public inbox for io-uring@vger.kernel.org
 help / color / mirror / Atom feed
From: Alexey Charkov <alchark@gmail.com>
To: Pavel Begunkov <asml.silence@gmail.com>
Cc: io-uring@vger.kernel.org, David Wei <dw@davidwei.uk>,
	netdev@vger.kernel.org,  Jamal Hadi Salim <jhs@mojatatu.com>,
	Pedro Tammela <pctammela@mojatatu.com>,
	 Victor Nogueira <victor@mojatatu.com>
Subject: Re: [PATCH io_uring 0/5] Add dmabuf support for io_uring zcrx
Date: Tue, 6 May 2025 18:34:06 +0400	[thread overview]
Message-ID: <CABjd4YzAJqvLiNid7RoVpLospTrAFzrBpTcFHuem2-JxfkzpmA@mail.gmail.com> (raw)
In-Reply-To: <cover.1746097431.git.asml.silence@gmail.com>

On Tue, May 6, 2025 at 6:29 PM Pavel Begunkov <asml.silence@gmail.com> wrote:
>
> Currently, io_uring zcrx uses regular user pages to populate the
> area for page pools, this series allows the user to pass a dmabuf
> instead.
>
> Patches 1-4 are preparatory and do code shuffling. All dmabuf
> touching changes are in the last patch. A basic example can be
> found at:
>
> https://github.com/isilence/liburing/tree/zcrx-dmabuf
> https://github.com/isilence/liburing.git zcrx-dmabuf
>
> Pavel Begunkov (5):
>   io_uring/zcrx: improve area validation
>   io_uring/zcrx: resolve netdev before area creation
>   io_uring/zcrx: split out memory holders from area
>   io_uring/zcrx: split common area map/unmap parts
>   io_uring/zcrx: dmabuf backed zerocopy receive
>
>  include/uapi/linux/io_uring.h |   6 +-
>  io_uring/rsrc.c               |  27 ++--
>  io_uring/rsrc.h               |   2 +-
>  io_uring/zcrx.c               | 260 +++++++++++++++++++++++++++-------
>  io_uring/zcrx.h               |  18 ++-
>  5 files changed, 248 insertions(+), 65 deletions(-)

Hi Pavel,

Looks like another "depends" line might be needed in io_uring/Kconfig:

diff --git a/io_uring/Kconfig b/io_uring/Kconfig
index 4b949c42c0bf..9fa2cf502940 100644
--- a/io_uring/Kconfig
+++ b/io_uring/Kconfig
@@ -9,3 +9,4 @@ config IO_URING_ZCRX
        depends on PAGE_POOL
        depends on INET
        depends on NET_RX_BUSY_POLL
+       depends on DMA_SHARED_BUFFER

Otherwise I'm having trouble compiling the next-20250506 kernel for
VT8500, which doesn't select DMA_BUF by default. The following linking
error appears at the very end:

armv7a-unknown-linux-gnueabihf-ld: io_uring/zcrx.o: in function
`io_release_dmabuf':
zcrx.c:(.text+0x1c): undefined reference to `dma_buf_unmap_attachment_unlocked'
armv7a-unknown-linux-gnueabihf-ld: zcrx.c:(.text+0x30): undefined
reference to `dma_buf_detach'
armv7a-unknown-linux-gnueabihf-ld: zcrx.c:(.text+0x40): undefined
reference to `dma_buf_put'
armv7a-unknown-linux-gnueabihf-ld: io_uring/zcrx.o: in function
`io_register_zcrx_ifq':
zcrx.c:(.text+0x15cc): undefined reference to `dma_buf_get'
armv7a-unknown-linux-gnueabihf-ld: zcrx.c:(.text+0x15e8): undefined
reference to `dma_buf_attach'
armv7a-unknown-linux-gnueabihf-ld: zcrx.c:(.text+0x1604): undefined
reference to `dma_buf_map_attachment_unlocked'
make[2]: *** [scripts/Makefile.vmlinux:91: vmlinux] Error 1
make[1]: *** [/home/alchark/linux/Makefile:1242: vmlinux] Error 2
make: *** [Makefile:248: __sub-make] Error 2

Best regards,
Alexey

  parent reply	other threads:[~2025-05-06 14:34 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-05-01 12:17 [PATCH io_uring 0/5] Add dmabuf support for io_uring zcrx Pavel Begunkov
2025-05-01 12:17 ` [PATCH io_uring 1/5] io_uring/zcrx: improve area validation Pavel Begunkov
2025-05-01 12:17 ` [PATCH io_uring 2/5] io_uring/zcrx: resolve netdev before area creation Pavel Begunkov
2025-05-01 12:17 ` [PATCH io_uring 3/5] io_uring/zcrx: split out memory holders from area Pavel Begunkov
2025-05-01 12:17 ` [PATCH io_uring 4/5] io_uring/zcrx: split common area map/unmap parts Pavel Begunkov
2025-05-01 12:17 ` [PATCH io_uring 5/5] io_uring/zcrx: dmabuf backed zerocopy receive Pavel Begunkov
2025-05-02 15:25 ` [PATCH io_uring 0/5] Add dmabuf support for io_uring zcrx Jens Axboe
2025-05-06 14:34 ` Alexey Charkov [this message]
2025-05-06 15:32   ` Pavel Begunkov

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=CABjd4YzAJqvLiNid7RoVpLospTrAFzrBpTcFHuem2-JxfkzpmA@mail.gmail.com \
    --to=alchark@gmail.com \
    --cc=asml.silence@gmail.com \
    --cc=dw@davidwei.uk \
    --cc=io-uring@vger.kernel.org \
    --cc=jhs@mojatatu.com \
    --cc=netdev@vger.kernel.org \
    --cc=pctammela@mojatatu.com \
    --cc=victor@mojatatu.com \
    /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