Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <242daeb2-b96b-d0dd-5597-ebf5fb2dfeca@intel.com>
Date:   Thu, 15 Dec 2022 13:53:07 -0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:102.0) Gecko/20100101
 Thunderbird/102.5.1
Subject: Re: [GIT PULL] x86/mm for 6.2
Content-Language: en-US
To:     Linus Torvalds <torvalds@linux-foundation.org>,
        kirill.shutemov@linux.intel.com
Cc:     Dave Hansen <dave.hansen@linux.intel.com>,
        linux-kernel@vger.kernel.org, x86@kernel.org,
        Andy Lutomirski <luto@kernel.org>,
        Thomas Gleixner <tglx@linutronix.de>,
        Borislav Petkov <bp@alien8.de>
References: <20221213174234.688534-1-dave.hansen@linux.intel.com>
 <CAHk-=wi=TY3Kte5Z1_nvfcsEh+rcz86pYnzeASw=pbG9QtpJEQ@mail.gmail.com>
 <20221215123007.cagd7qiidehqd77k@box.shutemov.name>
 <CAHk-=whC_ixb3FDdMhW_wiKw7-bB700kvUyqN+_cPUNp=1hNsQ@mail.gmail.com>
 <CAHk-=wjyy3iKS0B=A-yAPqjE3xiUU_5AiXApCasuYKTNu842dQ@mail.gmail.com>
From:   Dave Hansen <dave.hansen@intel.com>
In-Reply-To: <CAHk-=wjyy3iKS0B=A-yAPqjE3xiUU_5AiXApCasuYKTNu842dQ@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 7bit
Precedence: bulk

On 12/15/22 09:26, Linus Torvalds wrote:
> But if you feel like all threads have to share the same LAM state, it
> does seem a lot simpler if you just say "you need to set that state
> before you start any threads". No?

That would be a lot simpler.  I have one bit of hesitation there, though.

Back in the MPX days, we had some users pop up and tell us that MPX
wasn't working for them on certain threads.  Those threads ended up
having been clone()'d even _before_ MPX got enabled which was via some
pre-main() startup code that the compiler inserted.  Since these early
threads' FPU state was copied before MPX setup was done, they never got
MPX-enabled.

Right or wrong, since then I've basically assumed that someone might be
creating threads behind my back.

Maybe we call those "early thread" folks too crazy to get LAM.  Maybe I
need to forget it ever happened, but they were actual users that got
bitten and cared enough to report it.  Or, heck, maybe I'm just
delusional because I can't find any trace of this conversation in the
list archives.

LAM _is_ slightly different than what MPX hit, though, since instead of
totally silent breakage the app can whine about the LAM prctl() having
failed.

Anyway, message heard loud and clear about the untagged_addr() races and
the interfaces.  We'll find some way to fix those up for the next merge
window.