3fSk5KZ+Un
GILch5aINlPRUQMFUidoHJNF62er2/43sjdbVr87MmlIWedxkIbcetMVH6y2LUdo
FoViytyM0JmP/z0wqIrCOgaw0vSJZrjIhc+wruLWLAVobhAxGbTbTsedd1V/lC/m
I8meU2YNG2uqOE/x5Slc/VcemBOu+CdeFNEmPNkNJNgkSk5FPDkByagekTQ4bARl
+efdyF2o/SRzQQJeiJBOVra8afl88OXMy3YRDfKavbXOC3y1f9gfMdkNCy4gqMA=
=02Tl
-----END PGP SIGNATURE-----