gcc

[Thorsten Glaser <tg@mirbsd.de>]

| gcc-4.4 (4.4.5-5) unstable; urgency=medium
| .
|   * Update to SVN 20101021 from the gcc-4_4-branch (r165755, 4.4.5 release
|     Fix PR fortran/42169, PR middle-end/46019 (wrong code), PR fortran/33595.
| .
|   [ Thorsten Glaser ]
|   * Add preliminary m68k support. Closes: #600835.
|   * debian/patches/gcc-multiarch.diff: Add m68k support.
|   * debian/rules2: Disable multi{arch,lib} on m68k for now.

Now we’ve got to discuss which multiarch targets to build
eventually. The m68k targets are:
  { "m68040", "m68040-linux-gnu" },
  { "m68060", "m68060-linux-gnu" },
  { "mcpu32", "m68360-linux-gnu" },
  { "mfidoa", "fido-linux-gnu" },

I don’t know enough about them, and last time I tried, mfidoa
failed (but that may have been fixed since in libc headers).

|   * debian/patches/pr41302.diff: new, backported from trunk.
|   * debian/patches/pr43804.diff: new, taken from GCC BZ.

The latter must be ACK’d, and both should pass the testsuite
some time – but I’d like to get eglibc built first.

|   * debian/rules.patch: Add pr41302, pr43804 on m68k.
| .
|   [ Matthias Klose ]
|   * ARM: Fix Thumb-1 reload ICE with nested functions (Julian Brown),
|     taken from the trunk.
|   * Fix earlyclobbers on some arm.md DImode shifts (may miscompile "x >> 1"),
|     taken from the trunk. Closes: #600888.

bye,
//mirabilos
-- 
Solange man keine schmutzigen Tricks macht, und ich meine *wirklich*
schmutzige Tricks, wie bei einer doppelt verketteten Liste beide
Pointer XORen und in nur einem Word speichern, funktioniert Boehm ganz
hervorragend.		-- Andreas Bogk über boehm-gc in d.a.s.r