[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <1332236676.7847.83.camel@joe2Laptop>
Date: Tue, 20 Mar 2012 02:44:36 -0700
From: Joe Perches <joe@...ches.com>
To: Jiri Slaby <jslaby@...e.cz>
Cc: Valdis.Kletnieks@...edu, tytso@....edu, anca.emanuel@...il.com,
adilger.kernel@...ger.ca, linux-ext4@...r.kernel.org,
David Miller <davem@...emloft.net>,
linux-kernel@...r.kernel.org
Subject: Re: [PATCH 2/9] ext4: Use pr_fmt and pr_<level>
On Tue, 2012-03-20 at 09:47 +0100, Jiri Slaby wrote:
> Joe writes: "notify a particular set of per subsystem messages that
> pr_<level> could easily provide"
>
> Maybe the concept is not explained well enough that we do not follow?
Basically, when pr_fmt(fmt) is #defined to something
other than "#define pr_fmt(fmt) fmt", it could be used
as a mechanism to do more than standardize prefixes.
__builtin_strcmp is pretty useful.
You can do things like:
(I'm just typing, this may have errors)
#define _first_macro_arg(arg1) arg1
#define first_macro_arg(arg1, ...) _first_macro_arg(arg1)
#define pr_info(fmt, ...) \
do { \
if (!__builtin_strcmp(first_macro_arg(pr_fmt(fmt)), \
KBUILD_MODNAME ": " fmt))) \
advanced_printk(fmt, ##__VA_ARGS__); \
else \
printk(KERN_INFO pr_fmt(fmt), ##__VA_ARGS__); \
while (0)
etc...
Anyway, it's been a couple of years since I prototyped
something like that and tested file sizes.
The size delta of an allyesconfig was:
-rwxr-xr-x 1 joe joe 926968471 2010-06-29 10:39 vmlinux.allyesconfig.printk_level_with_module
-rwxr-xr-x 1 joe joe 927196127 2010-06-23 02:40 vmlinux.allyesconfig.pr_level
It reduced an allyesconfig by .1%
It wasn't a huge saving.
I'll play with it again and send an rfc
patch in a couple of days.
Maybe I'll prototype a pr_notify mechanism too.
cheers, Joe
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists