netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Eric Dumazet <dada1@cosmosbay.com>
To: Stephen Hemminger <stephen.hemminger@vyatta.com>
Cc: David Miller <davem@davemloft.net>,
	Robert Olsson <robert.olsson@its.uu.se>,
	netdev@vger.kernel.org
Subject: Re: [PATCH 9/9] fix sparse warnings
Date: Sat, 12 Jan 2008 12:16:13 +0100	[thread overview]
Message-ID: <4788A17D.5070903@cosmosbay.com> (raw)
In-Reply-To: <20080112064646.659443238@linux-foundation.org>

[-- Attachment #1: Type: text/plain, Size: 1541 bytes --]

Stephen Hemminger a écrit :
> Make FIB TRIE go through sparse checker without warnings.
> 
> Signed-off-by: Stephen Hemminger <stephen.hemminger@vyatta.com>

Hi Stephen

While reviewing your patches (and fib code) I had some questions :

1) I was wondering isn't trie_collect_stats() a potential cpu hog
(big latency) ?

2) struct tnode layout
    If tnode->bits is large enough, we allocate a big area
    of memory but roughly use only first half of it.
    We could use a better scheme with an extra indirection. For small
    nodes, we use space right after tnode, but for big nodes, we allocate
    a power of two set of pages, to exactly match the memory need.

3) 'pos' and 'bits' fields of 'struct tnode' might be converted to
    plain uchar, instead of 5-bits fields, to reduce complexity for
    generated code.

4) full_children & empty_children being 'unsigned short',
    we probably are limited to 2^15 elements, but I could not
    find this limit enforced somewhere.

[FIB]: Reduce text size of net/ipv4/fib_trie.o

In struct tnode, we use two fields of 5 bits for 'pos' and 'bits'.
Switching to plain 'unsigned char' (8 bits) take the same space
because of compiler alignments, and reduce text size by 435 bytes
on i386.

On i386 :
$ size net/ipv4/fib_trie.o.before_patch net/ipv4/fib_trie.o
    text    data     bss     dec     hex filename
   13714       4      64   13782    35d6 net/ipv4/fib_trie.o.before
   13279       4      64   13347    3423 net/ipv4/fib_trie.o

Signed-off-by: Eric Dumazet <dada1@cosmosbay.com>


[-- Attachment #2: fib_trie.patch --]
[-- Type: text/plain, Size: 605 bytes --]

diff --git a/net/ipv4/fib_trie.c b/net/ipv4/fib_trie.c
index 2832610..4e91532 100644
--- a/net/ipv4/fib_trie.c
+++ b/net/ipv4/fib_trie.c
@@ -119,8 +119,8 @@ struct leaf_info {
 struct tnode {
 	t_key key;
 	unsigned long parent;
-	unsigned short pos:5;		/* 2log(KEYLENGTH) bits needed */
-	unsigned short bits:5;		/* 2log(KEYLENGTH) bits needed */
+	unsigned char pos;		/* 2log(KEYLENGTH) bits needed */
+	unsigned char bits;		/* 2log(KEYLENGTH) bits needed */
 	unsigned short full_children;	/* KEYLENGTH bits needed */
 	unsigned short empty_children;	/* KEYLENGTH bits needed */
 	struct rcu_head rcu;

  parent reply	other threads:[~2008-01-12 11:16 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20080112064513.803976049@linux-foundation.org>
     [not found] ` <20080112064646.056241123@linux-foundation.org>
2008-01-13  4:49   ` [PATCH 1/9] get rid of trie_init David Miller
     [not found] ` <20080112064646.132747871@linux-foundation.org>
2008-01-13  4:50   ` [PATCH 2/9] get rid of unused revision element David Miller
2008-01-14 11:44     ` Robert Olsson
2008-01-14 12:06       ` David Miller
2008-01-14 16:35         ` Stephen Hemminger
2008-01-15  7:07           ` David Miller
     [not found] ` <20080112064646.207183428@linux-foundation.org>
2008-01-13  4:53   ` [PATCH 3/9] move size information to pr_debug() David Miller
     [not found] ` <20080112064646.282104074@linux-foundation.org>
2008-01-13  4:55   ` [PATCH 4/9] statistics improvements David Miller
2008-01-13  5:33     ` Stephen Hemminger
2008-01-13  5:44       ` David Miller
2008-01-14 20:57         ` [PATCH] [IPV4] fib_trie: size and statistics Stephen Hemminger
     [not found]           ` <20080114164450.55f8c9b2@deepthought>
2008-01-15  0:46             ` [PATCH 3/6] [IPV4] trie: put leaf nodes in a slab cache Stephen Hemminger
2008-01-15  0:47               ` [PATCH 4/6] [IPV4] fib_trie style cleanup Stephen Hemminger
2008-01-15  2:58                 ` [PATCH 5/6] [IPV4] fib_trie: checkleaf calling convention Stephen Hemminger
2008-01-15  5:07                   ` [RFC 6/6] fib_trie: combine leaf and info Stephen Hemminger
2008-01-15  6:12                     ` Eric Dumazet
2008-01-15  6:16                       ` Eric Dumazet
2008-01-15 16:19                         ` Stephen Hemminger
2008-01-15 16:44                           ` Robert Olsson
2008-01-15 17:25                             ` Eric Dumazet
2008-01-15 17:47                               ` Stephen Hemminger
2008-01-15 18:10                                 ` Eric Dumazet
2008-01-15 18:15                                   ` Stephen Hemminger
2008-01-15 18:32                                     ` Eric Dumazet
2008-01-15 20:18                                 ` Robert Olsson
2008-01-15 21:16                                   ` Eric Dumazet
2008-01-15 17:59                               ` Robert Olsson
2008-01-15  6:49               ` [PATCH 3/6] [IPV4] trie: put leaf nodes in a slab cache Eric Dumazet
2008-01-15  7:29               ` David Miller
2008-01-15  5:00           ` [PATCH 2/6] [IPV4] fib hash|trie initialization Stephen Hemminger
2008-01-15  7:14             ` David Miller
2008-01-15  6:55           ` [PATCH] [IPV4] fib_trie: size and statistics Eric Dumazet
2008-01-15  7:28             ` David Miller
2008-01-15  7:12           ` David Miller
     [not found] ` <20080112064646.356466158@linux-foundation.org>
2008-01-13  4:56   ` [PATCH 5/9] use %u for unsigned printfs David Miller
     [not found] ` <20080112064646.432200237@linux-foundation.org>
2008-01-13  4:57   ` [PATCH 6/9] : fib_insert_node cleanup David Miller
     [not found] ` <20080112064646.507015655@linux-foundation.org>
2008-01-13  4:58   ` [PATCH 7/9] printk related cleanups David Miller
     [not found] ` <20080112064646.583836190@linux-foundation.org>
2008-01-13  5:23   ` [PATCH 8/9] add statistics David Miller
     [not found] ` <20080112064646.659443238@linux-foundation.org>
2008-01-12 11:16   ` Eric Dumazet [this message]
2008-01-12 11:28     ` [PATCH 9/9] fix sparse warnings David Miller
2008-01-12 21:08     ` Stephen Hemminger
2008-01-14 11:07       ` Robert Olsson
2008-01-14 17:34         ` Eric Dumazet
2008-01-14 17:59           ` Robert Olsson
2008-01-14 19:27             ` [FIB]: Avoid using static variables without proper locking Eric Dumazet
2008-01-15  7:10               ` David Miller
2008-01-12 21:09     ` [PATCH 9/9] fix sparse warnings Stephen Hemminger
2008-01-13  5:28       ` David Miller
2008-01-13 18:30     ` [FIB]: full_children & empty_children should be uint, not ushort Eric Dumazet
2008-01-13 22:02       ` Robert Olsson
2008-01-14  6:32         ` David Miller
2008-01-13  5:25   ` [PATCH 9/9] fix sparse warnings David Miller

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=4788A17D.5070903@cosmosbay.com \
    --to=dada1@cosmosbay.com \
    --cc=davem@davemloft.net \
    --cc=netdev@vger.kernel.org \
    --cc=robert.olsson@its.uu.se \
    --cc=stephen.hemminger@vyatta.com \
    /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).