------- Additional Comments From ebotcazou at gcc dot gnu dot org 2004-10-14 18:57
-------
Quick summary of my investigation so far:
- ICE on g-awk.adb at -O0: pb in the RTL expander relating to ADDR_EXPR
- ICE on g-socket.adb at -O0: consistency check, again ADDR_EXPR
- infinite loop on make.adb at -O2: PHI nodes sharing pb (ok at -O1)
After quick fix for both ICE + workaround for infinite loop:
=== acats tests ===
FAIL: c330001
FAIL: c380004
FAIL: c391002
FAIL: c43214c
FAIL: c45273a
FAIL: c45347a
FAIL: c45347b
FAIL: c45347c
FAIL: c74004a
FAIL: c940006
FAIL: c951001
FAIL: c954023
FAIL: c97114a
FAIL: c97117a
FAIL: ca11005
FAIL: cc1221d
FAIL: cd10002
FAIL: cd2a21e
FAIL: cd2a22i
FAIL: cd2a31a
FAIL: cxaa010
FAIL: cxaa017
FAIL: cxb4004
FAIL: cxh3002
=== acats Summary ===
# of expected passes 2298
# of unexpected failures 24
--
http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17746
