All of lore.kernel.org
 help / color / mirror / Atom feed
* missing bit for thread_info-next-to-task_struct patch
@ 2003-06-19  1:02 David Mosberger
  2003-06-19  5:00 ` Christoph Hellwig
  0 siblings, 1 reply; 3+ messages in thread
From: David Mosberger @ 2003-06-19  1:02 UTC (permalink / raw)
  To: torvalds; +Cc: linux-kernel

The attached patch is needed to really allow the thread_info to live
in the same chunk of memory as task structure.  I missed it in my last
patch because the fix was originally keyed on INIT_THREAD_SIZE, which
was wrong but resulted in a one-liner patch which was easy to miss.
The patch below is cleaner.  I suppose it would be nice if we could
get rid of INIT_THREAD_SIZE entirely, but it looks like user-mode
Linux still relies on it.

	--david

diff -Nru a/include/linux/sched.h b/include/linux/sched.h
--- a/include/linux/sched.h	Wed Jun 18 18:01:04 2003
+++ b/include/linux/sched.h	Wed Jun 18 18:01:04 2003
@@ -504,9 +509,10 @@
  */
 extern struct exec_domain	default_exec_domain;
 
-#ifndef INIT_THREAD_SIZE
-# define INIT_THREAD_SIZE	2048*sizeof(long)
-#endif
+#ifndef __HAVE_ARCH_TASK_STRUCT_ALLOCATOR
+# ifndef INIT_THREAD_SIZE
+#  define INIT_THREAD_SIZE	2048*sizeof(long)
+# endif
 
 union thread_union {
 	struct thread_info thread_info;
@@ -514,6 +520,9 @@
 };
 
 extern union thread_union init_thread_union;
+
+#endif /* !__HAVE_ARCH_TASK_STRUCT_ALLOCATOR */
+
 extern struct task_struct init_task;
 
 extern struct   mm_struct init_mm;

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

* Re: missing bit for thread_info-next-to-task_struct patch
  2003-06-19  1:02 missing bit for thread_info-next-to-task_struct patch David Mosberger
@ 2003-06-19  5:00 ` Christoph Hellwig
  2003-06-19  5:06   ` David Mosberger
  0 siblings, 1 reply; 3+ messages in thread
From: Christoph Hellwig @ 2003-06-19  5:00 UTC (permalink / raw)
  To: davidm; +Cc: torvalds, linux-kernel

On Wed, Jun 18, 2003 at 06:02:20PM -0700, David Mosberger wrote:
> +++ b/include/linux/sched.h	Wed Jun 18 18:01:04 2003
> @@ -504,9 +509,10 @@
>   */
>  extern struct exec_domain	default_exec_domain;
>  
> -#ifndef INIT_THREAD_SIZE
> -# define INIT_THREAD_SIZE	2048*sizeof(long)
> -#endif
> +#ifndef __HAVE_ARCH_TASK_STRUCT_ALLOCATOR
> +# ifndef INIT_THREAD_SIZE
> +#  define INIT_THREAD_SIZE	2048*sizeof(long)
> +# endif

This looks strange.  Either you move the ifndef INIT_THREAD_SIZE
outside the other ifdef or maybe remove it comepltly it it's
not needed otherwise..


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

* Re: missing bit for thread_info-next-to-task_struct patch
  2003-06-19  5:00 ` Christoph Hellwig
@ 2003-06-19  5:06   ` David Mosberger
  0 siblings, 0 replies; 3+ messages in thread
From: David Mosberger @ 2003-06-19  5:06 UTC (permalink / raw)
  To: Christoph Hellwig; +Cc: davidm, torvalds, linux-kernel

>>>>> On Thu, 19 Jun 2003 06:00:59 +0100, Christoph Hellwig <hch@infradead.org> said:

  Christoph> On Wed, Jun 18, 2003 at 06:02:20PM -0700, David Mosberger wrote:
  >> +++ b/include/linux/sched.h	Wed Jun 18 18:01:04 2003
  >> @@ -504,9 +509,10 @@
  >> */
  >> extern struct exec_domain	default_exec_domain;
  >> 
  >> -#ifndef INIT_THREAD_SIZE
  >> -# define INIT_THREAD_SIZE	2048*sizeof(long)
  >> -#endif
  >> +#ifndef __HAVE_ARCH_TASK_STRUCT_ALLOCATOR
  >> +# ifndef INIT_THREAD_SIZE
  >> +#  define INIT_THREAD_SIZE	2048*sizeof(long)
  >> +# endif

  Christoph> This looks strange.  Either you move the ifndef INIT_THREAD_SIZE
  Christoph> outside the other ifdef or maybe remove it comepltly it it's
  Christoph> not needed otherwise..

The INIT_THREAD_SIZE cleanup can be done as a separate patch.  I don't
want to hack (and probably break) UM for no good reason.

	--david

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

end of thread, other threads:[~2003-06-19  4:52 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-06-19  1:02 missing bit for thread_info-next-to-task_struct patch David Mosberger
2003-06-19  5:00 ` Christoph Hellwig
2003-06-19  5:06   ` David Mosberger

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.