From: Lorenzo Bianconi <lorenzo@kernel.org>
To: Jakub Kicinski <kuba@kernel.org>
Cc: bpf@vger.kernel.org, netdev@vger.kernel.org,
lorenzo.bianconi@redhat.com, davem@davemloft.net, ast@kernel.org,
daniel@iogearbox.net, shayagr@amazon.com,
john.fastabend@gmail.com, dsahern@kernel.org, brouer@redhat.com,
echaudro@redhat.com, jasowang@redhat.com,
alexander.duyck@gmail.com, saeed@kernel.org,
maciej.fijalkowski@intel.com, magnus.karlsson@intel.com,
tirthendu.sarkar@intel.com, toke@redhat.com
Subject: Re: [PATCH v17 bpf-next 13/23] bpf: add multi-buffer support to xdp copy helpers
Date: Mon, 8 Nov 2021 14:59:59 +0100 [thread overview]
Message-ID: <YYktX7swvejoYdnN@lore-desk> (raw)
In-Reply-To: <20211105162933.113ce3c5@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
[-- Attachment #1: Type: text/plain, Size: 2809 bytes --]
> On Thu, 4 Nov 2021 18:35:33 +0100 Lorenzo Bianconi wrote:
> > -static unsigned long bpf_xdp_copy(void *dst_buff, const void *src_buff,
> > +static unsigned long bpf_xdp_copy(void *dst_buff, const void *ctx,
> > unsigned long off, unsigned long len)
> > {
> > - memcpy(dst_buff, src_buff + off, len);
> > + unsigned long base_len, copy_len, frag_off_total;
> > + struct xdp_buff *xdp = (struct xdp_buff *)ctx;
> > + struct skb_shared_info *sinfo;
> > + int i;
> > +
> > + if (likely(!xdp_buff_is_mb(xdp))) {
>
> Would it be better to do
>
> if (xdp->data_end - xdp->data >= off + len)
>
> ?
Hi Jakub,
I am fine with the patch (just a typo inline), thx :)
I will let Eelco to comment since he wrote the original code.
If there is no objections, I will integrate it in v18.
Regards,
Lorenzo
>
> > + memcpy(dst_buff, xdp->data + off, len);
> > + return 0;
> > + }
> > +
> > + base_len = xdp->data_end - xdp->data;
> > + frag_off_total = base_len;
> > + sinfo = xdp_get_shared_info_from_buff(xdp);
> > +
> > + /* If we need to copy data from the base buffer do it */
> > + if (off < base_len) {
> > + copy_len = min(len, base_len - off);
> > + memcpy(dst_buff, xdp->data + off, copy_len);
> > +
> > + off += copy_len;
> > + len -= copy_len;
> > + dst_buff += copy_len;
> > + }
> > +
> > + /* Copy any remaining data from the fragments */
> > + for (i = 0; len && i < sinfo->nr_frags; i++) {
> > + skb_frag_t *frag = &sinfo->frags[i];
> > + unsigned long frag_len, frag_off;
> > +
> > + frag_len = skb_frag_size(frag);
> > + frag_off = off - frag_off_total;
> > + if (frag_off < frag_len) {
> > + copy_len = min(len, frag_len - frag_off);
> > + memcpy(dst_buff,
> > + skb_frag_address(frag) + frag_off, copy_len);
> > +
> > + off += copy_len;
> > + len -= copy_len;
> > + dst_buff += copy_len;
> > + }
> > + frag_off_total += frag_len;
> > + }
> > +
>
> nit: can't help but feel that you can merge base copy and frag copy:
>
> sinfo = xdp_get_shared_info_from_buff(xdp);
> next_frag = &sinfo->frags[0];
> end_frag = &sinfo->frags[sinfo->nr_frags];
>
> ptr_off = 0;
> ptr_buf = xdp->data;
> ptr_len = xdp->data_end - xdp->data;
>
> while (true) {
> if (off < ptr_off + ptr_len) {
> copy_off = ptr_off - off;
I guess here should be:
copy_off = off - ptr_off;
> copy_len = min(len, ptr_len - copy_off);
> memcpy(dst_buff, ptr_buf + copy_off, copy_len);
>
> off += copy_len;
> len -= copy_len;
> dst_buff += copy_len;
> }
>
> if (!len || next_frag == end_frag)
> break;
>
> ptr_off += ptr_len;
> ptr_buf = skb_frag_address(next_frag);
> ptr_len = skb_frag_size(next_frag);
> next_frag++;
> }
>
> Up to you.
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 228 bytes --]
next prev parent reply other threads:[~2021-11-08 14:00 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-11-04 17:35 [PATCH v17 bpf-next 00/23] mvneta: introduce XDP multi-buffer support Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 01/23] net: skbuff: add size metadata to skb_shared_info for xdp Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 02/23] xdp: introduce flags field in xdp_buff/xdp_frame Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 03/23] net: mvneta: update mb bit before passing the xdp buffer to eBPF layer Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 04/23] net: mvneta: simplify mvneta_swbm_add_rx_fragment management Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 05/23] net: xdp: add xdp_update_skb_shared_info utility routine Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 06/23] net: marvell: rely on " Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 07/23] xdp: add multi-buff support to xdp_return_{buff/frame} Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 08/23] net: mvneta: add multi buffer support to XDP_TX Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 09/23] bpf: introduce BPF_F_XDP_MB flag in prog_flags loading the ebpf program Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 10/23] net: mvneta: enable jumbo frames if the loaded XDP program support mb Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 11/23] bpf: introduce bpf_xdp_get_buff_len helper Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 12/23] bpf: add multi-buff support to the bpf_xdp_adjust_tail() API Lorenzo Bianconi
2021-11-05 2:16 ` Jakub Kicinski
2021-11-05 13:57 ` Lorenzo Bianconi
2021-11-05 23:29 ` Jakub Kicinski
2021-11-08 16:55 ` Lorenzo Bianconi
2021-11-08 18:08 ` Toke Høiland-Jørgensen
2021-11-08 19:06 ` Lorenzo Bianconi
2021-11-08 20:46 ` Toke Høiland-Jørgensen
2021-11-08 21:40 ` Jakub Kicinski
2021-11-09 21:58 ` Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 13/23] bpf: add multi-buffer support to xdp copy helpers Lorenzo Bianconi
2021-11-05 23:29 ` Jakub Kicinski
2021-11-08 13:59 ` Lorenzo Bianconi [this message]
2021-11-11 7:50 ` Eelco Chaudron
2021-11-04 17:35 ` [PATCH v17 bpf-next 14/23] bpf: move user_size out of bpf_test_init Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 15/23] bpf: introduce multibuff support to bpf_prog_test_run_xdp() Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 16/23] bpf: test_run: add xdp_shared_info pointer in bpf_test_finish signature Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 17/23] bpf: selftests: update xdp_adjust_tail selftest to include multi-buffer Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 18/23] libbpf: Add SEC name for xdp_mb programs Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 19/23] bpf: generalise tail call map Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 20/23] net: xdp: introduce bpf_xdp_pointer utility routine Lorenzo Bianconi
2021-11-05 23:29 ` Jakub Kicinski
2021-11-08 16:48 ` Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 21/23] bpf: selftests: introduce bpf_xdp_{load,store}_bytes selftest Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 22/23] bpf: selftests: add CPUMAP/DEVMAP selftests for xdp multi-buff Lorenzo Bianconi
2021-11-04 17:35 ` [PATCH v17 bpf-next 23/23] xdp: disable XDP_REDIRECT " Lorenzo Bianconi
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=YYktX7swvejoYdnN@lore-desk \
--to=lorenzo@kernel.org \
--cc=alexander.duyck@gmail.com \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=brouer@redhat.com \
--cc=daniel@iogearbox.net \
--cc=davem@davemloft.net \
--cc=dsahern@kernel.org \
--cc=echaudro@redhat.com \
--cc=jasowang@redhat.com \
--cc=john.fastabend@gmail.com \
--cc=kuba@kernel.org \
--cc=lorenzo.bianconi@redhat.com \
--cc=maciej.fijalkowski@intel.com \
--cc=magnus.karlsson@intel.com \
--cc=netdev@vger.kernel.org \
--cc=saeed@kernel.org \
--cc=shayagr@amazon.com \
--cc=tirthendu.sarkar@intel.com \
--cc=toke@redhat.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;
as well as URLs for NNTP newsgroup(s).