Re: [PATCH] mm/slab: improve kmem_cache_alloc_bulk

[Vlastimil Babka (SUSE)]

On 6/3/26 13:13, Rob Clark wrote:
> On Wed, Jun 3, 2026 at 2:17 AM Vlastimil Babka (SUSE) <vbabka@xxxxxxxxxx> wrote:
>>
>> We know p->pages is NULL in this case, right? Because it was allocated by
>> vm_bind_job_create() using kzalloc().
>> And the job can't be reused with a leftover value?
>> (msm_iommu_pagetable_prealloc_cleanup doesn't set p->pages to zero).
>> Or should we set p->pages to NULL here.
> 
> Correct, the job is not reused.  But I suppose setting p->pages to
> NULL would make things more obvious, so no objection to that.

OK, did that, just in case. Thanks.

> BR,
> -R
> 
>> > +
>> >         p->pages = kvmalloc_objs(*p->pages, p->count);
>> >         if (!p->pages)
>> >                 return -ENOMEM;
>> >
>> >         ret = kmem_cache_alloc_bulk(pt_cache, GFP_KERNEL, p->count, p->pages);
>> >         if (ret != p->count) {
>> > -               kfree(p->pages);
>> > +               kvfree(p->pages);
>> >                 p->pages = NULL;
>> >                 p->count = ret;
>> >                 return -ENOMEM;
>>