sepavloff added inline comments.
================
Comment at: compiler-rt/test/builtins/Unit/divdf3_test.c:29-34
+ // qNaN / any = qNaN
+ if (test__divdf3(makeQNaN64(), 3., UINT64_C(0x7ff8000000000000)))
+ return 1;
+ // NaN / any = NaN
+ if (test__divdf3(makeNaN64(UINT64_C(0x123)), 3.,
UINT64_C(0x7ff8000000000000)))
+ return 1;
----------------
Expressions `any/nan' also could be checked.
================
Comment at: compiler-rt/test/builtins/Unit/divdf3_test.c:49-54
+ // Inf / Inf = NaN
+ if (test__divdf3(makeInf64(), makeInf64(), UINT64_C(0x7ff8000000000000)))
+ return 1;
+ // 0.0 / 0.0 = NaN
+ if (test__divdf3(+0x0.0p+0, +0x0.0p+0, UINT64_C(0x7ff8000000000000)))
+ return 1;
----------------
Expression `Inf/0' also could be checked.
================
Comment at: compiler-rt/test/builtins/Unit/divdf3_test.c:80
+ // divisor is 1.0 as UQ1.31
+ if (test__divdf3(0x1.0p+0, 0x1.00000001p+0, UINT64_C(0x3fefffffffe00000)))
return 1;
----------------
Is 0x1.00000001p+0 equal to 1.0 in UQ1.31?
================
Comment at: compiler-rt/test/builtins/Unit/divsf3_test.c:27
{
+ // Returned NaNs are assumed to be qNaN by default
+
----------------
Similar notes as for double precision checks.
================
Comment at: compiler-rt/test/builtins/Unit/divtf3_test.c:34
{
#if __LDBL_MANT_DIG__ == 113
+ // Returned NaNs are assumed to be qNaN by default
----------------
Similar notes as double precision checks.
Repository:
rG LLVM Github Monorepo
CHANGES SINCE LAST ACTION
https://reviews.llvm.org/D84932/new/
https://reviews.llvm.org/D84932
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
