On Mon, 13 Nov 2023, Nicholas Piggin wrote:
On Wed Nov 1, 2023 at 8:44 PM AEST, BALATON Zoltan wrote:
On Tue, 24 Oct 2023, BALATON Zoltan wrote:
These are some small clean ups for target/ppc/excp_helper.c trying to
make this code a bit simpler. No functional change is intended. This
series was submitted before but only partially merged due to freeze
and conflicting series os thia was postponed then to avoid conflicts.
Ping?
May just leave this for next release, sorry.
No problem, these aren't that important. I've tried to optimise exception
handling a bit as these seem to happen very often with some guests.
Especially sc is common so that's why I tried to avoid moving nip around
unnecessarily. But these did not have a great impact (unlike the ppc440
tlbwe patches) so they are just clean ups now to alow further profiling
and optimisation later. It can wait until the next devel window.
Regards,
BALATON Zoltan
I still didn't like the change to logging -- that's not intended to
print some machine implementation detail, but the address of the
instruction that caused the syscall/hcall. That could be changed
easily enough.
But I am also now in two minds about the change to nip too.
Synchronous interrupt is today handled here with nip at the address
of the instruction that caused it. That's *also* a nice invaraint to
have.
Other patches seem okay.
Thanks,
Nick
