LLVM Developers Have Been Reviewing Work To Offset The Performance Hit From Intel JCC

[phoronix]

Phoronix: LLVM Developers Have Been Reviewing Work To Offset The Performance Hit From Intel JCC

With the Jump Conditional Code (JCC) Erratum that was made public earlier this month and ushered in new Intel microcode to mitigate this Skylake to Cascade Lake design defect, compiler/toolchain patches have been in the works to help offset the performance cost incurred from the updated microcode. Besides the GNU Assembler work we've talked about several times since JCC came to light, the LLVM folks have also been reviewing their comparable changes...

http://www.phoronix.com/scan.php?page=news_item&px=LLVM-Review-JCC-Offset

[Xicronic]

If the work would decrease performance of other hardware, even by 1%... they should seek other means or just not do anything. I think the last thing anyone wants is patches for a few Intel generations draining performance from literally every other CPU.

[Weasel]

Nobody gives a shit about the crappy affected CPUs. Stop this insanity at once. Stop wasting man hours on this crap. STOP.

The microcode already "fixes" the bug so it's not like it will be disastrous, just slower for affected CPUs only. And since the CPUs are already FLAWED then they deserved the performance penalty and leave everyone else alone.

Why should everyone else suffer?!? It's like in school when 1 guy fucks something up and then the entire class gets punished for it. What the fuck?

[numacross]

Nobody gives a shit about the crappy affected CPUs. Stop this insanity at once. Stop wasting man hours on this crap. STOP.

The microcode already "fixes" the bug so it's not like it will be disastrous, just slower for affected CPUs only. And since the CPUs are already FLAWED then they deserved the performance penalty and leave everyone else alone.

Why should everyone else suffer?!? It's like in school when 1 guy fucks something up and then the entire class gets punished for it. What the fuck?

It's KPTI all over again, initially the Intel patches enabled that for all x86.

Can't have AMD get an unfair advantage now can we?