Skip to contentSkip to conversation

rename the ksoftirqd kernel thread.

DISCUSSION Martin Hicks 2 participants 3 messages May 15, 2003 – May 16, 2003

lore
Share & tools
lore
#0 lore raw

Marcelo,

Please consider this patch for 2.4.22-pre

It just renames the ksoftirqd kernel thread to be the same as in 2.5.

The side effect is that on machines with > 100 processors the last number in the thread name doesn't get truncated.

The patch is against linux-2.4 bk.

thanks
mh

Signature · 28 lines
Wild Open Source Inc.                  [email protected]


# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
#	           ChangeSet	1.1210  -> 1.1211 
#	    kernel/softirq.c	1.11    -> 1.12   
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/05/15	[email protected]	1.1211
# Rename the ksoftirqd thread to be the same as in 2.5.
# --------------------------------------------
#
diff -Nru a/kernel/softirq.c b/kernel/softirq.c
--- a/kernel/softirq.c	Thu May 15 17:13:08 2003
+++ b/kernel/softirq.c	Thu May 15 17:13:08 2003
@@ -372,7 +372,7 @@
 	while (smp_processor_id() != cpu)
 		schedule();
 
-	sprintf(current->comm, "ksoftirqd_CPU%d", bind_cpu);
+	sprintf(current->comm, "ksoftirqd/%d", bind_cpu);
 
 	__set_current_state(TASK_INTERRUPTIBLE);
 	mb();
#1 lore raw
Quoted 36 lines · Martin Hicks

Marcelo,

Please consider this patch for 2.4.22-pre

It just renames the ksoftirqd kernel thread to be the same as in 2.5.

The side effect is that on machines with > 100 processors the last number in the thread name doesn't get truncated.

The patch is against linux-2.4 bk.

thanks
mh

Signature · 17 lines
Wild Open Source Inc.                  [email protected]


# This is a BitKeeper generated patch for the following project:
# Project Name: Linux kernel tree
# This patch format is intended for GNU patch command version 2.5 or higher.
# This patch includes the following deltas:
#	           ChangeSet	1.1210  -> 1.1211 
#	    kernel/softirq.c	1.11    -> 1.12   
#
# The following is the BitKeeper ChangeSet Log
# --------------------------------------------
# 03/05/15	[email protected]	1.1211
# Rename the ksoftirqd thread to be the same as in 2.5.
# --------------------------------------------
#
diff -Nru a/kernel/softirq.c b/kernel/softirq.c

kernel/softirq.c line 378

Quoted 8 lines
kernel/softirq.c +1 -1
@@ -372,7 +372,7 @@
 	while (smp_processor_id() != cpu)
 		schedule();
 
-	sprintf(current->comm, "ksoftirqd_CPU%d", bind_cpu);
+	sprintf(current->comm, "ksoftirqd/%d", bind_cpu);
 
 	__set_current_state(TASK_INTERRUPTIBLE);
 	mb();

Standard Linux 2.4 only supports 32 CPUS (include/linux/threads.h). Wouldn't be useful to format it as %0.2d ?
Even in -aa, that supports 64 in 64-bits arches, it would be enough and you get rid of the jump from _CPU9 to _CPU10.

Signature · 4 lines
J.A. Magallon <[email protected]>      \                 Software is like sex:
werewolf.able.es                         \           It's better when it's free
Mandrake Linux release 9.2 (Cooker) for i586
Linux 2.4.21-rc2-jam1 (gcc 3.2.3 (Mandrake Linux 9.2 3.2.3-1mdk))
#2 lore raw
Quoted 5 lines · J.A. Magallon

Standard Linux 2.4 only supports 32 CPUS (include/linux/threads.h). Wouldn't be useful to format it as %0.2d ?
Even in -aa, that supports 64 in 64-bits arches, it would be enough and you get rid of the jump from _CPU9 to _CPU10.

I agree that currently there is only support for sizeof(long) CPUS. Including this small fix will make the diff to run > 64 CPUS smaller and will increase consistency between 2.4 and 2.5.

There are machines out there that are running 128 processors on 2.4

mh

Signature · 1 line
Wild Open Source Inc.                  [email protected]