https://sourceware.org/bugzilla/show_bug.cgi?id=29998
--- Comment #18 from Jan Janssen <medhefgo at web dot de> --- Compilation works now and the binary will run, but I get a miscompilation(?) wrt popcount builtin. Both cases are based on https://github.com/systemd/systemd/pull/26097, where I originally encountered them, btw. $ cat test1.c void test1() {} $ cat test2.c void test1(); static void test2() { test1(); } $ cat test3.c int run(void) { volatile unsigned i = 4278190080; return __builtin_popcount(i); } $ rm -f test.exe libtest.a test[123].o $ x86_64-w64-mingw32-gcc -c -flto test1.c test2.c test3.c $ x86_64-w64-mingw32-gcc-ar csrDT libtest.a test1.o $ x86_64-w64-mingw32-gcc -o test.exe -flto -nostdlib test2.o test3.o libtest.a -lgcc When I run the above build steps (with your path applied), they produce the same assembly for run(), except for this line: 14000102b: e8 00 00 00 00 call 140001030 <run+0x1c> vs this without -flto: 140001037: e8 24 00 00 00 call 140001060 <__popcountdi2> The lto build doesn't even include popcount and the EFI binary either produces a page fault or division error. -- You are receiving this mail because: You are on the CC list for the bug.
