Hi,
So, the original code that is now part of the #else part was:
memcpy_toio(offset, entry->buf, entry->len);
On Mon, Aug 22, 2016 at 7:25 PM, Nicholas Mc Guire <der.herr@xxxxxxx> wrote:
>
> Hi !
>
> A probably very basic sparse question - but I could not figure out a
> satisfying answer. while compile testing a patch for ntb_transfer.c I got
>
> CHECK drivers/ntb/ntb_transport.c
> drivers/ntb/ntb_transport.c:1583:43: warning: incorrect type in argument 1 (different address spaces)
> drivers/ntb/ntb_transport.c:1583:43: expected void *dst
> drivers/ntb/ntb_transport.c:1583:43: got void [noderef] <asn:2>*offset
> drivers/ntb/ntb_transport.c:1583:56: warning: incorrect type in argument 2 (different address spaces)
> drivers/ntb/ntb_transport.c:1583:56: expected void const [noderef] <asn:1>*src
> drivers/ntb/ntb_transport.c:1583:56: got void *buf
>
> looking at the offending line;
>
> static void ntb_memcpy_tx(struct ntb_queue_entry *entry, void __iomem *offset)
> {
> #ifdef ARCH_HAS_NOCACHE_UACCESS
> /*
> * Using non-temporal mov to improve performance on non-cached
> * writes, even though we aren't actually copying from user space.
> */
> __copy_from_user_inatomic_nocache(offset, entry->buf, entry->len);
>
> the absence of a __user in the buf argument seems intentional and the
> dst is not a kernel but __iomem address which triggers the second warning
> So the first warning seems to be a false positive here, the second one
> Im not clear about, but I guess its also a false positive. The question
> is if there is a clean way to anotate this to make sparse happy without
> any unwanted sideffects ?
>
> For the first warning using (void __user *) entry->buf should do
> and be side-effect free, but how could the second warning be fixed ?
>
> sparse message is from 4.8.0-rc2 (localversion-next is -next-20160822)
> sparse version is v0.5.0-44
>
> thx!
> hofrat
> --
> To unsubscribe from this list: send the line "unsubscribe linux-sparse" in
> the body of a message to majordomo@xxxxxxxxxxxxxxx
> More majordomo info at http://vger.kernel.org/majordomo-info.html
--
To unsubscribe from this list: send the line "unsubscribe linux-sparse" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
