On Sat, Jul 11, 2020 at 5:08 PM Andy Shevchenko
<andy.shevchenko@xxxxxxxxx> wrote:
>
> On Fri, Jul 10, 2020 at 7:17 PM Bartosz Golaszewski <brgl@xxxxxxxx> wrote:
> >
> > From: Bartosz Golaszewski <bgolaszewski@xxxxxxxxxxxx>
> >
> > Implement the managed variant of krealloc(). This function works with
> > all memory allocated by devm_kmalloc() (or devres functions using it
> > implicitly like devm_kmemdup(), devm_kstrdup() etc.).
> >
> > Managed realloc'ed chunks can be manually released with devm_kfree().
>
> ...
>
> > devm_kfree()
> > devm_kmalloc()
> > devm_kmalloc_array()
> > + devm_krealloc()
> > devm_kmemdup()
> > devm_kstrdup()
> > devm_kvasprintf()
>
> Order?
>
I didn't notice these were ordered alphabetically, I thought it was by
functionality.
> ...
>
> > +void *devm_krealloc(struct device *dev, void *ptr, size_t new_size, gfp_t gfp)
>
> Do we really need the 'new_' prefix in the parameter?
>
Yes, I think this is a good name for a function that's meant to modify
the size of a memory chunk.
> > +{
> > + struct devres *old_dr, *new_dr;
> > + struct list_head old_head;
> > + unsigned long flags;
> > + void *ret = NULL;
> > + size_t tot_size;
>
> tot -> total.
>
Meh, ok.
> > +
> > + if (unlikely(!new_size)) {
> > + devm_kfree(dev, ptr);
> > + return ZERO_SIZE_PTR;
> > + }
>
> I guess here we need a comment of the possibilities below to have
> ZERO_SIZE_PTR as input.
>
Better yet: I'll just add a kernel doc for this function.
> > + if (unlikely(ZERO_OR_NULL_PTR(ptr)))
> > + return devm_kmalloc(dev, new_size, gfp);
> > +
> > + if (WARN_ON(is_kernel_rodata((unsigned long)ptr)))
> > + /*
> > + * We cannot reliably realloc a const string returned by
> > + * devm_kstrdup_const().
> > + */
> > + return NULL;
> > +
> > + if (!check_dr_size(new_size, &tot_size))
> > + return NULL;
> > +
> > + spin_lock_irqsave(&dev->devres_lock, flags);
> > +
> > + old_dr = find_dr(dev, devm_kmalloc_release, devm_kmalloc_match, ptr);
>
> > + if (WARN_ON(!old_dr))
>
> Under spin lock? I would rather see spin unlock followed by WARN.
>
Yeah, can be done.
> > + /* Memory chunk not managed or managed by a different device. */
> > + goto out;
> > +
> > + old_head = old_dr->node.entry;
> > +
> > + new_dr = krealloc(old_dr, tot_size, gfp);
> > + if (!new_dr)
> > + goto out;
> > +
> > + if (new_dr != old_dr)
> > + list_replace(&old_head, &new_dr->node.entry);
> > +
> > + ret = new_dr->data;
> > +
> > +out:
> > + spin_unlock_irqrestore(&dev->devres_lock, flags);
> > + return ret;
> > +}
>
> --
> With Best Regards,
> Andy Shevchenko
Bartosz
