misc fixes

- Remove the IO error warning in end_buffer_io_sync(). Failed READA attempts trigger it.

- Emit a warning when an ext2 is mounting an ext3 filesystem.

We have had quite a few problem reports related to this, mainly arising from initrd problems. And mount(8) tends to report the fstype from /etc/fstab rather than reporting what has really happened.

Fixes some bogosity which I added to max_block():

- `size' doesn't need to be sector_t

- `retval' should not be initialised to "~0UL" because that is 0x00000000ffffffff with 64-bit sector_t.

- /proc/partitions shows wrong number of blocks (Badari Pulavarty)

- Allocate task_structs with GFP_KERNEL, as discussed.

- Convert the EXPORT_SYMBOL for generic_file_direct_IO() to EXPORT_SYMBOL_GPL. That was only exported as a practicality for the raw driver.

- Make the loop thread run balance_dirty_pages() after dirtying the backing file. So it will perform writeback of the backing file when dirty memory levels are high. Export balance_dirty_pages to GPL modules for this.

This makes loop work a lot better - I suspect it broke when callers of balance_dirty_pages() started writing back only their own queue.

There are many page allocation failures under heavy loop writeout. Coming from blk_queue_bounce()'s allocation from the page_pool mempool. So...

- Disable page allocation warnings around the initial atomic allocation attempt in mempool_alloc() - the one where __GFP_WAIT and __GFP_IO were turned off. That one can easily fail.

- Add some commentary in block_write_full_page()

 drivers/block/genhd.c |    2 +-
 drivers/block/loop.c  |    2 ++
 fs/block_dev.c        |    6 +++---
 fs/buffer.c           |   13 +++++++++++--
 fs/ext2/super.c       |    3 +++
 kernel/fork.c         |    5 +++--
 kernel/ksyms.c        |    2 +-
 mm/mempool.c          |    2 ++
 mm/page-writeback.c   |    1 +
 9 files changed, 27 insertions, 9 deletions

@@ -180,7 +180,10 @@ void end_buffer_io_sync(struct buffer_he
 	if (uptodate) {
 		set_buffer_uptodate(bh);
 	} else {
-		buffer_io_error(bh);
+		/*
+		 * This happens, due to failed READA attempts.
+		 * buffer_io_error(bh);
+		 */
 		clear_buffer_uptodate(bh);
 	}
 	unlock_buffer(bh);

@@ -2283,7 +2286,13 @@ int block_write_full_page(struct page *p
 		return -EIO;
 	}
 
-	/* The page straddles i_size */
+	/*
+	 * The page straddles i_size.  It must be zeroed out on each and every
+	 * writepage invokation because it may be mmapped.  "A file is mapped
+	 * in multiples of the page size.  For a file that is not a multiple of
+	 * the  page size, the remaining memory is zeroed when mapped, and
+	 * writes to that region are not written out to the file."
+	 */
 	kaddr = kmap(page);
 	memset(kaddr + offset, 0, PAGE_CACHE_SIZE - offset);
 	flush_dcache_page(page);

@@ -26,12 +26,12 @@
 
 static sector_t max_block(struct block_device *bdev)
 {
-	sector_t retval = ~0U;
+	sector_t retval = ~0;
 	loff_t sz = bdev->bd_inode->i_size;
 
 	if (sz) {
-		sector_t size = block_size(bdev);
-		unsigned sizebits = blksize_bits(size);
+		unsigned int size = block_size(bdev);
+		unsigned int sizebits = blksize_bits(size);
 		retval = (sz >> sizebits);
 	}
 	return retval;

@@ -698,6 +698,9 @@ static int ext2_fill_super(struct super_
 			printk(KERN_ERR "EXT2-fs: get root inode failed\n");
 		goto failed_mount2;
 	}
+	if (EXT2_HAS_COMPAT_FEATURE(sb, EXT3_FEATURE_COMPAT_HAS_JOURNAL))
+		ext2_warning(sb, __FUNCTION__,
+			"mounting ext3 filesystem as ext2\n");
 	ext2_setup_super (sb, es, sb->s_flags & MS_RDONLY);
 	return 0;
 failed_mount2:

@@ -166,7 +166,7 @@ static int show_partition(struct seq_fil
 			continue;
 		seq_printf(part, "%4d  %4d %10ld %s\n",
 			sgp->major, n + sgp->first_minor,
-			sgp->part[n].nr_sects << 1,
+			sgp->part[n].nr_sects >> 1,
 			disk_name(sgp, n + sgp->first_minor, buf));
 	}
 

@@ -106,9 +106,10 @@ static struct task_struct *dup_task_stru
 	struct thread_info *ti;
 
 	ti = alloc_thread_info();
-	if (!ti) return NULL;
+	if (!ti)
+		return NULL;
 
-	tsk = kmem_cache_alloc(task_struct_cachep,GFP_ATOMIC);
+	tsk = kmem_cache_alloc(task_struct_cachep, GFP_KERNEL);
 	if (!tsk) {
 		free_thread_info(ti);
 		return NULL;

@@ -342,7 +342,7 @@ EXPORT_SYMBOL(register_disk);
 EXPORT_SYMBOL(read_dev_sector);
 EXPORT_SYMBOL(init_buffer);
 EXPORT_SYMBOL(wipe_partitions);
-EXPORT_SYMBOL(generic_file_direct_IO);
+EXPORT_SYMBOL_GPL(generic_file_direct_IO);
 
 /* tty routines */
 EXPORT_SYMBOL(tty_hangup);

@@ -74,6 +74,7 @@
 #include <linux/slab.h>
 #include <linux/loop.h>
 #include <linux/suspend.h>
+#include <linux/writeback.h>
 #include <linux/buffer_head.h>		/* for invalidate_bdev() */
 
 #include <asm/uaccess.h>

@@ -235,6 +236,7 @@ do_lo_send(struct loop_device *lo, struc
 	up(&mapping->host->i_sem);
 out:
 	kunmap(bvec->bv_page);
+	balance_dirty_pages(mapping);
 	return ret;
 
 unlock:

@@ -133,6 +133,7 @@ void balance_dirty_pages(struct address_
 	if (!writeback_in_progress(bdi) && ps.nr_dirty > background_thresh)
 		pdflush_operation(background_writeout, 0);
 }
+EXPORT_SYMBOL_GPL(balance_dirty_pages);
 
 /**
  * balance_dirty_pages_ratelimited - balance dirty memory state

@@ -189,7 +189,9 @@ void * mempool_alloc(mempool_t *pool, in
 	int gfp_nowait = gfp_mask & ~(__GFP_WAIT | __GFP_IO);
 
 repeat_alloc:
+	current->flags |= PF_NOWARN;
 	element = pool->alloc(gfp_nowait, pool->pool_data);
+	current->flags &= ~PF_NOWARN;
 	if (likely(element != NULL))
 		return element;
 

.

From: Andrew Morton <[email protected]>
Date: Fri, 09 Aug 2002 17:55:24 -0700

- `retval' should not be initialised to "~0UL" because that is

     0x00000000ffffffff with 64-bit sector_t.

'0' defaults to 'int' on 32-bit systems so ~0 isn't right either.

What I think you really want is "~(sector_t) 0".

Actually, ~0 is fine. This is because of the associativity of events happening. Breaking it down into individual ops as the compiler sees it:

1) constant 0 -> int type
2) ~0 -> (int)-1
3) assign to larger type sector_t -> sign extend smaller type to size of larger type -> (sector_t)-1

Therefore not specifying anything after the initial constant definition _always_ works because of sign extension and the order of evaluation which is guaranteed by the C language.

OTOH if you hard code 0U or 0L or 0UL or whatever it breaks because it forces the compiler not to sign extend and assume this to be a hardvalue, which is zero extended.

Note I just tried all possible combinations with a little test program and both assignments and binary operators work as expected (with sign extension) when the constant is given without qualifiers and both break horribly (no sign extension) when the constant is given with qualifiers.

I did this testing for another reason, and that is PAGE_MASK and PAGE_CACHE_MASK which break when used for 64-bit values on 32-bit architectures exactly for the above outlined reasons.

If no one beats me to it I will submit a patch to change PAGE_SIZE (and thus automatically PAGE_CACHE_SIZE) on all architectures to no longer use the "UL" type qualifier which is responsible for the breakage. This supersedes my previous ugly patch introducing a PAGE_{CACHE_,}MASK_LL...

Best regards,

Anton

  "I've not lost my mind. It's backed up on tape somewhere." - Unknown
-- 
Anton Altaparmakov <aia21 at cantab.net> (replace at with @)
Linux NTFS Maintainer / IRC: #ntfs on irc.openprojects.net
WWW: http://linux-ntfs.sf.net/ & http://www-stu.christs.cam.ac.uk/~aia21/

Just for the record - If you actually try to compile the kernel with a C++ compiler it will warn you nicely about the particular issue here.