netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH 4/5] unix_diag: Report memory usage info
@ 2011-12-28 13:39 Pavel Emelyanov
  2011-12-28 18:55 ` Pavel Emelyanov
  0 siblings, 1 reply; 3+ messages in thread
From: Pavel Emelyanov @ 2011-12-28 13:39 UTC (permalink / raw)
  To: David Miller, Linux Netdev List

Symmetrical to inet sockets, but unix only have two meaningful
values -- the wmem allocation and sndbuf.

Signed-off-by: Pavel Emelyanov <xemul@parallels.com>

---
 include/linux/unix_diag.h |    7 +++++++
 net/unix/diag.c           |   18 ++++++++++++++++++
 2 files changed, 25 insertions(+), 0 deletions(-)

diff --git a/include/linux/unix_diag.h b/include/linux/unix_diag.h
index 3f7afb0..d62e562 100644
--- a/include/linux/unix_diag.h
+++ b/include/linux/unix_diag.h
@@ -16,6 +16,7 @@ struct unix_diag_req {
 #define UDIAG_SHOW_PEER		0x00000004	/* show peer socket info */
 #define UDIAG_SHOW_ICONS	0x00000008	/* show pending connections */
 #define UDIAG_SHOW_RQLEN	0x00000010	/* show skb receive queue len */
+#define UDIAG_SHOW_MEMINFO	0x00000020
 
 struct unix_diag_msg {
 	__u8	udiag_family;
@@ -33,6 +34,7 @@ enum {
 	UNIX_DIAG_PEER,
 	UNIX_DIAG_ICONS,
 	UNIX_DIAG_RQLEN,
+	UNIX_DIAG_MEMINFO,
 
 	UNIX_DIAG_MAX,
 };
@@ -42,4 +44,9 @@ struct unix_diag_vfs {
 	__u32	udiag_vfs_dev;
 };
 
+struct unix_diag_meminfo {
+	__u32	udiag_len;
+	__u32	udiag_limit;
+};
+
 #endif
diff --git a/net/unix/diag.c b/net/unix/diag.c
index c5bdbcb..dbae8a4 100644
--- a/net/unix/diag.c
+++ b/net/unix/diag.c
@@ -108,6 +108,20 @@ rtattr_failure:
 	return -EMSGSIZE;
 }
 
+static int sk_diag_show_meminfo(struct sock *sk, struct sk_buff *nlskb)
+{
+	struct unix_diag_meminfo *mi;
+
+	mi = UNIX_DIAG_PUT(nlskb, UNIX_DIAG_MEMINFO, sizeof(*mi));
+	mi->udiag_len = sk_wmem_alloc_get(sk);
+	mi->udiag_limit = sk->sk_sndbuf;
+
+	return 0;
+
+rtattr_failure:
+	return -EMSGSIZE;
+}
+
 static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, struct unix_diag_req *req,
 		u32 pid, u32 seq, u32 flags, int sk_ino)
 {
@@ -146,6 +160,10 @@ static int sk_diag_fill(struct sock *sk, struct sk_buff *skb, struct unix_diag_r
 			sk_diag_show_rqlen(sk, skb))
 		goto nlmsg_failure;
 
+	if ((req->udiag_show & UDIAG_SHOW_MEMINFO) &&
+			sk_diag_show_meminfo(sk, skb))
+		goto nlmsg_failure;
+
 	nlh->nlmsg_len = skb_tail_pointer(skb) - b;
 	return skb->len;
 
-- 
1.5.5.6

^ permalink raw reply related	[flat|nested] 3+ messages in thread

* Re: [PATCH 4/5] unix_diag: Report memory usage info
  2011-12-28 13:39 [PATCH 4/5] unix_diag: Report memory usage info Pavel Emelyanov
@ 2011-12-28 18:55 ` Pavel Emelyanov
  2011-12-28 19:00   ` David Miller
  0 siblings, 1 reply; 3+ messages in thread
From: Pavel Emelyanov @ 2011-12-28 18:55 UTC (permalink / raw)
  To: David Miller, Linux Netdev List

On 12/28/2011 05:39 PM, Pavel Emelyanov wrote:
> Symmetrical to inet sockets, but unix only have two meaningful
> values -- the wmem allocation and sndbuf.
> 
> Signed-off-by: Pavel Emelyanov <xemul@parallels.com>

David, taking into account you don't apply patch #3, please, ignore this and the 5th
ones (but your feedback about whether reporting the ack backlog info in the same place
whith the meminfo is still required). I will reimplement them according to the inet
meminfo rework.

Thanks,
Pavel

^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: [PATCH 4/5] unix_diag: Report memory usage info
  2011-12-28 18:55 ` Pavel Emelyanov
@ 2011-12-28 19:00   ` David Miller
  0 siblings, 0 replies; 3+ messages in thread
From: David Miller @ 2011-12-28 19:00 UTC (permalink / raw)
  To: xemul; +Cc: netdev

From: Pavel Emelyanov <xemul@parallels.com>
Date: Wed, 28 Dec 2011 22:55:59 +0400

> On 12/28/2011 05:39 PM, Pavel Emelyanov wrote:
>> Symmetrical to inet sockets, but unix only have two meaningful
>> values -- the wmem allocation and sndbuf.
>> 
>> Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
> 
> David, taking into account you don't apply patch #3, please, ignore this and the 5th
> ones

I always toss the entire set of patches when any of the individual
ones need changes.

> (but your feedback about whether reporting the ack backlog info in the same place
> whith the meminfo is still required)

I don't understand what you need to know from me.

The sk_max_ack_backlog needs to stay where it is right now, and I would advise
that you present it in the same way TCP does for other protocols.

What else do you need to know?

^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2011-12-28 19:00 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-12-28 13:39 [PATCH 4/5] unix_diag: Report memory usage info Pavel Emelyanov
2011-12-28 18:55 ` Pavel Emelyanov
2011-12-28 19:00   ` David Miller

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).