Re: [PATCH 1/1] mm: change inlined allocation helpers to account at the call site

From: Kent Overstreet
Date: Thu Apr 04 2024 - 18:29:23 EST


On Thu, Apr 04, 2024 at 03:17:43PM -0700, Suren Baghdasaryan wrote:
> On Thu, Apr 4, 2024 at 10:08 AM Suren Baghdasaryan <surenb@xxxxxxxxxx> wrote:
> >
> > On Thu, Apr 4, 2024 at 10:04 AM Matthew Wilcox <willy@xxxxxxxxxxxxx> wrote:
> > >
> > > On Thu, Apr 04, 2024 at 09:54:04AM -0700, Suren Baghdasaryan wrote:
> > > > +++ b/include/linux/dma-fence-chain.h
> > > > @@ -86,10 +86,7 @@ dma_fence_chain_contained(struct dma_fence *fence)
> > > > *
> > > > * Returns a new struct dma_fence_chain object or NULL on failure.
> > > > */
> > > > -static inline struct dma_fence_chain *dma_fence_chain_alloc(void)
> > > > -{
> > > > - return kmalloc(sizeof(struct dma_fence_chain), GFP_KERNEL);
> > > > -};
> > > > +#define dma_fence_chain_alloc() kmalloc(sizeof(struct dma_fence_chain), GFP_KERNEL)
> > >
> > > You've removed some typesafety here. Before, if I wrote:
> > >
> > > struct page *page = dma_fence_chain_alloc();
> > >
> > > the compiler would warn me that I've done something stupid. Now it
> > > can't tell. Suggest perhaps:
> > >
> > > #define dma_fence_chain_alloc() \
> > > (struct dma_fence_chain *)kmalloc(sizeof(struct dma_fence_chain), \
> > > GFP_KERNEL)
> > >
> > > but maybe there's a better way of doing that. There are a few other
> > > occurrences of the same problem in this monster patch.
> >
> > Got your point.
>
> Ironically, checkpatch generates warnings for these type casts:
>
> WARNING: unnecessary cast may hide bugs, see
> http://c-faq.com/malloc/mallocnocast.html
> #425: FILE: include/linux/dma-fence-chain.h:90:
> + ((struct dma_fence_chain *)kmalloc(sizeof(struct dma_fence_chain),
> GFP_KERNEL))
>
> I guess I can safely ignore them in this case (since we cast to the
> expected type)?

Correct, it's not hiding bugs in this case, it's adding type safety.

checkpatch is definitely not authoritative, you really have to use your
own judgement with it