On 09/02/2023 13:48, David Woodhouse wrote:
On Thu, 2023-02-09 at 12:10 +0000, David Woodhouse wrote:
On Thu, 2023-02-09 at 12:53 +0100, Thomas Gleixner wrote:
On Thu, Feb 09 2023 at 11:03, David Woodhouse wrote:
This one also fixes it for me. If we're happy with this approach, I'll
work it into Thomas's original patch (and hopefully eventually he'll be
happy enough with it and the commit message that he'll give us his
Signed-off-by for it.)
I'm happy enough by now, but I'm not sure how much of the original patch
is still left. Also you did the heavy lifting of making it work and
writing the nice changelog. So please make this:
From: David Woodhouse <dwmw2@xxxxxxxxxxxxx>
Co-developed-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Signed-off-by: David Woodhouse <dwmw2@xxxxxxxxxxxxx>
Thanks. I'll flip that to the Amazon address, of course. It's useless
for actual email (until I apply that LART some more) but I should still
use it for that.
I think I'm going to make one more change to that as I review it; in
the "should never happen" case of not finding the APIC ID in the
cpuid_to_apicid[] array it would just keep searching for ever. I don't
know if there's a better thing to do other than just dropping the
trampoline lock and 1:cli;hlt;jmp 1b but at least it's *attempting* to
handle the failure.
Patch below, and I'll update the tree shortly. There's a "what if
there's noise in the top 32 bits of %rcx" fix in there too.
All done and pushed out to parallel-6.2-rc7-part1 (and not -part1)
branches. Usama, are you able to redo the testing and take it from
here? Thanks for that; it's saving me a lot of time!
Thanks! Will retest and repost now!
Usama
I'm mostly done for the week now as by this time tomorrow, I need to
have the skis on the roof of the car and be ready to pick the family up
from school and start driving south...
