lkml.org 
[lkml]   [2000]   [Nov]   [30]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Linux 2.2.18pre24
On Thu, Nov 30, 2000 at 11:26:43AM -0700, Tom Rini wrote:
> Right. But the problem here was a new, unused sysctl-by-number, conflicted
> with an old-but-not-integrated sysctl-by-number that is used. :) The only

Who is using it? Not even the raid developers cared to take the
sysctl-by-number consistent between 2.4.0-test12-pre2 and 2.2.x raid 0.90
so nobody should be using it in first place.

Furthmore since the number 4 is the official one for raid/md, DEV_MAC_HID=3
isn't really colliding with the raid sysctl, but DEV_MAC_HID=3 is still wrong
because is it should be =5 to be consistent with 2.4.x...

2.2.x RAID 0.90:

enum {
DEV_CDROM=1,
- DEV_HWMON=2
+ DEV_HWMON=2,
+ DEV_MD=3
};
[..]
+/* /proc/sys/dev/md */
+enum {
+ DEV_MD_SPEED_LIMIT=1
};


2.2.18pre24:

enum {
DEV_CDROM=1,
DEV_HWMON=2,
DEV_MAC_HID=3
};

2.4.0-test12-pre2:

enum {
DEV_CDROM=1,
DEV_HWMON=2,
DEV_PARPORT=3,
DEV_RAID=4,
DEV_MAC_HID=5
};
[..]
/* /proc/sys/dev/raid */
enum {
DEV_RAID_SPEED_LIMIT_MIN=1,
DEV_RAID_SPEED_LIMIT_MAX=2
};

As we can clearly see nobody cares about the sysctl-by-number interface because
it generates collisions too easily so it should be declared obsolete and nobody
should use it anymore. sysctl-by-name is less performant but it doesn't
generate binary-level collisions so easily and in turn it's a big win for open
source projects where everybody has some tons of unofficial patches applied
(raid 0.90 in this case).

Andrea
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 12:47    [W:0.051 / U:1.176 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site