User-Agent: Cyrus-JMAP/3.5.0-alpha0-4458-g51a91c06b2-fm-20211130.004-g51a91c06
Mime-Version: 1.0
Message-Id: <d662bbd4-ed39-47ef-b2a4-012c765ec4ad@www.fastmail.com>
In-Reply-To: <e4d3d685-eddd-a240-eea5-949637809fed@intel.com>
References: <20211203003636.11417-1-jiaxun.yang@flygoat.com>
 <20211203003636.11417-8-jiaxun.yang@flygoat.com>
 <343c2c64-4d0b-6e21-80e0-834d0b7147aa@intel.com>
 <57665273-e56e-4b66-9dbd-003d592eb65d@www.fastmail.com>
 <e4d3d685-eddd-a240-eea5-949637809fed@intel.com>
Date:   Fri, 03 Dec 2021 15:51:26 +0000
From:   "Jiaxun Yang" <jiaxun.yang@flygoat.com>
To:     "Dave Hansen" <dave.hansen@intel.com>, x86@kernel.org
Cc:     "Thomas Gleixner" <tglx@linutronix.de>,
        "Ingo Molnar" <mingo@redhat.com>, "Borislav Petkov" <bp@alien8.de>,
        dave.hansen@linux.intel.com, hpa@zytor.com,
        chang.seok.bae@intel.com, linux-kernel@vger.kernel.org,
        "Jiaxun Yang" <j.yang-87@sms.ed.ac.uk>
Subject: Re: [RFC PATCH 07/10] x86/fpu: Rellocate fpstate on save_fpregs_to_fpstate
Content-Type: text/plain;charset=utf-8
Content-Transfer-Encoding: quoted-printable
Precedence: bulk


=E5=9C=A82021=E5=B9=B412=E6=9C=883=E6=97=A5=E5=8D=81=E4=BA=8C=E6=9C=88 =E4=
=B8=8B=E5=8D=883:18=EF=BC=8CDave Hansen=E5=86=99=E9=81=93=EF=BC=9A
> On 12/3/21 3:39 AM, Jiaxun Yang wrote:
>>>>  	if (likely(use_xsave())) {
>>>> +		xstate_update_size(fpu);
>>>>  		os_xsave(fpu->fpstate);
>>>>  		update_avx_timestamp(fpu);
>>>>  		return;
>>> Have you considered what exactly happens when you hit that WARN_ON_F=
PU()
>>> which otherwise ignores the allocation error?  Have you considered w=
hat
>>> happens on the os_xsave() that follows it immediately?  How about wh=
at
>>> happens the next time this task runs after that failure?
>> Thank you for the catch.
>> This is a few questions that I don't have answer, so it's a RFC.
>>=20
>> I thought it is unlikely to happen as kmalloc has emergency pool.
>> But in case it happens, I guess the best way to handle it is just
>> send SIGILL to corresponding user process or panic if it's kernel
>> fpu use?
>
> We've thought a *LOT* about this exact problem over the past few years.
>
> Intel even added hardware (XFD) to prevent the situation where you land
> in the context switch code, fail a memory allocation, and have to
> destroy user data in registers.  Without XFD, there are also zero ways
> to avoid this happening to apps, *other* than preallocating the memory
> in the first place.
>
> I don't think there is *any* viable path forward with this series.

Hmm, actually I can come up some ways to workaround it.
Like we can have some sort of preallocated emergency pool
with max_feature and utilize them in case of allocation failure during c=
ontext switch.
We'll get some chance to fulfill the pool again after going back from in=
terrupt context :-)

But maybe you are right, it's not for me, a first year undergraduate stu=
dent,
to comment on solutions from thousands of brilliant brains at Intel.

Appreciate for your comments to let me understand the nature of the prob=
lem.

Thanks.
--=20
- Jiaxun