linux-kernel - Re: [PATCH] livepatch: introduce klp

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]

Message-ID: <alpine.LSU.2.21.2405311603260.8344@pobox.suse.cz>
Date: Fri, 31 May 2024 16:06:46 +0200 (CEST)
From: Miroslav Benes <mbenes@...e.cz>
To: zhang warden <zhangwarden@...il.com>
cc: Petr Mladek <pmladek@...e.com>, Josh Poimboeuf <jpoimboe@...nel.org>, 
    Jiri Kosina <jikos@...nel.org>, Joe Lawrence <joe.lawrence@...hat.com>, 
    live-patching@...r.kernel.org, LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] livepatch: introduce klp_func called interface

> And for the unlikely branch, isn’t the complier will compile this branch 
> into a cold branch that will do no harm to the function performance?

The test (cmp insn or something like that) still needs to be there. Since 
there is only a simple assignment in the branch, the compiler may just 
choose not to have a cold branch in this case. The only difference is in 
which case you would jump here. You can see for yourself (and prove me 
wrong if it comes to it).

Miroslav