what is the state about "[v2] ppc64 boot: Wait for boot cpu to show up if nr_cpus limit is about to hit"

3 messages, 3 authors, 2017-12-06 · open the first message on its own page

what is the state about "[v2] ppc64 boot: Wait for boot cpu to show up if nr_cpus limit is about to hit"

From: Liu ping fan <hidden>
Date: 2017-11-28 06:03:19

Hi,

I can not find the history about:
https://patchwork.ozlabs.org/patch/577193/


Can we have this patch?

Thanks,
Pingfan

Re: what is the state about "[v2] ppc64 boot: Wait for boot cpu to show up if nr_cpus limit is about to hit"

From: Michael Ellerman <mpe@ellerman.id.au>
Date: 2017-11-28 11:05:41

Liu ping fan [off-list ref] writes:
Hi,

I can not find the history about:
https://patchwork.ozlabs.org/patch/577193/


Can we have this patch?
I strongly dislike it.

Our CPU discovery code is already a big mess, in two separate places,
and this makes it worse.

In theory we have a split between logical and hardware CPU numbers, so
we should just be able to call the whatever CPU we boot on CPU 0. Why
does that not work? And should we fix that?

cheers

Re: what is the state about "[v2] ppc64 boot: Wait for boot cpu to show up if nr_cpus limit is about to hit"

From: Guilherme G. Piccoli <hidden>
Date: 2017-12-06 16:25:25


On 11/28/2017 04:03 AM, Liu ping fan wrote:
Hi,

I can not find the history about:
https://patchwork.ozlabs.org/patch/577193/


Can we have this patch?

Thanks,
Pingfan
Cross-referencing 2 threads on the same topic:
https://patchwork.ozlabs.org/patch/834860

Cascardo recently rebased this patch against latest kernel and tried to
get v3 merged, on the above linked thread.

Let's choose one of the threads to further discuss the solution to this
long-term bug!

Thanks,


Guilherme
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help