public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Nick Piggin <nickpiggin@yahoo.com.au>
To: Andrew Morton <akpm@osdl.org>,
	linux-kernel <linux-kernel@vger.kernel.org>
Subject: [PATCH 1/2] fix PID hash sizing
Date: Sun, 22 Aug 2004 14:44:46 +1000	[thread overview]
Message-ID: <412824BE.4040801@yahoo.com.au> (raw)

[-- Attachment #1: Type: text/plain, Size: 89 bytes --]

I see PID hash sizing problems on an Opteron.
I thought this got fixed a while ago? Hm.


[-- Attachment #2: pid-hash-alloc.patch --]
[-- Type: text/x-patch, Size: 2180 bytes --]



Export nr_kernel_pages, nr_all_pages. Use nr_kernel_pages when sizing
the PID hash. This fixes a sizing problem I'm seeing with the x86-64 kernel
on an Opteron.


---

 linux-2.6-npiggin/include/linux/bootmem.h |    3 +++
 linux-2.6-npiggin/kernel/pid.c            |    2 +-
 linux-2.6-npiggin/mm/page_alloc.c         |    4 ++--
 3 files changed, 6 insertions(+), 3 deletions(-)

diff -puN kernel/pid.c~pid-hash-alloc kernel/pid.c
--- linux-2.6/kernel/pid.c~pid-hash-alloc	2004-08-22 14:28:46.000000000 +1000
+++ linux-2.6-npiggin/kernel/pid.c	2004-08-22 14:34:13.000000000 +1000
@@ -276,7 +276,7 @@ int kgdb_pid_init_done; /* so we don't c
 void __init pidhash_init(void)
 {
 	int i, j, pidhash_size;
-	unsigned long megabytes = max_pfn >> (20 - PAGE_SHIFT);
+	unsigned long megabytes = nr_kernel_pages >> (20 - PAGE_SHIFT);
 
 	pidhash_shift = max(4, fls(megabytes * 4));
 	pidhash_shift = min(12, pidhash_shift);
diff -puN include/linux/bootmem.h~pid-hash-alloc include/linux/bootmem.h
--- linux-2.6/include/linux/bootmem.h~pid-hash-alloc	2004-08-22 14:28:46.000000000 +1000
+++ linux-2.6-npiggin/include/linux/bootmem.h	2004-08-22 14:28:46.000000000 +1000
@@ -67,6 +67,9 @@ extern void * __init __alloc_bootmem_nod
 	__alloc_bootmem_node((pgdat), (x), PAGE_SIZE, 0)
 #endif /* !CONFIG_HAVE_ARCH_BOOTMEM_NODE */
 
+extern unsigned long __initdata nr_kernel_pages;
+extern unsigned long __initdata nr_all_pages;
+
 extern void *__init alloc_large_system_hash(const char *tablename,
 					    unsigned long bucketsize,
 					    unsigned long numentries,
diff -puN mm/page_alloc.c~pid-hash-alloc mm/page_alloc.c
--- linux-2.6/mm/page_alloc.c~pid-hash-alloc	2004-08-22 14:29:02.000000000 +1000
+++ linux-2.6-npiggin/mm/page_alloc.c	2004-08-22 13:41:11.000000000 +1000
@@ -55,8 +55,8 @@ EXPORT_SYMBOL(zone_table);
 static char *zone_names[MAX_NR_ZONES] = { "DMA", "Normal", "HighMem" };
 int min_free_kbytes = 1024;
 
-static unsigned long __initdata nr_kernel_pages;
-static unsigned long __initdata nr_all_pages;
+unsigned long __initdata nr_kernel_pages;
+unsigned long __initdata nr_all_pages;
 
 /*
  * Temporary debugging check for pages not lying within a given zone.

_

             reply	other threads:[~2004-08-22  4:48 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-08-22  4:44 Nick Piggin [this message]
2004-08-22  4:46 ` [PATCH 2/2] use hlist for pid hash Nick Piggin
2004-08-22  5:00   ` David S. Miller
2004-08-22  5:31     ` Nick Piggin
2004-08-22  5:25   ` Ryan Cumming
2004-08-22  5:26     ` Nick Piggin
2004-09-06 11:35   ` [PATCH] Allocate correct amount of memory " Anton Blanchard
2004-09-06 11:41     ` Nick Piggin
2004-08-22 12:32 ` [PATCH 1/2] fix PID hash sizing William Lee Irwin III
2004-08-23  0:16   ` Nick Piggin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=412824BE.4040801@yahoo.com.au \
    --to=nickpiggin@yahoo.com.au \
    --cc=akpm@osdl.org \
    --cc=linux-kernel@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox