linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Willem de Bruijn <willemdebruijn.kernel@gmail.com>
To: Breno Leitao <leitao@debian.org>,
	 Andrew Lunn <andrew+netdev@lunn.ch>,
	 "David S. Miller" <davem@davemloft.net>,
	 Eric Dumazet <edumazet@google.com>,
	 Jakub Kicinski <kuba@kernel.org>,
	 Paolo Abeni <pabeni@redhat.com>,
	 Simon Horman <horms@kernel.org>,
	 Sebastian Andrzej Siewior <bigeasy@linutronix.de>,
	 Clark Williams <clrkwllms@kernel.org>,
	 Steven Rostedt <rostedt@goodmis.org>
Cc: netdev@vger.kernel.org,  linux-kernel@vger.kernel.org,
	 linux-rt-devel@lists.linux.dev,  kernel-team@meta.com,
	 efault@gmx.de,  calvin@wbinvd.org,
	 Breno Leitao <leitao@debian.org>
Subject: Re: [PATCH 5/7] netpoll: Move SKBs pool to netconsole side
Date: Tue, 02 Sep 2025 18:56:13 -0400	[thread overview]
Message-ID: <willemdebruijn.kernel.a0f67bb6112a@gmail.com> (raw)
In-Reply-To: <20250902-netpoll_untangle_v3-v1-5-51a03d6411be@debian.org>

Breno Leitao wrote:
> Since netconsole is the sole user of the SKBs pool within netpoll, move
> the pool management into the netconsole driver.
> 
> This change prevents other netpoll users from allocating and holding
> onto skb pool memory unnecessarily, thereby reducing memory usage when
> the pool is not required (which is all the cases except netconsole).
> 
> The skb poll struct is still attached to the netpoll, but, eventually
> this should move to the netconsole target, since it has nothing to do
> with netpoll.
> 
> Signed-off-by: Breno Leitao <leitao@debian.org>
> ---
>  drivers/net/netconsole.c | 58 ++++++++++++++++++++++++++++++++++++++++++++++--
>  net/core/netpoll.c       | 44 ------------------------------------
>  2 files changed, 56 insertions(+), 46 deletions(-)
> 
> diff --git a/drivers/net/netconsole.c b/drivers/net/netconsole.c
> index 90e359b87469a..3fe55db07cfe5 100644
> --- a/drivers/net/netconsole.c
> +++ b/drivers/net/netconsole.c
> @@ -57,6 +57,19 @@ MODULE_LICENSE("GPL");
>  #define MAX_EXTRADATA_ITEMS		16
>  #define MAX_PRINT_CHUNK			1000
>  
> +/*
> + * We maintain a small pool of fully-sized skbs, to make sure the
> + * message gets out even in extreme OOM situations.
> + */
> +
> +#define MAX_SKBS 32
> +#define MAX_UDP_CHUNK 1460
> +#define MAX_SKB_SIZE							\
> +	(sizeof(struct ethhdr) +					\
> +	 sizeof(struct iphdr) +						\
> +	 sizeof(struct udphdr) +					\
> +	 MAX_UDP_CHUNK)
> +
>  static char config[MAX_PARAM_LENGTH];
>  module_param_string(netconsole, config, MAX_PARAM_LENGTH, 0);
>  MODULE_PARM_DESC(netconsole, " netconsole=[src-port]@[src-ip]/[dev],[tgt-port]@<tgt-ip>/[tgt-macaddr]");
> @@ -172,6 +185,33 @@ struct netconsole_target {
>  	char			buf[MAX_PRINT_CHUNK];
>  };
>  
> +static void refill_skbs(struct netpoll *np)
> +{
> +	struct sk_buff_head *skb_pool;
> +	struct sk_buff *skb;
> +	unsigned long flags;
> +
> +	skb_pool = &np->skb_pool;
> +
> +	spin_lock_irqsave(&skb_pool->lock, flags);
> +	while (skb_pool->qlen < MAX_SKBS) {
> +		skb = alloc_skb(MAX_SKB_SIZE, GFP_ATOMIC);
> +		if (!skb)
> +			break;
> +
> +		__skb_queue_tail(skb_pool, skb);
> +	}
> +	spin_unlock_irqrestore(&skb_pool->lock, flags);
> +}
> +
> +static void refill_skbs_work_handler(struct work_struct *work)
> +{
> +	struct netpoll *np =
> +		container_of(work, struct netpoll, refill_wq);
> +
> +	refill_skbs(np);
> +}
> +
>  #ifdef	CONFIG_NETCONSOLE_DYNAMIC
>  
>  static struct configfs_subsystem netconsole_subsys;
> @@ -341,6 +381,20 @@ static int netpoll_parse_ip_addr(const char *str, union inet_addr *addr)
>  	return -1;
>  }
>  
> +static int setup_netpoll(struct netpoll *np)

Having both netpoll_setup and setup_netpoll is a bit confusing.
Maybe netconsole_setup_netpoll?

> +{
> +	int err;
> +
> +	err = netpoll_setup(np);
> +	if (err)
> +		return err;
> +
> +	refill_skbs(np);
> +	INIT_WORK(&np->refill_wq, refill_skbs_work_handler);
> +
> +	return 0;
> +}
> +
>  #ifdef	CONFIG_NETCONSOLE_DYNAMIC
>  
>  /*
> @@ -615,7 +669,7 @@ static ssize_t enabled_store(struct config_item *item,
>  		 */
>  		netconsole_print_banner(&nt->np);
>  
> -		ret = netpoll_setup(&nt->np);
> +		ret = setup_netpoll(&nt->np);
>  		if (ret)
>  			goto out_unlock;
>  
> @@ -2036,7 +2090,7 @@ static struct netconsole_target *alloc_param_target(char *target_config,
>  	if (err)
>  		goto fail;
>  
> -	err = netpoll_setup(&nt->np);
> +	err = setup_netpoll(&nt->np);
>  	if (err) {
>  		pr_err("Not enabling netconsole for %s%d. Netpoll setup failed\n",
>  		       NETCONSOLE_PARAM_TARGET_PREFIX, cmdline_count);

  reply	other threads:[~2025-09-02 22:56 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-09-02 14:36 [PATCH 0/7] netpoll: Untangle netpoll and netconsole Breno Leitao
2025-09-02 14:36 ` [PATCH 1/7] netconsole: Split UDP message building and sending operations Breno Leitao
2025-09-02 22:41   ` Willem de Bruijn
2025-09-02 14:36 ` [PATCH 2/7] netpoll: move prepare skb functions to netconsole Breno Leitao
2025-09-02 22:44   ` Willem de Bruijn
2025-09-02 14:36 ` [PATCH 3/7] netpoll: Move netpoll_cleanup implementation " Breno Leitao
2025-09-02 22:49   ` Willem de Bruijn
2025-09-03 16:44     ` Breno Leitao
2025-09-03 17:13       ` Willem de Bruijn
2025-09-02 14:36 ` [PATCH 4/7] netpoll: Export zap_completion_queue Breno Leitao
2025-09-02 22:50   ` Willem de Bruijn
2025-09-03 16:51     ` Breno Leitao
2025-09-03 17:16       ` Willem de Bruijn
2025-09-02 14:36 ` [PATCH 5/7] netpoll: Move SKBs pool to netconsole side Breno Leitao
2025-09-02 22:56   ` Willem de Bruijn [this message]
2025-09-02 14:36 ` [PATCH 6/7] netpoll: Move find_skb() to netconsole and make it static Breno Leitao
2025-09-02 23:07   ` Willem de Bruijn
2025-09-02 14:36 ` [PATCH 7/7] netpoll: Flush skb_pool as part of netconsole cleanup Breno Leitao
2025-09-02 23:09   ` Willem de Bruijn
2025-09-03  0:09   ` Jakub Kicinski
2025-09-03 16:55     ` Breno Leitao
2025-09-02 15:23 ` [PATCH 0/7] netpoll: Untangle netpoll and netconsole Breno Leitao

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=willemdebruijn.kernel.a0f67bb6112a@gmail.com \
    --to=willemdebruijn.kernel@gmail.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=bigeasy@linutronix.de \
    --cc=calvin@wbinvd.org \
    --cc=clrkwllms@kernel.org \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=efault@gmx.de \
    --cc=horms@kernel.org \
    --cc=kernel-team@meta.com \
    --cc=kuba@kernel.org \
    --cc=leitao@debian.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rt-devel@lists.linux.dev \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=rostedt@goodmis.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 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).