From: Jesper Dangaard Brouer <brouer@redhat.com>
To: Lorenzo Bianconi <lorenzo@kernel.org>
Cc: netdev@vger.kernel.org, lorenzo.bianconi@redhat.com,
davem@davemloft.net, thomas.petazzoni@bootlin.com,
ilias.apalodimas@linaro.org, matteo.croce@redhat.com,
brouer@redhat.com
Subject: Re: [PATCH net-next 2/3] net: page_pool: add the possibility to sync DMA memory for non-coherent devices
Date: Wed, 13 Nov 2019 09:29:07 +0100 [thread overview]
Message-ID: <20191113092907.569f6b8e@carbon> (raw)
In-Reply-To: <68229f90060d01c1457ac945b2f6524e2aa27d05.1573383212.git.lorenzo@kernel.org>
On Sun, 10 Nov 2019 14:09:09 +0200
Lorenzo Bianconi <lorenzo@kernel.org> wrote:
> diff --git a/include/net/page_pool.h b/include/net/page_pool.h
> index 2cbcdbdec254..defbfd90ab46 100644
[...]
> @@ -150,8 +153,8 @@ static inline void page_pool_destroy(struct page_pool *pool)
> }
>
> /* Never call this directly, use helpers below */
> -void __page_pool_put_page(struct page_pool *pool,
> - struct page *page, bool allow_direct);
> +void __page_pool_put_page(struct page_pool *pool, struct page *page,
> + unsigned int dma_sync_size, bool allow_direct);
>
> static inline void page_pool_put_page(struct page_pool *pool,
> struct page *page, bool allow_direct)
> @@ -160,14 +163,14 @@ static inline void page_pool_put_page(struct page_pool *pool,
> * allow registering MEM_TYPE_PAGE_POOL, but shield linker.
> */
> #ifdef CONFIG_PAGE_POOL
> - __page_pool_put_page(pool, page, allow_direct);
> + __page_pool_put_page(pool, page, 0, allow_direct);
> #endif
> }
> /* Very limited use-cases allow recycle direct */
> static inline void page_pool_recycle_direct(struct page_pool *pool,
> struct page *page)
> {
> - __page_pool_put_page(pool, page, true);
> + __page_pool_put_page(pool, page, 0, true);
> }
We need to use another "default" value than zero for 'dma_sync_size' in
above calls. I suggest either 0xFFFFFFFF or -1 (which unsigned is
0xFFFFFFFF).
Point is that in case caller doesn't know the length (the CPU have had
access to) then page_pool will need to sync with pool->p.max_len.
If choosing a larger value here default value your code below takes
care of it via min(dma_sync_size, pool->p.max_len).
> /* API user MUST have disconnected alloc-side (not allowed to call
> diff --git a/net/core/page_pool.c b/net/core/page_pool.c
> index 5bc65587f1c4..af9514c2d15b 100644
> --- a/net/core/page_pool.c
> +++ b/net/core/page_pool.c
> @@ -112,6 +112,17 @@ static struct page
> *__page_pool_get_cached(struct page_pool *pool) return page;
> }
>
> +/* Used for non-coherent devices */
> +static void page_pool_dma_sync_for_device(struct page_pool *pool,
> + struct page *page,
> + unsigned int dma_sync_size)
> +{
> + dma_sync_size = min(dma_sync_size, pool->p.max_len);
> + dma_sync_single_range_for_device(pool->p.dev, page->dma_addr,
> + pool->p.offset, dma_sync_size,
> + pool->p.dma_dir);
> +}
--
Best regards,
Jesper Dangaard Brouer
MSc.CS, Principal Kernel Engineer at Red Hat
LinkedIn: http://www.linkedin.com/in/brouer
next prev parent reply other threads:[~2019-11-13 8:29 UTC|newest]
Thread overview: 21+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-11-10 12:09 [PATCH net-next 0/3] add DMA sync capability to page_pool API Lorenzo Bianconi
2019-11-10 12:09 ` [PATCH net-next 1/3] net: mvneta: rely on page_pool_recycle_direct in mvneta_run_xdp Lorenzo Bianconi
2019-11-10 12:09 ` [PATCH net-next 2/3] net: page_pool: add the possibility to sync DMA memory for non-coherent devices Lorenzo Bianconi
2019-11-11 16:48 ` Jesper Dangaard Brouer
2019-11-11 19:11 ` Lorenzo Bianconi
2019-11-13 8:29 ` Jesper Dangaard Brouer [this message]
2019-11-14 18:48 ` Jonathan Lemon
2019-11-14 18:53 ` Ilias Apalodimas
2019-11-14 20:27 ` Jonathan Lemon
2019-11-14 20:42 ` Ilias Apalodimas
2019-11-14 21:04 ` Jonathan Lemon
2019-11-14 21:43 ` Jesper Dangaard Brouer
2019-11-15 7:05 ` Ilias Apalodimas
2019-11-15 7:49 ` Lorenzo Bianconi
2019-11-15 8:03 ` Ilias Apalodimas
2019-11-15 16:47 ` Jonathan Lemon
2019-11-15 16:53 ` Lorenzo Bianconi
2019-11-15 7:17 ` Ilias Apalodimas
2019-11-10 12:09 ` [PATCH net-next 3/3] net: mvneta: get rid of huge DMA sync in mvneta_rx_refill Lorenzo Bianconi
2019-11-14 18:14 ` Jonathan Lemon
2019-11-14 18:18 ` Ilias Apalodimas
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=20191113092907.569f6b8e@carbon \
--to=brouer@redhat.com \
--cc=davem@davemloft.net \
--cc=ilias.apalodimas@linaro.org \
--cc=lorenzo.bianconi@redhat.com \
--cc=lorenzo@kernel.org \
--cc=matteo.croce@redhat.com \
--cc=netdev@vger.kernel.org \
--cc=thomas.petazzoni@bootlin.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).