linux-kernel - Re: [RFC PATCH] mm: count zram read/write into PSI_IO

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Wed, 8 Dec 2021 17:45:22 +0000
From:   Chris Down <chris@...isdown.name>
To:     Zhaoyang Huang <huangzhaoyang@...il.com>
Cc:     Johannes Weiner <hannes@...xchg.org>,
        Nitin Gupta <ngupta@...are.org>,
        Sergey Senozhatsky <senozhatsky@...omium.org>,
        Jens Axboe <axboe@...nel.dk>, Minchan Kim <minchan@...nel.org>,
        Zhaoyang Huang <zhaoyang.huang@...soc.com>,
        "open list:MEMORY MANAGEMENT" <linux-mm@...ck.org>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [RFC PATCH] mm: count zram read/write into PSI_IO_WAIT

Zhaoyang Huang writes:
>No. Block device related D-state will be counted in via
>psi_dequeue(io_wait). What I am proposing here is do NOT ignore the
>influence on non-productive time by huge numbers of in-context swap
>in/out (zram like). This can help to make IO pressure more accurate
>and coordinate with the number of PSWPIN/OUT. It is like counting the
>IO time within filemap_fault->wait_on_page_bit_common into
>psi_mem_stall, which introduces memory pressure high by IO.

I think part of the confusion here is that the name "io" doesn't really just 
mean "io", it means "disk I/O". As in, we are targeting real, physical or 
network disk I/O. Of course, we can only do what's reasonable if the device 
we're accounting for is layers upon layers eventually leading to a 
memory-backed device, but _intentionally_ polluting that with more memory-bound 
accesses doesn't make any sense when we already have separate accounting for 
memory. Why would anyone want that?

I'm with Johannes here, I think this would actively make memory pressure 
monitoring less useful. This is a NAK from my perspective as someone who 
actually uses these things in production.