arsenm added inline comments.
================ Comment at: clang/lib/CodeGen/CGExpr.cpp:5602 + // source are correctly rounded. + SetFPAccuracy(Val, 2.5); + } ---------------- arsenm wrote: > yaxunl wrote: > > the spec says sqrt relative error is 3ULP > > https://registry.khronos.org/OpenCL/specs/2.2/html/OpenCL_C.html#relative-error-as-ulps > Did that change between versions? In any case I don’t want to change the > currently used threshold in this patch. We only need 1.0 anyway Oh, I see the threshold is 2.5 for fdiv and 3.0 for sqrt. CHANGES SINCE LAST ACTION https://reviews.llvm.org/D154495/new/ https://reviews.llvm.org/D154495 _______________________________________________ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
