Path ping...

Jan Hubicka hubicka@ucw.cz
Sun Nov 26 18:25:00 GMT 2006


Hi,
following patches seems to've got stuck in reviewing.  It would be great
if someone could take a look at them ;)

Random stuff:
http://gcc.gnu.org/ml/gcc-patches/2006-11/msg00741.html
   Remove NOTE_INSN_FUNCTION_END
http://gcc.gnu.org/ml/gcc-patches/2006-11/msg00270.html
   Limit stack usage growth caused by inliner
http://gcc.gnu.org/ml/gcc-patches/2006-07/msg01244.html
   Add statistics to bitmaps
   While Daniel, perhaps correctly, complained on usefulness of the
   code counting number of trips over the linked lists, there was few
   scenarios, where this patch has proved to be useful to figure out
   where the memory is going (especially after the fwprob merging).
   I can prepare stripped down patch not measuring the list walking
   overhead at all.

Memcpy stuff:
http://gcc.gnu.org/ml/gcc-patches/2006-11/msg01352.html
 Rewrite i386 string operation expansion
http://gcc.gnu.org/ml/gcc-patches/2006-10/msg01357.html
 More memcpy folding
http://gcc.gnu.org/ml/gcc-patches/2006-10/msg01371.html
 Pass info about hotness of instruction to RTL expanders
http://gcc.gnu.org/ml/gcc-patches/2006-10/msg01358.html
 Infrastructure for passing memcpy/memset profiles to backend

I would like to proceed with the histograms being discussed with Paolo,
the i386 backend stuff and symmetric folding changes to memset as I did
for memcpy.


Thanks,
Honza



More information about the Gcc-patches mailing list