Received-SPF: pass (google.com: domain of linux-kernel+bounces-31212-linux.lists.archive=gmail.com@vger.kernel.org designates 139.178.88.99 as permitted sender) client-ip=139.178.88.99;
From: Andi Kleen <ak@linux.intel.com>
To: Lance Yang <ioworker0@gmail.com>
Cc: akpm@linux-foundation.org,  zokeefe@google.com,
  songmuchun@bytedance.com,  linux-kernel@vger.kernel.org
Subject: Re: [PATCH v1 1/2] mm/madvise: introduce MADV_TRY_COLLAPSE for
 attempted synchronous hugepage collapse
In-Reply-To: <20240117050217.43610-1-ioworker0@gmail.com> (Lance Yang's
	message of "Wed, 17 Jan 2024 13:02:16 +0800")
References: <20240117050217.43610-1-ioworker0@gmail.com>
Date: Fri, 19 Jan 2024 05:41:08 -0800
Message-ID: <87ededl9u3.fsf@linux.intel.com>
User-Agent: Gnus/5.13 (Gnus v5.13)
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain

Lance Yang <ioworker0@gmail.com> writes:

> This idea was inspired by MADV_COLLAPSE introduced by Zach O'Keefe[1].
>
> Introduce a new madvise mode, MADV_TRY_COLLAPSE, that allows users to
> make a least-effort attempt at a synchronous collapse of memory at
> their own expense.
>
> The only difference from MADV_COLLAPSE is that the new hugepage allocation
> avoids direct reclaim and/or compaction, quickly failing on allocation errors.
>
> The benefits of this approach are:
>
> * CPU is charged to the process that wants to spend the cycles for the THP
> * Avoid unpredictable timing of khugepaged collapse
> * Prevent unpredictable stalls caused by direct reclaim and/or
> compaction

I haven't completely followed the discussion, but it seem your second
and third point could be addressed by a asynchronous THP fault without
any new APIs: allocate 2MB while failing quickly, then on failure get
a 4K page and provide it to the process, while asking khugepaged to
convert the page ASAP in the background, but only after
it managed to allocate a fresh 2MB page to minimize the process visible
down time.

I suppose that would be much more predictable, although there would be a
slightly risk of overwhelming khugepaged. The later could be
addressed by using a scalable workqueue that allocates more threads
when needed.

-Andi