On Fri, Jan 12, 2024 at 06:30:44PM +, Haiyang Zhang wrote:
>
>
> > -Original Message-
> > From: Michael Kelley
> > Sent: Friday, January 12, 2024 11:37 AM
> > To: Souradeep Chakrabarti
> > Cc: Yury Norov ; KY Srinivasan ;
> > Haiyang Zhang ; wei@kernel.org; Dexuan Cui
> > ; da..
From: Edgecombe, Rick P Sent: Friday, January 12,
2024 9:17 AM
>
> On Fri, 2024-01-12 at 15:07 +, Michael Kelley wrote:
> > The comment is Kirill Shutemov's suggestion based on comments in
> > an earlier version of the patch series. See [1]. The intent is to
> > prevent
> > some later rev
> -Original Message-
> From: Michael Kelley
> Sent: Friday, January 12, 2024 11:37 AM
> To: Souradeep Chakrabarti
> Cc: Yury Norov ; KY Srinivasan ;
> Haiyang Zhang ; wei@kernel.org; Dexuan Cui
> ; da...@davemloft.net; eduma...@google.com;
> k...@kernel.org; pab...@redhat.com; Long
On Fri, 2024-01-12 at 15:07 +, Michael Kelley wrote:
> The comment is Kirill Shutemov's suggestion based on comments in
> an earlier version of the patch series. See [1]. The intent is to
> prevent
> some later revision to slow_virt_to_phys() from adding a check for
> the
> present bit and b
From: Souradeep Chakrabarti Sent: Wednesday,
January 10, 2024 10:13 PM
>
> The test topology was used to check the performance between
> cpu_local_spread() and the new approach is :
> Case 1
> IRQ Nodes Cores CPUs
> 0 1 0 0-1
> 1 1 1 2-3
> 2 1 2
From: Markus Elfring Sent: Friday, January 12, 2024
12:06 AM
>
> …
> > Eliminate the duplication by making minor tweaks to the logic and
> > associated comments. While here, simplify the handling of memory
> > allocation errors, and use umin() instead of open coding it.
> …
>
> I got the impres
From: Edgecombe, Rick P Sent: Thursday, January
11, 2024 5:20 PM
>
> On Fri, 2024-01-05 at 10:30 -0800, mhkelle...@gmail.com wrote:
> > + * It is also used in callbacks for CoCo VM page transitions between
> > private
> > + * and shared because it works when the PRESENT bit is not set in the le
…
> Eliminate the duplication by making minor tweaks to the logic and
> associated comments. While here, simplify the handling of memory
> allocation errors, and use umin() instead of open coding it.
…
I got the impression that the adjustment for the mentioned macro
should be performed in a separa
