DPDK patches and discussions
 help / color / mirror / Atom feed
From: Krzysztof Kanas <kkanas@marvell.com>
To: Ferruh Yigit <ferruh.yigit@intel.com>
Cc: Krzysztof Kanas <kkanas@marvell.com>,
	"dev@dpdk.org" <dev@dpdk.org>,
	"John W. Linville" <linville@tuxdriver.com>
Subject: Re: [dpdk-dev] [EXT] Re: [PATCH] net/af_packet: set default blocksize to pagesize
Date: Wed, 3 Jul 2019 09:42:41 +0000	[thread overview]
Message-ID: <20190703094238.GA13142@kk-box-0> (raw)
In-Reply-To: <aa85cbed-3f83-a4fa-db17-b53e4b8b428b@intel.com>

On 19-06-27 19:25, Ferruh Yigit wrote:
> External Email
> 
> ----------------------------------------------------------------------
> On 6/24/2019 3:32 PM, kkanas@marvell.com wrote:
> > From: Krzysztof Kanas <kkanas@marvell.com>
> > 
> > Kernel validates block size to be aligned to page size. Code works on
> > platforms with page size 4k, but on others e.g. Arm64 fails with default
> > parameters.
> 
> What is the pagesize in arm64?

On aarch64 it is configurable. Linux kernel, supports sizes 4k, 16k, 64k 
see ARM64_64K_PAGES in arch/arm64/Kconfig.

> 
> > 
> > Signed-off-by: Krzysztof Kanas <kkanas@marvell.com>
> > ---
> >  drivers/net/af_packet/rte_eth_af_packet.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/net/af_packet/rte_eth_af_packet.c b/drivers/net/af_packet/rte_eth_af_packet.c
> > index 4a660d5f013f..ec3d2cbb5202 100644
> > --- a/drivers/net/af_packet/rte_eth_af_packet.c
> > +++ b/drivers/net/af_packet/rte_eth_af_packet.c
> > @@ -32,7 +32,6 @@
> >  #define ETH_AF_PACKET_FRAMECOUNT_ARG	"framecnt"
> >  #define ETH_AF_PACKET_QDISC_BYPASS_ARG	"qdisc_bypass"
> >  
> > -#define DFLT_BLOCK_SIZE		(1 << 12)
> >  #define DFLT_FRAME_SIZE		(1 << 11)
> >  #define DFLT_FRAME_COUNT	(1 << 9)
> >  
> > @@ -811,7 +810,7 @@ rte_eth_from_packet(struct rte_vdev_device *dev,
> >  	struct rte_kvargs_pair *pair = NULL;
> >  	unsigned k_idx;
> >  	unsigned int blockcount;
> > -	unsigned int blocksize = DFLT_BLOCK_SIZE;
> > +	unsigned int blocksize;
> >  	unsigned int framesize = DFLT_FRAME_SIZE;
> >  	unsigned int framecount = DFLT_FRAME_COUNT;
> >  	unsigned int qpairs = 1;
> > @@ -821,6 +820,7 @@ rte_eth_from_packet(struct rte_vdev_device *dev,
> >  	if (*sockfd < 0)
> >  		return -1;
> >  
> > +	blocksize = getpagesize();
> >  	/*
> >  	 * Walk arguments for configurable settings
> >  	 */
> > 
> 

-- 
-
Regards,
Krzysztof(Chris) Kanas

  reply	other threads:[~2019-07-03  9:43 UTC|newest]

Thread overview: 4+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-24 14:32 [dpdk-dev] " kkanas
2019-06-27 18:25 ` Ferruh Yigit
2019-07-03  9:42   ` Krzysztof Kanas [this message]
2019-07-03 17:23 ` Ferruh Yigit

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=20190703094238.GA13142@kk-box-0 \
    --to=kkanas@marvell.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@intel.com \
    --cc=linville@tuxdriver.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).