linux-kernel - Re: [PATCH] vmalloc: add __alloc_vm

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: <a7dd5f4e-5a63-3129-4b42-924ae2166d36@yandex-team.ru>
Date:   Wed, 11 Oct 2017 16:32:10 +0300
From:   Konstantin Khlebnikov <khlebnikov@...dex-team.ru>
To:     Christoph Hellwig <hch@...radead.org>
Cc:     linux-mm@...ck.org, Andrew Morton <akpm@...ux-foundation.org>,
        linux-kernel@...r.kernel.org, Andy Lutomirski <luto@...nel.org>
Subject: Re: [PATCH] vmalloc: add __alloc_vm_area() for optimizing vmap stack

On 08.10.2017 12:16, Christoph Hellwig wrote:
> This looks fine in general, but a few comments:
> 
>   - can you split adding the new function from switching over the fork
>     codeok

>   - at least kasan and vmalloc_user/vmalloc_32_user use very similar
>     patterns, can you switch them over as well?

I don't see why VM_USERMAP cannot be set right at allocation.

I'll add vm_flags argument to __vmalloc_node() and
pass here VM_USERMAP from vmalloc_user/vmalloc_32_user
in separate patch.

KASAN is different: it allocates shadow area for area allocated for module.
Pointer to module area must be pushed from module_alloc().
This isn't worth optimization.

>   - the new __alloc_vm_area looks very different from alloc_vm_area,
>     maybe it needs a better name?  vmalloc_range_area for example?

__vmalloc_area() is vacant - this most low-level, so I'll keep "__".

>   - when you split an existing function please keep the more low-level
>     function on top of the higher level one that calls it.ok