linux-kernel - Re: [PATCH v4 0/3] targeted TLB sync IPIs for lockless page table

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: <13205d28-7478-4578-8bf3-ead18b4dc6b0@kernel.org>
Date: Thu, 5 Feb 2026 16:05:37 +0100
From: "David Hildenbrand (Arm)" <david@...nel.org>
To: Lance Yang <lance.yang@...ux.dev>, Peter Zijlstra <peterz@...radead.org>,
 dave.hansen@...el.com
Cc: Liam.Howlett@...cle.com, akpm@...ux-foundation.org,
 aneesh.kumar@...nel.org, arnd@...db.de, baohua@...nel.org,
 baolin.wang@...ux.alibaba.com, boris.ostrovsky@...cle.com, bp@...en8.de,
 dave.hansen@...ux.intel.com, dev.jain@....com, hpa@...or.com,
 hughd@...gle.com, ioworker0@...il.com, jannh@...gle.com, jgross@...e.com,
 kvm@...r.kernel.org, linux-arch@...r.kernel.org,
 linux-kernel@...r.kernel.org, linux-mm@...ck.org,
 lorenzo.stoakes@...cle.com, mingo@...hat.com, npache@...hat.com,
 npiggin@...il.com, pbonzini@...hat.com, riel@...riel.com,
 ryan.roberts@....com, seanjc@...gle.com, shy828301@...il.com,
 tglx@...utronix.de, virtualization@...ts.linux.dev, will@...nel.org,
 x86@...nel.org, ypodemsk@...hat.com, ziy@...dia.com
Subject: Re: [PATCH v4 0/3] targeted TLB sync IPIs for lockless page table

On 2/5/26 16:01, Lance Yang wrote:
> 
> 
> On 2026/2/5 21:25, David Hildenbrand (Arm) wrote:
>> On 2/2/26 16:52, Lance Yang wrote:
>>>
>>>
>>>
>>> Yep, we could replace the IPI with synchronize_rcu() on the sync side:
>>>
>>> - Currently: TLB flush → send IPI → wait for walkers to finish
>>> - With synchronize_rcu(): TLB flush → synchronize_rcu() -> waits for 
>>> grace period
>>>
>>> Lockless walkers (e.g. GUP-fast) use local_irq_disable(); 
>>> synchronize_rcu() also
>>> waits for regions with preemption/interrupts disabled, so it should 
>>> work, IIUC.
>>>
>>> And then, the trade-off would be:
>>> - Read side: zero cost (no per-CPU tracking)
>>> - Write side: wait for RCU grace period (potentially slower)
>>>
>>> For collapse/unshare, that write-side latency might be acceptable :)
>>>
>>> @David, what do you think?
>>
>> Given that we just fixed the write-side latency from breaking Oracle's 
>> databases completely, we have to be a bit careful here :)
> 
> Yep, agreed.
> 
>>
>> The thing is: on many x86 configs we don't need *any* TLB flushed or 
>> RCU syncs.
> 
> Right. Looks like that is low-hanging fruit. I'll send that out 
> separately :)
> 
>>
>> So "how much slower" are we talking about, especially on bigger/loaded 
>> systems?
> 
> Unfortunately the numbers are pretry bad. On an x86-64 64-core system
> under high load, each synchronize_rcu() is about *22.9* ms on average ...
> 
> So for now, neither approach looks good: tracking on the read side adss
> cost to GUP-fast, and syncing on the write side e.g. synchronize_rcu()
> is too slow on large systems.

GUP-fast is 3%, right? Any way we can reduce that to 1% and call it 
noise? :)

-- 
Cheers,

David