linux-kernel - Re: [PATCH 08/11] x86/mm: enable broadcast TLB invalidation for multi-threaded processes

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 for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-Id: <4F173A6B-B09B-40D0-B118-94B08B9ACC32@gmail.com>
Date: Thu, 26 Dec 2024 01:22:10 +0200
From: Nadav Amit <nadav.amit@...il.com>
To: Rik van Riel <riel@...riel.com>
Cc: the arch/x86 maintainers <x86@...nel.org>,
 Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
 kernel-team@...a.com,
 Dave Hansen <dave.hansen@...ux.intel.com>,
 luto@...nel.org,
 peterz@...radead.org,
 Thomas Gleixner <tglx@...utronix.de>,
 Ingo Molnar <mingo@...hat.com>,
 Borislav Petkov <bp@...en8.de>,
 "H. Peter Anvin" <hpa@...or.com>,
 Andrew Morton <akpm@...ux-foundation.org>,
 "open list:MEMORY MANAGEMENT" <linux-mm@...ck.org>
Subject: Re: [PATCH 08/11] x86/mm: enable broadcast TLB invalidation for
 multi-threaded processes

On 23 Dec 2024, at 4:55, Rik van Riel <riel@...riel.com> wrote:

> +static int mm_active_cpus(struct mm_struct *mm)
> +{
> +	int count = 0;
> +	int cpu;
> +
> +	for_each_cpu(cpu, mm_cpumask(mm)) {
> +		/* Skip the CPUs that aren't really running this process. */
> +		if (per_cpu(cpu_tlbstate.loaded_mm, cpu) != mm)
> +			continue;
> +
> +		if (per_cpu(cpu_tlbstate_shared.is_lazy, cpu))
> +			continue;
> +
> +		count++;
> +	}
> +	return count;
> +}

Since you are only interested in checking whether the number of “mm active
CPUs" is greater than a certain threshold, don’t you want to add some
checks for early termination? This can allow to avoid cachelines of
cpu_tlbstate traversing back and forth.

For instance, by running cpumask_weight() first, if the weight is lower than
the threshold, no need to loop. Similarly, if inside the loop the threshold
has already been crossed, no need for more iterations.