linux.git/mm/filemap.c, branch v2.6.15

[PATCH] x86_64: Remove obsolete ARCH_HAS_ATOMIC_UNSIGNED and page_flags_t

2005-11-15T03:55:14+00:00

Has been introduced for x86-64 at some point to save memory
in struct page, but has been obsolete for some time. Just
remove it.

Signed-off-by: Andi Kleen 
Signed-off-by: Linus Torvalds

[PATCH] fs: error case fix in __generic_file_aio_read

2005-10-31T01:37:22+00:00

When __generic_file_aio_read() hits an error during reading, it reports the
error iff nothing has successfully been read yet.  This is condition - when
an error occurs, if nothing has been read/written, report the error code;
otherwise, report the amount of bytes successfully transferred upto that
point.

This corner case can be exposed by performing readv(2) with the following
iov.

 iov[0] = len0 @ ptr0
 iov[1] = len1 @ NULL (or any other invalid pointer)
 iov[2] = len2 @ ptr2

When file size is enough, performing above readv(2) results in

 len0 bytes from file_pos @ ptr0
 len2 bytes from file_pos + len0 @ ptr2

And the return value is len0 + len2.  Test program is attached to this
mail.

This patch makes __generic_file_aio_read()'s error handling identical to
other functions.

#include 
#include 
#include 
#include 
#include 
#include 
#include 
#include 
#include 

int main(int argc, char **argv)
{
	const char *path;
	struct stat stbuf;
	size_t len0, len1;
	void *buf0, *buf1;
	struct iovec iov[3];
	int fd, i;
	ssize_t ret;

	if (argc < 2) {
		fprintf(stderr, "Usage: testreadv path (better be a "
			"small text file)\n");
		return 1;
	}
	path = argv[1];

	if (stat(path, &stbuf) < 0) {
		perror("stat");
		return 1;
	}

	len0 = stbuf.st_size / 2;
	len1 = stbuf.st_size - len0;

	if (!len0 || !len1) {
		fprintf(stderr, "Dude, file is too small\n");
		return 1;
	}

	if ((fd = open(path, O_RDONLY)) < 0) {
		perror("open");
		return 1;
	}

	if (!(buf0 = malloc(len0)) || !(buf1 = malloc(len1))) {
		perror("malloc");
		return 1;
	}

	memset(buf0, 0, len0);
	memset(buf1, 0, len1);

	iov[0].iov_base = buf0;
	iov[0].iov_len = len0;
	iov[1].iov_base = NULL;
	iov[1].iov_len = len1;
	iov[2].iov_base = buf1;
	iov[2].iov_len = len1;

	printf("vector ");
	for (i = 0; i < 3; i++)
		printf("%p:%zu ", iov[i].iov_base, iov[i].iov_len);
	printf("\n");

	ret = readv(fd, iov, 3);
	if (ret < 0)
		perror("readv");

	printf("readv returned %zd\nbuf0 = [%s]\nbuf1 = [%s]\n",
	       ret, (char *)buf0, (char *)buf1);

	return 0;
}

Signed-off-by: Tejun Heo 
Cc: Benjamin LaHaise 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] mm/filemap.c:filemap_populate(): move export.

2005-10-30T04:40:45+00:00

move EXPORT_SYMBOL(filemap_populate) to the proper place: just after
function itself: it's easy to miss that function is exported otherwise.

Signed-off-by: Nikita Danilov 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] mm: update comments to pte lock

2005-10-30T04:40:42+00:00

Updated several references to page_table_lock in common code comments.

Signed-off-by: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] mm: split page table lock

2005-10-30T04:40:42+00:00

Christoph Lameter demonstrated very poor scalability on the SGI 512-way, with
a many-threaded application which concurrently initializes different parts of
a large anonymous area.

This patch corrects that, by using a separate spinlock per page table page, to
guard the page table entries in that page, instead of using the mm's single
page_table_lock.  (But even then, page_table_lock is still used to guard page
table allocation, and anon_vma allocation.)

In this implementation, the spinlock is tucked inside the struct page of the
page table page: with a BUILD_BUG_ON in case it overflows - which it would in
the case of 32-bit PA-RISC with spinlock debugging enabled.

Splitting the lock is not quite for free: another cacheline access.  Ideally,
I suppose we would use split ptlock only for multi-threaded processes on
multi-cpu machines; but deciding that dynamically would have its own costs.
So for now enable it by config, at some number of cpus - since the Kconfig
language doesn't support inequalities, let preprocessor compare that with
NR_CPUS.  But I don't think it's worth being user-configurable: for good
testing of both split and unsplit configs, split now at 4 cpus, and perhaps
change that to 8 later.

There is a benefit even for singly threaded processes: kswapd can be attacking
one part of the mm while another part is busy faulting.

Signed-off-by: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] mm: page fault handlers tidyup

2005-10-30T04:40:37+00:00

Impose a little more consistency on the page fault handlers do_wp_page,
do_swap_page, do_anonymous_page, do_no_page, do_file_page: why not pass their
arguments in the same order, called the same names?

break_cow is all very well, but what it did was inlined elsewhere: easier to
compare if it's brought back into do_wp_page.

do_file_page's fallback to do_no_page dates from a time when we were testing
pte_file by using it wherever possible: currently it's peculiar to nonlinear
vmas, so just check that.  BUG_ON if not?  Better not, it's probably page
table corruption, so just show the pte: hmm, there's a pte_ERROR macro, let's
use that for do_wp_page's invalid pfn too.

Hah!  Someone in the ppc64 world noticed pte_ERROR was unused so removed it:
restored (and say "pud" not "pmd" in its pud_ERROR).

Signed-off-by: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] gfp_t: mm/* (easy parts)

2005-10-28T15:16:47+00:00

Signed-off-by: Al Viro 
Signed-off-by: Linus Torvalds

[PATCH] mm/filemap.c: make two functions static

2005-09-10T17:06:25+00:00

With Nick Piggin 

Give some things static scope.

Signed-off-by: Adrian Bunk 
Signed-off-by: Nick Piggin 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] shmem_populate: avoid an useless check, and some comments

2005-09-05T07:05:45+00:00

Either shmem_getpage returns a failure, or it found a page, or it was told
it couldn't do any I/O.  So it's useless to check nonblock in the else
branch.  We could add a BUG() there but I preferred to comment the
offending function.

This was taken out from one Ingo Molnar's old patch I'm resurrecting.

Signed-off-by: Paolo 'Blaisorblade' Giarrusso 
Cc: Ingo Molnar 
Cc: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] swap: swap_lock replace list+device

2005-09-05T07:05:42+00:00

The idea of a swap_device_lock per device, and a swap_list_lock over them all,
is appealing; but in practice almost every holder of swap_device_lock must
already hold swap_list_lock, which defeats the purpose of the split.

The only exceptions have been swap_duplicate, valid_swaphandles and an
untrodden path in try_to_unuse (plus a few places added in this series).
valid_swaphandles doesn't show up high in profiles, but swap_duplicate does
demand attention.  However, with the hold time in get_swap_pages so much
reduced, I've not yet found a load and set of swap device priorities to show
even swap_duplicate benefitting from the split.  Certainly the split is mere
overhead in the common case of a single swap device.

So, replace swap_list_lock and swap_device_lock by spinlock_t swap_lock
(generally we seem to prefer an _ in the name, and not hide in a macro).

If someone can show a regression in swap_duplicate, then probably we should
add a hashlock for the swap_map entries alone (shorts being anatomic), so as
to help the case of the single swap device too.

Signed-off-by: Hugh Dickins 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds