| From: |
| Thomas Schlichter <schlicht@uni-mannheim.de> |
| To: |
| Andrew Morton <akpm@digeo.com> |
| Subject: |
| [RFC] second try for swap prefetch (does Oops!) |
| Date: |
| Thu, 17 Apr 2003 18:02:13 +0200 |
| Cc: |
| linux-kernel@vger.kernel.org |
diff -urP linux-2.5.67/arch/i386/Kconfig linux-2.5.67_patched/arch/i386/Kconfig
--- linux-2.5.67/arch/i386/Kconfig Thu Apr 10 19:25:40 2003
+++ linux-2.5.67_patched/arch/i386/Kconfig Thu Apr 10 17:47:36 2003
@@ -373,6 +373,13 @@
depends on MK8 || MPENTIUM4
default y
+config SWAP_PREFETCH
+ tristate "Prefetch swapped memory"
+ depends on SWAP
+ help
+ This option enables the kernel to prefetch swapped memory pages
+ when idle.
+
config HUGETLB_PAGE
bool "Huge TLB Page Support"
help
diff -urP linux-2.5.67/fs/inode.c linux-2.5.67_patched/fs/inode.c
--- linux-2.5.67/fs/inode.c Mon Apr 7 19:32:58 2003
+++ linux-2.5.67_patched/fs/inode.c Sat Apr 12 03:30:39 2003
@@ -180,6 +180,7 @@
INIT_LIST_HEAD(&inode->i_dentry);
INIT_LIST_HEAD(&inode->i_devices);
sema_init(&inode->i_sem, 1);
+ INIT_RADIX_TREE(&inode->i_data.swap_tree, GFP_ATOMIC);
INIT_RADIX_TREE(&inode->i_data.page_tree, GFP_ATOMIC);
rwlock_init(&inode->i_data.page_lock);
init_MUTEX(&inode->i_data.i_shared_sem);
diff -urP linux-2.5.67/include/linux/fs.h linux-2.5.67_patched/include/linux/fs.h
--- linux-2.5.67/include/linux/fs.h Mon Apr 7 19:30:58 2003
+++ linux-2.5.67_patched/include/linux/fs.h Sat Apr 12 03:31:24 2003
@@ -312,6 +312,7 @@
struct backing_dev_info;
struct address_space {
struct inode *host; /* owner: inode, block_device */
+ struct radix_tree_root swap_tree; /* radix tree of swapped pages */
struct radix_tree_root page_tree; /* radix tree of all pages */
rwlock_t page_lock; /* and rwlock protecting it */
struct list_head clean_pages; /* list of clean pages */
diff -urP linux-2.5.67/include/linux/swap.h linux-2.5.67_patched/include/linux/swap.h
--- linux-2.5.67/include/linux/swap.h Thu Apr 10 19:25:40 2003
+++ linux-2.5.67_patched/include/linux/swap.h Thu Apr 10 18:36:33 2003
@@ -155,6 +155,8 @@
extern unsigned int nr_free_pages_pgdat(pg_data_t *pgdat);
extern unsigned int nr_free_buffer_pages(void);
extern unsigned int nr_free_pagecache_pages(void);
+extern unsigned int nr_avail_buffer_pages(void);
+extern unsigned int nr_avail_pagecache_pages(void);
/* linux/mm/swap.c */
extern void FASTCALL(lru_cache_add(struct page *));
diff -urP linux-2.5.67/include/linux/swap_prefetch.h linux-2.5.67_patched/include/linux/swap_prefetch.h
--- linux-2.5.67/include/linux/swap_prefetch.h Thu Jan 1 01:00:00 1970
+++ linux-2.5.67_patched/include/linux/swap_prefetch.h Wed Apr 16 16:00:09 2003
@@ -0,0 +1,57 @@
+#ifndef _LINUX_SWAP_PREFETCH_H
+#define _LINUX_SWAP_PREFETCH_H
+
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/radix-tree.h>
+
+struct swapped_entry_t {
+ struct list_head head;
+ swp_entry_t swp_entry;
+ struct address_space *mapping;
+};
+
+struct swapped_list_t {
+ spinlock_t lock;
+ struct list_head head;
+ kmem_cache_t *cache;
+};
+
+extern struct swapped_list_t swapped_list;
+
+static inline void add_to_swapped_list(struct address_space *mapping,
+ unsigned long index)
+{
+ struct swapped_entry_t *entry;
+ int error;
+
+ entry = kmem_cache_alloc(swapped_list.cache, GFP_ATOMIC);
+ if(entry) {
+ entry->swp_entry.val = index;
+ entry->mapping = mapping;
+
+ spin_lock(&swapped_list.lock);
+ error = radix_tree_insert(&mapping->swap_tree, index, entry);
+ if(!error)
+ list_add(&entry->head, &swapped_list.head);
+ else
+ kmem_cache_free(swapped_list.cache, entry);
+ spin_unlock(&swapped_list.lock);
+ }
+}
+
+static inline void remove_from_swapped_list(struct address_space *mapping,
+ unsigned long index)
+{
+ struct swapped_entry_t *entry;
+
+ spin_lock(&swapped_list.lock);
+ entry = radix_tree_delete(&mapping->swap_tree, index);
+ if(entry) {
+ list_del(&entry->head);
+ kmem_cache_free(swapped_list.cache, entry);
+ }
+ spin_unlock(&swapped_list.lock);
+}
+
+#endif /* _LINUX_SWAP_PREFETCH_H */
diff -urP linux-2.5.67/kernel/ksyms.c linux-2.5.67_patched/kernel/ksyms.c
--- linux-2.5.67/kernel/ksyms.c Thu Apr 10 19:25:40 2003
+++ linux-2.5.67_patched/kernel/ksyms.c Mon Apr 14 01:51:51 2003
@@ -58,6 +58,7 @@
#include <linux/ptrace.h>
#include <linux/time.h>
#include <linux/backing-dev.h>
+#include <linux/swap_prefetch.h>
#include <asm/checksum.h>
#if defined(CONFIG_PROC_FS)
@@ -70,6 +71,13 @@
extern struct timezone sys_tz;
extern int panic_timeout;
+
+/* needed for swap prefetch support */
+EXPORT_SYMBOL(swapped_list);
+EXPORT_SYMBOL(swapper_space);
+EXPORT_SYMBOL(swapin_readahead);
+EXPORT_SYMBOL(do_page_cache_readahead);
+EXPORT_SYMBOL(nr_avail_pagecache_pages);
/* process memory management */
EXPORT_SYMBOL(do_mmap_pgoff);
diff -urP linux-2.5.67/mm/Makefile linux-2.5.67_patched/mm/Makefile
--- linux-2.5.67/mm/Makefile Thu Apr 10 19:25:40 2003
+++ linux-2.5.67_patched/mm/Makefile Thu Apr 10 17:47:36 2003
@@ -12,3 +12,5 @@
slab.o swap.o truncate.o vcache.o vmscan.o $(mmu-y)
obj-$(CONFIG_SWAP) += page_io.o swap_state.o swapfile.o
+
+obj-$(CONFIG_SWAP_PREFETCH) += swap_prefetch.o
diff -urP linux-2.5.67/mm/filemap.c linux-2.5.67_patched/mm/filemap.c
--- linux-2.5.67/mm/filemap.c Mon Apr 7 19:31:02 2003
+++ linux-2.5.67_patched/mm/filemap.c Wed Apr 16 16:04:40 2003
@@ -16,8 +16,7 @@
#include <linux/fs.h>
#include <linux/aio.h>
#include <linux/kernel_stat.h>
-#include <linux/mm.h>
-#include <linux/swap.h>
+#include <linux/swap_prefetch.h>
#include <linux/mman.h>
#include <linux/pagemap.h>
#include <linux/file.h>
@@ -84,6 +83,7 @@
BUG_ON(PageDirty(page) && !PageSwapCache(page));
+ remove_from_swapped_list(mapping, page->index);
radix_tree_delete(&mapping->page_tree, page->index);
list_del(&page->list);
page->mapping = NULL;
@@ -223,8 +223,11 @@
int add_to_page_cache(struct page *page, struct address_space *mapping,
pgoff_t offset, int gfp_mask)
{
- int error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM);
+ int error;
+ remove_from_swapped_list(mapping, offset);
+
+ error = radix_tree_preload(gfp_mask & ~__GFP_HIGHMEM);
if (error == 0) {
page_cache_get(page);
write_lock(&mapping->page_lock);
diff -urP linux-2.5.67/mm/page_alloc.c linux-2.5.67_patched/mm/page_alloc.c
--- linux-2.5.67/mm/page_alloc.c Thu Apr 10 19:25:40 2003
+++ linux-2.5.67_patched/mm/page_alloc.c Thu Apr 10 17:47:36 2003
@@ -787,6 +787,48 @@
}
#endif
+static unsigned int nr_avail_zone_pages(int offset)
+{
+ pg_data_t *pgdat;
+ unsigned long avail = 0;
+
+ for_each_pgdat(pgdat) {
+ struct zonelist *zonelist = pgdat->node_zonelists + offset;
+ struct zone **zonep = zonelist->zones;
+ struct zone *zone;
+ unsigned long low = 0;
+
+ for (zone = *zonep++; zone; zone = *zonep++) {
+ unsigned long local_free = zone->free_pages;
+ unsigned long local_low = zone->pages_low;
+
+ low += local_low;
+ if (local_free > low) {
+ avail = max(avail, local_free - low);
+ }
+ low += local_low * sysctl_lower_zone_protection;
+ }
+ }
+
+ return avail;
+}
+
+/*
+ * Amount of available RAM allocatable within ZONE_DMA and ZONE_NORMAL
+ */
+unsigned int nr_avail_buffer_pages(void)
+{
+ return nr_avail_zone_pages(GFP_USER & GFP_ZONEMASK);
+}
+
+/*
+ * Amount of available RAM allocatable within all zones
+ */
+unsigned int nr_avail_pagecache_pages(void)
+{
+ return nr_avail_zone_pages(GFP_HIGHUSER & GFP_ZONEMASK);
+}
+
#ifdef CONFIG_NUMA
static void show_node(struct zone *zone)
{
diff -urP linux-2.5.67/mm/swap.c linux-2.5.67_patched/mm/swap.c
--- linux-2.5.67/mm/swap.c Mon Apr 7 19:31:05 2003
+++ linux-2.5.67_patched/mm/swap.c Sat Apr 12 03:19:53 2003
@@ -13,9 +13,8 @@
* Buffermem limits added 12.3.98, Rik van Riel.
*/
-#include <linux/mm.h>
#include <linux/kernel_stat.h>
-#include <linux/swap.h>
+#include <linux/swap_prefetch.h>
#include <linux/mman.h>
#include <linux/pagemap.h>
#include <linux/pagevec.h>
@@ -24,6 +23,11 @@
#include <linux/buffer_head.h>
#include <linux/percpu.h>
+struct swapped_list_t swapped_list = {
+ .lock = SPIN_LOCK_UNLOCKED,
+ .head = LIST_HEAD_INIT(swapped_list.head),
+};
+
/* How many pages do we try to swap or page in/out together? */
int page_cluster;
@@ -390,4 +394,12 @@
* Right now other parts of the system means that we
* _really_ don't want to cluster much more
*/
+
+ /*
+ * Create kmem cache for swapped entries
+ */
+ swapped_list.cache = kmem_cache_create("swapped_entry",
+ sizeof(struct swapped_entry_t), 0, 0, NULL, NULL);
+ if(!swapped_list.cache)
+ panic("swap_setup(): cannot create swapped_entry SLAB cache");
}
diff -urP linux-2.5.67/mm/swap_prefetch.c linux-2.5.67_patched/mm/swap_prefetch.c
--- linux-2.5.67/mm/swap_prefetch.c Thu Jan 1 01:00:00 1970
+++ linux-2.5.67_patched/mm/swap_prefetch.c Thu Apr 17 00:29:40 2003
@@ -0,0 +1,88 @@
+#include <linux/init.h>
+#include <linux/module.h>
+#include <linux/swap_prefetch.h>
+
+#define RESERVED_PAGES 50 /* let 200 kByte of pagecache free */
+#define INTERVAL 60 /* (secs) Default is 1 minute */
+
+static int reserved_pages = RESERVED_PAGES;
+static int interval = INTERVAL;
+
+MODULE_PARM(reserved_pages,"i");
+MODULE_PARM_DESC(reserved_pages,
+ "count of pagechache pages to let free (default 50)");
+
+MODULE_PARM(interval,"i");
+MODULE_PARM_DESC(interval,
+ "delay in seconds to wait between memory checks (default 60)");
+
+MODULE_LICENSE("GPL");
+MODULE_AUTHOR("Thomas Schlichter <thomas.schlichter@web.de>");
+MODULE_DESCRIPTION("prefetches swap pages when there is free memory");
+
+/*
+ * Our timer
+ */
+static void prefetch_timer_handler(unsigned long data);
+static struct timer_list prefetch_timer =
+ TIMER_INITIALIZER(prefetch_timer_handler, 0, 0);
+
+/*
+ * Our work
+ */
+static void prefetch_work_handler(void *data);
+static DECLARE_WORK(prefetch_work, prefetch_work_handler, 0);
+
+/*
+ * If the timer expires..
+ */
+static void prefetch_timer_handler(unsigned long data)
+{
+ schedule_work(&prefetch_work);
+ prefetch_timer.expires = jiffies + interval * HZ;
+ add_timer(&prefetch_timer);
+}
+
+/*
+ * ..do the work
+ */
+static void prefetch_work_handler(void *data)
+{
+ printk(KERN_INFO "Available pages before: %d\n", nr_avail_pagecache_pages());
+
+ while(nr_avail_pagecache_pages() > reserved_pages) {
+ struct swapped_entry_t *entry;
+
+ spin_lock(&swapped_list.lock);
+ if(list_empty(&swapped_list.head)) {
+ spin_unlock(&swapped_list.lock);
+ break;
+ }
+ entry = list_entry(swapped_list.head.next, struct swapped_entry_t, head);
+ radix_tree_delete(&entry->mapping->swap_tree, entry->swp_entry.val);
+ list_del(&entry->head);
+ spin_unlock(&swapped_list.lock);
+
+ if(entry->mapping == &swapper_space)
+ swapin_readahead(entry->swp_entry);
+ else
+ do_page_cache_readahead(entry->mapping, NULL, entry->swp_entry.val, 1);
+ kmem_cache_free(swapped_list.cache, entry);
+ }
+
+ printk(KERN_INFO "Available pages after: %d\n", nr_avail_pagecache_pages());
+}
+
+static int __init prefetch_init(void)
+{
+ prefetch_timer_handler(0);
+ return 0;
+}
+
+static void __exit prefetch_exit(void)
+{
+ del_timer(&prefetch_timer);
+}
+
+module_init(prefetch_init);
+module_exit(prefetch_exit);
diff -urP linux-2.5.67/mm/swap_state.c linux-2.5.67_patched/mm/swap_state.c
--- linux-2.5.67/mm/swap_state.c Thu Apr 10 19:25:40 2003
+++ linux-2.5.67_patched/mm/swap_state.c Sat Apr 12 03:29:59 2003
@@ -33,6 +33,7 @@
extern struct address_space_operations swap_aops;
struct address_space swapper_space = {
+ .swap_tree = RADIX_TREE_INIT(GFP_ATOMIC),
.page_tree = RADIX_TREE_INIT(GFP_ATOMIC),
.page_lock = RW_LOCK_UNLOCKED,
.clean_pages = LIST_HEAD_INIT(swapper_space.clean_pages),
diff -urP linux-2.5.67/mm/vmscan.c linux-2.5.67_patched/mm/vmscan.c
--- linux-2.5.67/mm/vmscan.c Thu Apr 10 19:25:40 2003
+++ linux-2.5.67_patched/mm/vmscan.c Thu Apr 17 14:41:17 2003
@@ -11,10 +11,9 @@
* Multiqueue VM started 5.8.00, Rik van Riel.
*/
-#include <linux/mm.h>
#include <linux/slab.h>
#include <linux/kernel_stat.h>
-#include <linux/swap.h>
+#include <linux/swap_prefetch.h>
#include <linux/pagemap.h>
#include <linux/init.h>
#include <linux/highmem.h>
@@ -417,6 +416,9 @@
ret++;
if (!pagevec_add(&freed_pvec, page))
__pagevec_release_nonlru(&freed_pvec);
+ if (mapping)
+// if (mapping == &swapper_space)
+ add_to_swapped_list(mapping, page->index);
continue;
activate_locked:
ksymoops 2.4.2 on i586 2.5.67. Options used
-V (default)
-k /proc/ksyms (default)
-l /proc/modules (default)
-o /lib/modules/2.5.67/ (default)
-m /boot/System.map-2.5.67 (default)
Warning: You did not tell me where to find symbol information. I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc. ksymoops -h explains the options.
Error (regular_file): read_ksyms stat /proc/ksyms failed
No modules in ksyms, skipping objects
No ksyms, skipping lsmod
Unable to handle kernel paging request at virtual address 64815c28
c02018e8
*pde = 00000000
Oops: 0000 [#1]
CPU: 0
EIP: 0060:[<c02018e8>] Tainted: P
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010283
eax: 64815c28 ebx: c12b9f3c ecx: 8b21ca84 edx: 64815c24
esi: 8b21ca7e edi: 00000001 ebp: c12b9f6c esp: c12b9f20
ds: 007b es: 007b ss: 0068
Stack: c2e34c6c c03840e0 c12b8000 c1daf716 00000000 00000000 ceeb5570 64815c24
64815c28 00000282 c2e34c50 c2e34c50 c03840e0 c12b8000 00000000 00000000
000003d4 00002a76 c12b9f74 c12b9f88 d4b0415f ceeb5568 00000001 c12b8000
[<d4b0415f>] prefetch_work_handler+0x12f/0x210 [swap_prefetch]
[<d4b04d80>] prefetch_work+0x0/0x60 [swap_prefetch]
[<c0139dcf>] worker_thread+0x28f/0x438
[<c0139b40>] worker_thread+0x0/0x438
[<d4b04030>] prefetch_work_handler+0x0/0x210 [swap_prefetch]
[<c0122d38>] default_wake_function+0x0/0x18
[<c0122d38>] default_wake_function+0x0/0x18
[<c01081e5>] kernel_thread_helper+0x5/0xc
Code: 8b 10 85 d2 74 6a 89 f8 89 f1 d3 e8 83 e0 3f 8d 44 82 04 89
>>EIP; c02018e8 <radix_tree_delete+4c/c8> <=====
Code; c02018e8 <radix_tree_delete+4c/c8>
00000000 <_EIP>:
Code; c02018e8 <radix_tree_delete+4c/c8> <=====
0: 8b 10 mov (%eax),%edx <=====
Code; c02018ea <radix_tree_delete+4e/c8>
2: 85 d2 test %edx,%edx
Code; c02018ec <radix_tree_delete+50/c8>
4: 74 6a je 70 <_EIP+0x70> c0201958 <radix_tree_delete+bc/c8>
Code; c02018ee <radix_tree_delete+52/c8>
6: 89 f8 mov %edi,%eax
Code; c02018f0 <radix_tree_delete+54/c8>
8: 89 f1 mov %esi,%ecx
Code; c02018f2 <radix_tree_delete+56/c8>
a: d3 e8 shr %cl,%eax
Code; c02018f4 <radix_tree_delete+58/c8>
c: 83 e0 3f and $0x3f,%eax
Code; c02018f6 <radix_tree_delete+5a/c8>
f: 8d 44 82 04 lea 0x4(%edx,%eax,4),%eax
Code; c02018fa <radix_tree_delete+5e/c8>
13: 89 00 mov %eax,(%eax)
1 warning and 1 error issued. Results may not be reliable.
