linux-kernel - Re: [PATCH bpf] bpf: invalidate unused part of bpf_prog

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 for Android: free password hash cracker in your pocket

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

Message-ID: <4EA80DBB-4194-4FC4-866F-7088C2107E17@fb.com>
Date:   Fri, 22 Apr 2022 01:31:17 +0000
From:   Song Liu <songliubraving@...com>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
CC:     Peter Zijlstra <peterz@...radead.org>, Song Liu <song@...nel.org>,
        Alexei Starovoitov <alexei.starovoitov@...il.com>,
        bpf <bpf@...r.kernel.org>, Linux-MM <linux-mm@...ck.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Kernel Team <Kernel-team@...com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        "Edgecombe, Rick P" <rick.p.edgecombe@...el.com>,
        Christoph Hellwig <hch@...radead.org>,
        Andrii Nakryiko <andrii@...nel.org>
Subject: Re: [PATCH bpf] bpf: invalidate unused part of bpf_prog_pack

Hi Linus,

> On Apr 21, 2022, at 4:10 PM, Linus Torvalds <torvalds@...ux-foundation.org> wrote:
> 
> On Thu, Apr 21, 2022 at 3:52 PM Song Liu <songliubraving@...com> wrote:
>> 
>> I think this won’t work, as set_memory_ro makes all the aliases of
>> these pages read only.
> 
> Argh. I thought we only did that for the whole memory type thing
> (history: nasty machine checks possible on some hardware if you mix
> memory types for the same physical page with virtual mappings), but if
> we do it for RO too, then yeah.
> 
> It's sad to use that horrid machinery for basically non-live code, but
> whatever.

I guess we will stick with bpf_arch_text_copy(), and we will keep the 
Invalidation at BPF program free time?

I will reorder and resend pending patches. Then we can decide which ones
to ship with 5.18. 

Thanks,
Song