From: fengchengwen <fengchengwen@huawei.com>
To: Bruce Richardson <bruce.richardson@intel.com>
Cc: <thomas@monjalon.net>, <ferruh.yigit@amd.com>,
<andrew.rybchenko@oktetlabs.ru>, <dev@dpdk.org>,
<ciara.power@intel.com>, <kevin.laatz@intel.com>
Subject: Re: [PATCH v2 5/5] ethdev: telemetry xstats support hide zero
Date: Fri, 20 Jan 2023 11:51:50 +0800 [thread overview]
Message-ID: <96a22fa7-be51-70b2-188f-c9ac3803aee3@huawei.com> (raw)
In-Reply-To: <Y77C+A9k8AZLeEY9@bricha3-MOBL.ger.corp.intel.com>
Hi Bruce,
On 2023/1/11 22:08, Bruce Richardson wrote:
> On Wed, Jan 11, 2023 at 12:06:30PM +0000, Chengwen Feng wrote:
>> The number of xstats may be large, after the hide zero option is added,
>> only non-zero values can be displayed.
>>
>> Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
>> ---
>> lib/ethdev/rte_ethdev.c | 28 ++++++++++++++++++++++------
>> 1 file changed, 22 insertions(+), 6 deletions(-)
>>
>> diff --git a/lib/ethdev/rte_ethdev.c b/lib/ethdev/rte_ethdev.c
>> index 2fc593b865..77cacc0829 100644
>> --- a/lib/ethdev/rte_ethdev.c
>> +++ b/lib/ethdev/rte_ethdev.c
>> @@ -5870,20 +5870,33 @@ eth_dev_handle_port_xstats(const char *cmd __rte_unused,
>> {
>> struct rte_eth_xstat *eth_xstats;
>> struct rte_eth_xstat_name *xstat_names;
>> + char *end_param, *hide_param;
>> int port_id, num_xstats;
>> + int hide_zero = 0;
>> int i, ret;
>> - char *end_param;
>>
>> if (params == NULL || strlen(params) == 0 || !isdigit(*params))
>> return -1;
>>
>> port_id = strtoul(params, &end_param, 0);
>> - if (*end_param != '\0')
>> - RTE_ETHDEV_LOG(NOTICE,
>> - "Extra parameters passed to ethdev telemetry command, ignoring\n");
>> if (!rte_eth_dev_is_valid_port(port_id))
>> return -1;
>>
>> + if (*end_param != '\0') {
>> + hide_param = strtok(end_param, ",");
>> + if (!hide_param || strlen(hide_param) == 0 || !isdigit(*hide_param))
>> + return -EINVAL;
>> + hide_zero = strtoul(hide_param, &end_param, 0);
>> + if (*end_param != '\0')
>> + RTE_ETHDEV_LOG(NOTICE,
>> + "Extra parameters passed to ethdev telemetry command, ignoring\n");
>> + if (hide_zero != 0 && hide_zero != 1) {
>> + hide_zero = !!hide_zero;
>> + RTE_ETHDEV_LOG(NOTICE,
>> + "Hide zero parameter is non-boolean, cast to boolean\n");
>> + }
>> + }
>> +
>
> I'm not sure about this adding of an extra flag as a 0/1 value. That would
> seem to make it confusing with a queue number or extra port number. For
> such an optional parameter, I think a string value would be clearer. A
> number of alternatives I'd suggest - in order of my preference (least
> preferred to most preferred):
>
> * have the extra parameter as a literal string "hide_zeros" which is
> matched against rather than looking for 0/1, or alternatively
add string "hide_zeros" requires more characters, it may not user friendly.
>
> * add a new command /ethdev/xstats_nonzero which does this, the same
> callback can be reusued, just checking the command given, or finally,
The new command may cause confusion.
>
> * if this is primarily for the benefit of users using the telemetry script
> to interactively view stats, then the functionality could be implemented
> in the script itself rather than in the backend. We could add some
> setting, or extra flag to the display code, to possibly filter out zeros
> in the display.
This may causes the display logic to be coupled with the command name.
>
> Thoughts?
I prefer add optional parameter, and the optional parameter should be simpler, just a number here.
And for clearer expression, I reword the help string in V4:
Returns the extended stats for a port. Parameters: int port_id, bool hide_zero (Optional, non-zero indicates hide zero xstats)
Thanks.
>
> /Bruce
> .
>
next prev parent reply other threads:[~2023-01-20 3:51 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-19 9:07 [PATCH v5 0/5] support dmadev/ethdev stats reset Chengwen Feng
2022-12-19 9:07 ` [PATCH v5 1/5] dmadev: support stats reset telemetry command Chengwen Feng
2022-12-19 9:07 ` [PATCH v5 2/5] telemetry: fix repeated display when callback don't set dict Chengwen Feng
2022-12-19 9:33 ` Bruce Richardson
2022-12-26 4:53 ` fengchengwen
2023-01-06 16:07 ` Bruce Richardson
2023-01-06 17:33 ` Bruce Richardson
2023-01-11 12:38 ` fengchengwen
2022-12-19 9:07 ` [PATCH v5 3/5] ethdev: support xstats reset telemetry command Chengwen Feng
2022-12-19 9:07 ` [PATCH v5 4/5] ethdev: telemetry xstats support hide zero Chengwen Feng
2022-12-19 9:07 ` [PATCH v5 5/5] ethdev: add newline to telemetry log string Chengwen Feng
2022-12-26 4:55 ` [PATCH v5 0/5] support dmadev/ethdev stats reset fengchengwen
2023-01-11 12:02 ` [PATCH v2 " Chengwen Feng
2023-01-11 12:02 ` [PATCH v2 1/5] dmadev: support stats reset telemetry command Chengwen Feng
2023-01-11 12:02 ` [PATCH v2 2/5] telemetry: fix repeat display when callback don't set dict Chengwen Feng
2023-01-11 12:02 ` [PATCH v2 3/5] ethdev: add newline to telemetry log string Chengwen Feng
2023-01-11 12:02 ` [PATCH v2 4/5] ethdev: support xstats reset telemetry command Chengwen Feng
2023-01-11 12:02 ` [PATCH v2 5/5] ethdev: telemetry xstats support hide zero Chengwen Feng
2023-01-11 12:06 ` [PATCH v2 0/5] support dmadev/ethdev stats reset Chengwen Feng
2023-01-11 12:06 ` [PATCH v2 1/5] dmadev: support stats reset telemetry command Chengwen Feng
2023-01-11 13:58 ` Bruce Richardson
2023-01-11 12:06 ` [PATCH v2 2/5] telemetry: fix repeat display when callback don't set dict Chengwen Feng
2023-01-11 14:01 ` Bruce Richardson
2023-01-11 12:06 ` [PATCH v2 3/5] ethdev: add newline to telemetry log string Chengwen Feng
2023-01-11 12:06 ` [PATCH v2 4/5] ethdev: support xstats reset telemetry command Chengwen Feng
2023-01-11 12:06 ` [PATCH v2 5/5] ethdev: telemetry xstats support hide zero Chengwen Feng
2023-01-11 14:08 ` Bruce Richardson
2023-01-20 3:51 ` fengchengwen [this message]
2023-01-11 12:36 ` [PATCH v2 0/5] support dmadev/ethdev stats reset fengchengwen
2023-01-20 3:25 ` [PATCH v3 " Chengwen Feng
2023-01-20 3:25 ` [PATCH v3 1/5] dmadev: support stats reset telemetry command Chengwen Feng
2023-01-20 3:25 ` [PATCH v3 2/5] telemetry: fix repeat display when callback don't init dict Chengwen Feng
2023-01-20 3:25 ` [PATCH v3 3/5] ethdev: add newline to telemetry log string Chengwen Feng
2023-01-20 3:25 ` [PATCH v3 4/5] ethdev: support xstats reset telemetry command Chengwen Feng
2023-01-20 3:25 ` [PATCH v3 5/5] ethdev: telemetry xstats support hide zero Chengwen Feng
2023-01-20 3:34 ` [PATCH v4 0/5] support dmadev/ethdev stats reset Chengwen Feng
2023-01-20 3:34 ` [PATCH v4 1/5] dmadev: support stats reset telemetry command Chengwen Feng
2023-01-20 3:34 ` [PATCH v4 2/5] telemetry: fix repeat display when callback don't init dict Chengwen Feng
2023-02-08 14:15 ` Bruce Richardson
2023-02-09 1:33 ` fengchengwen
2023-01-20 3:34 ` [PATCH v4 3/5] ethdev: add newline to telemetry log string Chengwen Feng
2023-01-20 3:34 ` [PATCH v4 4/5] ethdev: support xstats reset telemetry command Chengwen Feng
2023-01-20 3:34 ` [PATCH v4 5/5] ethdev: telemetry xstats support hide zero Chengwen Feng
2023-02-06 8:39 ` [PATCH v4 0/5] support dmadev/ethdev stats reset Thomas Monjalon
2023-02-08 14:17 ` Bruce Richardson
2023-02-09 1:45 ` fengchengwen
2023-02-09 2:32 ` [PATCH v5 0/2] " Chengwen Feng
2023-02-09 2:32 ` [PATCH v5 1/2] dmadev: support stats reset telemetry command Chengwen Feng
2023-02-09 2:32 ` [PATCH v5 2/2] ethdev: support xstats " Chengwen Feng
2023-02-15 3:19 ` Dongdong Liu
2023-02-16 11:53 ` fengchengwen
2023-02-16 12:06 ` Ferruh Yigit
2023-02-16 12:42 ` fengchengwen
2023-02-16 12:54 ` Bruce Richardson
2023-02-16 12:55 ` Bruce Richardson
2023-02-17 9:44 ` fengchengwen
2023-02-20 13:05 ` Thomas Monjalon
2023-07-03 3:58 ` fengchengwen
2023-07-03 7:20 ` Thomas Monjalon
2023-07-03 13:44 ` Morten Brørup
2023-07-04 6:41 ` fengchengwen
2024-10-03 18:51 ` Stephen Hemminger
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=96a22fa7-be51-70b2-188f-c9ac3803aee3@huawei.com \
--to=fengchengwen@huawei.com \
--cc=andrew.rybchenko@oktetlabs.ru \
--cc=bruce.richardson@intel.com \
--cc=ciara.power@intel.com \
--cc=dev@dpdk.org \
--cc=ferruh.yigit@amd.com \
--cc=kevin.laatz@intel.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).