Re: [PATCH v3 0/4] optimize memory hotplug

[Pavel Tatashin]

Hi Ingo,

Thank you very much for your review. I will address spelling issues,
and will also try to split the patch #4.  Regarding runtime concern
for patch #3: the extra checking is only performed when the both of
the following CONFIGs are enabled:

CONFIG_DEBUG_VM=y
CONFIG_DEBUG_VM_PGFLAGS=y

I do not expect either of these to be ever enabled on a production systems.

Thank you,
Pavel

On Wed, Feb 14, 2018 at 3:09 AM, Ingo Molnar <mingo@xxxxxxxxxx> wrote:
>
> * Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> wrote:
>
>> On Tue, 13 Feb 2018 14:31:55 -0500 Pavel Tatashin <pasha.tatashin@xxxxxxxxxx> wrote:
>>
>> > This patchset:
>> > - Improves hotplug performance by eliminating a number of
>> > struct page traverses during memory hotplug.
>> >
>> > - Fixes some issues with hotplugging, where boundaries
>> > were not properly checked. And on x86 block size was not properly aligned
>> > with end of memory
>> >
>> > - Also, potentially improves boot performance by eliminating condition from
>> >   __init_single_page().
>> >
>> > - Adds robustness by verifying that that struct pages are correctly
>> >   poisoned when flags are accessed.
>>
>> I'm now attempting to get a 100% review rate on MM patches, which is
>> why I started adding my Reviewed-by: when I do that thing.
>>
>> I'm not familiar enough with this code to add my own Reviewed-by:, and
>> we'll need to figure out what to do in such cases.  I shall be sending
>> out periodic review-status summaries.
>>
>> If you're able to identify a suitable reviewer for this work and to
>> offer them beer, that would help.  Let's see what happens as the weeks
>> unfold.
>
> The largest patch, fix patch #2, looks good to me and fixes a real bug.
> Patch #1 and #3 also look good to me (assuming the runtime overhead
> added by patch #3 is OK to you):
>
>   Reviewed-by: Ingo Molnar <mingo@xxxxxxxxxx>
>
> (I suspect patch #1 and patch #2 should also get a Cc: stable.)
>
> Patch #4 is too large to review IMO: it should be split up into as many patches as
> practically possible. That will also help bisectability, should anything break.
>
> Before applying these patches please fix changelog and code comment spelling.
>
> But it's all good stuff AFAICS!
>
> Thanks,
>
>         Ingo
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@xxxxxxxxxx  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>