Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Date:   Thu, 6 Feb 2020 12:39:24 +0800
From:   Baoquan He <bhe@redhat.com>
To:     Wei Yang <richard.weiyang@gmail.com>
Cc:     Wei Yang <richardw.yang@linux.intel.com>,
        David Hildenbrand <david@redhat.com>,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org,
        Segher Boessenkool <segher@kernel.crashing.org>,
        Andrew Morton <akpm@linux-foundation.org>,
        Michal Hocko <mhocko@kernel.org>,
        Oscar Salvador <osalvador@suse.de>
Subject: Re: [PATCH v1] mm/memory_hotplug: Easier calculation to get pages to
 next section boundary
Message-ID: <20200206043924.GM8965@MiWiFi-R3L-srv>
References: <20200205135251.37488-1-david@redhat.com>
 <20200205231945.GB28446@richard>
 <20200205235007.GA28870@richard>
 <20200206001317.GH8965@MiWiFi-R3L-srv>
 <20200206003736.GI8965@MiWiFi-R3L-srv>
 <20200206022644.6u7pxf7by2w5trmi@master>
 <20200206024816.GK8965@MiWiFi-R3L-srv>
 <20200206043401.22i2cucwqctsrtps@master>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20200206043401.22i2cucwqctsrtps@master>
User-Agent: Mutt/1.10.1 (2018-07-13)
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

On 02/06/20 at 04:34am, Wei Yang wrote:
> On Thu, Feb 06, 2020 at 10:48:16AM +0800, Baoquan He wrote:
> >On 02/06/20 at 02:26am, Wei Yang wrote:
> >> On Thu, Feb 06, 2020 at 08:37:36AM +0800, Baoquan He wrote:
> >> >On 02/06/20 at 08:13am, Baoquan He wrote:
> >> >> On 02/06/20 at 07:50am, Wei Yang wrote:
> >> >> > On Thu, Feb 06, 2020 at 07:19:45AM +0800, Wei Yang wrote:
> >> >> > >On Wed, Feb 05, 2020 at 02:52:51PM +0100, David Hildenbrand wrote:
> >> >> > >>Let's use a calculation that's easier to understand and calculates the
> >> >> > >>same result. Reusing existing macros makes this look nicer.
> >> >> > >>
> >> >> > >>We always want to have the number of pages (> 0) to the next section
> >> >> > >>boundary, starting from the current pfn.
> >> >> > >>
> >> >> > >>Suggested-by: Segher Boessenkool <segher@kernel.crashing.org>
> >> >> > >>Cc: Andrew Morton <akpm@linux-foundation.org>
> >> >> > >>Cc: Michal Hocko <mhocko@kernel.org>
> >> >> > >>Cc: Oscar Salvador <osalvador@suse.de>
> >> >> > >>Cc: Baoquan He <bhe@redhat.com>
> >> >> > >>Cc: Wei Yang <richardw.yang@linux.intel.com>
> >> >> > >>Signed-off-by: David Hildenbrand <david@redhat.com>
> >> >> > >
> >> >> > >Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
> >> >> > >
> >> >> > >BTW, I got one question about hotplug size requirement.
> >> >> > >
> >> >> > >I thought the hotplug range should be section size aligned, while taking a
> >> >> > >look into current code function check_hotplug_memory_range() guard the range.
> >> >> 
> >> >> A good question. The current code should be block size aligned. I
> >> >> remember in some places we assume each block comprise all the sections.
> >> >> Can't imagine one or some of them are half section filled.
> >> >
> >> >I could be wrong, half filled block may not cause problem. 
> >> >
> >> 
> >> David must be angry about our flooding the mail list :-)
> >
> >Believe he won't, :-) If you like, we can talk off line.
> >
> >> 
> >> Check the code again, there are two memory range check:
> >> 
> >>   * check_hotplug_memory_range(), block/section aligned
> >>   * check_pfn_span(), subsection aligned
> >> 
> >> The second check, check_pfn_span() in __add_pages(), enable the capability to
> >> add a memory range with subsection size.
> >> 
> >> This means hotplug still keeps section alignment.
> >
> >memremap_pages() also call add_pages(), it doesn't have the
> >check_hotplug_memory_range() invocation. check_pfn_span() is made for
> >it specifically.
> >
> 
> If my understanding is correct, memremap_pages() is used to add some dev
> memory to system. This is the use case which Dan want to enable for
> sub-section. Since memremap_pages() is not called in mem-hotplug path, this
> doesn't affect the hotplug range alignment.

Yeah, we are on the same page.

> 
> >> 
> >> BTW, __add_pages() share the same logic as __remove_pages(). Why not change it
> >> too? Do I miss something or I don't have the latest source code?
> >
> >Good question, and I think it need. Just David is refactoring/cleaning
> >up the remove_pages() code path, this is found out by Segher from patch
> >reviewing.
> 
> Ah, we may need a following cleanup :-)

Agree. See what David will say. Fold it into this patch, or anyone post
a new one.