linux-kernel - Re: [RFC PATCH v2 1/4] rseq: Add sched

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: <c1064e76-70f8-4a5f-2624-95a1e2dc9d64@efficios.com>
Date:   Tue, 30 May 2023 11:13:41 -0400
From:   Mathieu Desnoyers <mathieu.desnoyers@...icios.com>
To:     Florian Weimer <fweimer@...hat.com>
Cc:     Peter Zijlstra <peterz@...radead.org>,
        linux-kernel@...r.kernel.org, Thomas Gleixner <tglx@...utronix.de>,
        "Paul E . McKenney" <paulmck@...nel.org>,
        Boqun Feng <boqun.feng@...il.com>,
        "H . Peter Anvin" <hpa@...or.com>, Paul Turner <pjt@...gle.com>,
        linux-api@...r.kernel.org, Christian Brauner <brauner@...nel.org>,
        David.Laight@...LAB.COM, carlos@...hat.com,
        Peter Oskolkov <posk@...k.io>,
        Alexander Mikhalitsyn <alexander@...alicyn.com>,
        Chris Kennelly <ckennelly@...gle.com>,
        Ingo Molnar <mingo@...hat.com>,
        Darren Hart <dvhart@...radead.org>,
        Davidlohr Bueso <dave@...olabs.net>,
        André Almeida <andrealmeid@...lia.com>,
        libc-alpha@...rceware.org, Steven Rostedt <rostedt@...dmis.org>,
        Jonathan Corbet <corbet@....net>,
        Noah Goldstein <goldstein.w.n@...il.com>, longman@...hat.com,
        Olivier Dion <odion@...icios.com>
Subject: Re: [RFC PATCH v2 1/4] rseq: Add sched_state field to struct rseq

On 5/30/23 10:25, Mathieu Desnoyers wrote:
> On 5/30/23 04:20, Florian Weimer wrote:
[...]
>>
>> The challenge is that once we put stuff at fixed offsets, we can't
>> transparently fix it later.  It would need more auxv entries with
>> further offsets, or accessing this data through some indirection,
>> perhaps via vDSO helpers.
> 
> Perhaps this is more flexibility/complexity than we really need. One 
> possible approach would be to split struct rseq into sub-structures, e.g.:
> 
> rseq_len = overall size of all sub-structures.
> auxv AT_RSEQ_ALIGN = 256
> 
> auxv AT_RSEQ_FEATURE_SIZE = size of first portion of struct rseq,
>                              at most 256 bytes, meant to contain fields
>                              stored/loaded from the thread doing the
>                              registration.
> auxv AT_RSEQ_SHARED_FEATURE_SIZE =
>                              size of 2nd portion of struct rseq,
>                              starts at offset 256, at most 256 bytes,
>                              meant to contain fields stored/loaded by
>                              any thread.
> 
> Then we have this layout:
> 
> struct rseq {
>    struct rseq_local {
>      /* Fields accessed from local thread. */
> 
>    } __attribute__((aligned((256));
>    struct rseq_shared {
>      /* Shared fields. */
> 
>    } __attribute__((aligned(256));
> } __attribute__((aligned(256));
> 
> And if someday AT_RSEQ_FEATURE_SIZE needs to grow over 256 bytes
> (32 * u64), we can still extend with a new auxv entry after the "shared"
> features.

Actually, after giving it some more thoughts, I think we can do better:

- Add a sys_rseq() rseq_flag RSEQ_FLAG_SHARED, which changes the 
behavior of sys_rseq() to expect an additional "struct rseq_shared *" 
argument.

- Introduce auxv AT_RSEQ_SHARED_FEATURE_SIZE.

This way, it's up to the libc to decide how to allocate its private vs 
shared rseq structures.

The auxv "AT_RSEQ_ALIGN" would dictate the minimal alignment required 
for both private and shared rseq structures.

I don't think we need to express the size of the rseq_shared memory area 
allocated by libc because we know that it needs to be large enough to 
handle the shared feature size.

Thoughts ?

Thanks,

Mathieu

-- 
Mathieu Desnoyers
EfficiOS Inc.
https://www.efficios.com