lkml.org 
[lkml]   [1999]   [Nov]   [18]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: [patch] zoned-2.3.28-K4

[ Ok, back from comdex, looking through my horrible backlog of email.. ]

On Wed, 17 Nov 1999, Ingo Molnar wrote:
>
> the latest patchset is at:
>
> http://www.redhat.com/~mingo/zoned-2.3.28-K4

Ugh. I don't like some of this:

- the "task->refcount" thing is just silly, considering that any
architecture that uses pages for task allocation would have been better
off just using the memory management refcount like we used to.

Why not do this by just doing a per-architecture "get_task_struct()"
and "put_task_struct()", to go with the already existing "alloc" and
"free" functionality?

Thus, on the 99% of all machines where the page allocator already does
the work for us, we'd just have

#define get_task_struct(tsk) \
atomic_inc(&mem_map[MAP_NR(tsk)].count)
#define put_task_struct(tsk) \
free_task_struct(tsk)

which is just obviously right and cleans up the need to have that silly
refcount. Then ARM can just have a refcount in its own definition of
get_task_struct() and friends in task->thread.refcount..

- I don't see the reason for renaming page_address() to page_vaddr()?

- "flush_cache_all()" etc in "pgalloc.h"? Senseless. I can see pgalloc.h
containing the page table allocation routines. But why did you move the
other stuff there? Looks like a cut-and-paste bug to me. Why a new
header file at all?

Would you mind explaining or re-doing?

Linus


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

\
 
 \ /
  Last update: 2005-03-22 13:55    [W:0.077 / U:0.160 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site