From: Jason Gunthorpe <jgg@nvidia.com>
To: Linus Walleij <linus.walleij@linaro.org>
Cc: Leon Romanovsky <leonro@nvidia.com>,
Bernard Metzler <BMT@zurich.ibm.com>,
linux-rdma@vger.kernel.org
Subject: Re: [PATCH] RDMA/rxe: Pass a pointer to virt_to_page()
Date: Fri, 24 Mar 2023 10:59:56 -0300 [thread overview]
Message-ID: <ZB2s3GeaN/FBpR5K@nvidia.com> (raw)
In-Reply-To: <20230324103252.712107-1-linus.walleij@linaro.org>
On Fri, Mar 24, 2023 at 11:32:52AM +0100, Linus Walleij wrote:
> Like the other calls in this function virt_to_page() expects
> a pointer, not an integer.
>
> However since many architectures implement virt_to_pfn() as
> a macro, this function becomes polymorphic and accepts both a
> (unsigned long) and a (void *).
>
> Fix this up with an explicit cast.
>
> Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
> ---
> drivers/infiniband/sw/rxe/rxe_mr.c | 8 ++++----
> 1 file changed, 4 insertions(+), 4 deletions(-)
>
> diff --git a/drivers/infiniband/sw/rxe/rxe_mr.c b/drivers/infiniband/sw/rxe/rxe_mr.c
> index b10aa1580a64..5c90d83002f0 100644
> --- a/drivers/infiniband/sw/rxe/rxe_mr.c
> +++ b/drivers/infiniband/sw/rxe/rxe_mr.c
> @@ -213,7 +213,7 @@ int rxe_mr_init_fast(int max_pages, struct rxe_mr *mr)
> static int rxe_set_page(struct ib_mr *ibmr, u64 iova)
> {
All these functions have the wrong names, they are kva not IOVA.
> @@ -288,7 +288,7 @@ static void rxe_mr_copy_dma(struct rxe_mr *mr, u64 iova, void *addr,
> u8 *va;
> while (length) {
> - page = virt_to_page(iova & mr->page_mask);
> + page = virt_to_page((void *)(iova & mr->page_mask));
> bytes = min_t(unsigned int, length,
> PAGE_SIZE - page_offset);
This is actually a bug, this function is only called on IB_MR_TYPE_DMA
and in that case 'iova' is actually a phys addr
So iova should be called phys and the above should be:
page = pfn_to_page(phys >> PAGE_SHIFT);
> @@ -488,7 +488,7 @@ int rxe_mr_do_atomic_op(struct rxe_mr *mr, u64 iova, int opcode,
>
> if (mr->ibmr.type == IB_MR_TYPE_DMA) {
> page_offset = iova & (PAGE_SIZE - 1);
> - page = virt_to_page(iova & PAGE_MASK);
> + page = virt_to_page((void *)(iova & PAGE_MASK));
These masks against PAGE_MASK are not needed, virt_to_page does them
already.
> } else {
> unsigned long index;
> int err;
> @@ -545,7 +545,7 @@ int rxe_mr_do_atomic_write(struct rxe_mr *mr, u64 iova, u64 value)
>
> if (mr->ibmr.type == IB_MR_TYPE_DMA) {
> page_offset = iova & (PAGE_SIZE - 1);
> - page = virt_to_page(iova & PAGE_MASK);
> + page = virt_to_page((void *)(iova & PAGE_MASK));
pfn_to_page here as well
Jason
next prev parent reply other threads:[~2023-03-24 14:01 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-03-24 10:32 [PATCH] RDMA/rxe: Pass a pointer to virt_to_page() Linus Walleij
2023-03-24 12:27 ` kernel test robot
2023-03-24 13:59 ` Jason Gunthorpe [this message]
2023-03-29 14:28 ` Linus Walleij
2023-03-29 20:17 ` Bob Pearson
2023-03-29 23:16 ` Jason Gunthorpe
2023-03-30 2:45 ` Bob Pearson
2023-03-30 12:06 ` Jason Gunthorpe
2023-03-30 15:19 ` Linus Walleij
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ZB2s3GeaN/FBpR5K@nvidia.com \
--to=jgg@nvidia.com \
--cc=BMT@zurich.ibm.com \
--cc=leonro@nvidia.com \
--cc=linus.walleij@linaro.org \
--cc=linux-rdma@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox