On Sun, Aug 23, 2020 at 11:29 AM Linus Torvalds <[email protected]> wrote: > > On Sun, Aug 23, 2020 at 1:26 AM Thomas Gleixner <[email protected]> wrote: > > > > Remove the RDPID optimization, which is not even > > backed by numbers from the paranoid entry path instead. > > Ugh, that's sad. I'd expect the LSL to be quite a bit slower than the > RDPID on raw hardware, since LSL has to go out to the GDT. > > And I don't think we need the GDT for anything else normally, so it's > not even going to be cached.
Every interrupt is going to load the CS and SS descriptor cache lines. Every IRET to user mode will get the user CS cache line. Because x86 is optimized to be as convoluted as possible and to have as much garbage in microcode as possible! --Andy
