linux-kernel - Re: [PATCH v2 09/11] powerpc/smp: Optimize update_mask_by

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: <f848a6761de05d655d847130e77b23b2bb39aa26.camel@redhat.com>
Date:   Wed, 07 Oct 2020 09:05:42 -0400
From:   Qian Cai <cai@...hat.com>
To:     Srikar Dronamraju <srikar@...ux.vnet.ibm.com>,
        Michael Ellerman <mpe@...erman.id.au>
Cc:     Nathan Lynch <nathanl@...ux.ibm.com>,
        Gautham R Shenoy <ego@...ux.vnet.ibm.com>,
        Michael Neuling <mikey@...ling.org>,
        Peter Zijlstra <peterz@...radead.org>,
        LKML <linux-kernel@...r.kernel.org>,
        Nicholas Piggin <npiggin@...il.com>,
        Valentin Schneider <valentin.schneider@....com>,
        Oliver O'Halloran <oohall@...il.com>,
        Satheesh Rajendran <sathnaga@...ux.vnet.ibm.com>,
        linuxppc-dev <linuxppc-dev@...ts.ozlabs.org>,
        Ingo Molnar <mingo@...nel.org>
Subject: Re: [PATCH v2 09/11] powerpc/smp: Optimize update_mask_by_l2

On Mon, 2020-09-21 at 15:26 +0530, Srikar Dronamraju wrote:
> All threads of a SMT4 core can either be part of this CPU's l2-cache
> mask or not related to this CPU l2-cache mask. Use this relation to
> reduce the number of iterations needed to find all the CPUs that share
> the same l2-cache.
> 
> Use a temporary mask to iterate through the CPUs that may share l2_cache
> mask. Also instead of setting one CPU at a time into cpu_l2_cache_mask,
> copy the SMT4/sub mask at one shot.
> 
...
>  static bool update_mask_by_l2(int cpu)
>  {
> +	struct cpumask *(*submask_fn)(int) = cpu_sibling_mask;
>  	struct device_node *l2_cache, *np;
> +	cpumask_var_t mask;
>  	int i;
>  
>  	l2_cache = cpu_to_l2cache(cpu);
> @@ -1240,22 +1264,37 @@ static bool update_mask_by_l2(int cpu)
>  		return false;
>  	}
>  
> -	cpumask_set_cpu(cpu, cpu_l2_cache_mask(cpu));
> -	for_each_cpu_and(i, cpu_online_mask, cpu_cpu_mask(cpu)) {
> +	alloc_cpumask_var_node(&mask, GFP_KERNEL, cpu_to_node(cpu));

Shouldn't this be GFP_ATOMIC? Otherwise, during the CPU hotplugging, we have,

(irqs were disabled in do_idle())

[  335.420001][    T0] BUG: sleeping function called from invalid context at mm/slab.h:494
[  335.420003][    T0] in_atomic(): 1, irqs_disabled(): 1, non_block: 0, pid: 0, name: swapper/88
[  335.420005][    T0] no locks held by swapper/88/0.
[  335.420007][    T0] irq event stamp: 18074448
[  335.420015][    T0] hardirqs last  enabled at (18074447): [<c0000000001a2a7c>] tick_nohz_idle_enter+0x9c/0x110
[  335.420019][    T0] hardirqs last disabled at (18074448): [<c000000000106798>] do_idle+0x138/0x3b0
do_idle at kernel/sched/idle.c:253 (discriminator 1)
[  335.420023][    T0] softirqs last  enabled at (18074440): [<c0000000000bbec4>] irq_enter_rcu+0x94/0xa0
[  335.420026][    T0] softirqs last disabled at (18074439): [<c0000000000bbea0>] irq_enter_rcu+0x70/0xa0
[  335.420030][    T0] CPU: 88 PID: 0 Comm: swapper/88 Tainted: G        W         5.9.0-rc8-next-20201007 #1
[  335.420032][    T0] Call Trace:
[  335.420037][    T0] [c00020000a4bfcf0] [c000000000649e98] dump_stack+0xec/0x144 (unreliable)
[  335.420043][    T0] [c00020000a4bfd30] [c0000000000f6c34] ___might_sleep+0x2f4/0x310
[  335.420048][    T0] [c00020000a4bfdb0] [c000000000354f94] slab_pre_alloc_hook.constprop.82+0x124/0x190
[  335.420051][    T0] [c00020000a4bfe00] [c00000000035e9e8] __kmalloc_node+0x88/0x3a0
slab_alloc_node at mm/slub.c:2817
(inlined by) __kmalloc_node at mm/slub.c:4013
[  335.420054][    T0] [c00020000a4bfe80] [c0000000006494d8] alloc_cpumask_var_node+0x38/0x80
kmalloc_node at include/linux/slab.h:577
(inlined by) alloc_cpumask_var_node at lib/cpumask.c:116
[  335.420060][    T0] [c00020000a4bfef0] [c00000000003eedc] start_secondary+0x27c/0x800
update_mask_by_l2 at arch/powerpc/kernel/smp.c:1267
(inlined by) add_cpu_to_masks at arch/powerpc/kernel/smp.c:1387
(inlined by) start_secondary at arch/powerpc/kernel/smp.c:1420
[  335.420063][    T0] [c00020000a4bff90] [c00000000000c468] start_secondary_resume+0x10/0x14

> +	cpumask_and(mask, cpu_online_mask, cpu_cpu_mask(cpu));
> +
> +	if (has_big_cores)
> +		submask_fn = cpu_smallcore_mask;
> +
> +	/* Update l2-cache mask with all the CPUs that are part of submask */
> +	or_cpumasks_related(cpu, cpu, submask_fn, cpu_l2_cache_mask);
> +
> +	/* Skip all CPUs already part of current CPU l2-cache mask */
> +	cpumask_andnot(mask, mask, cpu_l2_cache_mask(cpu));
> +
> +	for_each_cpu(i, mask) {
>  		/*
>  		 * when updating the marks the current CPU has not been marked
>  		 * online, but we need to update the cache masks
>  		 */
>  		np = cpu_to_l2cache(i);
> -		if (!np)
> -			continue;
>  
> -		if (np == l2_cache)
> -			set_cpus_related(cpu, i, cpu_l2_cache_mask);
> +		/* Skip all CPUs already part of current CPU l2-cache */
> +		if (np == l2_cache) {
> +			or_cpumasks_related(cpu, i, submask_fn,
> cpu_l2_cache_mask);
> +			cpumask_andnot(mask, mask, submask_fn(i));
> +		} else {
> +			cpumask_andnot(mask, mask, cpu_l2_cache_mask(i));
> +		}
>  
>  		of_node_put(np);
>  	}
>  	of_node_put(l2_cache);
> +	free_cpumask_var(mask);
>  
>  	return true;
>  }