DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ori Kam <orika@nvidia.com>
To: Guy Kaneti <guyk@marvell.com>
Cc: NBU-Contact-Thomas Monjalon <thomas@monjalon.net>,
	Smadar Fuks <smadarf@marvell.com>, "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v1] app/regex: change default qp configuration flags
Date: Tue, 20 Oct 2020 05:59:20 +0000	[thread overview]
Message-ID: <DM6PR12MB4987C4FE741AC203EAAC92E9D61F0@DM6PR12MB4987.namprd12.prod.outlook.com> (raw)
In-Reply-To: <DM6PR18MB2410BFC5336EE076609C591AD81E0@DM6PR18MB2410.namprd18.prod.outlook.com>



> -----Original Message-----
> From: Guy Kaneti <guyk@marvell.com>
> Sent: Monday, October 19, 2020 8:12 PM
> Subject: RE: [PATCH v1] app/regex: change default qp configuration flags
> 
> Hi Ori,
> 
> 
> > -----Original Message-----
> > From: Ori Kam <orika@nvidia.com>
> > Sent: Monday, October 19, 2020 6:03 PM
> > Subject: [EXT] RE: [PATCH v1] app/regex: change default qp configuration
> > flags
> >
> > External Email
> >
> > ----------------------------------------------------------------------
> > Hi Guy,
> >
> > > -----Original Message-----
> > > From: guyk@marvell.com <guyk@marvell.com>
> > > Sent: Monday, October 19, 2020 12:27 PM
> > > Subject: [PATCH v1] app/regex: change default qp configuration flags
> > >
> > > From: Guy Kaneti <guyk@marvell.com>
> > >
> > > By default qp configuration flags was set with
> > > RTE_REGEX_QUEUE_PAIR_CFG_OOS_F.
> > > Changed flags default to 0. In order to set OOS flag use new argument
> > > --cfg_oos to config queue pair out of order scan.
> > >
> > > Signed-off-by: Guy Kaneti <guyk@marvell.com>
> > > ---
> > >  app/test-regex/main.c | 24 +++++++++++++++++-------
> > >  1 file changed, 17 insertions(+), 7 deletions(-)
> > >
> > > diff --git a/app/test-regex/main.c b/app/test-regex/main.c index
> > > 0d35f4583..9b19031ea 100644
> > > --- a/app/test-regex/main.c
> > > +++ b/app/test-regex/main.c
> > > @@ -33,6 +33,7 @@ enum app_args {
> > >  	ARG_NUM_OF_JOBS,
> > >  	ARG_PERF_MODE,
> > >  	ARG_NUM_OF_ITERATIONS,
> > > +	ARG_CFG_OOS,
> > >  };
> > >
> > >  static void
> > > @@ -43,13 +44,15 @@ usage(const char *prog_name)
> > >  		" --data NAME: data file to use\n"
> > >  		" --nb_jobs: number of jobs to use\n"
> > >  		" --perf N: only outputs the performance data\n"
> > > -		" --nb_iter N: number of iteration to run\n",
> > > +		" --nb_iter N: number of iteration to run\n"
> > > +		" --cfg_oos: configure regexdev queue flag out of order
> > > scan\n",
> > >  		prog_name);
> > >  }
> > >
> > >  static void
> > >  args_parse(int argc, char **argv, char *rules_file, char *data_file,
> > > -	   uint32_t *nb_jobs, bool *perf_mode, uint32_t *nb_iterations)
> > > +	   uint32_t *nb_jobs, bool *perf_mode, uint32_t *nb_iterations,
> > > +	   uint32_t *qp_conf_flags)
> > >  {
> > >  	char **argvopt;
> > >  	int opt;
> > > @@ -66,7 +69,9 @@ args_parse(int argc, char **argv, char *rules_file,
> > > char *data_file,
> > >  		/* Perf test only */
> > >  		{ "perf", 0, 0, ARG_PERF_MODE},
> > >  		/* Number of iterations to run with perf test */
> > > -		{ "nb_iter", 1, 0, ARG_NUM_OF_ITERATIONS}
> > > +		{ "nb_iter", 1, 0, ARG_NUM_OF_ITERATIONS},
> > > +		/* configure out of order scan */
> > > +		{ "cfg_oos", 0, 0, ARG_CFG_OOS}
> > >  	};
> > >
> > >  	argvopt = argv;
> > > @@ -98,6 +103,9 @@ args_parse(int argc, char **argv, char *rules_file,
> > > char *data_file,
> > >  		case ARG_NUM_OF_ITERATIONS:
> > >  			*nb_iterations = atoi(optarg);
> > >  			break;
> > > +		case ARG_CFG_OOS:
> > > +			*qp_conf_flags =
> > > RTE_REGEX_QUEUE_PAIR_CFG_OOS_F;
> > > +			break;
> > >  		case ARG_HELP:
> > >  			usage("RegEx test app");
> > >  			break;
> > > @@ -158,7 +166,8 @@ read_file(char *file, char **buf)
> > >
> > >  static int
> > >  init_port(struct rte_mempool **mbuf_mp, uint32_t nb_jobs,
> > > -	  uint16_t *nb_max_payload, char *rules_file, uint8_t
> > > *nb_max_matches)
> > > +	  uint16_t *nb_max_payload, char *rules_file, uint8_t
> > > *nb_max_matches,
> > > +	  uint32_t qp_conf_flags)
> > >  {
> > >  	uint16_t id;
> > >  	uint16_t num_devs;
> > > @@ -171,7 +180,7 @@ init_port(struct rte_mempool **mbuf_mp,
> > uint32_t
> > > nb_jobs,
> > >  	};
> > >  	struct rte_regexdev_qp_conf qp_conf = {
> > >  		.nb_desc = 1024,
> > > -		.qp_conf_flags = RTE_REGEX_QUEUE_PAIR_CFG_OOS_F,
> > > +		.qp_conf_flags = qp_conf_flags,
> > >  	};
> > >  	int res = 0;
> > >
> > > @@ -407,6 +416,7 @@ main(int argc, char **argv)
> > >  	bool perf_mode = 0;
> > >  	uint32_t nb_iterations = 0;
> > >  	uint8_t nb_max_matches = 0;
> > > +	uint32_t qp_conf_flags = 0;
> > >  	int ret;
> > >
> > >  	ret = rte_eal_init(argc, argv);
> > > @@ -416,10 +426,10 @@ main(int argc, char **argv)
> > >  	argv += ret;
> > >  	if (argc > 1)
> > >  		args_parse(argc, argv, rules_file, data_file, &nb_jobs,
> > > -			   &perf_mode, &nb_iterations);
> > > +			   &perf_mode, &nb_iterations, &qp_conf_flags);
> > >
> > >  	ret = init_port(&mbuf_mp, nb_jobs, &nb_max_payload, rules_file,
> > > -			&nb_max_matches);
> > > +			&nb_max_matches, qp_conf_flags);
> > >  	if (ret < 0)
> > >  		rte_exit(EXIT_FAILURE, "init port failed\n");
> > >  	ret = run_regex(mbuf_mp, nb_jobs, nb_max_payload, perf_mode,
> > > --
> > > 2.28.0
> >
> > General comment I think we should add
> > capability to the rte_regexdev to check if the PMD support OOF. and have
> > the default based on this capability.
> 
> I agree.
> Should it be returned by rte_regexdev_info_get()
> And be defined as RTE_REGEXDEV_CAPA_ ?
> 
Yes this is my thinking.

> >
> > Best,
> > Ori

  reply	other threads:[~2020-10-20  5:59 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-10-19  9:26 [dpdk-dev] [PATCH] " guyk
2020-10-19  9:26 ` [dpdk-dev] [PATCH v1] " guyk
2020-10-19 15:03   ` Ori Kam
2020-10-19 17:12     ` Guy Kaneti
2020-10-20  5:59       ` Ori Kam [this message]
2020-10-19  9:30 ` [dpdk-dev] [PATCH] " Guy Kaneti

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=DM6PR12MB4987C4FE741AC203EAAC92E9D61F0@DM6PR12MB4987.namprd12.prod.outlook.com \
    --to=orika@nvidia.com \
    --cc=dev@dpdk.org \
    --cc=guyk@marvell.com \
    --cc=smadarf@marvell.com \
    --cc=thomas@monjalon.net \
    /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).