linux-kernel - Re: [PATCH v3 08/11] slub: Replace cmpxchg

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: <d2e02a4e-2a73-7b07-1bfb-ddf3f2a53ff1@suse.cz>
Date:   Wed, 24 May 2023 12:13:09 +0200
From:   Vlastimil Babka <vbabka@...e.cz>
To:     Peter Zijlstra <peterz@...radead.org>,
        torvalds@...ux-foundation.org
Cc:     corbet@....net, will@...nel.org, boqun.feng@...il.com,
        mark.rutland@....com, catalin.marinas@....com, dennis@...nel.org,
        tj@...nel.org, cl@...ux.com, hca@...ux.ibm.com, gor@...ux.ibm.com,
        agordeev@...ux.ibm.com, borntraeger@...ux.ibm.com,
        svens@...ux.ibm.com, tglx@...utronix.de, mingo@...hat.com,
        bp@...en8.de, dave.hansen@...ux.intel.com, x86@...nel.org,
        hpa@...or.com, joro@...tes.org, suravee.suthikulpanit@....com,
        robin.murphy@....com, dwmw2@...radead.org,
        baolu.lu@...ux.intel.com, Arnd Bergmann <arnd@...db.de>,
        Herbert Xu <herbert@...dor.apana.org.au>, davem@...emloft.net,
        penberg@...nel.org, rientjes@...gle.com, iamjoonsoo.kim@....com,
        Andrew Morton <akpm@...ux-foundation.org>,
        roman.gushchin@...ux.dev, 42.hyeyoo@...il.com,
        linux-doc@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-mm@...ck.org, linux-s390@...r.kernel.org,
        iommu@...ts.linux.dev, linux-arch@...r.kernel.org,
        linux-crypto@...r.kernel.org, sfr@...b.auug.org.au,
        mpe@...erman.id.au
Subject: Re: [PATCH v3 08/11] slub: Replace cmpxchg_double()

On 5/24/23 11:32, Peter Zijlstra wrote:
> On Mon, May 15, 2023 at 09:57:07AM +0200, Peter Zijlstra wrote:
> 
>> @@ -3008,6 +3029,22 @@ static inline bool pfmemalloc_match(stru
>>  }
>>  
>>  #ifndef CONFIG_SLUB_TINY
>> +static inline bool
>> +__update_cpu_freelist_fast(struct kmem_cache *s,
>> +			   void *freelist_old, void *freelist_new,
>> +			   unsigned long tid)
>> +{
>> +#ifdef system_has_freelist_aba
>> +	freelist_aba_t old = { .freelist = freelist_old, .counter = tid };
>> +	freelist_aba_t new = { .freelist = freelist_new, .counter = next_tid(tid) };
>> +
>> +	return this_cpu_cmpxchg_freelist(s->cpu_slab->freelist_tid.full,
>> +					 old.full, new.full) == old.full;
>> +#else
>> +	return false;
>> +#endif
>> +}
>> +
>>  /*
>>   * Check the slab->freelist and either transfer the freelist to the
>>   * per cpu freelist or deactivate the slab.
>> @@ -3359,11 +3396,7 @@ static __always_inline void *__slab_allo
>>  		 * against code executing on this cpu *not* from access by
>>  		 * other cpus.
>>  		 */
>> -		if (unlikely(!this_cpu_cmpxchg_double(
>> -				s->cpu_slab->freelist, s->cpu_slab->tid,
>> -				object, tid,
>> -				next_object, next_tid(tid)))) {
>> -
>> +		if (unlikely(!__update_cpu_freelist_fast(s, object, next_object, tid))) {
>>  			note_cmpxchg_failure("slab_alloc", s, tid);
>>  			goto redo;
>>  		}
>> @@ -3736,11 +3769,7 @@ static __always_inline void do_slab_free
>>  
>>  		set_freepointer(s, tail_obj, freelist);
>>  
>> -		if (unlikely(!this_cpu_cmpxchg_double(
>> -				s->cpu_slab->freelist, s->cpu_slab->tid,
>> -				freelist, tid,
>> -				head, next_tid(tid)))) {
>> -
>> +		if (unlikely(!__update_cpu_freelist_fast(s, freelist, head, tid))) {
>>  			note_cmpxchg_failure("slab_free", s, tid);
>>  			goto redo;
>>  		}
> 
> This isn't right; the this_cpu_cmpxchg_double() was unconditional and
> relied on the local_irq_save() fallback when no native cmpxchg128 is
> present.
> 
> The below delta makes things boot again when system_has_cmpxchg128 is
> not defined.

Right, that should do.

> I'm going to zap these patches from tip/locking/core for a few days and
> fold the below back into the series and let it run through the robots
> again.

I noticed some comments in mm/slub.c still mention "cmpxchg_double", dunno
how much you want to clean it right now or can be postponed. Also some sysfs
stats files for CONFIG_SLUB_STATS (not widely used) which we probably might
try renaming without breaking anyone, but it's not guaranteed.