LinuxLists.cc - 2.5.73-mm2

2003-06-28 03:07:38

Subject: 2.5.73-mm2

ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.5/2.5.73/2.5.73-mm2/

Just bits and pieces.

Changes since 2.5.73-mm1:

linus.patch

Latest Linus tree

-show_stack-fix.patch
-pci-1.patch
-pci-2.patch
-pci-3.patch
-pci-4.patch
-pci-5.patch
-alsa-pnp-fix.patch
-setscheduler-fix.patch
-ide_setting_sem-fix.patch
-misc6.patch
-AT_SECURE-auxv-entry.patch
-common-kernel-DSO-name.patch
-get_unmapped_area-speedup.patch
-d_invalidate-fix.patch
-nfs_unlink-d_count-fix.patch
-hpfs-d_count-fix.patch
-smbfs-oops-workaround.patch
-enable-cardbus-bursting.patch
-n_tty-column-counting-fix.patch
-numa-normalised-node-load.patch
-enable-local-apic-on-p4.patch
-knfsd-umount-fix.patch
-getrlimit-ifdef-fix.patch
-amd64-monotonic-clock.patch

Merged

+kgdb-ga-docco-fixes.patch

kgdb documentation fixes

+pppoe-revert.patch

Back out broken pppoe changes until it gets fixed up.

+move_vma-VM_LOCKED-fix.patch

Fix the mremap use-after-free fix

+ipcsem-speedup.patch

Speed up sysv semaphore operations

+feral-fix.patch

Fix the linux_isp qlogic driver for non-odular builds.

-cfq-2.patch

Dropped. I'm firming up the IO scheduler rework for a merge and the -mm
CFQ inplementation is way out of date.

+blk-batching-cleanups.patch

Minor touchups

+truncate-pagefault-race-fix-fix.patch

Tighten up the MP synchronisation

-security-vm_enough_memory.patch
+security_vm_enough_memory.patch

New version

+nbd-remove-blksize-bits.patch
+nbd-kobject-oops-fix.patch
+nbd-paranioa-cleanups.patch
+nbd-locking-fixes.patch

More NBD work

-nr_running-speedup.patch

Dropped. It seemed to have no net benefit.

+lowmem_page_address-cleanup.patch

Simplify lowmem_page_address()

+numa-memory-reporting-fix.patch

Fix NUMA memory reporting (needs more work)

+syslog-efault-reporting.patch

check copy_*_user return values

+acpismp-fix.patch

Fix `acpismp=force'

+div64-cleanup.patch

Consolidate and fix div64 code

+init_timer-debug-trap.patch

Debug code to catch people running init_timer() against a running timer.

+dvd-ram-rw-fix.patch

ide-scsi RW mount fix

+mixcomwd-update.patch
+arc-rimi-race-fix.patch

minor fixups

+slab-drain-all-objects-fix.patch

Make slab free up all the objects when destroying caches.

+ext3-remove-version.patch

Remove the version information from ext3.

+cdrom-eject-hang-fix.patch

Fix mount-time hangs caused by CDROM eject commands.

All 138 patches:

linus.patch
cset-1.1348.16.4-to-1.1516.txt.gz

mm.patch
add -mmN to EXTRAVERSION

kgdb-ga.patch
kgdb stub for ia32 (George Anzinger's one)

kgdb-use-ggdb.patch

kgdb-ga-docco-fixes.patch
kgdb doc. edits/corrections

HZ-100.patch

handle-no-readpage-2.patch
check for presence of readpage() in the readahead code

pppoe-revert.patch
PPPOE reversion

config_spinline.patch
uninline spinlocks for profiling accuracy.

ppc64-fixes-2.patch
Maek ppc64 compile

ppc64-bat-initialisation-fix.patch
ppc64: BAT initialisation fix

ppc64-pci-update.patch

ppc64-reloc_hide.patch

ppc64-semaphore-reimplementation.patch
ppc64: use the ia32 semaphore implementation

sym-do-160.patch
make the SYM driver do 160 MB/sec

x86_64-fixes.patch
x86_64 fixes

irqreturn-snd-via-fix.patch
via sound irqreturn fix

config-PAGE_OFFSET.patch
Configurable kenrel/user memory split

lru_cache_add-check.patch
lru_cache_add debug check

delay-ksoftirqd-fallback.patch
Try harded in IRQ context before falling back to ksoftirqd

fb-image-depth-fix.patch
fbdev image depth fix

move_vma-VM_LOCKED-fix.patch
move_vma() make_pages_present() fix

ds-09-vicam-usercopy-fix.patch
vicam usercopy fix

buffer-debug.patch
buffer.c debugging

reiserfs-unmapped-buffer-fix.patch
Fix reiserfs BUG

e100-use-after-free-fix.patch

3-unmap-page-debugging.patch
page unmappng debug patch

VM_RESERVED-check.patch
VM_RESERVED check

ipcsem-speedup.patch
ipc semaphore optimization

rcu-stats.patch
RCU statistics reporting

mtrr-hang-fix.patch
Fix mtrr-related hang

reslabify-pgds-and-pmds.patch
re-slabify i386 pgd's and pmd's

linux-isp.patch

isp-update-1.patch

isp-remove-pci_detect.patch

feral-fix.patch
linux-isp fix

list_del-debug.patch
list_del debug check

airo-schedule-fix.patch
airo.c: don't sleep in atomic regions

resurrect-batch_requests.patch
bring back the batch_requests function

kblockd.patch
Create `kblockd' workqueue

cfq-infrastructure.patch

elevator-completion-api.patch
elevator completion API

as-iosched.patch
anticipatory I/O scheduler
AS: pgbench improvement
AS: discrete read fifo batches
AS sync/async batches
AS: hash removal fix
AS jumbo patch (for SCSI and TCQ)
AS: fix stupid thinko
AS: no batch-antic-limit
AS: autotune write batches
AS: divide by zero fix
AS: more HZ != 1000 fixes
AS: update_write_batch tuning
AS locking
AS HZ fixes

as-double-free-and-debug.patch
AS: fix a leak + more debugging

as-fix-seek-estimation.patch
AS: maybe repair performance drop of random read O_DIRECT

as-fix-seeky-loads.patch
AS: fix IBM's seek load

unplug-use-kblockd.patch
Use kblockd for running request queues

per-queue-nr_requests.patch
per queue nr_requests

blk-invert-watermarks.patch
blk_congestion_wait threshold cleanup

blk-as-hint.patch
blk-as-hint

get_request_wait-oom-fix.patch
handle OOM in get_request_wait().

blk-fair-batches.patch
blk-fair-batches

blk-fair-batches-2.patch
blk fair batches #2

generic-io-contexts.patch
generic io contexts

blk-request-batching.patch
block request batching

get_io_context-fix.patch
get_io_context fixes

blk-allocation-commentary.patch
block allocation comments

blk-batching-throttle-fix.patch
blk batch requests fix

blk-batching-cleanups.patch
block batching cleanups

print-build-options-on-oops.patch
print a few config options on oops

mmap-prefault.patch
prefault of executable mmaps

bio-debug-trap.patch
BIO debugging patch

sound-irq-hack.patch

show_task-free-stack-fix.patch
show_task() fix and cleanup

put_task_struct-debug.patch

ia32-mknod64.patch
mknod64 for ia32

ext2-64-bit-special-inodes.patch
ext2: support for 64-bit device nodes

ext3-64-bit-special-inodes.patch
ext3: support for 64-bit device nodes

64-bit-dev_t-kdev_t.patch
64-bit dev_t and kdev_t

oops-dump-preceding-code.patch
i386 oops output: dump preceding code

lockmeter.patch

invalidate_mmap_range.patch
Interface to invalidate regions of mmaps

aio-mm-refcounting-fix.patch
fix /proc mm_struct refcounting bug

aio-01-retry.patch
AIO: Core retry infrastructure

io_submit_one-EINVAL-fix.patch
Fix aio process hang on EINVAL

aio-02-lockpage_wq.patch
AIO: Async page wait

aio-03-fs_read.patch
AIO: Filesystem aio read

aio-04-buffer_wq.patch
AIO: Async buffer wait

aio-05-fs_write.patch
AIO: Filesystem aio write

aio-05-fs_write-fix.patch

aio-06-bread_wq.patch
AIO: Async block read

aio-06-bread_wq-fix.patch

aio-07-ext2getblk_wq.patch
AIO: Async get block for ext2

O_SYNC-speedup-2.patch
speed up O_SYNC writes

aio-09-o_sync.patch
aio O_SYNC

aio-10-BUG-fix.patch
AIO: fix a BUG

aio-11-workqueue-flush.patch
AIO: flush workqueues before destroying ioctx'es

aio-12-readahead.patch
AIO: readahead fixes

lock_buffer_wq-fix.patch
lock_buffer_wq fix

unuse_mm-locked.patch
AIO: hold the context lock across unuse_mm

aio-take-task_lock.patch
From: Suparna Bhattacharya <[email protected]>
Subject: Re: 2.5.72-mm1 - Under heavy testing with AIO,.. vmstat seems to blow the kernel

vfsmount_lock.patch
From: Maneesh Soni <[email protected]>
Subject: [patch 1/2] vfsmount_lock

sched-hot-balancing-fix.patch
fix for CPU scheduler load distribution

truncate-pagefault-race-fix.patch
Fix vmtruncate race and distributed filesystem race

truncate-pagefault-race-fix-fix.patch
Make sure truncate fix has no race

sleepometer.patch
sleep instrumentation

time-goes-backwards.patch
demonstrate do_gettimeofday() going backwards

skip-apic-ids-on-boot.patch
skip apicids on boot

printk-oops-mangle-fix.patch
disentangle printk's whilst oopsing on SMP

20-odirect_enable.patch

21-odirect_cruft.patch

22-read_proc.patch

23-write_proc.patch

24-commit_proc.patch

25-odirect.patch

nfs-O_DIRECT-always-enabled.patch
Force CONFIG_NFS_DIRECTIO

seqcount-locking.patch
i_size atomic access: infrastructure

i_size-atomic-access.patch
i_size atomic access

aha152x-oops-fix.patch
aha152X oops fixes

security_vm_enough_memory.patch
Security hook for vm_enough_memory

nbd-cleanups.patch
NBD: cosmetic cleanups

nbd-enhanced-diagnostics.patch
nbd: enhanced diagnostics support

nbd-remove-blksize-bits.patch
nbd: remove unneeded blksize_bits field

nbd-kobject-oops-fix.patch
nbd: initialise the embedded kobject

nbd-paranioa-cleanups.patch
nbd: cleanup PARANOIA usage & code

nbd-locking-fixes.patch
nbd: fix locking issues with ioctl UI

pcmcia-event-20030623-1.patch

pcmcia-event-20030623-2.patch

pcmcia-event-20030623-3.patch

pcmcia-event-20030623-4.patch

pcmcia-event-20030623-5.patch

pcmcia-event-20030623-6.patch

sym2-bus_addr-fix.patch
sym53c8xx_2 bus_addr fix

lost-tick-speedstep-fix.patch
Fix lost tick detection for speedstep

sym2-remove-broken-bios-check.patch
remove a bogus check in sym2 driver

rename-timer.patch
timer cleanups

lowmem_page_address-cleanup.patch
cleanup and generalise lowmem_page_address

numa-memory-reporting-fix.patch
fix NUMA memory reporting ... again

syslog-efault-reporting.patch
Fix syslog(2) EFAULT reporting

acpismp-fix.patch
ACPI_HT_ONLY acpismp=force

div64-cleanup.patch
Kill div64.h dupes and parenthesize do_div() parameters

init_timer-debug-trap.patch
init_timer debugging

dvd-ram-rw-fix.patch
2.5.73 can't mount DVD-RAM via ide-scsi

mixcomwd-update.patch
Remove check_region and MOD_*_USE_COUNT from mixcomwd.c

arc-rimi-race-fix.patch
Remove racy check_mem_region() call from arc-rimi.c

slab-drain-all-objects-fix.patch
kmem_cache_destroy() forgets to drain all objects

ext3-remove-version.patch
ext3: remove the version number

cdrom-eject-hang-fix.patch
cdrom eject scribbles on the request flags

2003-06-28 08:42:45

by William Lee Irwin III

[permalink] [raw]

Subject: Re: 2.5.73-mm2

On Fri, Jun 27, 2003 at 08:21:30PM -0700, Andrew Morton wrote:
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.5/2.5.73/2.5.73-mm2/
> Just bits and pieces.

This could almost be sent through rusty, but I think in general BKL
removal patches should go through whatever ringer -mm et al provide.

-- wli

Remove spurious BKL acquisitions in /proc/. The BKL is not required to
access nr_threads for reporting, and get_locks_status() takes it
internally, wrapping all operations with it.

diff -prauN wli-2.5.73-3/fs/proc/proc_misc.c wli-2.5.73-4/fs/proc/proc_misc.c
--- wli-2.5.73-3/fs/proc/proc_misc.c 2003-06-23 10:29:54.000000000 -0700
+++ wli-2.5.73-4/fs/proc/proc_misc.c 2003-06-23 10:32:25.000000000 -0700
@@ -497,11 +497,10 @@ static int ds1286_read_proc(char *page,
static int locks_read_proc(char *page, char **start, off_t off,
int count, int *eof, void *data)
{
- int len;
- lock_kernel();
- len = get_locks_status(page, start, off, count);
- unlock_kernel();
- if (len < count) *eof = 1;
+ int len = get_locks_status(page, start, off, count);
+
+ if (len < count)
+ *eof = 1;
return len;
}

diff -prauN wli-2.5.73-3/fs/proc/root.c wli-2.5.73-4/fs/proc/root.c
--- wli-2.5.73-3/fs/proc/root.c 2003-06-22 11:33:07.000000000 -0700
+++ wli-2.5.73-4/fs/proc/root.c 2003-06-23 10:32:25.000000000 -0700
@@ -81,11 +81,13 @@ void __init proc_root_init(void)

static struct dentry *proc_root_lookup(struct inode * dir, struct dentry * dentry)
{
- if (dir->i_ino == PROC_ROOT_INO) { /* check for safety... */
- lock_kernel();
+ /*
+ * nr_threads is actually protected by the tasklist_lock;
+ * however, it's conventional to do reads, especially for
+ * reporting, without any locking whatsoever.
+ */
+ if (dir->i_ino == PROC_ROOT_INO) /* check for safety... */
dir->i_nlink = proc_root.nlink + nr_threads;
- unlock_kernel();
- }

if (!proc_lookup(dir, dentry)) {
return NULL;

2003-06-28 15:41:35

by William Lee Irwin III

[permalink] [raw]

Subject: Re: 2.5.73-mm2

On Fri, Jun 27, 2003 at 08:21:30PM -0700, Andrew Morton wrote:
> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.5/2.5.73/2.5.73-mm2/
> Just bits and pieces.

Here's highpmd. This allocates L2 pagetables from highmem, decreasing
the per-process lowmem overhead on CONFIG_HIGHMEM64G from 20KB to 8KB.
Some attempts were made to update non-i386 architectures to the new
API's, though they're entirely untested. It's been tested for a while
in -wli on i386 machines, both lowmem and highmem boxen.

-- wli

diff -prauN mm2-2.5.73-1/arch/i386/Kconfig mm2-2.5.73-2/arch/i386/Kconfig
--- mm2-2.5.73-1/arch/i386/Kconfig 2003-06-28 03:09:46.000000000 -0700
+++ mm2-2.5.73-2/arch/i386/Kconfig 2003-06-28 03:11:37.000000000 -0700
@@ -765,6 +765,15 @@ config HIGHPTE
low memory. Setting this option will put user-space page table
entries in high memory.

+config HIGHPMD
+ bool "Allocate 2nd-level pagetables from highmem"
+ depends on HIGHMEM64G
+ help
+ The VM uses one pmd entry for each pagetable page of physical
+ memory allocated. For systems with extreme amounts of highmem,
+ this cannot be tolerated. Setting this option will put
+ userspace 2nd-level pagetables in highmem.
+
config MATH_EMULATION
bool "Math emulation"
---help---
diff -prauN mm2-2.5.73-1/arch/i386/kernel/vm86.c mm2-2.5.73-2/arch/i386/kernel/vm86.c
--- mm2-2.5.73-1/arch/i386/kernel/vm86.c 2003-06-22 11:32:33.000000000 -0700
+++ mm2-2.5.73-2/arch/i386/kernel/vm86.c 2003-06-28 03:11:37.000000000 -0700
@@ -144,12 +144,14 @@ static void mark_screen_rdonly(struct ta
pgd_clear(pgd);
goto out;
}
- pmd = pmd_offset(pgd, 0xA0000);
- if (pmd_none(*pmd))
+ pmd = pmd_offset_map(pgd, 0xA0000);
+ if (pmd_none(*pmd)) {
+ pmd_unmap(pmd);
goto out;
- if (pmd_bad(*pmd)) {
+ } else if (pmd_bad(*pmd)) {
pmd_ERROR(*pmd);
pmd_clear(pmd);
+ pmd_unmap(pmd);
goto out;
}
pte = mapped = pte_offset_map(pmd, 0xA0000);
@@ -159,6 +161,7 @@ static void mark_screen_rdonly(struct ta
pte++;
}
pte_unmap(mapped);
+ pmd_unmap(pmd);
out:
spin_unlock(&tsk->mm->page_table_lock);
preempt_enable();
diff -prauN mm2-2.5.73-1/arch/i386/mm/fault.c mm2-2.5.73-2/arch/i386/mm/fault.c
--- mm2-2.5.73-1/arch/i386/mm/fault.c 2003-06-28 03:09:46.000000000 -0700
+++ mm2-2.5.73-2/arch/i386/mm/fault.c 2003-06-28 03:28:31.000000000 -0700
@@ -253,6 +253,7 @@ no_context:
printk(" printing eip:\n");
printk("%08lx\n", regs->eip);
asm("movl %%cr3,%0":"=r" (page));
+#ifndef CONFIG_HIGHPMD /* Oh boy. Error reporting is going to blow major goats. */
page = ((unsigned long *) __va(page))[address >> 22];
printk(KERN_ALERT "*pde = %08lx\n", page);
/*
@@ -268,7 +269,14 @@ no_context:
page = ((unsigned long *) __va(page))[address >> PAGE_SHIFT];
printk(KERN_ALERT "*pte = %08lx\n", page);
}
-#endif
+#endif /* !CONFIG_HIGHPTE */
+#else /* CONFIG_HIGHPMD */
+ printk(KERN_ALERT "%%cr3 = 0x%lx\n", page);
+ /* Mask off flag bits. It should end up 32B-aligned. */
+ page &= ~(PTRS_PER_PGD*sizeof(pgd_t) - 1);
+ printk(KERN_ALERT "*pdpte = 0x%Lx\n",
+ pgd_val(((pgd_t *)__va(page))[address >> PGDIR_SHIFT]));
+#endif /* CONFIG_HIGHPMD */
die("Oops", regs, error_code);
bust_spinlocks(0);
do_exit(SIGKILL);
@@ -336,8 +344,8 @@ vmalloc_fault:
* and redundant with the set_pmd() on non-PAE.
*/

- pmd = pmd_offset(pgd, address);
- pmd_k = pmd_offset(pgd_k, address);
+ pmd = pmd_offset_kernel(pgd, address);
+ pmd_k = pmd_offset_kernel(pgd_k, address);
if (!pmd_present(*pmd_k))
goto no_context;
set_pmd(pmd, *pmd_k);
diff -prauN mm2-2.5.73-1/arch/i386/mm/hugetlbpage.c mm2-2.5.73-2/arch/i386/mm/hugetlbpage.c
--- mm2-2.5.73-1/arch/i386/mm/hugetlbpage.c 2003-06-22 11:33:17.000000000 -0700
+++ mm2-2.5.73-2/arch/i386/mm/hugetlbpage.c 2003-06-28 03:11:37.000000000 -0700
@@ -87,8 +87,8 @@ static pte_t *huge_pte_alloc(struct mm_s
pmd_t *pmd = NULL;

pgd = pgd_offset(mm, addr);
- pmd = pmd_alloc(mm, pgd, addr);
- return (pte_t *) pmd;
+ pmd = pmd_alloc_map(mm, pgd, addr);
+ return (pte_t *)pmd;
}

static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
@@ -97,8 +97,8 @@ static pte_t *huge_pte_offset(struct mm_
pmd_t *pmd = NULL;

pgd = pgd_offset(mm, addr);
- pmd = pmd_offset(pgd, addr);
- return (pte_t *) pmd;
+ pmd = pmd_offset_map(pgd, addr);
+ return (pte_t *)pmd;
}

static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, struct page *page, pte_t * page_table, int write_access)
@@ -145,6 +145,8 @@ int copy_hugetlb_page_range(struct mm_st
ptepage = pte_page(entry);
get_page(ptepage);
set_pte(dst_pte, entry);
+ pmd_unmap(dst_pte);
+ pmd_unmap_nested(src_pte);
dst->rss += (HPAGE_SIZE / PAGE_SIZE);
addr += HPAGE_SIZE;
}
@@ -182,6 +184,7 @@ follow_hugetlb_page(struct mm_struct *mm

get_page(page);
pages[i] = page;
+ pmd_unmap(pte);
}

if (vmas)
@@ -271,6 +274,7 @@ follow_huge_pmd(struct mm_struct *mm, un
page += ((address & ~HPAGE_MASK) >> PAGE_SHIFT);
get_page(page);
}
+ pmd_unmap(pmd);
return page;
}
#endif
@@ -314,6 +318,7 @@ void unmap_hugepage_range(struct vm_area
page = pte_page(*pte);
huge_page_release(page);
pte_clear(pte);
+ pmd_unmap(pte);
}
mm->rss -= (end - start) >> PAGE_SHIFT;
flush_tlb_range(vma, start, end);
@@ -358,16 +363,19 @@ int hugetlb_prefault(struct address_spac
page = alloc_hugetlb_page();
if (!page) {
ret = -ENOMEM;
+ pmd_unmap(pte);
goto out;
}
ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
unlock_page(page);
if (ret) {
free_huge_page(page);
+ pmd_unmap(pte);
goto out;
}
}
set_huge_pte(mm, vma, page, pte, vma->vm_flags & VM_WRITE);
+ pmd_unmap(pte);
}
out:
spin_unlock(&mm->page_table_lock);
diff -prauN mm2-2.5.73-1/arch/i386/mm/init.c mm2-2.5.73-2/arch/i386/mm/init.c
--- mm2-2.5.73-1/arch/i386/mm/init.c 2003-06-28 03:09:46.000000000 -0700
+++ mm2-2.5.73-2/arch/i386/mm/init.c 2003-06-28 03:22:00.000000000 -0700
@@ -59,10 +59,10 @@ static pmd_t * __init one_md_table_init(
#ifdef CONFIG_X86_PAE
pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT));
- if (pmd_table != pmd_offset(pgd, 0))
+ if (pmd_table != pmd_offset_kernel(pgd, 0))
BUG();
#else
- pmd_table = pmd_offset(pgd, 0);
+ pmd_table = pmd_offset_kernel(pgd, 0);
#endif

return pmd_table;
@@ -113,7 +113,7 @@ static void __init page_table_range_init
if (pgd_none(*pgd))
one_md_table_init(pgd);

- pmd = pmd_offset(pgd, vaddr);
+ pmd = pmd_offset_kernel(pgd, vaddr);
for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); pmd++, pmd_idx++) {
if (pmd_none(*pmd))
one_page_table_init(pmd);
@@ -194,7 +194,7 @@ pte_t *kmap_pte;
pgprot_t kmap_prot;

#define kmap_get_fixmap_pte(vaddr) \
- pte_offset_kernel(pmd_offset(pgd_offset_k(vaddr), (vaddr)), (vaddr))
+ pte_offset_kernel(pmd_offset_kernel(pgd_offset_k(vaddr), (vaddr)), (vaddr))

void __init kmap_init(void)
{
@@ -218,7 +218,7 @@ void __init permanent_kmaps_init(pgd_t *
page_table_range_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, pgd_base);

pgd = swapper_pg_dir + pgd_index(vaddr);
- pmd = pmd_offset(pgd, vaddr);
+ pmd = pmd_offset_kernel(pgd, vaddr);
pte = pte_offset_kernel(pmd, vaddr);
pkmap_page_table = pte;
}
@@ -513,20 +513,9 @@ void __init mem_init(void)
}

kmem_cache_t *pgd_cache;
-kmem_cache_t *pmd_cache;

void __init pgtable_cache_init(void)
{
- if (PTRS_PER_PMD > 1) {
- pmd_cache = kmem_cache_create("pmd",
- PTRS_PER_PMD*sizeof(pmd_t),
- 0,
- SLAB_HWCACHE_ALIGN | SLAB_MUST_HWCACHE_ALIGN,
- pmd_ctor,
- NULL);
- if (!pmd_cache)
- panic("pgtable_cache_init(): cannot create pmd cache");
- }
pgd_cache = kmem_cache_create("pgd",
PTRS_PER_PGD*sizeof(pgd_t),
0,
diff -prauN mm2-2.5.73-1/arch/i386/mm/ioremap.c mm2-2.5.73-2/arch/i386/mm/ioremap.c
--- mm2-2.5.73-1/arch/i386/mm/ioremap.c 2003-06-22 11:32:38.000000000 -0700
+++ mm2-2.5.73-2/arch/i386/mm/ioremap.c 2003-06-28 03:11:37.000000000 -0700
@@ -82,7 +82,7 @@ static int remap_area_pages(unsigned lon
spin_lock(&init_mm.page_table_lock);
do {
pmd_t *pmd;
- pmd = pmd_alloc(&init_mm, dir, address);
+ pmd = pmd_alloc_kernel(&init_mm, dir, address);
error = -ENOMEM;
if (!pmd)
break;
diff -prauN mm2-2.5.73-1/arch/i386/mm/pageattr.c mm2-2.5.73-2/arch/i386/mm/pageattr.c
--- mm2-2.5.73-1/arch/i386/mm/pageattr.c 2003-06-28 03:09:46.000000000 -0700
+++ mm2-2.5.73-2/arch/i386/mm/pageattr.c 2003-06-28 03:12:16.000000000 -0700
@@ -23,7 +23,7 @@ static inline pte_t *lookup_address(unsi
pmd_t *pmd;
if (pgd_none(*pgd))
return NULL;
- pmd = pmd_offset(pgd, address);
+ pmd = pmd_offset_kernel(pgd, address);
if (pmd_none(*pmd))
return NULL;
if (pmd_large(*pmd))
@@ -79,7 +79,7 @@ static void set_pmd_pte(pte_t *kpte, uns
pgd_t *pgd;
pmd_t *pmd;
pgd = (pgd_t *)page_address(page) + pgd_index(address);
- pmd = pmd_offset(pgd, address);
+ pmd = pmd_offset_kernel(pgd, address);
set_pte_atomic((pte_t *)pmd, pte);
}
spin_unlock_irqrestore(&pgd_lock, flags);
@@ -92,7 +92,7 @@ static void set_pmd_pte(pte_t *kpte, uns
static inline void revert_page(struct page *kpte_page, unsigned long address)
{
pte_t *linear = (pte_t *)
- pmd_offset(pgd_offset(&init_mm, address), address);
+ pmd_offset_kernel(pgd_offset_k(address), address);
set_pmd_pte(linear, address,
pfn_pte((__pa(address) & LARGE_PAGE_MASK) >> PAGE_SHIFT,
PAGE_KERNEL_LARGE));
diff -prauN mm2-2.5.73-1/arch/i386/mm/pgtable.c mm2-2.5.73-2/arch/i386/mm/pgtable.c
--- mm2-2.5.73-1/arch/i386/mm/pgtable.c 2003-06-28 03:09:46.000000000 -0700
+++ mm2-2.5.73-2/arch/i386/mm/pgtable.c 2003-06-28 08:20:29.000000000 -0700
@@ -70,7 +70,7 @@ static void set_pte_pfn(unsigned long va
BUG();
return;
}
- pmd = pmd_offset(pgd, vaddr);
+ pmd = pmd_offset_kernel(pgd, vaddr);
if (pmd_none(*pmd)) {
BUG();
return;
@@ -110,7 +110,7 @@ void set_pmd_pfn(unsigned long vaddr, un
printk ("set_pmd_pfn: pgd_none\n");
return; /* BUG(); */
}
- pmd = pmd_offset(pgd, vaddr);
+ pmd = pmd_offset_kernel(pgd, vaddr);
set_pmd(pmd, pfn_pmd(pfn, flags));
/*
* It's enough to flush this one mapping.
@@ -152,11 +152,6 @@ struct page *pte_alloc_one(struct mm_str
return pte;
}

-void pmd_ctor(void *pmd, kmem_cache_t *cache, unsigned long flags)
-{
- memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t));
-}
-
/*
* List of all pgd's needed for non-PAE so it can invalidate entries
* in both cached and uncached pgd's; not needed for PAE since the
@@ -203,6 +198,12 @@ void pgd_dtor(void *pgd, kmem_cache_t *c
spin_unlock_irqrestore(&pgd_lock, flags);
}

+#ifdef CONFIG_HIGHPMD
+#define GFP_PMD (__GFP_REPEAT|__GFP_HIGHMEM|GFP_KERNEL)
+#else
+#define GFP_PMD (__GFP_REPEAT|GFP_KERNEL)
+#endif
+
pgd_t *pgd_alloc(struct mm_struct *mm)
{
int i;
@@ -212,16 +213,17 @@ pgd_t *pgd_alloc(struct mm_struct *mm)
return pgd;

for (i = 0; i < USER_PTRS_PER_PGD; ++i) {
- pmd_t *pmd = kmem_cache_alloc(pmd_cache, GFP_KERNEL);
+ struct page *pmd = alloc_page(GFP_PMD);
if (!pmd)
goto out_oom;
- set_pgd(&pgd[i], __pgd(1 + __pa((u64)((u32)pmd))));
+ clear_highpage(pmd);
+ set_pgd(&pgd[i], __pgd(1ULL | (u64)page_to_pfn(pmd) << PAGE_SHIFT));
}
return pgd;

out_oom:
for (i--; i >= 0; i--)
- kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1));
+ __free_page(pgd_page(pgd[i]));
kmem_cache_free(pgd_cache, pgd);
return NULL;
}
@@ -233,7 +235,7 @@ void pgd_free(pgd_t *pgd)
/* in the PAE case user pgd entries are overwritten before usage */
if (PTRS_PER_PMD > 1)
for (i = 0; i < USER_PTRS_PER_PGD; ++i)
- kmem_cache_free(pmd_cache, (void *)__va(pgd_val(pgd[i])-1));
+ __free_page(pgd_page(pgd[i]));
/* in the non-PAE case, clear_page_tables() clears user pgd entries */
kmem_cache_free(pgd_cache, pgd);
}
diff -prauN mm2-2.5.73-1/arch/sparc/mm/srmmu.c mm2-2.5.73-2/arch/sparc/mm/srmmu.c
--- mm2-2.5.73-1/arch/sparc/mm/srmmu.c 2003-06-22 11:32:56.000000000 -0700
+++ mm2-2.5.73-2/arch/sparc/mm/srmmu.c 2003-06-28 03:11:37.000000000 -0700
@@ -2180,7 +2180,7 @@ void __init ld_mmu_srmmu(void)

BTFIXUPSET_CALL(pte_pfn, srmmu_pte_pfn, BTFIXUPCALL_NORM);
BTFIXUPSET_CALL(pmd_page, srmmu_pmd_page, BTFIXUPCALL_NORM);
- BTFIXUPSET_CALL(pgd_page, srmmu_pgd_page, BTFIXUPCALL_NORM);
+ BTFIXUPSET_CALL(__pgd_page, srmmu_pgd_page, BTFIXUPCALL_NORM);

BTFIXUPSET_SETHI(none_mask, 0xF0000000);

diff -prauN mm2-2.5.73-1/arch/sparc/mm/sun4c.c mm2-2.5.73-2/arch/sparc/mm/sun4c.c
--- mm2-2.5.73-1/arch/sparc/mm/sun4c.c 2003-06-22 11:33:06.000000000 -0700
+++ mm2-2.5.73-2/arch/sparc/mm/sun4c.c 2003-06-28 03:11:37.000000000 -0700
@@ -2252,5 +2252,5 @@ void __init ld_mmu_sun4c(void)

/* These should _never_ get called with two level tables. */
BTFIXUPSET_CALL(pgd_set, sun4c_pgd_set, BTFIXUPCALL_NOP);
- BTFIXUPSET_CALL(pgd_page, sun4c_pgd_page, BTFIXUPCALL_RETO0);
+ BTFIXUPSET_CALL(__pgd_page, sun4c_pgd_page, BTFIXUPCALL_RETO0);
}
diff -prauN mm2-2.5.73-1/drivers/char/drm/drm_memory.h mm2-2.5.73-2/drivers/char/drm/drm_memory.h
--- mm2-2.5.73-1/drivers/char/drm/drm_memory.h 2003-06-22 11:32:35.000000000 -0700
+++ mm2-2.5.73-2/drivers/char/drm/drm_memory.h 2003-06-28 03:11:37.000000000 -0700
@@ -123,7 +123,7 @@ static inline unsigned long
drm_follow_page (void *vaddr)
{
pgd_t *pgd = pgd_offset_k((unsigned long) vaddr);
- pmd_t *pmd = pmd_offset(pgd, (unsigned long) vaddr);
+ pmd_t *pmd = pmd_offset_kernel(pgd, (unsigned long)vaddr);
pte_t *ptep = pte_offset_kernel(pmd, (unsigned long) vaddr);
return pte_pfn(*ptep) << PAGE_SHIFT;
}
diff -prauN mm2-2.5.73-1/fs/exec.c mm2-2.5.73-2/fs/exec.c
--- mm2-2.5.73-1/fs/exec.c 2003-06-28 03:09:53.000000000 -0700
+++ mm2-2.5.73-2/fs/exec.c 2003-06-28 03:11:37.000000000 -0700
@@ -304,10 +304,10 @@ void put_dirty_page(struct task_struct *
if (!pte_chain)
goto out_sig;
spin_lock(&tsk->mm->page_table_lock);
- pmd = pmd_alloc(tsk->mm, pgd, address);
+ pmd = pmd_alloc_map(tsk->mm, pgd, address);
if (!pmd)
goto out;
- pte = pte_alloc_map(tsk->mm, pmd, address);
+ pte = pte_alloc_map(tsk->mm, &pmd, address);
if (!pte)
goto out;
if (!pte_none(*pte)) {
@@ -319,6 +319,7 @@ void put_dirty_page(struct task_struct *
set_pte(pte, pte_mkdirty(pte_mkwrite(mk_pte(page, prot))));
pte_chain = page_add_rmap(page, pte, pte_chain);
pte_unmap(pte);
+ pmd_unmap(pmd);
tsk->mm->rss++;
spin_unlock(&tsk->mm->page_table_lock);

@@ -326,6 +327,8 @@ void put_dirty_page(struct task_struct *
pte_chain_free(pte_chain);
return;
out:
+ if (pmd)
+ pmd_unmap(pmd);
spin_unlock(&tsk->mm->page_table_lock);
out_sig:
__free_page(page);
diff -prauN mm2-2.5.73-1/include/asm-alpha/pgtable.h mm2-2.5.73-2/include/asm-alpha/pgtable.h
--- mm2-2.5.73-1/include/asm-alpha/pgtable.h 2003-06-22 11:32:38.000000000 -0700
+++ mm2-2.5.73-2/include/asm-alpha/pgtable.h 2003-06-28 08:20:41.000000000 -0700
@@ -229,9 +229,11 @@ pmd_page_kernel(pmd_t pmd)
#define pmd_page(pmd) (mem_map + ((pmd_val(pmd) & _PFN_MASK) >> 32))
#endif

-extern inline unsigned long pgd_page(pgd_t pgd)
+extern inline unsigned long __pgd_page(pgd_t pgd)
{ return PAGE_OFFSET + ((pgd_val(pgd) & _PFN_MASK) >> (32-PAGE_SHIFT)); }

+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))
+
extern inline int pte_none(pte_t pte) { return !pte_val(pte); }
extern inline int pte_present(pte_t pte) { return pte_val(pte) & _PAGE_VALID; }
extern inline void pte_clear(pte_t *ptep) { pte_val(*ptep) = 0; }
@@ -280,7 +282,7 @@ extern inline pte_t pte_mkyoung(pte_t pt
/* Find an entry in the second-level page table.. */
extern inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
{
- return (pmd_t *) pgd_page(*dir) + ((address >> PMD_SHIFT) & (PTRS_PER_PAGE - 1));
+ return (pmd_t *)__pgd_page(*dir) + ((address >> PMD_SHIFT) & (PTRS_PER_PAGE - 1));
}

/* Find an entry in the third-level page table.. */
diff -prauN mm2-2.5.73-1/include/asm-arm/pgtable.h mm2-2.5.73-2/include/asm-arm/pgtable.h
--- mm2-2.5.73-1/include/asm-arm/pgtable.h 2003-06-22 11:32:38.000000000 -0700
+++ mm2-2.5.73-2/include/asm-arm/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -125,6 +125,11 @@ extern struct page *empty_zero_page;

/* Find an entry in the second-level page table.. */
#define pmd_offset(dir, addr) ((pmd_t *)(dir))
+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)

/* Find an entry in the third-level page table.. */
#define __pte_index(addr) (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
diff -prauN mm2-2.5.73-1/include/asm-arm26/pgtable.h mm2-2.5.73-2/include/asm-arm26/pgtable.h
--- mm2-2.5.73-1/include/asm-arm26/pgtable.h 2003-06-22 11:32:32.000000000 -0700
+++ mm2-2.5.73-2/include/asm-arm26/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -189,6 +189,12 @@ extern struct page *empty_zero_page;
#define pte_unmap(pte) do { } while (0)
#define pte_unmap_nested(pte) do { } while (0)

+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pgd, addr) do { } while (0)
+#define pmd_unmap_nested(pgd, addr) do { } while (0)
+

#define _PAGE_PRESENT 0x01
#define _PAGE_READONLY 0x02
diff -prauN mm2-2.5.73-1/include/asm-h8300/pgtable.h mm2-2.5.73-2/include/asm-h8300/pgtable.h
--- mm2-2.5.73-1/include/asm-h8300/pgtable.h 2003-06-22 11:32:42.000000000 -0700
+++ mm2-2.5.73-2/include/asm-h8300/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -15,6 +15,11 @@ typedef pte_t *pte_addr_t;
#define pgd_clear(pgdp)
#define kern_addr_valid(addr) (1)
#define pmd_offset(a, b) ((void *)0)
+#define pmd_offset_kernel(a,b) pmd_offset(a,b)
+#define pmd_offset_map(a,b) pmd_offset(a,b)
+#define pmd_offset_map_nested(a,b) pmd_offset(a,b)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)

#define PAGE_NONE __pgprot(0) /* these mean nothing to NO_MM */
#define PAGE_SHARED __pgprot(0) /* these mean nothing to NO_MM */
diff -prauN mm2-2.5.73-1/include/asm-i386/kmap_types.h mm2-2.5.73-2/include/asm-i386/kmap_types.h
--- mm2-2.5.73-1/include/asm-i386/kmap_types.h 2003-06-22 11:33:01.000000000 -0700
+++ mm2-2.5.73-2/include/asm-i386/kmap_types.h 2003-06-28 03:11:37.000000000 -0700
@@ -17,14 +17,16 @@ D(3) KM_USER0,
D(4) KM_USER1,
D(5) KM_BIO_SRC_IRQ,
D(6) KM_BIO_DST_IRQ,
-D(7) KM_PTE0,
-D(8) KM_PTE1,
-D(9) KM_PTE2,
-D(10) KM_IRQ0,
-D(11) KM_IRQ1,
-D(12) KM_SOFTIRQ0,
-D(13) KM_SOFTIRQ1,
-D(14) KM_TYPE_NR
+D(7) KM_PMD0,
+D(8) KM_PMD1,
+D(9) KM_PTE0,
+D(10) KM_PTE1,
+D(11) KM_PTE2,
+D(12) KM_IRQ0,
+D(13) KM_IRQ1,
+D(14) KM_SOFTIRQ0,
+D(15) KM_SOFTIRQ1,
+D(16) KM_TYPE_NR
};

#undef D
diff -prauN mm2-2.5.73-1/include/asm-i386/pgalloc.h mm2-2.5.73-2/include/asm-i386/pgalloc.h
--- mm2-2.5.73-1/include/asm-i386/pgalloc.h 2003-06-22 11:32:31.000000000 -0700
+++ mm2-2.5.73-2/include/asm-i386/pgalloc.h 2003-06-28 08:06:24.000000000 -0700
@@ -46,6 +46,7 @@ static inline void pte_free(struct page
*/

#define pmd_alloc_one(mm, addr) ({ BUG(); ((pmd_t *)2); })
+#define pmd_alloc_one_kernel(mm, addr) ({ BUG(); ((pmd_t *)2); })
#define pmd_free(x) do { } while (0)
#define __pmd_free_tlb(tlb,x) do { } while (0)
#define pgd_populate(mm, pmd, pte) BUG()
diff -prauN mm2-2.5.73-1/include/asm-i386/pgtable-2level.h mm2-2.5.73-2/include/asm-i386/pgtable-2level.h
--- mm2-2.5.73-1/include/asm-i386/pgtable-2level.h 2003-06-22 11:32:55.000000000 -0700
+++ mm2-2.5.73-2/include/asm-i386/pgtable-2level.h 2003-06-28 03:11:37.000000000 -0700
@@ -48,13 +48,15 @@ static inline int pgd_present(pgd_t pgd)
#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
#define set_pgd(pgdptr, pgdval) (*(pgdptr) = pgdval)

-#define pgd_page(pgd) \
-((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
+#define pgd_page(pgd) pfn_to_page(pgd_val(pgd) >> PAGE_SHIFT)
+
+#define pmd_offset_map(pgd, addr) ({ (pmd_t *)(pgd); })
+#define pmd_offset_map_nested(pgd, addr) pmd_offset_map(pgd, addr)
+#define pmd_offset_kernel(pgd, addr) pmd_offset_map(pgd, addr)
+
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)

-static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
-{
- return (pmd_t *) dir;
-}
#define ptep_get_and_clear(xp) __pte(xchg(&(xp)->pte_low, 0))
#define pte_same(a, b) ((a).pte_low == (b).pte_low)
#define pte_page(x) pfn_to_page(pte_pfn(x))
diff -prauN mm2-2.5.73-1/include/asm-i386/pgtable-3level.h mm2-2.5.73-2/include/asm-i386/pgtable-3level.h
--- mm2-2.5.73-1/include/asm-i386/pgtable-3level.h 2003-06-28 03:09:54.000000000 -0700
+++ mm2-2.5.73-2/include/asm-i386/pgtable-3level.h 2003-06-28 08:21:14.000000000 -0700
@@ -64,12 +64,32 @@ static inline void set_pte(pte_t *ptep,
*/
static inline void pgd_clear (pgd_t * pgd) { }

-#define pgd_page(pgd) \
-((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
+static inline unsigned long pgd_pfn(pgd_t pgd)
+{
+ return pgd_val(pgd) >> PAGE_SHIFT;
+}
+
+#define pgd_page(pgd) pfn_to_page(pgd_pfn(pgd))
+
+#define pmd_offset_kernel(pgd, addr) \
+ ((pmd_t *)__va(pgd_val(*(pgd)) & PAGE_MASK) + pmd_index(addr))

/* Find an entry in the second-level page table.. */
-#define pmd_offset(dir, address) ((pmd_t *) pgd_page(*(dir)) + \
- pmd_index(address))
+#ifdef CONFIG_HIGHPMD
+#define __pmd_offset(pgd, addr, type) \
+ ((pmd_t *)kmap_atomic(pgd_page(*(pgd)), type) + pmd_index(addr))
+#define __pmd_unmap(pmd, type) kunmap_atomic(pmd, type)
+#else
+#define __pmd_offset(pgd, addr, type) \
+ ((pmd_t *)__va(pgd_val(*(pgd)) & PAGE_MASK) + pmd_index(addr))
+#define __pmd_unmap(pmd, type) do { } while (0)
+#endif
+
+#define pmd_offset_map(pgd, addr) __pmd_offset(pgd, addr, KM_PMD0)
+#define pmd_offset_map_nested(pgd, addr) __pmd_offset(pgd, addr, KM_PMD1)
+
+#define pmd_unmap(pmd) __pmd_unmap(pmd, KM_PMD0);
+#define pmd_unmap_nested(pmd) __pmd_unmap(pmd, KM_PMD1);

static inline pte_t ptep_get_and_clear(pte_t *ptep)
{
diff -prauN mm2-2.5.73-1/include/asm-i386/pgtable.h mm2-2.5.73-2/include/asm-i386/pgtable.h
--- mm2-2.5.73-1/include/asm-i386/pgtable.h 2003-06-28 03:09:54.000000000 -0700
+++ mm2-2.5.73-2/include/asm-i386/pgtable.h 2003-06-28 08:18:44.000000000 -0700
@@ -33,11 +33,9 @@
extern unsigned long empty_zero_page[1024];
extern pgd_t swapper_pg_dir[1024];
extern kmem_cache_t *pgd_cache;
-extern kmem_cache_t *pmd_cache;
extern spinlock_t pgd_lock;
extern struct list_head pgd_list;

-void pmd_ctor(void *, kmem_cache_t *, unsigned long);
void pgd_ctor(void *, kmem_cache_t *, unsigned long);
void pgd_dtor(void *, kmem_cache_t *, unsigned long);
void pgtable_cache_init(void);
diff -prauN mm2-2.5.73-1/include/asm-ia64/pgtable.h mm2-2.5.73-2/include/asm-ia64/pgtable.h
--- mm2-2.5.73-1/include/asm-ia64/pgtable.h 2003-06-22 11:32:39.000000000 -0700
+++ mm2-2.5.73-2/include/asm-ia64/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -257,7 +257,8 @@ ia64_phys_addr_valid (unsigned long addr
#define pgd_bad(pgd) (!ia64_phys_addr_valid(pgd_val(pgd)))
#define pgd_present(pgd) (pgd_val(pgd) != 0UL)
#define pgd_clear(pgdp) (pgd_val(*(pgdp)) = 0UL)
-#define pgd_page(pgd) ((unsigned long) __va(pgd_val(pgd) & _PFN_MASK))
+#define __pgd_page(pgd) ((unsigned long)__va(pgd_val(pgd) & _PFN_MASK))
+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))

/*
* The following have defined behavior only work if pte_present() is true.
@@ -326,7 +327,13 @@ pgd_offset (struct mm_struct *mm, unsign

/* Find an entry in the second-level page table.. */
#define pmd_offset(dir,addr) \
- ((pmd_t *) pgd_page(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
+ ((pmd_t *)__pgd_page(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
+
+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)

/*
* Find an entry in the third-level page table. This looks more complicated than it
diff -prauN mm2-2.5.73-1/include/asm-m68k/motorola_pgtable.h mm2-2.5.73-2/include/asm-m68k/motorola_pgtable.h
--- mm2-2.5.73-1/include/asm-m68k/motorola_pgtable.h 2003-06-22 11:32:57.000000000 -0700
+++ mm2-2.5.73-2/include/asm-m68k/motorola_pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -115,6 +115,7 @@ extern inline void pgd_set(pgd_t * pgdp,
#define __pte_page(pte) ((unsigned long)__va(pte_val(pte) & PAGE_MASK))
#define __pmd_page(pmd) ((unsigned long)__va(pmd_val(pmd) & _TABLE_MASK))
#define __pgd_page(pgd) ((unsigned long)__va(pgd_val(pgd) & _TABLE_MASK))
+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))

#define pte_none(pte) (!pte_val(pte))
diff -prauN mm2-2.5.73-1/include/asm-m68knommu/pgtable.h mm2-2.5.73-2/include/asm-m68knommu/pgtable.h
--- mm2-2.5.73-1/include/asm-m68knommu/pgtable.h 2003-06-22 11:32:56.000000000 -0700
+++ mm2-2.5.73-2/include/asm-m68knommu/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -21,7 +21,12 @@ typedef pte_t *pte_addr_t;
#define pgd_bad(pgd) (0)
#define pgd_clear(pgdp)
#define kern_addr_valid(addr) (1)
-#define pmd_offset(a, b) ((void *)0)
+#define pmd_offset(a, b) ((void *)0)
+#define pmd_offset_kernel(a, b) pmd_offset(a, b)
+#define pmd_offset_map(a, b) pmd_offset(a, b)
+#define pmd_offset_map_nested(a, b) pmd_offset(a, b)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)

#define PAGE_NONE __pgprot(0)
#define PAGE_SHARED __pgprot(0)
diff -prauN mm2-2.5.73-1/include/asm-mips64/pgtable.h mm2-2.5.73-2/include/asm-mips64/pgtable.h
--- mm2-2.5.73-1/include/asm-mips64/pgtable.h 2003-06-28 03:09:55.000000000 -0700
+++ mm2-2.5.73-2/include/asm-mips64/pgtable.h 2003-06-28 03:16:19.000000000 -0700
@@ -155,11 +155,13 @@ extern pmd_t empty_bad_pmd_table[2*PAGE_
#define pmd_page(pmd) (pfn_to_page(pmd_phys(pmd) >> PAGE_SHIFT))
#define pmd_page_kernel(pmd) pmd_val(pmd)

-static inline unsigned long pgd_page(pgd_t pgd)
+static inline unsigned long __pgd_page(pgd_t pgd)
{
return pgd_val(pgd);
}

+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))
+
static inline int pte_none(pte_t pte)
{
return !(pte_val(pte) & ~_PAGE_GLOBAL);
@@ -397,7 +399,7 @@ static inline pte_t pte_modify(pte_t pte
/* Find an entry in the second-level page table.. */
static inline pmd_t *pmd_offset(pgd_t * dir, unsigned long address)
{
- return (pmd_t *) pgd_page(*dir) +
+ return (pmd_t *)__pgd_page(*dir) +
((address >> PMD_SHIFT) & (PTRS_PER_PMD - 1));
}

diff -prauN mm2-2.5.73-1/include/asm-parisc/pgtable.h mm2-2.5.73-2/include/asm-parisc/pgtable.h
--- mm2-2.5.73-1/include/asm-parisc/pgtable.h 2003-06-22 11:33:15.000000000 -0700
+++ mm2-2.5.73-2/include/asm-parisc/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -242,7 +242,8 @@ extern unsigned long *empty_zero_page;

#ifdef __LP64__
-#define pgd_page(pgd) ((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
+#define __pgd_page(pgd) ((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))

/* For 64 bit we have three level tables */

@@ -339,11 +340,17 @@ extern inline pte_t pte_modify(pte_t pte

#ifdef __LP64__
#define pmd_offset(dir,address) \
-((pmd_t *) pgd_page(*(dir)) + (((address)>>PMD_SHIFT) & (PTRS_PER_PMD-1)))
+((pmd_t *)__pgd_page(*(dir)) + (((address)>>PMD_SHIFT) & (PTRS_PER_PMD-1)))
#else
#define pmd_offset(dir,addr) ((pmd_t *) dir)
#endif

+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)
+
/* Find an entry in the third-level page table.. */
#define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE-1))
#define pte_offset_kernel(pmd, address) \
diff -prauN mm2-2.5.73-1/include/asm-ppc/pgtable.h mm2-2.5.73-2/include/asm-ppc/pgtable.h
--- mm2-2.5.73-1/include/asm-ppc/pgtable.h 2003-06-22 11:32:37.000000000 -0700
+++ mm2-2.5.73-2/include/asm-ppc/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -370,8 +370,9 @@ static inline int pgd_bad(pgd_t pgd) {
static inline int pgd_present(pgd_t pgd) { return 1; }
#define pgd_clear(xp) do { } while (0)

-#define pgd_page(pgd) \
+#define __pgd_page(pgd) \
((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))

/*
* The following only work if pte_present() is true.
diff -prauN mm2-2.5.73-1/include/asm-ppc64/pgtable.h mm2-2.5.73-2/include/asm-ppc64/pgtable.h
--- mm2-2.5.73-1/include/asm-ppc64/pgtable.h 2003-06-22 11:33:18.000000000 -0700
+++ mm2-2.5.73-2/include/asm-ppc64/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -190,7 +190,8 @@ extern unsigned long empty_zero_page[PAG
#define pgd_bad(pgd) ((pgd_val(pgd)) == 0)
#define pgd_present(pgd) (pgd_val(pgd) != 0UL)
#define pgd_clear(pgdp) (pgd_val(*(pgdp)) = 0UL)
-#define pgd_page(pgd) (__bpn_to_ba(pgd_val(pgd)))
+#define __pgd_page(pgd) (__bpn_to_ba(pgd_val(pgd)))
+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))

/*
* Find an entry in a page-table-directory. We combine the address region
@@ -203,12 +204,18 @@ extern unsigned long empty_zero_page[PAG

/* Find an entry in the second-level page table.. */
#define pmd_offset(dir,addr) \
- ((pmd_t *) pgd_page(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))
+ ((pmd_t *)__pgd_page(*(dir)) + (((addr) >> PMD_SHIFT) & (PTRS_PER_PMD - 1)))

/* Find an entry in the third-level page table.. */
#define pte_offset_kernel(dir,addr) \
((pte_t *) pmd_page_kernel(*(dir)) + (((addr) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1)))

+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)
+
#define pte_offset_map(dir,addr) pte_offset_kernel((dir), (addr))
#define pte_offset_map_nested(dir,addr) pte_offset_kernel((dir), (addr))
#define pte_unmap(pte) do { } while(0)
diff -prauN mm2-2.5.73-1/include/asm-s390/pgtable.h mm2-2.5.73-2/include/asm-s390/pgtable.h
--- mm2-2.5.73-1/include/asm-s390/pgtable.h 2003-06-22 11:33:07.000000000 -0700
+++ mm2-2.5.73-2/include/asm-s390/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -613,6 +613,7 @@ static inline pte_t mk_pte_phys(unsigned
/* to find an entry in a page-table-directory */
#define pgd_index(address) ((address >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
#define pgd_offset(mm, address) ((mm)->pgd+pgd_index(address))
+#define pgd_page(pgd) virt_to_page(pgd_page_kernel(pgd))

/* to find an entry in a kernel page-table-directory */
#define pgd_offset_k(address) pgd_offset(&init_mm, address)
@@ -634,6 +635,12 @@ extern inline pmd_t * pmd_offset(pgd_t *

#endif /* __s390x__ */

+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)
+
/* Find an entry in the third-level page table.. */
#define pte_index(address) (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE-1))
#define pte_offset_kernel(pmd, address) \
diff -prauN mm2-2.5.73-1/include/asm-sh/pgtable-2level.h mm2-2.5.73-2/include/asm-sh/pgtable-2level.h
--- mm2-2.5.73-1/include/asm-sh/pgtable-2level.h 2003-06-22 11:33:32.000000000 -0700
+++ mm2-2.5.73-2/include/asm-sh/pgtable-2level.h 2003-06-28 03:11:37.000000000 -0700
@@ -48,8 +48,9 @@ static inline void pgd_clear (pgd_t * pg
#define set_pmd(pmdptr, pmdval) (*(pmdptr) = pmdval)
#define set_pgd(pgdptr, pgdval) (*(pgdptr) = pgdval)

-#define pgd_page(pgd) \
+#define __pgd_page(pgd) \
((unsigned long) __va(pgd_val(pgd) & PAGE_MASK))
+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))

static inline pmd_t * pmd_offset(pgd_t * dir, unsigned long address)
{
diff -prauN mm2-2.5.73-1/include/asm-sparc/pgtable.h mm2-2.5.73-2/include/asm-sparc/pgtable.h
--- mm2-2.5.73-1/include/asm-sparc/pgtable.h 2003-06-22 11:32:56.000000000 -0700
+++ mm2-2.5.73-2/include/asm-sparc/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -202,10 +202,11 @@ extern unsigned long empty_zero_page;
/*
*/
BTFIXUPDEF_CALL_CONST(struct page *, pmd_page, pmd_t)
-BTFIXUPDEF_CALL_CONST(unsigned long, pgd_page, pgd_t)
+BTFIXUPDEF_CALL_CONST(unsigned long, __pgd_page, pgd_t)

#define pmd_page(pmd) BTFIXUP_CALL(pmd_page)(pmd)
-#define pgd_page(pgd) BTFIXUP_CALL(pgd_page)(pgd)
+#define __pgd_page(pgd) BTFIXUP_CALL(__pgd_page)(pgd)
+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))

BTFIXUPDEF_SETHI(none_mask)
BTFIXUPDEF_CALL_CONST(int, pte_present, pte_t)
@@ -352,6 +353,11 @@ extern __inline__ pte_t pte_modify(pte_t
/* Find an entry in the second-level page table.. */
BTFIXUPDEF_CALL(pmd_t *, pmd_offset, pgd_t *, unsigned long)
#define pmd_offset(dir,addr) BTFIXUP_CALL(pmd_offset)(dir,addr)
+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)

/* Find an entry in the third-level page table.. */
BTFIXUPDEF_CALL(pte_t *, pte_offset_kernel, pmd_t *, unsigned long)
diff -prauN mm2-2.5.73-1/include/asm-sparc64/pgtable.h mm2-2.5.73-2/include/asm-sparc64/pgtable.h
--- mm2-2.5.73-1/include/asm-sparc64/pgtable.h 2003-06-22 11:32:31.000000000 -0700
+++ mm2-2.5.73-2/include/asm-sparc64/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -228,7 +228,8 @@ static inline pte_t pte_modify(pte_t ori
(pgd_val(*(pgdp)) = (__pa((unsigned long) (pmdp)) >> 11UL))
#define __pmd_page(pmd) ((unsigned long) __va((pmd_val(pmd)<<11UL)))
#define pmd_page(pmd) virt_to_page((void *)__pmd_page(pmd))
-#define pgd_page(pgd) ((unsigned long) __va((pgd_val(pgd)<<11UL)))
+#define __pgd_page(pgd) ((unsigned long) __va((pgd_val(pgd)<<11UL)))
+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))
#define pte_none(pte) (!pte_val(pte))
#define pte_present(pte) (pte_val(pte) & _PAGE_PRESENT)
#define pte_clear(pte) (pte_val(*(pte)) = 0UL)
@@ -270,8 +271,13 @@ static inline pte_t pte_modify(pte_t ori
#define pgd_offset_k(address) pgd_offset(&init_mm, address)

/* Find an entry in the second-level page table.. */
-#define pmd_offset(dir, address) ((pmd_t *) pgd_page(*(dir)) + \
+#define pmd_offset(dir, address) ((pmd_t *)__pgd_page(*(dir)) + \
((address >> PMD_SHIFT) & (REAL_PTRS_PER_PMD-1)))
+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)

/* Find an entry in the third-level page table.. */
#define pte_index(dir, address) ((pte_t *) __pmd_page(*(dir)) + \
diff -prauN mm2-2.5.73-1/include/asm-v850/pgtable.h mm2-2.5.73-2/include/asm-v850/pgtable.h
--- mm2-2.5.73-1/include/asm-v850/pgtable.h 2003-06-22 11:32:58.000000000 -0700
+++ mm2-2.5.73-2/include/asm-v850/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -13,6 +13,11 @@ typedef pte_t *pte_addr_t;
#define pgd_clear(pgdp) ((void)0)

#define pmd_offset(a, b) ((void *)0)
+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)

#define kern_addr_valid(addr) (1)

diff -prauN mm2-2.5.73-1/include/asm-x86_64/pgtable.h mm2-2.5.73-2/include/asm-x86_64/pgtable.h
--- mm2-2.5.73-1/include/asm-x86_64/pgtable.h 2003-06-28 03:09:57.000000000 -0700
+++ mm2-2.5.73-2/include/asm-x86_64/pgtable.h 2003-06-28 03:11:37.000000000 -0700
@@ -98,8 +98,9 @@ static inline void set_pml4(pml4_t *dst,
pml4_val(*dst) = pml4_val(val);
}

-#define pgd_page(pgd) \
+#define __pgd_page(pgd) \
((unsigned long) __va(pgd_val(pgd) & PHYSICAL_PAGE_MASK))
+#define pgd_page(pgd) virt_to_page(__pgd_page(pgd))

#define ptep_get_and_clear(xp) __pte(xchg(&(xp)->pte, 0))
#define pte_same(a, b) ((a).pte == (b).pte)
@@ -332,8 +333,13 @@ static inline pgd_t *current_pgd_offset_
#define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))

#define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
-#define pmd_offset(dir, address) ((pmd_t *) pgd_page(*(dir)) + \
+#define pmd_offset(dir, address) ((pmd_t *)__pgd_page(*(dir)) + \
pmd_index(address))
+#define pmd_offset_kernel(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_offset_map_nested(pgd, addr) pmd_offset(pgd, addr)
+#define pmd_unmap(pmd) do { } while (0)
+#define pmd_unmap_nested(pmd) do { } while (0)
#define pmd_none(x) (!pmd_val(x))
#define pmd_present(x) (pmd_val(x) & _PAGE_PRESENT)
#define pmd_clear(xp) do { set_pmd(xp, __pmd(0)); } while (0)
diff -prauN mm2-2.5.73-1/include/linux/mm.h mm2-2.5.73-2/include/linux/mm.h
--- mm2-2.5.73-1/include/linux/mm.h 2003-06-28 03:09:57.000000000 -0700
+++ mm2-2.5.73-2/include/linux/mm.h 2003-06-28 03:11:37.000000000 -0700
@@ -426,8 +426,9 @@ extern void invalidate_mmap_range(struct
loff_t const holelen);
extern int vmtruncate(struct inode * inode, loff_t offset);
extern pmd_t *FASTCALL(__pmd_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address));
+pmd_t *FASTCALL(__pmd_alloc_kernel(struct mm_struct *mm, pgd_t *pmd, unsigned long address));
extern pte_t *FASTCALL(pte_alloc_kernel(struct mm_struct *mm, pmd_t *pmd, unsigned long address));
-extern pte_t *FASTCALL(pte_alloc_map(struct mm_struct *mm, pmd_t *pmd, unsigned long address));
+pte_t *FASTCALL(pte_alloc_map(struct mm_struct *mm, pmd_t **pmd, unsigned long address));
extern int install_page(struct mm_struct *mm, struct vm_area_struct *vma, unsigned long addr, struct page *page, pgprot_t prot);
extern int handle_mm_fault(struct mm_struct *mm,struct vm_area_struct *vma, unsigned long address, int write_access);
extern int make_pages_present(unsigned long addr, unsigned long end);
@@ -488,12 +489,11 @@ static inline int set_page_dirty(struct
* inlining and the symmetry break with pte_alloc_map() that does all
* of this out-of-line.
*/
-static inline pmd_t *pmd_alloc(struct mm_struct *mm, pgd_t *pgd, unsigned long address)
-{
- if (pgd_none(*pgd))
- return __pmd_alloc(mm, pgd, address);
- return pmd_offset(pgd, address);
-}
+#define pmd_alloc_map(mm, pgd, addr) \
+ (pgd_none(*(pgd))? __pmd_alloc(mm,pgd,addr): pmd_offset_map(pgd,addr))
+
+#define pmd_alloc_kernel(mm, pgd, addr) \
+ (pgd_none(*(pgd))? __pmd_alloc_kernel(mm,pgd,addr): pmd_offset_kernel(pgd,addr))

extern void free_area_init(unsigned long * zones_size);
extern void free_area_init_node(int nid, pg_data_t *pgdat, struct page *pmap,
diff -prauN mm2-2.5.73-1/mm/fremap.c mm2-2.5.73-2/mm/fremap.c
--- mm2-2.5.73-1/mm/fremap.c 2003-06-22 11:32:31.000000000 -0700
+++ mm2-2.5.73-2/mm/fremap.c 2003-06-28 03:11:37.000000000 -0700
@@ -67,11 +67,11 @@ int install_page(struct mm_struct *mm, s
pgd = pgd_offset(mm, addr);
spin_lock(&mm->page_table_lock);

- pmd = pmd_alloc(mm, pgd, addr);
+ pmd = pmd_alloc_map(mm, pgd, addr);
if (!pmd)
goto err_unlock;

- pte = pte_alloc_map(mm, pmd, addr);
+ pte = pte_alloc_map(mm, &pmd, addr);
if (!pte)
goto err_unlock;

@@ -82,6 +82,7 @@ int install_page(struct mm_struct *mm, s
set_pte(pte, mk_pte(page, prot));
pte_chain = page_add_rmap(page, pte, pte_chain);
pte_unmap(pte);
+ pmd_unmap(pmd);
if (flush)
flush_tlb_page(vma, addr);
update_mmu_cache(vma, addr, *pte);
diff -prauN mm2-2.5.73-1/mm/memory.c mm2-2.5.73-2/mm/memory.c
--- mm2-2.5.73-1/mm/memory.c 2003-06-28 03:09:58.000000000 -0700
+++ mm2-2.5.73-2/mm/memory.c 2003-06-28 08:35:17.000000000 -0700
@@ -103,6 +103,7 @@ static inline void free_one_pmd(struct m
static inline void free_one_pgd(struct mmu_gather *tlb, pgd_t * dir)
{
pmd_t * pmd, * md, * emd;
+ struct page *page;

if (pgd_none(*dir))
return;
@@ -111,7 +112,8 @@ static inline void free_one_pgd(struct m
pgd_clear(dir);
return;
}
- pmd = pmd_offset(dir, 0);
+ page = pgd_page(*dir);
+ pmd = pmd_offset_map(dir, 0);
pgd_clear(dir);
/*
* Beware if changing the loop below. It once used int j,
@@ -128,7 +130,8 @@ static inline void free_one_pgd(struct m
*/
for (md = pmd, emd = pmd + PTRS_PER_PMD; md < emd; md++)
free_one_pmd(tlb,md);
- pmd_free_tlb(tlb, pmd);
+ pmd_unmap(pmd);
+ pmd_free_tlb(tlb, page);
}

/*
@@ -148,30 +151,40 @@ void clear_page_tables(struct mmu_gather
} while (--nr);
}

-pte_t * pte_alloc_map(struct mm_struct *mm, pmd_t *pmd, unsigned long address)
+/*
+ * error return happens with pmd unmapped
+ */
+pte_t *pte_alloc_map(struct mm_struct *mm, pmd_t **pmd, unsigned long address)
{
- if (!pmd_present(*pmd)) {
+ if (!pmd_present(**pmd)) {
+ pgd_t *pgd;
struct page *new;

+ pmd_unmap(*pmd);
spin_unlock(&mm->page_table_lock);
new = pte_alloc_one(mm, address);
spin_lock(&mm->page_table_lock);
- if (!new)
+ if (!new) {
+ *pmd = NULL;
return NULL;
+ }
+
+ pgd = pgd_offset(mm, address);
+ *pmd = pmd_offset_map(pgd, address);

/*
* Because we dropped the lock, we should re-check the
* entry, as somebody else could have populated it..
*/
- if (pmd_present(*pmd)) {
+ if (pmd_present(**pmd)) {
pte_free(new);
goto out;
}
pgtable_add_rmap(new, mm, address);
- pmd_populate(mm, pmd, new);
+ pmd_populate(mm, *pmd, new);
}
out:
- return pte_offset_map(pmd, address);
+ return pte_offset_map(*pmd, address);
}

pte_t * pte_alloc_kernel(struct mm_struct *mm, pmd_t *pmd, unsigned long address)
@@ -256,11 +269,10 @@ skip_copy_pmd_range: address = (address
continue;
}

- src_pmd = pmd_offset(src_pgd, address);
- dst_pmd = pmd_alloc(dst, dst_pgd, address);
+ dst_pmd = pmd_alloc_map(dst, dst_pgd, address);
if (!dst_pmd)
goto nomem;
-
+ src_pmd = pmd_offset_map_nested(src_pgd, address);
do {
pte_t * src_pte, * dst_pte;

@@ -273,15 +285,20 @@ skip_copy_pmd_range: address = (address
pmd_clear(src_pmd);
skip_copy_pte_range:
address = (address + PMD_SIZE) & PMD_MASK;
- if (address >= end)
+ if (address >= end) {
+ pmd_unmap(dst_pmd);
+ pmd_unmap_nested(src_pmd);
goto out;
+ }
goto cont_copy_pmd_range;
}

- dst_pte = pte_alloc_map(dst, dst_pmd, address);
+ pmd_unmap_nested(src_pmd);
+ dst_pte = pte_alloc_map(dst, &dst_pmd, address);
if (!dst_pte)
goto nomem;
spin_lock(&src->page_table_lock);
+ src_pmd = pmd_offset_map_nested(src_pgd, address);
src_pte = pte_offset_map_nested(src_pmd, address);
do {
pte_t pte = *src_pte;
@@ -348,6 +365,8 @@ skip_copy_pte_range:
*/
pte_unmap_nested(src_pte);
pte_unmap(dst_pte);
+ pmd_unmap_nested(src_pmd);
+ pmd_unmap(dst_pmd);
spin_unlock(&src->page_table_lock);
spin_unlock(&dst->page_table_lock);
pte_chain = pte_chain_alloc(GFP_KERNEL);
@@ -355,12 +374,16 @@ skip_copy_pte_range:
if (!pte_chain)
goto nomem;
spin_lock(&src->page_table_lock);
+ dst_pmd = pmd_offset_map(dst_pgd, address);
+ src_pmd = pmd_offset_map_nested(src_pgd, address);
dst_pte = pte_offset_map(dst_pmd, address);
src_pte = pte_offset_map_nested(src_pmd,
address);
cont_copy_pte_range_noset:
address += PAGE_SIZE;
if (address >= end) {
+ pmd_unmap(dst_pmd);
+ pmd_unmap_nested(src_pmd);
pte_unmap_nested(src_pte);
pte_unmap(dst_pte);
goto out_unlock;
@@ -376,6 +399,8 @@ cont_copy_pmd_range:
src_pmd++;
dst_pmd++;
} while ((unsigned long)src_pmd & PMD_TABLE_MASK);
+ pmd_unmap_nested(src_pmd-1);
+ pmd_unmap(dst_pmd-1);
}
out_unlock:
spin_unlock(&src->page_table_lock);
@@ -451,7 +476,7 @@ zap_pmd_range(struct mmu_gather *tlb, pg
pgd_clear(dir);
return;
}
- pmd = pmd_offset(dir, address);
+ pmd = pmd_offset_map(dir, address);
end = address + size;
if (end > ((address + PGDIR_SIZE) & PGDIR_MASK))
end = ((address + PGDIR_SIZE) & PGDIR_MASK);
@@ -460,6 +485,7 @@ zap_pmd_range(struct mmu_gather *tlb, pg
address = (address + PMD_SIZE) & PMD_MASK;
pmd++;
} while (address < end);
+ pmd_unmap(pmd - 1);
}

void unmap_page_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
@@ -641,20 +667,27 @@ follow_page(struct mm_struct *mm, unsign
if (pgd_none(*pgd) || pgd_bad(*pgd))
goto out;

- pmd = pmd_offset(pgd, address);
+ pmd = pmd_offset_map(pgd, address);
if (pmd_none(*pmd))
- goto out;
- if (pmd_huge(*pmd))
- return follow_huge_pmd(mm, address, pmd, write);
- if (pmd_bad(*pmd))
- goto out;
+ goto out_unmap;
+ if (pmd_bad(*pmd)) {
+ pmd_ERROR(*pmd);
+ pmd_clear(pmd);
+ goto out_unmap;
+ }
+ if (pmd_huge(*pmd)) {
+ struct page *page = follow_huge_pmd(mm, address, pmd, write);
+ pmd_unmap(pmd);
+ return page;
+ }

ptep = pte_offset_map(pmd, address);
if (!ptep)
- goto out;
+ goto out_unmap;

pte = *ptep;
pte_unmap(ptep);
+ pmd_unmap(pmd);
if (pte_present(pte)) {
if (!write || (pte_write(pte) && pte_dirty(pte))) {
pfn = pte_pfn(pte);
@@ -665,6 +698,9 @@ follow_page(struct mm_struct *mm, unsign

out:
return NULL;
+out_unmap:
+ pmd_unmap(pmd);
+ goto out;
}

/*
@@ -723,7 +759,7 @@ int get_user_pages(struct task_struct *t
pgd = pgd_offset_k(pg);
if (!pgd)
return i ? : -EFAULT;
- pmd = pmd_offset(pgd, pg);
+ pmd = pmd_offset_kernel(pgd, pg);
if (!pmd)
return i ? : -EFAULT;
pte = pte_offset_kernel(pmd, pg);
@@ -815,8 +851,8 @@ static void zeromap_pte_range(pte_t * pt
} while (address && (address < end));
}

-static inline int zeromap_pmd_range(struct mm_struct *mm, pmd_t * pmd, unsigned long address,
- unsigned long size, pgprot_t prot)
+static inline int zeromap_pmd_range(struct mm_struct *mm, pmd_t **pmd,
+ unsigned long address, unsigned long size, pgprot_t prot)
{
unsigned long end;

@@ -825,13 +861,13 @@ static inline int zeromap_pmd_range(stru
if (end > PGDIR_SIZE)
end = PGDIR_SIZE;
do {
- pte_t * pte = pte_alloc_map(mm, pmd, address);
+ pte_t *pte = pte_alloc_map(mm, pmd, address);
if (!pte)
return -ENOMEM;
zeromap_pte_range(pte, address, end - address, prot);
pte_unmap(pte);
address = (address + PMD_SIZE) & PMD_MASK;
- pmd++;
+ (*pmd)++;
} while (address && (address < end));
return 0;
}
@@ -851,13 +887,14 @@ int zeromap_page_range(struct vm_area_st

spin_lock(&mm->page_table_lock);
do {
- pmd_t *pmd = pmd_alloc(mm, dir, address);
+ pmd_t *pmd = pmd_alloc_map(mm, dir, address);
error = -ENOMEM;
if (!pmd)
break;
- error = zeromap_pmd_range(mm, pmd, address, end - address, prot);
+ error = zeromap_pmd_range(mm, &pmd, address, end - address, prot);
if (error)
break;
+ pmd_unmap(pmd - 1);
address = (address + PGDIR_SIZE) & PGDIR_MASK;
dir++;
} while (address && (address < end));
@@ -892,8 +929,9 @@ static inline void remap_pte_range(pte_t
} while (address && (address < end));
}

-static inline int remap_pmd_range(struct mm_struct *mm, pmd_t * pmd, unsigned long address, unsigned long size,
- unsigned long phys_addr, pgprot_t prot)
+static inline int remap_pmd_range(struct mm_struct *mm, pmd_t **pmd,
+ unsigned long address, unsigned long size,
+ unsigned long phys_addr, pgprot_t prot)
{
unsigned long base, end;

@@ -904,13 +942,13 @@ static inline int remap_pmd_range(struct
end = PGDIR_SIZE;
phys_addr -= address;
do {
- pte_t * pte = pte_alloc_map(mm, pmd, base + address);
+ pte_t *pte = pte_alloc_map(mm, pmd, base + address);
if (!pte)
return -ENOMEM;
remap_pte_range(pte, base + address, end - address, address + phys_addr, prot);
pte_unmap(pte);
address = (address + PMD_SIZE) & PMD_MASK;
- pmd++;
+ (*pmd)++;
} while (address && (address < end));
return 0;
}
@@ -932,13 +970,14 @@ int remap_page_range(struct vm_area_stru

spin_lock(&mm->page_table_lock);
do {
- pmd_t *pmd = pmd_alloc(mm, dir, from);
+ pmd_t *pmd = pmd_alloc_map(mm, dir, from);
error = -ENOMEM;
if (!pmd)
break;
- error = remap_pmd_range(mm, pmd, from, end - from, phys_addr + from, prot);
+ error = remap_pmd_range(mm, &pmd, from, end - from, phys_addr + from, prot);
if (error)
break;
+ pmd_unmap(pmd - 1);
from = (from + PGDIR_SIZE) & PGDIR_MASK;
dir++;
} while (from && (from < end));
@@ -1008,6 +1047,7 @@ static int do_wp_page(struct mm_struct *
* data, but for the moment just pretend this is OOM.
*/
pte_unmap(page_table);
+ pmd_unmap(pmd);
printk(KERN_ERR "do_wp_page: bogus page at address %08lx\n",
address);
goto oom;
@@ -1022,11 +1062,13 @@ static int do_wp_page(struct mm_struct *
establish_pte(vma, address, page_table,
pte_mkyoung(pte_mkdirty(pte_mkwrite(pte))));
pte_unmap(page_table);
+ pmd_unmap(pmd);
ret = VM_FAULT_MINOR;
goto out;
}
}
pte_unmap(page_table);
+ pmd_unmap(pmd);

/*
* Ok, we need to copy. Oh, well..
@@ -1046,6 +1088,7 @@ static int do_wp_page(struct mm_struct *
* Re-check the pte - we dropped the lock
*/
spin_lock(&mm->page_table_lock);
+ pmd = pmd_offset_map(pgd_offset(mm, address), address);
page_table = pte_offset_map(pmd, address);
if (pte_same(*page_table, pte)) {
if (PageReserved(old_page))
@@ -1059,6 +1102,7 @@ static int do_wp_page(struct mm_struct *
new_page = old_page;
}
pte_unmap(page_table);
+ pmd_unmap(pmd);
page_cache_release(new_page);
page_cache_release(old_page);
ret = VM_FAULT_MINOR;
@@ -1227,6 +1271,7 @@ static int do_swap_page(struct mm_struct
struct pte_chain *pte_chain = NULL;

pte_unmap(page_table);
+ pmd_unmap(pmd);
spin_unlock(&mm->page_table_lock);
page = lookup_swap_cache(entry);
if (!page) {
@@ -1238,12 +1283,14 @@ static int do_swap_page(struct mm_struct
* we released the page table lock.
*/
spin_lock(&mm->page_table_lock);
+ pmd = pmd_offset_map(pgd_offset(mm, address), address);
page_table = pte_offset_map(pmd, address);
if (pte_same(*page_table, orig_pte))
ret = VM_FAULT_OOM;
else
ret = VM_FAULT_MINOR;
pte_unmap(page_table);
+ pmd_unmap(pmd);
spin_unlock(&mm->page_table_lock);
goto out;
}
@@ -1266,9 +1313,11 @@ static int do_swap_page(struct mm_struct
* released the page table lock.
*/
spin_lock(&mm->page_table_lock);
+ pmd = pmd_offset_map(pgd_offset(mm, address), address);
page_table = pte_offset_map(pmd, address);
if (!pte_same(*page_table, orig_pte)) {
pte_unmap(page_table);
+ pmd_unmap(pmd);
spin_unlock(&mm->page_table_lock);
unlock_page(page);
page_cache_release(page);
@@ -1294,6 +1343,7 @@ static int do_swap_page(struct mm_struct

/* No need to invalidate - it was non-present before */
update_mmu_cache(vma, address, pte);
+ pmd_unmap(pmd);
pte_unmap(page_table);
spin_unlock(&mm->page_table_lock);
out:
@@ -1319,11 +1369,13 @@ do_anonymous_page(struct mm_struct *mm,
pte_chain = pte_chain_alloc(GFP_ATOMIC);
if (!pte_chain) {
pte_unmap(page_table);
+ pmd_unmap(pmd);
spin_unlock(&mm->page_table_lock);
pte_chain = pte_chain_alloc(GFP_KERNEL);
if (!pte_chain)
goto no_mem;
spin_lock(&mm->page_table_lock);
+ pmd = pmd_offset_map(pgd_offset(mm, addr), addr);
page_table = pte_offset_map(pmd, addr);
}

@@ -1334,6 +1386,7 @@ do_anonymous_page(struct mm_struct *mm,
if (write_access) {
/* Allocate our own private page. */
pte_unmap(page_table);
+ pmd_unmap(pmd);
spin_unlock(&mm->page_table_lock);

page = alloc_page(GFP_HIGHUSER);
@@ -1342,9 +1395,11 @@ do_anonymous_page(struct mm_struct *mm,
clear_user_highpage(page, addr);

spin_lock(&mm->page_table_lock);
+ pmd = pmd_offset_map(pgd_offset(mm, addr), addr);
page_table = pte_offset_map(pmd, addr);

if (!pte_none(*page_table)) {
+ pmd_unmap(pmd);
pte_unmap(page_table);
page_cache_release(page);
spin_unlock(&mm->page_table_lock);
@@ -1360,6 +1415,7 @@ do_anonymous_page(struct mm_struct *mm,
set_pte(page_table, entry);
/* ignores ZERO_PAGE */
pte_chain = page_add_rmap(page, page_table, pte_chain);
+ pmd_unmap(pmd);
pte_unmap(page_table);

/* No need to invalidate - it was non-present before */
@@ -1402,6 +1458,7 @@ do_no_page(struct mm_struct *mm, struct
return do_anonymous_page(mm, vma, page_table,
pmd, write_access, address);
pte_unmap(page_table);
+ pmd_unmap(pmd);

mapping = vma->vm_file->f_dentry->d_inode->i_mapping;
sequence = atomic_read(&mapping->truncate_count);
@@ -1446,6 +1503,7 @@ retry:
page_cache_release(new_page);
goto retry;
}
+ pmd = pmd_offset_map(pgd_offset(mm, address), address);
page_table = pte_offset_map(pmd, address);

/*
@@ -1468,9 +1526,11 @@ retry:
set_pte(page_table, entry);
pte_chain = page_add_rmap(new_page, page_table, pte_chain);
pte_unmap(page_table);
+ pmd_unmap(pmd);
} else {
/* One of our sibling threads was faster, back out. */
pte_unmap(page_table);
+ pmd_unmap(pmd);
page_cache_release(new_page);
spin_unlock(&mm->page_table_lock);
ret = VM_FAULT_MINOR;
@@ -1514,6 +1574,7 @@ static int do_file_page(struct mm_struct
pgoff = pte_to_pgoff(*pte);

pte_unmap(pte);
+ pmd_unmap(pmd);
spin_unlock(&mm->page_table_lock);

err = vma->vm_ops->populate(vma, address & PAGE_MASK, PAGE_SIZE, vma->vm_page_prot, pgoff, 0);
@@ -1574,6 +1635,7 @@ static inline int handle_pte_fault(struc
entry = pte_mkyoung(entry);
establish_pte(vma, address, pte, entry);
pte_unmap(pte);
+ pmd_unmap(pmd);
spin_unlock(&mm->page_table_lock);
return VM_FAULT_MINOR;
}
@@ -1600,10 +1662,10 @@ int handle_mm_fault(struct mm_struct *mm
* and the SMP-safe atomic PTE updates.
*/
spin_lock(&mm->page_table_lock);
- pmd = pmd_alloc(mm, pgd, address);
+ pmd = pmd_alloc_map(mm, pgd, address);

if (pmd) {
- pte_t * pte = pte_alloc_map(mm, pmd, address);
+ pte_t *pte = pte_alloc_map(mm, &pmd, address);
if (pte)
return handle_pte_fault(mm, vma, address, write_access, pte, pmd);
}
@@ -1640,7 +1702,30 @@ pmd_t *__pmd_alloc(struct mm_struct *mm,
}
pgd_populate(mm, pgd, new);
out:
- return pmd_offset(pgd, address);
+ return pmd_offset_map(pgd, address);
+}
+
+pmd_t *__pmd_alloc_kernel(struct mm_struct *mm, pgd_t *pgd, unsigned long address)
+{
+ pmd_t *new;
+
+ spin_unlock(&mm->page_table_lock);
+ new = pmd_alloc_one_kernel(mm, address);
+ spin_lock(&mm->page_table_lock);
+ if (!new)
+ return NULL;
+
+ /*
+ * Because we dropped the lock, we should re-check the
+ * entry, as somebody else could have populated it..
+ */
+ if (pgd_present(*pgd)) {
+ pmd_free(new);
+ goto out;
+ }
+ pgd_populate(mm, pgd, new);
+out:
+ return pmd_offset_kernel(pgd, address);
}

int make_pages_present(unsigned long addr, unsigned long end)
@@ -1672,7 +1757,7 @@ struct page * vmalloc_to_page(void * vma
pte_t *ptep, pte;

if (!pgd_none(*pgd)) {
- pmd = pmd_offset(pgd, addr);
+ pmd = pmd_offset_map(pgd, addr);
if (!pmd_none(*pmd)) {
preempt_disable();
ptep = pte_offset_map(pmd, addr);
@@ -1682,6 +1767,7 @@ struct page * vmalloc_to_page(void * vma
pte_unmap(ptep);
preempt_enable();
}
+ pmd_unmap(pmd);
}
return page;
}
diff -prauN mm2-2.5.73-1/mm/mprotect.c mm2-2.5.73-2/mm/mprotect.c
--- mm2-2.5.73-1/mm/mprotect.c 2003-06-28 03:09:58.000000000 -0700
+++ mm2-2.5.73-2/mm/mprotect.c 2003-06-28 03:11:37.000000000 -0700
@@ -73,7 +73,7 @@ change_pmd_range(pgd_t *pgd, unsigned lo
pgd_clear(pgd);
return;
}
- pmd = pmd_offset(pgd, address);
+ pmd = pmd_offset_map(pgd, address);
address &= ~PGDIR_MASK;
end = address + size;
if (end > PGDIR_SIZE)
@@ -83,6 +83,7 @@ change_pmd_range(pgd_t *pgd, unsigned lo
address = (address + PMD_SIZE) & PMD_MASK;
pmd++;
} while (address && (address < end));
+ pmd_unmap(pmd - 1);
}

static void
diff -prauN mm2-2.5.73-1/mm/mremap.c mm2-2.5.73-2/mm/mremap.c
--- mm2-2.5.73-1/mm/mremap.c 2003-06-28 03:09:58.000000000 -0700
+++ mm2-2.5.73-2/mm/mremap.c 2003-06-28 03:11:37.000000000 -0700
@@ -38,7 +38,7 @@ static pte_t *get_one_pte_map_nested(str
goto end;
}

- pmd = pmd_offset(pgd, addr);
+ pmd = pmd_offset_map_nested(pgd, addr);
if (pmd_none(*pmd))
goto end;
if (pmd_bad(*pmd)) {
@@ -53,6 +53,7 @@ static pte_t *get_one_pte_map_nested(str
pte = NULL;
}
end:
+ pmd_unmap_nested(pmd);
return pte;
}

@@ -61,12 +62,15 @@ static inline int page_table_present(str
{
pgd_t *pgd;
pmd_t *pmd;
+ int ret;

pgd = pgd_offset(mm, addr);
if (pgd_none(*pgd))
return 0;
- pmd = pmd_offset(pgd, addr);
- return pmd_present(*pmd);
+ pmd = pmd_offset_map(pgd, addr);
+ ret = pmd_present(*pmd);
+ pmd_unmap(pmd);
+ return ret != 0;
}
#else
#define page_table_present(mm, addr) (1)
@@ -77,9 +81,10 @@ static inline pte_t *alloc_one_pte_map(s
pmd_t *pmd;
pte_t *pte = NULL;

- pmd = pmd_alloc(mm, pgd_offset(mm, addr), addr);
+ pmd = pmd_alloc_map(mm, pgd_offset(mm, addr), addr);
if (pmd)
- pte = pte_alloc_map(mm, pmd, addr);
+ pte = pte_alloc_map(mm, &pmd, addr);
+ pmd_unmap(pmd);
return pte;
}

diff -prauN mm2-2.5.73-1/mm/msync.c mm2-2.5.73-2/mm/msync.c
--- mm2-2.5.73-1/mm/msync.c 2003-06-22 11:32:42.000000000 -0700
+++ mm2-2.5.73-2/mm/msync.c 2003-06-28 03:11:37.000000000 -0700
@@ -82,7 +82,7 @@ static inline int filemap_sync_pmd_range
pgd_clear(pgd);
return 0;
}
- pmd = pmd_offset(pgd, address);
+ pmd = pmd_offset_map(pgd, address);
if ((address & PGDIR_MASK) != (end & PGDIR_MASK))
end = (address & PGDIR_MASK) + PGDIR_SIZE;
error = 0;
@@ -91,6 +91,7 @@ static inline int filemap_sync_pmd_range
address = (address + PMD_SIZE) & PMD_MASK;
pmd++;
} while (address && (address < end));
+ pmd_unmap(pmd - 1);
return error;
}

diff -prauN mm2-2.5.73-1/mm/slab.c mm2-2.5.73-2/mm/slab.c
--- mm2-2.5.73-1/mm/slab.c 2003-06-28 03:09:58.000000000 -0700
+++ mm2-2.5.73-2/mm/slab.c 2003-06-28 03:27:00.000000000 -0700
@@ -2717,7 +2717,7 @@ void ptrinfo(unsigned long addr)
printk("No pgd.\n");
break;
}
- pmd = pmd_offset(pgd, addr);
+ pmd = pmd_offset_kernel(pgd, addr);
if (pmd_none(*pmd)) {
printk("No pmd.\n");
break;
diff -prauN mm2-2.5.73-1/mm/swapfile.c mm2-2.5.73-2/mm/swapfile.c
--- mm2-2.5.73-1/mm/swapfile.c 2003-06-28 03:09:58.000000000 -0700
+++ mm2-2.5.73-2/mm/swapfile.c 2003-06-28 03:11:37.000000000 -0700
@@ -448,7 +448,7 @@ static int unuse_pgd(struct vm_area_stru
pgd_clear(dir);
return 0;
}
- pmd = pmd_offset(dir, address);
+ pmd = pmd_offset_map(dir, address);
offset = address & PGDIR_MASK;
address &= ~PGDIR_MASK;
end = address + size;
@@ -463,6 +463,7 @@ static int unuse_pgd(struct vm_area_stru
address = (address + PMD_SIZE) & PMD_MASK;
pmd++;
} while (address && (address < end));
+ pmd_unmap(pmd - 1);
return 0;
}

diff -prauN mm2-2.5.73-1/mm/vmalloc.c mm2-2.5.73-2/mm/vmalloc.c
--- mm2-2.5.73-1/mm/vmalloc.c 2003-06-22 11:32:56.000000000 -0700
+++ mm2-2.5.73-2/mm/vmalloc.c 2003-06-28 03:11:37.000000000 -0700
@@ -70,7 +70,7 @@ static void unmap_area_pmd(pgd_t *dir, u
return;
}

- pmd = pmd_offset(dir, address);
+ pmd = pmd_offset_kernel(dir, address);
address &= ~PGDIR_MASK;
end = address + size;
if (end > PGDIR_SIZE)
@@ -159,7 +159,7 @@ int map_vm_area(struct vm_struct *area,
dir = pgd_offset_k(address);
spin_lock(&init_mm.page_table_lock);
do {
- pmd_t *pmd = pmd_alloc(&init_mm, dir, address);
+ pmd_t *pmd = pmd_alloc_kernel(&init_mm, dir, address);
if (!pmd) {
err = -ENOMEM;
break;

2003-06-28 15:54:25

by Christoph Hellwig

[permalink] [raw]

Subject: Re: 2.5.73-mm2

On Sat, Jun 28, 2003 at 08:54:36AM -0700, William Lee Irwin III wrote:
> +config HIGHPMD
> + bool "Allocate 2nd-level pagetables from highmem"
> + depends on HIGHMEM64G
> + help
> + The VM uses one pmd entry for each pagetable page of physical
> + memory allocated. For systems with extreme amounts of highmem,
> + this cannot be tolerated. Setting this option will put
> + userspace 2nd-level pagetables in highmem.

Does this make sense for !HIGHPTE? In fact does it make sense to
carry along HIGHPTE as an option still? ..

> +#ifndef CONFIG_HIGHPMD /* Oh boy. Error reporting is going to blow major goats. */

Any chance you can rearragne the code to avoid the ifndef in favour
of an ifdef?

> set_pte(dst_pte, entry);
> + pmd_unmap(dst_pte);
> + pmd_unmap_nested(src_pte);

<Lots more pmd_unmap* calls snipped>

Looks like you changed some API so that pmds are now returned mapped?
It might make sense to change their names into foo_map then so the
breakage is at the API level if someone misses updates for the changes.

> +#ifdef CONFIG_HIGHPMD
> +#define GFP_PMD (__GFP_REPEAT|__GFP_HIGHMEM|GFP_KERNEL)
> +#else
> +#define GFP_PMD (__GFP_REPEAT|GFP_KERNEL)
> +#endif

So what? Do you want to use a space or tab after the #define? :)

Also Given that GFP_PMD is used just once it's argueable whether it makes
sense to get rid of the defintion and use the expanded values directly.

Otherwise the patch looks fine to me and should allow to get some more
free lowmem on those insanely big 32bit machines.. :)

2003-06-28 20:36:15

William Lee Irwin III <[email protected]> wrote:
>> @@ -217,9 +217,9 @@ void out_of_memory(void)
>> unsigned long now, since;
>>
>> /*
>> - * Enough swap space left? Not OOM.
>> + * Enough swap space and ZONE_NORMAL left? Not OOM.
>> */
>> - if (nr_swap_pages > 0)
>> + if (nr_swap_pages > 0 && nr_free_buffer_pages() + nr_used_low_pages() > 0)
>> return;

On Mon, Jun 30, 2003 at 07:14:56PM -0700, Andrew Morton wrote:
> a) if someone is trying to allocate some ZONE_DMA pages and there are
> still swappable or free ZONE_NORMAL pages, nobody gets killed.

This is yet another problem for the method above. =(

On Mon, Jun 30, 2003 at 07:14:56PM -0700, Andrew Morton wrote:
> b) If there are free ZONE_NORMAL pages then why on earth did we call
> out_of_memory()? Does nr_free_buffer_pages() ever return non-zero in
> here? It will do so for a ZONE_DMA allocation, but you're not doing
> them...

Allocations will enter this path if free memory is below the minimum
page thresholds, since the allocation will be sort of artificially
failed. Basically, with this in place it's more likely to livelock than
to go on killing sprees. There's a small amount of empirical evidence
suggesting this avoids livelocking in some common scenarios, though
that really isn't good enough for this kind of affair.

On Mon, Jun 30, 2003 at 07:14:56PM -0700, Andrew Morton wrote:
> Generally, I'm thinking that this test should just be removed. It is
> the responsibility of try_to_free_pages() to work out whether the
> allocation can succeed.
> If try_to_free_pages() calls out_of_memory() when there are still
> swappable, reclaimable or free pages in the relevant zones then
> try_to_free_pages() goofed, and needs mending. out_of_memory()
> shouldn't be cleaning up after try_to_free_pages()'s mistakes.
> I have a bad feeling that it _will_ goof. A long time ago I looked
> at the amount of scanning we're doing in there and decided that it
> was way overkill and reduced it by a lot. I may have gone overboard.
> So how's about I and thy take that test out, see how things get along?

I'm not particularly attached to the method, only the result, so I'm game.

-- wli

2003-07-01 05:43:47

On Fri, Jun 27, 2003 at 08:21:30PM -0700, Andrew Morton wrote:
>> ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.5/2.5.73/2.5.73-mm2/
>> Just bits and pieces.

On Sat, Jun 28, 2003 at 08:54:36AM -0700, William Lee Irwin III wrote:
> Here's highpmd. This allocates L2 pagetables from highmem, decreasing
> the per-process lowmem overhead on CONFIG_HIGHMEM64G from 20KB to 8KB.
> Some attempts were made to update non-i386 architectures to the new
> API's, though they're entirely untested. It's been tested for a while
> in -wli on i386 machines, both lowmem and highmem boxen.

Here's highpmd again, but with the bash-shared-mappings oops fixed.
Some missing s/pmd_alloc()/pmd_alloc_map()/ conversions in non-i386
code are also included in the update.

Included as a MIME attachment to prevent MTA's from barfing on its size.

-- wli

Attachments:

(No filename) (848.00 B)
highpmd-2.5.73-mm2-2.bz2 (14.06 kB)
highpmd-2.5.73-mm2-2.bz2 Download all attachments