* [PATCH] convert uses of ZONE_HIGHMEM to is_highmem
@ 2004-06-25 22:03 Andy Whitcroft
2004-06-25 22:27 ` Andrew Morton
0 siblings, 1 reply; 4+ messages in thread
From: Andy Whitcroft @ 2004-06-25 22:03 UTC (permalink / raw)
To: linux-kernel; +Cc: akpm, apw
As the comments in mmzone.h indicate is_highmem() is designed to
reduce the proliferation of the constant ZONE_HIGHMEM. This patch
updates three references to ZONE_HIGHMEM to use is_highmem().
None appear to be on critical paths.
Revision: $Rev: 305 $
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
---
arch/i386/mm/discontig.c | 17 +++++++++++------
mm/page_alloc.c | 9 +++++----
2 files changed, 16 insertions(+), 10 deletions(-)
diff -upN reference/arch/i386/mm/discontig.c current/arch/i386/mm/discontig.c
--- reference/arch/i386/mm/discontig.c 2004-06-25 22:26:08.000000000 +0100
+++ current/arch/i386/mm/discontig.c 2004-06-25 22:26:50.000000000 +0100
@@ -411,17 +411,22 @@ void __init zone_sizes_init(void)
void __init set_highmem_pages_init(int bad_ppro)
{
#ifdef CONFIG_HIGHMEM
- int nid;
+ struct zone *zone;
- for (nid = 0; nid < numnodes; nid++) {
+ for_each_zone(zone) {
unsigned long node_pfn, node_high_size, zone_start_pfn;
struct page * zone_mem_map;
- node_high_size = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].spanned_pages;
- zone_mem_map = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].zone_mem_map;
- zone_start_pfn = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].zone_start_pfn;
+ if (!is_highmem(zone))
+ continue;
+
+ printk("Initializing %s for node %d\n", zone->name,
+ zone->zone_pgdat->node_id);
+
+ node_high_size = zone->spanned_pages;
+ zone_mem_map = zone->zone_mem_map;
+ zone_start_pfn = zone->zone_start_pfn;
- printk("Initializing highpages for node %d\n", nid);
for (node_pfn = 0; node_pfn < node_high_size; node_pfn++) {
one_highpage_init((struct page *)(zone_mem_map + node_pfn),
zone_start_pfn + node_pfn, bad_ppro);
diff -upN reference/mm/page_alloc.c current/mm/page_alloc.c
--- reference/mm/page_alloc.c 2004-06-25 22:26:08.000000000 +0100
+++ current/mm/page_alloc.c 2004-06-25 22:26:50.000000000 +0100
@@ -930,11 +930,12 @@ unsigned int nr_free_pagecache_pages(voi
#ifdef CONFIG_HIGHMEM
unsigned int nr_free_highpages (void)
{
- pg_data_t *pgdat;
+ struct zone *zone;
unsigned int pages = 0;
- for_each_pgdat(pgdat)
- pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages;
+ for_each_zone(zone)
+ if (is_highmem(zone))
+ pages += zone->free_pages;
return pages;
}
@@ -1422,7 +1423,7 @@ void __init memmap_init_zone(struct page
INIT_LIST_HEAD(&page->lru);
#ifdef WANT_PAGE_VIRTUAL
/* The shift won't overflow because ZONE_NORMAL is below 4G. */
- if (zone != ZONE_HIGHMEM)
+ if (!is_highmem(zone))
set_page_address(page, __va(start_pfn << PAGE_SHIFT));
#endif
start_pfn++;
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] convert uses of ZONE_HIGHMEM to is_highmem
2004-06-25 22:03 [PATCH] convert uses of ZONE_HIGHMEM to is_highmem Andy Whitcroft
@ 2004-06-25 22:27 ` Andrew Morton
2004-06-25 22:46 ` Andy Whitcroft
2004-07-07 16:25 ` Andy Whitcroft
0 siblings, 2 replies; 4+ messages in thread
From: Andrew Morton @ 2004-06-25 22:27 UTC (permalink / raw)
To: Andy Whitcroft; +Cc: linux-kernel, apw
Andy Whitcroft <apw@shadowen.org> wrote:
>
> As the comments in mmzone.h indicate is_highmem() is designed to
> reduce the proliferation of the constant ZONE_HIGHMEM. This patch
> updates three references to ZONE_HIGHMEM to use is_highmem().
> None appear to be on critical paths.
>
> Revision: $Rev: 305 $
>
> void __init set_highmem_pages_init(int bad_ppro)
> {
> #ifdef CONFIG_HIGHMEM
> - int nid;
> + struct zone *zone;
>
> - for (nid = 0; nid < numnodes; nid++) {
> + for_each_zone(zone) {
> unsigned long node_pfn, node_high_size, zone_start_pfn;
> struct page * zone_mem_map;
>
> - node_high_size = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].spanned_pages;
> - zone_mem_map = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].zone_mem_map;
> - zone_start_pfn = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].zone_start_pfn;
> + if (!is_highmem(zone))
> + continue;
> +
> + printk("Initializing %s for node %d\n", zone->name,
> + zone->zone_pgdat->node_id);
> +
> + node_high_size = zone->spanned_pages;
> + zone_mem_map = zone->zone_mem_map;
> + zone_start_pfn = zone->zone_start_pfn;
>
> - printk("Initializing highpages for node %d\n", nid);
> for (node_pfn = 0; node_pfn < node_high_size; node_pfn++) {
> one_highpage_init((struct page *)(zone_mem_map + node_pfn),
> zone_start_pfn + node_pfn, bad_ppro);
Fair enough.
> @@ -930,11 +930,12 @@ unsigned int nr_free_pagecache_pages(voi
> #ifdef CONFIG_HIGHMEM
> unsigned int nr_free_highpages (void)
> {
> - pg_data_t *pgdat;
> + struct zone *zone;
> unsigned int pages = 0;
>
> - for_each_pgdat(pgdat)
> - pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages;
> + for_each_zone(zone)
> + if (is_highmem(zone))
> + pages += zone->free_pages;
but that's slower.
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] convert uses of ZONE_HIGHMEM to is_highmem
2004-06-25 22:27 ` Andrew Morton
@ 2004-06-25 22:46 ` Andy Whitcroft
2004-07-07 16:25 ` Andy Whitcroft
1 sibling, 0 replies; 4+ messages in thread
From: Andy Whitcroft @ 2004-06-25 22:46 UTC (permalink / raw)
To: Andrew Morton; +Cc: linux-kernel
--On 25 June 2004 15:27 -0700 Andrew Morton <akpm@digeo.com> wrote:
> Andy Whitcroft <apw@shadowen.org> wrote:
>>
>> As the comments in mmzone.h indicate is_highmem() is designed to
>> reduce the proliferation of the constant ZONE_HIGHMEM. This patch
>> updates three references to ZONE_HIGHMEM to use is_highmem().
>> None appear to be on critical paths.
>>
>> Revision: $Rev: 305 $
>>
>> void __init set_highmem_pages_init(int bad_ppro)
>> {
>> # ifdef CONFIG_HIGHMEM
>> - int nid;
>> + struct zone *zone;
>>
>> - for (nid = 0; nid < numnodes; nid++) {
>> + for_each_zone(zone) {
>> unsigned long node_pfn, node_high_size, zone_start_pfn;
>> struct page * zone_mem_map;
>>
>> - node_high_size =
>> NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].spanned_pages; - zone_mem_map
>> = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].zone_mem_map; -
>> zone_start_pfn =
>> NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].zone_start_pfn; + if
>> (!is_highmem(zone))
>> + continue;
>> +
>> + printk("Initializing %s for node %d\n", zone->name,
>> + zone->zone_pgdat->node_id);
>> +
>> + node_high_size = zone->spanned_pages;
>> + zone_mem_map = zone->zone_mem_map;
>> + zone_start_pfn = zone->zone_start_pfn;
>>
>> - printk("Initializing highpages for node %d\n", nid);
>> for (node_pfn = 0; node_pfn < node_high_size; node_pfn++) {
>> one_highpage_init((struct page *)(zone_mem_map + node_pfn),
>> zone_start_pfn + node_pfn, bad_ppro);
>
> Fair enough.
>
>> @@ -930,11 +930,12 @@ unsigned int nr_free_pagecache_pages(voi
>> # ifdef CONFIG_HIGHMEM
>> unsigned int nr_free_highpages (void)
>> {
>> - pg_data_t *pgdat;
>> + struct zone *zone;
>> unsigned int pages = 0;
>>
>> - for_each_pgdat(pgdat)
>> - pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages;
>> + for_each_zone(zone)
>> + if (is_highmem(zone))
>> + pages += zone->free_pages;
>
> but that's slower.
Yes. Although I didn't think any of the users were particularly
performance critical. The routine counts the number of free pages in the
zones which count as highmem, ie memory not direct mapped into KVA. In
theory there could be more than one zone per node which was highmem,
ZONE_HIGHMEM is currently the only one.
I'll have a think if there is a more performant way to 'fix' that.
-apw
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [PATCH] convert uses of ZONE_HIGHMEM to is_highmem
2004-06-25 22:27 ` Andrew Morton
2004-06-25 22:46 ` Andy Whitcroft
@ 2004-07-07 16:25 ` Andy Whitcroft
1 sibling, 0 replies; 4+ messages in thread
From: Andy Whitcroft @ 2004-07-07 16:25 UTC (permalink / raw)
To: akpm, apw; +Cc: linux-kernel
akpm@osdl.org wrote:
>
> Andy Whitcroft <apw@shadowen.org> wrote:
> >
> > As the comments in mmzone.h indicate is_highmem() is designed to
> > reduce the proliferation of the constant ZONE_HIGHMEM. This patch
> > updates three references to ZONE_HIGHMEM to use is_highmem().
> > None appear to be on critical paths.
> >
> > Revision: $Rev: 305 $
[...]
> Fair enough.
>
> > #ifdef CONFIG_HIGHMEM
> > unsigned int nr_free_highpages (void)
> > {
> > - pg_data_t *pgdat;
> > + struct zone *zone;
> > unsigned int pages = 0;
> >
> > - for_each_pgdat(pgdat)
> > - pages += pgdat->node_zones[ZONE_HIGHMEM].free_pages;
> > + for_each_zone(zone)
> > + if (is_highmem(zone))
> > + pages += zone->free_pages;
>
> but that's slower.
Fair point. This patch only changes the first which is inialisation
only and the last which is the same code via inline. The other
I'll hold onto and think about more.
-apw
=== 8< ===
As the comments in mmzone.h indicate is_highmem() is designed to
reduce the proliferation of the constant ZONE_HIGHMEM. This patch
updates references to ZONE_HIGHMEM to use is_highmem(). None appear
to be on critical paths.
Revision: $Rev: 377 $
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
---
arch/i386/mm/discontig.c | 17 +++++++++++------
mm/page_alloc.c | 2 +-
2 files changed, 12 insertions(+), 7 deletions(-)
diff -upN reference/arch/i386/mm/discontig.c current/arch/i386/mm/discontig.c
--- reference/arch/i386/mm/discontig.c 2004-06-25 22:26:03.000000000 +0100
+++ current/arch/i386/mm/discontig.c 2004-07-07 18:10:15.000000000 +0100
@@ -409,17 +409,22 @@ void __init zone_sizes_init(void)
void __init set_highmem_pages_init(int bad_ppro)
{
#ifdef CONFIG_HIGHMEM
- int nid;
+ struct zone *zone;
- for (nid = 0; nid < numnodes; nid++) {
+ for_each_zone(zone) {
unsigned long node_pfn, node_high_size, zone_start_pfn;
struct page * zone_mem_map;
- node_high_size = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].spanned_pages;
- zone_mem_map = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].zone_mem_map;
- zone_start_pfn = NODE_DATA(nid)->node_zones[ZONE_HIGHMEM].zone_start_pfn;
+ if (!is_highmem(zone))
+ continue;
+
+ printk("Initializing %s for node %d\n", zone->name,
+ zone->zone_pgdat->node_id);
+
+ node_high_size = zone->spanned_pages;
+ zone_mem_map = zone->zone_mem_map;
+ zone_start_pfn = zone->zone_start_pfn;
- printk("Initializing highpages for node %d\n", nid);
for (node_pfn = 0; node_pfn < node_high_size; node_pfn++) {
one_highpage_init((struct page *)(zone_mem_map + node_pfn),
zone_start_pfn + node_pfn, bad_ppro);
diff -upN reference/mm/page_alloc.c current/mm/page_alloc.c
--- reference/mm/page_alloc.c 2004-07-07 18:08:56.000000000 +0100
+++ current/mm/page_alloc.c 2004-07-07 18:10:15.000000000 +0100
@@ -1421,7 +1421,7 @@ void __init memmap_init_zone(struct page
INIT_LIST_HEAD(&page->lru);
#ifdef WANT_PAGE_VIRTUAL
/* The shift won't overflow because ZONE_NORMAL is below 4G. */
- if (zone != ZONE_HIGHMEM)
+ if (!is_highmem(zone))
set_page_address(page, __va(start_pfn << PAGE_SHIFT));
#endif
start_pfn++;
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2004-07-07 16:26 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2004-06-25 22:03 [PATCH] convert uses of ZONE_HIGHMEM to is_highmem Andy Whitcroft
2004-06-25 22:27 ` Andrew Morton
2004-06-25 22:46 ` Andy Whitcroft
2004-07-07 16:25 ` Andy Whitcroft
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®