Unanalyzed PRs possibly relevant for 3.3

Unanalyzed PRs possibly relevant for 3.3

[Wolfgang Bangerth]

Our view of how ready we are for 3.3 relies mostly on analyzed PRs in 
GNATS. However, there are many reports that don't get analyzed since those 
who usually do that don't have access to these platforms or are lacking 
knowledge in certain areas. Below is a list of PRs that by briefly looking 
at I thought might be relevant to 3.3, but are not analyzed. Can I ask the 
relevant port maintainers to at least check these PRs? Whether they need 
to be fixed is another matter, but it would be good to know at least where 
we are.

Mark, can you if necessary make sure that at least some of them are 
getting reviewed before we go to 3.3?

There are also a number of PRs in the target class, which I haven't 
reviewed for the list below. Most of them now have a prefix in their 
synopsis indicating the name of the target, to make it easier for port 
maintainers to identify what is relevant to them.

Finally, there are also significant problems in the area of debug
information, but I cannot review these PRs due to lack of knowledge of the
field. From my own often very frustrating experiences in this area, I
would very much appreciate people looking at these reports.

Apart from that, I see great progress recently in fixing bugs, and I'd 
like to publicly thanks those doing that! Great work!

W.

----------------------------------------------------------------------

x86:
  optimization/9798: miscompilation of SPEC's 176.gcc benchmark

ia-64:
  c/10201, c++/10202: ICE in int_mode_for_mode
  optimization/10116: problem in ifcvt.c (with patch)
  optimization/7823: illegal instruction
  optimization/6686: incorrect optimization
  c++/6685: wrong setting of REG_POINTER

mips:
  optimization/9301: ICE in try_forward_edges (likely a regression)
  c++/7519: crash with large structures

powerpc-aix:
  optimization/7011: ICE in elimination_effects

powerpc-macosx:
  c/10039: infinite loop in gcc

powerpc-linux:
  c++/9377: miscompilation in 64bit mode

hp-ux:
  optimization/7260: ICE with -O3
  c++/8512: bootstrap error on hppa64-hpux

m68k:
  target/10021: loop optimization (regression)
  middle-end/7594: ICE in simplify_rtx

parisc-linux:
  optimization/10130: ICE (likely a regression)

arc-elf32
  optimization/6324: wrong code for simple function

sh:
  optimization/8867: wrong code

windows/mingw/cygwin:
  c/10148: ICE
  c/9738: ICE in rest_of_compilation
  c/9237: ICE with dllimport
  c/8859: Segfault
  c++/8378: ICE
  c/7910: Segfault
  c++/6670: ICE with dllimport

-------------------------------------------------------------------------
Wolfgang Bangerth             email:            bangerth@ticam.utexas.edu
                              www: http://www.ticam.utexas.edu/~bangerth/