DPDK patches and discussions
 help / color / mirror / Atom feed
From: Jerin Jacob <jerin.jacob@caviumnetworks.com>
To: Bruce Richardson <bruce.richardson@intel.com>
Cc: <dev@dpdk.org>, <thomas.monjalon@6wind.com>
Subject: Re: [dpdk-dev] [PATCH] ethdev: make struct rte_eth_dev cache aligned
Date: Tue, 3 May 2016 17:40:57 +0530	[thread overview]
Message-ID: <20160503121056.GA17000@localhost.localdomain> (raw)
In-Reply-To: <20160503094052.GA21300@bricha3-MOBL3>

On Tue, May 03, 2016 at 10:40:53AM +0100, Bruce Richardson wrote:
> On Mon, May 02, 2016 at 01:37:45PM +0530, Jerin Jacob wrote:
> > Elements of struct rte_eth_dev used in the fast path.
> > Make struct rte_eth_dev cache aligned to avoid the cases where
> > rte_eth_dev elements share the same cache line with other structures.
> > 
> > Signed-off-by: Jerin Jacob <jerin.jacob@caviumnetworks.com>
> > ---
> >  lib/librte_ether/rte_ethdev.c | 2 +-
> >  lib/librte_ether/rte_ethdev.h | 2 +-
> >  2 files changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/lib/librte_ether/rte_ethdev.c b/lib/librte_ether/rte_ethdev.c
> > index a31018e..04f492d 100644
> > --- a/lib/librte_ether/rte_ethdev.c
> > +++ b/lib/librte_ether/rte_ethdev.c
> > @@ -70,7 +70,7 @@
> >  #include "rte_ethdev.h"
> >  
> >  static const char *MZ_RTE_ETH_DEV_DATA = "rte_eth_dev_data";
> > -struct rte_eth_dev rte_eth_devices[RTE_MAX_ETHPORTS];
> > +struct rte_eth_dev rte_eth_devices[RTE_MAX_ETHPORTS] __rte_cache_aligned;
> >  static struct rte_eth_dev_data *rte_eth_dev_data;
> >  static uint8_t nb_ports;
> >  
> > diff --git a/lib/librte_ether/rte_ethdev.h b/lib/librte_ether/rte_ethdev.h
> > index 8519ff6..e359dda 100644
> > --- a/lib/librte_ether/rte_ethdev.h
> > +++ b/lib/librte_ether/rte_ethdev.h
> > @@ -1630,7 +1630,7 @@ struct rte_eth_dev {
> >  	struct rte_eth_rxtx_callback *pre_tx_burst_cbs[RTE_MAX_QUEUES_PER_PORT];
> >  	uint8_t attached; /**< Flag indicating the port is attached */
> >  	enum rte_eth_dev_type dev_type; /**< Flag indicating the device type */
> > -};
> > +} __rte_cache_aligned;
> >  
> >  struct rte_eth_dev_sriov {
> >  	uint8_t active;               /**< SRIOV is active with 16, 32 or 64 pools */
> > -- 
> Adding cache aligned in two places is overkill, I think. If the structure is
> marked as being cache aligned, there is no need to have the rte_eth_devices
> marked that way too. I suggest therefore just keeping the structure alignment.

OK. Will fix it in v2

> 
> Regards,
> /Bruce

  reply	other threads:[~2016-05-03 12:11 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-05-02  8:07 Jerin Jacob
2016-05-03  9:40 ` Bruce Richardson
2016-05-03 12:10   ` Jerin Jacob [this message]
2016-05-03 12:42 ` [dpdk-dev] [PATCH v2] " Jerin Jacob
2016-05-04 11:09   ` Bruce Richardson
2016-05-04 13:42     ` Jerin Jacob
2016-05-04 13:53       ` Richardson, Bruce
2016-05-04 15:19         ` Jerin Jacob
2016-05-04 15:48           ` Bruce Richardson
2016-06-22 21:20   ` Thomas Monjalon

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=20160503121056.GA17000@localhost.localdomain \
    --to=jerin.jacob@caviumnetworks.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=thomas.monjalon@6wind.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).