Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
MIME-Version: 1.0
References: <20221222061341.381903-1-yuanchu@google.com> <20221222061341.381903-2-yuanchu@google.com>
In-Reply-To: <20221222061341.381903-2-yuanchu@google.com>
From:   Yu Zhao <yuzhao@google.com>
Date:   Thu, 22 Dec 2022 14:37:00 -0700
Message-ID: <CAOUHufZhcJh8PdVtFuoOPWBWw_fWNAB61GndXoWjekYaubXTAQ@mail.gmail.com>
Subject: Re: [PATCH 2/2] mm: support POSIX_FADV_NOREUSE for generic fadvise handler
To:     Yuanchu Xie <yuanchu@google.com>
Cc:     Ivan Babrou <ivan@cloudflare.com>,
        Johannes Weiner <hannes@cmpxchg.org>,
        Alexander Viro <viro@zeniv.linux.org.uk>,
        Andrew Morton <akpm@linux-foundation.org>,
        Steven Barrett <steven@liquorix.net>,
        Brian Geffon <bgeffon@google.com>,
        Oleksandr Natalenko <oleksandr@natalenko.name>,
        Suren Baghdasaryan <surenb@google.com>,
        Arnd Bergmann <arnd@arndb.de>, Peter Xu <peterx@redhat.com>,
        Hugh Dickins <hughd@google.com>,
        Gaosheng Cui <cuigaosheng1@huawei.com>,
        linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org,
        linux-mm@kvack.org, Jens Axboe <axboe@kernel.dk>,
        Michael Larabel <michael@michaellarabel.com>
Content-Type: text/plain; charset="UTF-8"
Precedence: bulk

On Wed, Dec 21, 2022 at 11:13 PM Yuanchu Xie <yuanchu@google.com> wrote:

Thanks for following up on this.

> POSIX_FADV_NOREUSE allows an application to specify that accesses to
> file data does not follow LRU and is used only once. Since 2.6.18 this
> is a no-op. We add FMODE_NOREUSE, checked in vma_has_locality to prevent
> LRU activation.

This needs to include what you plan to write on the man page.

A few questions to answer:
1. Does this flag work with accesses via FDs?
2. If there is a random or sequential file VMA, should the user choose
this flag or the VMA flag or both? Consider a) how those flags affect
readahead; b) their scopes, i.e., per VMA or per file.

Please also follow up with Jens to add this flag to fio.

Micheal reported that SVT-AV1 regressed with MGLRU, which is the only
real one [1]. The following not only fixes the regression but also
improves the baseline. Please follow up on that as well.

--- a/Source/App/EncApp/EbAppMain.c
+++ b/Source/App/EncApp/EbAppMain.c
@@ -115,6 +115,7 @@ void init_memory_file_map(EbConfig* config) {
             fseeko(config->input_file, curr_loc, SEEK_SET); // seek
back to that location
 #ifndef _WIN32
             config->mmap.fd = fileno(config->input_file);
+            posix_fadvise(config->mmap.fd, 0, 0, POSIX_FADV_NOREUSE);
 #endif
         }
         config->mmap.file_frame_it = 0;

[1] https://openbenchmarking.org/result/2209259-PTS-MGLRU8GB57