DPDK patches and discussions
 help / color / mirror / Atom feed
From: David Marchand <david.marchand@redhat.com>
To: "Power, Ciara" <ciara.power@intel.com>,
	"Richardson, Bruce" <bruce.richardson@intel.com>
Cc: "jerinj@marvell.com" <jerinj@marvell.com>,
	"dev@dpdk.org" <dev@dpdk.org>,
	 "thomas@monjalon.net" <thomas@monjalon.net>
Subject: Re: [dpdk-dev] [PATCH] telemetry: remove internal symbol from public header
Date: Wed, 5 May 2021 10:44:17 +0200	[thread overview]
Message-ID: <CAJFAV8wr+=sYtt-PD_AknN8On5+pzpWZcWec9chmvfNCvqRbKg@mail.gmail.com> (raw)
In-Reply-To: <MN2PR11MB3821E8BBE89B016837D44D1DE65A9@MN2PR11MB3821.namprd11.prod.outlook.com>

On Tue, May 4, 2021 at 6:19 PM Power, Ciara <ciara.power@intel.com> wrote:
>
> >-----Original Message-----
> >From: Richardson, Bruce <bruce.richardson@intel.com>
> >Sent: Tuesday 4 May 2021 13:14
> >To: jerinj@marvell.com
> >Cc: Power, Ciara <ciara.power@intel.com>; dev@dpdk.org;
> >thomas@monjalon.net
> >Subject: Re: [PATCH] telemetry: remove internal symbol from public header
> >
> >On Mon, May 03, 2021 at 10:04:28PM +0530, jerinj@marvell.com wrote:
> >> From: Jerin Jacob <jerinj@marvell.com>
> >>
> >> Remove TELEMETRY_MAX_CALLBACKS symbol from public rte_telemetry.h
> >> header file.
> >>
> >> Signed-off-by: Jerin Jacob <jerinj@marvell.com>
> >> ---
> >Acked-by: Bruce Richardson <bruce.richardson@intel.com>
>
> Thanks,
> Acked-by: Ciara Power <ciara.power@intel.com>

I agree this define should be hidden.

Just, what do you think of using a dynamic allocation and remove the
limitation entirely?


diff --git a/lib/telemetry/rte_telemetry.h b/lib/telemetry/rte_telemetry.h
index 031db9e968..8776998b54 100644
--- a/lib/telemetry/rte_telemetry.h
+++ b/lib/telemetry/rte_telemetry.h
@@ -10,8 +10,6 @@
 #ifndef _RTE_TELEMETRY_H_
 #define _RTE_TELEMETRY_H_

-/** Maximum number of telemetry callbacks. */
-#define TELEMETRY_MAX_CALLBACKS 64
 /** Maximum length for string used in object. */
 #define RTE_TEL_MAX_STRING_LEN 64
 /** Maximum length of string. */
@@ -285,7 +283,7 @@ typedef void * (*handler)(void *sock_id);
  * @return
  *  -EINVAL for invalid parameters failure.
  *  @return
- *  -ENOENT if max callbacks limit has been reached.
+ *  -ENOMEM for mem allocation failure.
  */
 __rte_experimental
 int
diff --git a/lib/telemetry/telemetry.c b/lib/telemetry/telemetry.c
index 68b479e0e4..6baba57ec2 100644
--- a/lib/telemetry/telemetry.c
+++ b/lib/telemetry/telemetry.c
@@ -59,7 +59,7 @@ static uint32_t logtype;
         rte_log_ptr(RTE_LOG_ ## l, logtype, "TELEMETRY: " __VA_ARGS__)

 /* list of command callbacks, with one command registered by default */
-static struct cmd_callback callbacks[TELEMETRY_MAX_CALLBACKS];
+static struct cmd_callback *callbacks;
 static int num_callbacks; /* How many commands are registered */
 /* Used when accessing or modifying list of command callbacks */
 static rte_spinlock_t callback_sl = RTE_SPINLOCK_INITIALIZER;
@@ -70,15 +70,21 @@ static uint16_t v2_clients;
 int
 rte_telemetry_register_cmd(const char *cmd, telemetry_cb fn, const char *help)
 {
+       struct cmd_callback *new_callbacks;
        int i = 0;

        if (strlen(cmd) >= MAX_CMD_LEN || fn == NULL || cmd[0] != '/'
                        || strlen(help) >= MAX_HELP_LEN)
                return -EINVAL;
-       if (num_callbacks >= TELEMETRY_MAX_CALLBACKS)
-               return -ENOENT;

        rte_spinlock_lock(&callback_sl);
+       new_callbacks = realloc(callbacks, sizeof(callbacks[0]) *
(num_callbacks + 1));
+       if (new_callbacks == NULL) {
+               rte_spinlock_unlock(&callback_sl);
+               return -ENOMEM;
+       }
+       callbacks = new_callbacks;
+
        while (i < num_callbacks && strcmp(cmd, callbacks[i].cmd) > 0)
                i++;
        if (i != num_callbacks)


And there is a race to fix in list_commands() (which accesses the
callbacks array without taking the lock).

-- 
David Marchand


  reply	other threads:[~2021-05-05  8:44 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-03 16:34 jerinj
2021-05-04 12:14 ` Bruce Richardson
2021-05-04 16:19   ` Power, Ciara
2021-05-05  8:44     ` David Marchand [this message]
2021-05-05 10:24       ` Jerin Jacob
2021-05-05 14:12         ` Power, Ciara
2021-05-05 16:15     ` David Marchand

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='CAJFAV8wr+=sYtt-PD_AknN8On5+pzpWZcWec9chmvfNCvqRbKg@mail.gmail.com' \
    --to=david.marchand@redhat.com \
    --cc=bruce.richardson@intel.com \
    --cc=ciara.power@intel.com \
    --cc=dev@dpdk.org \
    --cc=jerinj@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).