All of lore.kernel.org
 help / color / mirror / Atom feed
From: Eric Dumazet <dada1@cosmosbay.com>
To: James Cammarata <jimi@sngx.net>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH updated] net: add ability to clear per-interface network statistics via procfs
Date: Fri, 16 May 2008 18:20:58 +0200	[thread overview]
Message-ID: <482DB46A.8020103@cosmosbay.com> (raw)
In-Reply-To: <482DA5B6.1020606@sngx.net>

James Cammarata a écrit :
> From: James Cammarata <jimi@sngx.net>
>
> Added the ability to write to /proc/net/dev in order to clear the 
> interface counters for a given interface.  The ability to zero out 
> counters (especially the error counters) is extremely useful when 
> troubleshooting interface issues.  Now diffed against 2.6.25.4, and 
> added some sanity checking.
>
Hello James

1) Every call to get_proc_net() should be paired with a call to 
put_net(), or else you  leak  struct net.

2) I am not sure this is the right way to do this... Did you consider to 
extend ethtool instead ?

> Syntax:
>  echo 'net clear-stats ifdev' > /proc/net/dev
> Where "ifdev" is the device name you wish to clear.
>
> This code is based mainly on the code found in drivers/scsi/scsi_proc.c
>
> Signed-off-by: James Cammarata <jimi@sngx.net>
> ---
>
> --- linux-2.6.25.4/net/core/dev.c       2008-05-15 10:00:12.000000000 
> -0500
> +++ linux-2.6.25.4-jcammara/net/core/dev.c      2008-05-16 
> 07:02:48.000000000 -0500
> @@ -2455,2 +2455,78 @@
>
> +/**
> + * proc_net_dev_write - handle writes to /proc/net/dev
> + * @file: not used
> + * @buf: buffer to write
> + * @length: length of buf, at most PAGE_SIZE
> + * @ppos: not used
> + *
> + * Description: this provides a mechanism to clear statistics on a
> + * per-interface basis
> + * "echo 'net clear-stats ifdev' >/proc/net/dev"
> + * with  "ifdev" replaced by the device name you wish to clear.
> + *
> + */
> +static ssize_t proc_net_dev_write(struct file *file, const char 
> __user *buf,
> +                                  size_t length, loff_t *ppos)
> +{
> +        char *buffer, *p;
> +        char devname[IFNAMSIZ];
> +        struct net *net;
> +        struct net_device *dev;
> +        int err;
> +
> +        if (!buf || length > PAGE_SIZE)
> +                return -EINVAL;
> +
> +        buffer = (char *)__get_free_page(GFP_KERNEL);
> +        if (!buffer)
> +                return -ENOMEM;
> +
> +        err = -EFAULT;
> +        if (copy_from_user(buffer, buf, length))
> +                goto out;
> +
> +        err = -EINVAL;
> +        if (length < PAGE_SIZE)
> +                buffer[length] = '\0';
> +        else if (buffer[PAGE_SIZE-1])
> +                goto out;
> +
> +        err = -ENXIO;
> +        net = get_proc_net(file->f_dentry->d_inode);
> +        if (!net)
> +                goto out;
> +
> +        /*
> +         * Usage: echo "net clear-stats ifdev" >/proc/net/dev
> +         * with  "ifdev" replaced by the device name you wish to clear.
> +         */
> +        if (!strncmp("net clear-stats",buffer,15)) {
> +                p = buffer + 16;
> +                if(sscanf(p,"%16s",devname)>0) {
> +                        dev = dev_get_by_name(net,devname);
> +                        if (dev) {
> +                                if (dev->get_stats) {
> +                                        struct net_device_stats *stats =
> +                                               dev->get_stats(dev);
> +                                        memset(stats,0,
> +                                               sizeof(struct 
> net_device_stats));
> +                                }
> +                                dev_put(dev);
> +                        }
> +                }
> +        }
> +
> +        /*
> +         * convert success returns so that we return the
> +         * number of bytes consumed.
> +         */
> +        if (!err)
> +                err = length;
> +
Here probably, you should add

    put_net(net);

> + out:
> +        free_page((unsigned long)buffer);
> +        return err;
> +}
> +
> static void *softnet_seq_start(struct seq_file *seq, loff_t *pos)
> @@ -2498,2 +2574,3 @@
>        .read    = seq_read,
> +        .write   = proc_net_dev_write,
>        .llseek  = seq_lseek,
>
>





  reply	other threads:[~2008-05-16 16:21 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2008-05-16 15:18 [PATCH updated] net: add ability to clear per-interface network statistics via procfs James Cammarata
2008-05-16 16:20 ` Eric Dumazet [this message]
2008-05-16 20:03   ` David Miller
2008-05-17 14:54   ` James Cammarata
2008-05-17 21:41     ` Eric Dumazet
2008-05-17 22:49       ` James Cammarata
2008-05-18  0:31         ` Ben Hutchings
2008-05-18  1:43           ` Patrick McHardy
2008-05-18  5:09           ` James Cammarata
2008-05-18 11:27             ` Ben Hutchings
2008-05-29  1:45             ` [PATCH] net: add ability to clear stats via ethtool - e1000/pcnet32 James Cammarata
2008-05-29  2:08               ` James Cammarata
2008-05-29  5:11               ` Andrew Morton
2008-05-29 12:34                 ` James Cammarata
2008-05-29 14:45                   ` Alan Cox
2008-05-29 17:15                     ` James Cammarata
2008-05-29 20:50                       ` David Miller
2008-05-29 21:18                         ` James Cammarata
2008-05-30 19:12                     ` Bill Fink
2008-05-30 22:14                       ` Rick Jones
2008-05-31  1:09                         ` Bill Fink
2008-05-31  2:41                           ` Stephen Hemminger
2008-05-31  2:41                             ` Stephen Hemminger
2008-05-31  4:47                             ` Bill Fink
2008-05-31 12:11                       ` Alan Cox
2008-05-31 23:57                         ` Bill Fink
2008-06-01  1:46                           ` Ben Hutchings
2008-06-01 20:46                             ` Bill Fink
2008-06-01 22:29                               ` Ben Hutchings
2008-06-02  3:55                                 ` Bill Fink
2008-06-02  5:39                               ` David Miller
2008-06-02 15:41                                 ` Bill Fink
2008-06-02  4:50                           ` Glen Turner
2008-06-02 16:10                             ` Bill Fink
2008-06-03 12:28                               ` James Cammarata
2008-06-03 12:35                                 ` Ben Hutchings
2008-06-03 14:46                                 ` Alan Cox
2008-06-04  3:05                                   ` James Cammarata
2008-05-29 14:48                   ` Chris Friesen
2008-05-16 20:00 ` [PATCH updated] net: add ability to clear per-interface network statistics via procfs David Miller
2008-05-16 20:09   ` Rick Jones
2008-05-17 15:06     ` James Cammarata

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=482DB46A.8020103@cosmosbay.com \
    --to=dada1@cosmosbay.com \
    --cc=jimi@sngx.net \
    --cc=linux-kernel@vger.kernel.org \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.