diff options
| -rw-r--r-- | libraries/ghc-prim/cbits/ctz.c | 18 |
1 files changed, 17 insertions, 1 deletions
diff --git a/libraries/ghc-prim/cbits/ctz.c b/libraries/ghc-prim/cbits/ctz.c index cc420b9acd..f68f628bd3 100644 --- a/libraries/ghc-prim/cbits/ctz.c +++ b/libraries/ghc-prim/cbits/ctz.c @@ -31,7 +31,23 @@ hs_ctz32(StgWord x) StgWord hs_ctz64(StgWord64 x) { -#if SIZEOF_UNSIGNED_LONG == 8 +#if defined(__GNUC__) && defined(i386_HOST_ARCH) + /* On Linux/i386, the 64bit `__builtin_ctzll()` instrinsic doesn't + get inlined by GCC but rather a short `__ctzdi2` runtime function + is inserted when needed into compiled object files. + + This workaround forces GCC on 32bit x86 to to express `hs_ctz64` in + terms of the 32bit `__builtin_ctz()` (this is no loss, as there's no + 64bit BSF instruction on i686 anyway) and thus avoid the problematic + out-of-line runtime function. + */ + + if (!x) return 64; + + return ((uint32_t)x ? __builtin_ctz((uint32_t)x) + : (__builtin_ctz(x >> 32) + 32)); + +#elif SIZEOF_UNSIGNED_LONG == 8 return x ? __builtin_ctzl(x) : 64; #elif SIZEOF_UNSIGNED_LONG_LONG == 8 return x ? __builtin_ctzll(x) : 64; |