From: David Miller <davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
To: willemdebruijn.kernel-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org
Cc: netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
willemb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org
Subject: Re: [PATCH net-next v3 02/13] sock: skb_copy_ubufs support for compound pages
Date: Thu, 22 Jun 2017 21:36:46 -0400 (EDT) [thread overview]
Message-ID: <20170622.213646.1398197704798474384.davem@davemloft.net> (raw)
In-Reply-To: <CAF=yD-+WQ1b9BytG4JuhgHCR8gAKvUVJPUfi4t-u_DqAFnVrkA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
From: Willem de Bruijn <willemdebruijn.kernel-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
Date: Thu, 22 Jun 2017 16:57:07 -0400
>>
>> Likewise.
>>
>>> + f_off = f->page_offset;
>>> + f_size = f->size;
>>> +
>>> + vaddr = kmap_atomic(skb_frag_page(f));
>>
>> I looked at some kmap_atomic() implementations and I do not think
>> it supports compound pages.
>
> Indeed. Thanks. It appears that I can do the obvious thing and
> kmap the individual page that is being copied inside the loop:
>
> kmap_atomic(skb_frag_page(f) + (f_off >> PAGE_SHIFT));
>
> This is similar to existing logic in copy_huge_page_from_user
> and __flush_dcache_page in arch/arm/mm/flush.c
>
> But, this also applies to other skb operations that call kmap_atomic,
> such as skb_copy_bits and __skb_checksum. Not all can be called
> from a codepath with a compound user page, but I have to address
> the ones that can.
Yeah that's quite a mess, it looks like this assumption that
kmap can handle compound pages exists in quite a few places.
next prev parent reply other threads:[~2017-06-23 1:36 UTC|newest]
Thread overview: 22+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-06-21 21:18 [PATCH net-next v3 00/13] socket sendmsg MSG_ZEROCOPY Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 01/13] sock: allocate skbs from optmem Willem de Bruijn
[not found] ` <20170621211816.53837-1-willemdebruijn.kernel-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2017-06-21 21:18 ` [PATCH net-next v3 02/13] sock: skb_copy_ubufs support for compound pages Willem de Bruijn
[not found] ` <20170621211816.53837-3-willemdebruijn.kernel-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2017-06-22 17:05 ` David Miller
[not found] ` <20170622.130528.1762873686654379973.davem-fT/PcQaiUtIeIZ0/mPfg9Q@public.gmane.org>
2017-06-22 20:57 ` Willem de Bruijn
[not found] ` <CAF=yD-+WQ1b9BytG4JuhgHCR8gAKvUVJPUfi4t-u_DqAFnVrkA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-06-23 1:36 ` David Miller [this message]
2017-06-23 3:59 ` Willem de Bruijn
[not found] ` <CAF=yD-JGekEkBsFfgZa+-TN3-QBPy4sfK0QLPx83Uh3DAoodvg-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-06-27 15:53 ` Willem de Bruijn
[not found] ` <CAF=yD-LJ-EvWbKtxqrcK1D=nRSzWzpPcFVOX1poZ3+P=aa0QfA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2017-06-29 15:54 ` Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 13/13] test: add msg_zerocopy test Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 03/13] sock: add MSG_ZEROCOPY Willem de Bruijn
[not found] ` <20170621211816.53837-4-willemdebruijn.kernel-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2017-06-22 17:06 ` David Miller
2017-06-21 21:18 ` [PATCH net-next v3 04/13] sock: add SOCK_ZEROCOPY sockopt Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 05/13] sock: enable MSG_ZEROCOPY Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 06/13] sock: MSG_ZEROCOPY notification coalescing Willem de Bruijn
[not found] ` <20170621211816.53837-7-willemdebruijn.kernel-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2017-06-22 17:07 ` David Miller
2017-06-21 21:18 ` [PATCH net-next v3 07/13] sock: add ee_code SO_EE_CODE_ZEROCOPY_COPIED Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 08/13] sock: ulimit on MSG_ZEROCOPY pages Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 09/13] tcp: enable MSG_ZEROCOPY Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 10/13] udp: " Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 11/13] raw: enable MSG_ZEROCOPY with IP_HDRINCL Willem de Bruijn
2017-06-21 21:18 ` [PATCH net-next v3 12/13] packet: enable MSG_ZEROCOPY Willem de Bruijn
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=20170622.213646.1398197704798474384.davem@davemloft.net \
--to=davem-ft/pcqaiutieiz0/mpfg9q@public.gmane.org \
--cc=linux-api-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=netdev-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=willemb-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=willemdebruijn.kernel-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.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;
as well as URLs for NNTP newsgroup(s).