http://gcc.gnu.org/bugzilla/show_bug.cgi?id=59062
Bug ID: 59062 Summary: poor code generated at -Os (affecting all gcc versions): 10+ sec vs. 1 sec at -O0 Product: gcc Version: 4.9.0 Status: UNCONFIRMED Severity: normal Priority: P3 Component: tree-optimization Assignee: unassigned at gcc dot gnu.org Reporter: su at cs dot ucdavis.edu The current gcc trunk (as well as 4.6, 4.7, and 4.8) produces poor code for the following testcase on x86_64-linux-gnu at -Os in both 32-bit and 64-bit modes. The produced code has the same or very similar size. $ gcc-trunk -v Using built-in specs. COLLECT_GCC=gcc-trunk COLLECT_LTO_WRAPPER=/usr/local/gcc-trunk/libexec/gcc/x86_64-unknown-linux-gnu/4.9.0/lto-wrapper Target: x86_64-unknown-linux-gnu Configured with: ../gcc-trunk/configure --enable-languages=c,c++,objc,obj-c++,fortran,lto --disable-werror --enable-checking=release --with-gmp=/usr/local/gcc-trunk --with-mpfr=/usr/local/gcc-trunk --with-mpc=/usr/local/gcc-trunk --with-cloog=/usr/local/gcc-trunk --prefix=/usr/local/gcc-trunk Thread model: posix gcc version 4.9.0 20131109 (experimental) [trunk revision 204611] (GCC) $ $ gcc-trunk -O0 small.c $ time a.out 1.02user 0.00system 0:01.05elapsed 97%CPU (0avgtext+0avgdata 1584maxresident)k 0inputs+0outputs (0major+142minor)pagefaults 0swaps $ gcc-trunk -O1 small.c $ time a.out 0.68user 0.00system 0:00.68elapsed 98%CPU (0avgtext+0avgdata 1568maxresident)k 0inputs+0outputs (0major+140minor)pagefaults 0swaps $ gcc-trunk -Os small.c $ time a.out 12.84user 0.00system 0:12.86elapsed 99%CPU (0avgtext+0avgdata 1568maxresident)k 0inputs+0outputs (0major+140minor)pagefaults 0swaps $ -------------------------------- #pragma pack(1) struct { int f0:23; unsigned int f1:23; unsigned int f2:7; } b, c; unsigned int a; int main () { for (; a != 14; a += 9) c = b; return 0; }