linux-kernel - Re: [PATCH 3/4] Documentation/vm: Remove "Using kmap-atomic" from highmem.rst.

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:   Fri, 22 Apr 2022 22:09:03 +0200
From:   "Fabio M. De Francesco" <fmdefrancesco@...il.com>
To:     Ira Weiny <ira.weiny@...el.com>
Cc:     Andrew Morton <akpm@...ux-foundation.org>,
        Catalin Marinas <catalin.marinas@....com>,
        "Matthew Wilcox (Oracle)" <willy@...radead.org>,
        Will Deacon <will@...nel.org>,
        Peter Collingbourne <pcc@...gle.com>,
        Vlastimil Babka <vbabka@...e.cz>,
        Sebastian Andrzej Siewior <bigeasy@...utronix.de>,
        linux-kernel@...r.kernel.org, Jonathan Corbet <corbet@....net>,
        linux-doc@...r.kernel.org, outreachy@...ts.linux.dev,
        Thomas Gleixner <tglx@...utronix.de>,
        Peter Zijlstra <peterz@...radead.org>
Subject: Re: [PATCH 3/4] Documentation/vm: Remove "Using kmap-atomic" from highmem.rst.

On venerdì 22 aprile 2022 20:38:09 CEST Ira Weiny wrote:
> On Thu, Apr 21, 2022 at 08:01:59PM +0200, Fabio M. De Francesco wrote:
> > The use of kmap_atomic() is deprecated in favor of kmap_local_page().
> 
> I'm not sure deprecated is the right word.

OK, in v2 I won't use "deprecated". Instead I'll say something about a 
strong preference to avoid its use. The reason why developers should avoid 
kmap_atomic() are explained in 4/4 (I've copy-pasted some lines here for 
your convenience):

+  Each call of kmap_atomic() in the kernel creates a non-preemptible 
section
+  and disable pagefaults. This could be a source of unwanted latency, so 
it
+  should be only used if it is absolutely required, otherwise 
kmap_local_page()
+  should be used where it is feasible.

> And I think the fact that this
> documentation is stale is a better reason for the patch as is.
> 
> This series should end up indicating the desire to stop growing kmap() 
and
> kmap_atomic() call sites and that their deprecation is on the horizon.  
I've
> not read the text in patch 4/4 yet.

I'll wait for your review of 4/4 before sending v2.

> 
> > For
> > this reason the "Using kmap_atomic" section in highmem.rst is obsolete 
and
> > unnecessary.
> 
> A lot of the text is obsolete (and redundant) but the example code might 
be
> useful.
> 
> Why not move the example and relevant bits into the kdoc for 
kmap_atomic()
> which is then automatically picked up via patch 2/4.

Yes, I agree with you. I'll take into account your suggestion for v2.

However, as I said above, I'll hold v2 until you have time to review 4/4 
for the purpose to not miss any changes that you might require for that 
patch too.

Furthermore, while working on v2, I think that I'll extend this series with 
one or two patch more, in order to address other issues I noticed. 

Thanks,

Fabio