I have built a 4.4.0 version of GCC and built newlib with that version and installed it in say /MYPREFIX and then go and try to build 4.5.0. I then build GCC with the same /MYPREFIX but not a combined build, the testsuite is not able to find the libc headers. I don't know if this a regression because this is the first time I hit it and I just happened to build 4.5 and was getting many failures this way. An example of a failure is: Executing on host: /home/apinski/src/local/gcc/objdir-spu/gcc/xgcc -B/home/apinski/src/local/gcc/objdir-spu/gcc/ -O0 -w -c -o 990625-1.o /home/apinski/src/local/gcc/gcc/testsuite/gcc.c-torture/compile/990625-1.c (timeout = 300) /home/apinski/src/local/gcc/gcc/testsuite/gcc.c-torture/compile/990625-1.c:2: fatal error: string.h: No such file or directory^M compilation terminated.^Mcompiler exited with status 1 output is: /home/apinski/src/local/gcc/gcc/testsuite/gcc.c-torture/compile/990625-1.c:2: fatal error: string.h: No such file or directory^Mcompilation terminated.^M FAIL: gcc.c-torture/compile/990625-1.c -O0 (test for excess errors)Excess errors:
This bug has been there for a long time, maybe too old to call it a regression (I believe older than the currently open branches). Look for changes in handling of GCC_EXEC_PREFIX internally (not just the test-suite) and for "relocated" toolchains.