Re: [PATCH v2] mm/uffd: UFFD_FEATURE_WP_UNPOPULATED

From: Muhammad Usama Anjum <usama.anjum@collabora.com>
To: David Hildenbrand <david@redhat.com>, Peter Xu <peterx@redhat.com>
Cc: Muhammad Usama Anjum <usama.anjum@collabora.com>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org,
	Andrea Arcangeli <aarcange@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Mike Rapoport <rppt@linux.vnet.ibm.com>,
	Axel Rasmussen <axelrasmussen@google.com>,
	Nadav Amit <nadav.amit@gmail.com>,
	"kernel@collabora.com" <kernel@collabora.com>
Subject: Re: [PATCH v2] mm/uffd: UFFD_FEATURE_WP_UNPOPULATED
Date: Thu, 2 Mar 2023 20:14:05 +0500	[thread overview]
Message-ID: <fad40511-3909-4362-5760-bcb57c94e534@collabora.com> (raw)
In-Reply-To: <982c1623-ac28-2743-d6a3-4faf42f72d47@redhat.com>

On 3/2/23 7:01 PM, David Hildenbrand wrote:
> On 02.03.23 14:57, Peter Xu wrote:
>> On Thu, Mar 02, 2023 at 10:37:44AM +0100, David Hildenbrand wrote:
>>> Especially for such large sparse VMAs, the current way of allocating
>>> pagetables to place markers/zeropages is far from optimal.
>>
>> IMHO that's not a generic workload.  As mentioned in the reply there, I
>> would suggest we go with simple then we have space to optimize it in the
>> future if necessary, because the API will be the same.
This is a good idea.

I'm trying to understand why aren't we going with most optimized
implementation. Why aren't we targeting it at this point in time?

>>
> 
> I disagree with "generic workload", we use sparse mmaps all over the place,
> and when blindly used by e.g., CRIU, we'll simply end up wasting memory and
> time.
I've heard about a use case where a file of size 10s of GBs can be mapped
to the memory and then accessed off and on. We need to handle this
correctly and efficiently.

> 
> But I already agreed that this optimization that is a separate thing to
> implement.
> 

-- 
BR,
Muhammad Usama Anjum