Am Dienstag, 21. Dezember 2021, 19:21:22 CET schrieb John David Anglin: > Correct completer in lws start. > > The completer in the "or,ev %r1,%r30,%r30" instruction is reversed, so we > are not clipping the LWS number when we are called from a 32-bit process > (W=0). We need to nulify the depdi instruction when the least-significant > bit of %r30 is 1. I'm curious: what effect has this bug? Since this is syscall code I guess it can somehow be exposed from userspace, but how? Maybe some sort of explanation like this can even be added to the commit message? Eike
Attachment:
signature.asc
Description: This is a digitally signed message part.
