From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mail-pz2-f1.google.com (mail-pz2-f1.google.com [74.125.228.1]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by smtp.subspace.kernel.org (Postfix) with ESMTPS id 53A36280309 for ; Thu, 6 Aug 2026 07:30:16 +0000 (UTC) Authentication-Results: smtp.subspace.kernel.org; arc=none smtp.client-ip=74.125.228.1 ARC-Seal:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1786001418; cv=none; b=r8GSzhcMvnaWUJ11Pg/8kSl98SUio39CwebjWNkz7XAPjZcDGcWUo2l7S86oe73QzwpLq9wqpWNbMguH6EZ8uVlU5eMUnsuInhj9ONuUHuLgnMUuU+ReHqpm6hK7nYf/5kyBBSVofa8ibCf4YkTytM6kEHalwVyFxRxpeQ0tW1k= ARC-Message-Signature:i=1; a=rsa-sha256; d=subspace.kernel.org; s=arc-20240116; t=1786001418; c=relaxed/simple; bh=luKJF3U56bE6sjj7461iEOHtFnsU45ZlM+Jl4IWRvGA=; h=From:To:Cc:Subject:Date:Message-ID:In-Reply-To:References: MIME-Version; b=lIvZ259fu/IMO3AZdX0iBiV0M5jgPxYerhnaa7GVSC6VclJ3jZ7glL2PBWwMVeU3UyH1CjQhomZ8TP9cG/3G84fTk8Mojgpjk/ZCPtHQqcPJv76PlLVOOVNrx6VEelQNgFDyPJV/zsHOEKXcQ4b8b2asyuNhGHrxhSff0F/wj20= ARC-Authentication-Results:i=1; smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com; spf=pass smtp.mailfrom=gmail.com; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b=UOCP2F2r; arc=none smtp.client-ip=74.125.228.1 Authentication-Results: smtp.subspace.kernel.org; dmarc=pass (p=none dis=none) header.from=gmail.com Authentication-Results: smtp.subspace.kernel.org; spf=pass smtp.mailfrom=gmail.com Authentication-Results: smtp.subspace.kernel.org; dkim=pass (2048-bit key) header.d=gmail.com header.i=@gmail.com header.b="UOCP2F2r" Received: by mail-pz2-f1.google.com with SMTP id 41be03b00d2f7-ca696c4b152so706979a12.1 for ; Thu, 06 Aug 2026 00:30:16 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20251104; t=1786001415; x=1786606215; darn=vger.kernel.org; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:from:to:cc:subject:date :message-id:reply-to:content-type; bh=irsV3UUmbsPIUvx/Xa2IxxMY0Ug7vOgp56JTztI0ESo=; b=UOCP2F2rgIMyZCHiDtBOTHWP6JHnBJhqY1QlFO3hxX6f4786OkiTF/CS6+M7YC6NaD N4QB+tIMMXxkPK90p0/mml9EVgcDrBkjilyyoaYLXoOi4K887Y/jmA7s3MS/QbWlZrJt 2EgHCx3KSzQ/QzuWyx3MlYvk+JCezk0ApFcAUKuW+zw8dM4Hh5/qR0EGkEwlRWjFlM5B 3MQgmy8S5l9P77MHPw+5OQZdN6JmcYqAJVWaR8ZKKvhnVBZgjyJ05nkBBou9gkVztswH JUFPaJhHQNhq0JTHnwEsXq4FpdYR1s3HKGC/l+OLDjKA1SM3OQN5iXIeJ9TCumh6PKrm L4lg== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20251104; t=1786001415; x=1786606215; h=content-transfer-encoding:mime-version:references:in-reply-to :message-id:date:subject:cc:to:from:x-gm-gg:x-gm-message-state:from :to:cc:subject:date:message-id:reply-to:content-type; bh=irsV3UUmbsPIUvx/Xa2IxxMY0Ug7vOgp56JTztI0ESo=; b=cncAyH+rM8U+DdXHyroQ2MiI8RkI7ydfJJ0o/Myx32CDtkcopJJ2DkUggdaExBWOty L2twLWX5CnehZSJEly594r3Su5sA4MzhuFMOSX+dYKx+pzlewOVYte18xuNWmdezOXUZ 1DMhWrLMTAEh2bJWWopF5HvT1/XD7RqN5gGOhRFbBemh5uNXe9leH9KqoNGmxcH1p40k YI1unvMuutRXHuU/XQ5QvfLF9o7nkYB25MTako3rQ0f6O+QppDXyH7VuK/A9sl+z8wCp btTIgJTXbCRh71s4uBkZdej1N5m3VxiHR8wxhZ9bfLKY5rNwJSsgcv3wgdBJInGkZlSv dhYQ== X-Forwarded-Encrypted: i=1; AHgh+Rq76eoiPxTRxMZzWlThO9y6rLI9NIanGn/PPZN+s3QTmpQoMnIIeTgCinep3/ztkzJ31IV8BAJbeL3YF8g=@vger.kernel.org X-Gm-Message-State: AOJu0Yz6fJdHXteWoCvW5tazBkZP4sulkp2GWV4oTN81eP/zYF0GwT6B 0sMZrU1PsfFiadzLaEwMlhlwVHPgrxyjbKveMjIQsuHuXpBQRr3Yc6EO X-Gm-Gg: AR+sD13ZEDyH7G6WyucjL9WEQAsGauuE9V0FW2s4tVgkkxRFmRDLywZyLvV2WswZRaE p95f8HBJpxOfEc8fZqOBFCaKiTETnSX5ev89N19sNaL9BL3paVSbf5vvTY37Tukeh1FqqeMcV/j 0wrQGaolqFgW01aqz0UutuGPUR711ej5V+IuMnMNdRnyqSFbVWXWgTByx2cFJV082q0T9pG5FbS RFnANcYqFbMREGMeScdLkpjFo0UWTLC4As8qVZlohukxhthri1v/mNX/IYvb2TwddJuj50RQG6/ hzpO0ARdge0e0Ute609SN/JN0C0E4Ag+v6S7Tb3nCYYHNaAiD4VN3t175B991lfPD588/RP/QON nN99g3LJGjEyq9HZ+IQ/JwCnB+GIsc1hKLFtLR3GGjhkOMWsTSnSG4sl8JjRM3djSerchRupCPG k395weh3Qy6tF1zKaC98ct1xDQ/mM/0EKNCaKsDk2M3yifUHpM5dgbgUC9QhSVH/95jeuRpiXRm 983De4FDZq/WVHp X-Received: by 2002:a05:6a00:1250:b0:847:97a4:d6b3 with SMTP id d2e1a72fcca58-84f2e05124dmr13179356b3a.16.1786001414418; Thu, 06 Aug 2026 00:30:14 -0700 (PDT) Received: from zhr-ThinkStation-K.mioffice.cn ([2408:8607:1b00:8:7dbc:3daf:8e91:7035]) by smtp.gmail.com with ESMTPSA id 41be03b00d2f7-cbe708e7860sm2534892a12.22.2026.08.06.00.30.04 (version=TLS1_3 cipher=TLS_AES_256_GCM_SHA384 bits=256/256); Thu, 06 Aug 2026 00:30:13 -0700 (PDT) From: Hongru Zhang X-Google-Original-From: Hongru Zhang To: baohua@kernel.org, ljs@kernel.org Cc: akpm@linux-foundation.org, david@kernel.org, liam@infradead.org, linux-kernel@vger.kernel.org, linux-mm@kvack.org, mhocko@suse.com, rppt@kernel.org, shakeel.butt@linux.dev, surenb@google.com, vbabka@kernel.org, willy@infradead.org, zhanghongru06@gmail.com, zhanghongru@xiaomi.com Subject: Re: [RFC PATCH v4 1/3] mm: allow page faults to request VMA-lock retry Date: Thu, 6 Aug 2026 15:29:34 +0800 Message-ID: <20260806072934.77976-1-zhanghongru@xiaomi.com> X-Mailer: git-send-email 2.43.0 In-Reply-To: References: Precedence: bulk X-Mailing-List: linux-kernel@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit > On Tue, Aug 4, 2026 at 8:32 PM Lorenzo Stoakes (ARM) wrote: > > > > On Tue, Aug 04, 2026 at 05:52:19PM +0800, Hongru Zhang wrote: > > > From: Hongru Zhang > > > > > > Page faults handled under the per-VMA lock currently fall back to the > > > mmap_lock path whenever handle_mm_fault() returns VM_FAULT_RETRY. This > > > means that lower-level fault handlers have no way to tell the > > > architecture fault handler that the retry can safely continue under the > > > per-VMA lock. > > > > > > Add VM_FAULT_MAY_USE_VMA_LOCK as an advisory bit that can be returned > > > > I don't love that name or that faulting retry behaviour is _modified_ by a > > value that indicates fault resolution state... ugh. > > > > It's kinda confusing things 'VM_FAULT_RETRY' is 'you have to retry this > > fault'. > > > > 'VM_FAULT_MAY_...' is starting to bring in effectively configuration > > options into it and that's kinda horrible. > > > > I mean is there any reason we shouldn't ALWAYS do this if a VMA lock was > > used? > > > > It's not too expensive to do a single retry with the VMA lock before > > falling back to the mmap lock. > > > > So maybe simplify like that? > > > > And like that this series becomes a single patch right? > > This is a brilliant idea. That's a genius insight, Lorenzo. > > I guess the conceptual model could simply be: > > diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c > index 45b99c3b1442..3592bcc9bbd7 100644 > --- a/arch/x86/mm/fault.c > +++ b/arch/x86/mm/fault.c > @@ -1222,6 +1222,7 @@ void do_user_addr_fault(struct pt_regs *regs, > struct mm_struct *mm; > vm_fault_t fault; > unsigned int flags = FAULT_FLAG_DEFAULT; > + bool vma_lock_retried = false; > > tsk = current; > mm = tsk->mm; > @@ -1331,6 +1332,7 @@ void do_user_addr_fault(struct pt_regs *regs, > if (!(flags & FAULT_FLAG_USER)) > goto lock_mmap; > > +vma_lock: > vma = lock_vma_under_rcu(mm, address); > if (!vma) > goto lock_mmap; > @@ -1352,6 +1354,11 @@ void do_user_addr_fault(struct pt_regs *regs, > if (fault & VM_FAULT_MAJOR) > flags |= FAULT_FLAG_TRIED; > > + if (!vma_lock_retried) { > + vma_lock_retried = true; > + goto vma_lock; > + } > + > /* Quick path to respond to signals */ > if (fault_signal_pending(fault, regs)) { > if (!user_mode(regs)) > > Nothing else needs to change then. I wonder if there is a cleaner > way to implement the idea, but it is really stunning. > > Best Regards > Barry Filemap Throughput (higher is better): +---------+------------+---------------------+---------------------+---------------------+ | Threads | Vanilla | RFC v4 | P1 | P2 | +---------+------------+---------------------+---------------------+---------------------+ | 40 | 1069.34 /s | 1404.47 /s (+31.3%) | 1400.13 /s (+30.9%) | 1412.38 /s (+32.1%) | +---------+------------+---------------------+---------------------+---------------------+ | 60 | 1038.12 /s | 1682.88 /s (+62.1%) | 1683.37 /s (+62.2%) | 1685.23 /s (+62.3%) | +---------+------------+---------------------+---------------------+---------------------+ | 80 | 1042.62 /s | 1766.72 /s (+69.5%) | 1767.83 /s (+69.6%) | 1771.73 /s (+69.9%) | +---------+------------+---------------------+---------------------+---------------------+ Swap Throughput (higher is better): +--------------+-------------+----------------------+----------------------+----------------------+ | mmap writers | Vanilla | RFC v4 | P1 | P2 | +--------------+-------------+----------------------+----------------------+----------------------+ | 0 | 17303.09 /s | 18394.51 /s (+6.3%) | 17899.48 /s (+3.4%) | 18337.30 /s (+6.0%) | +--------------+-------------+----------------------+----------------------+----------------------+ | 2 | 16728.04 /s | 18591.68 /s (+11.1%) | 18346.72 /s (+9.7%) | 18848.17 /s (+12.7%) | +--------------+-------------+----------------------+----------------------+----------------------+ | 4 | 12596.23 /s | 18534.00 /s (+47.1%) | 16095.20 /s (+27.8%) | 18507.62 /s (+46.9%) | +--------------+-------------+----------------------+----------------------+----------------------+ The key difference between P1 and P2 is how FAULT_FLAG_TRIED is handled: P1 keeps the existing major-fault-only setting before the retry, while P2 sets it before the VMA-lock retry for all retrying faults. In filemap throughput test, each reader thread operates on its own file. In swap throughput test, all reader threads fault the same memory area. P1: diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 45b99c3b1442..c3ab30d32a15 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -1222,6 +1222,7 @@ void do_user_addr_fault(struct pt_regs *regs, struct mm_struct *mm; vm_fault_t fault; unsigned int flags = FAULT_FLAG_DEFAULT; + bool vma_lock_retried = false; tsk = current; mm = tsk->mm; @@ -1331,6 +1332,7 @@ void do_user_addr_fault(struct pt_regs *regs, if (!(flags & FAULT_FLAG_USER)) goto lock_mmap; +lock_vma: vma = lock_vma_under_rcu(mm, address); if (!vma) goto lock_mmap; @@ -1360,6 +1362,12 @@ void do_user_addr_fault(struct pt_regs *regs, ARCH_DEFAULT_PKEY); return; } + + if (!vma_lock_retried) { + vma_lock_retried = true; + goto lock_vma; + } + lock_mmap: P2: diff --git a/arch/x86/mm/fault.c b/arch/x86/mm/fault.c index 45b99c3b1442..9507b8a0fe18 100644 --- a/arch/x86/mm/fault.c +++ b/arch/x86/mm/fault.c @@ -1222,6 +1222,7 @@ void do_user_addr_fault(struct pt_regs *regs, struct mm_struct *mm; vm_fault_t fault; unsigned int flags = FAULT_FLAG_DEFAULT; + bool vma_lock_retried = false; tsk = current; mm = tsk->mm; @@ -1331,6 +1332,7 @@ void do_user_addr_fault(struct pt_regs *regs, if (!(flags & FAULT_FLAG_USER)) goto lock_mmap; +lock_vma: vma = lock_vma_under_rcu(mm, address); if (!vma) goto lock_mmap; @@ -1349,8 +1351,6 @@ void do_user_addr_fault(struct pt_regs *regs, goto done; } count_vm_vma_lock_event(VMA_LOCK_RETRY); - if (fault & VM_FAULT_MAJOR) - flags |= FAULT_FLAG_TRIED; /* Quick path to respond to signals */ if (fault_signal_pending(fault, regs)) { @@ -1360,6 +1360,13 @@ void do_user_addr_fault(struct pt_regs *regs, ARCH_DEFAULT_PKEY); return; } + + if (!vma_lock_retried) { + flags |= FAULT_FLAG_TRIED; + vma_lock_retried = true; + goto lock_vma; + } + lock_mmap: