https://bugs.kde.org/show_bug.cgi?id=485148
--- Comment #10 from Petr <kobalicek.p...@gmail.com> --- BTW are you sure about the patch? Basically what SS operations do: DST[31..0] = result DST[127..32] = unchanged DST[vLen..128] = 0 So the high part of the YMM register should be zeroed, but the zeroing seems to be commented out. Similarly, SD operation would preserve the high part of XMM register, but would zero the high part of YMM register. -- You are receiving this mail because: You are watching all bug changes.
