Merge master.kernel.org:/pub/scm/linux/kernel/git/torvalds/linux-2.6

Conflicts:

	drivers/usb/input/Makefile
	drivers/usb/input/gtco.c

12 files changed, 330 insertions, 367 deletions

diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 5554adaa58f9..e61e0efe9ec7 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -863,9 +863,7 @@ static int bitmap_init_from_disk(struct bitmap *bitmap, sector_t start)
 
 	/* We need 4 bits per page, rounded up to a multiple of sizeof(unsigned long) */
 	bitmap->filemap_attr = kzalloc(
-		(((num_pages*4/8)+sizeof(unsigned long)-1)
-		 /sizeof(unsigned long))
-		*sizeof(unsigned long),
+		roundup( DIV_ROUND_UP(num_pages*4, 8), sizeof(unsigned long)),
 		GFP_KERNEL);
 	if (!bitmap->filemap_attr)
 		goto out;
diff --git a/drivers/md/dm-crypt.c b/drivers/md/dm-crypt.c
index 4c2471ee054a..d8121234c347 100644
--- a/drivers/md/dm-crypt.c
+++ b/drivers/md/dm-crypt.c
@@ -867,7 +867,7 @@ static int crypt_ctr(struct dm_target *ti, unsigned int argc, char **argv)
 		goto bad4;
 	}
 
-	cc->bs = bioset_create(MIN_IOS, MIN_IOS, 4);
+	cc->bs = bioset_create(MIN_IOS, MIN_IOS);
 	if (!cc->bs) {
 		ti->error = "Cannot allocate crypt bioset";
 		goto bad_bs;
diff --git a/drivers/md/dm-io.c b/drivers/md/dm-io.c
index 4eb73d395213..8bdc8a87b249 100644
--- a/drivers/md/dm-io.c
+++ b/drivers/md/dm-io.c
@@ -60,7 +60,7 @@ static int resize_pool(unsigned int new_ios)
 		if (!_io_pool)
 			return -ENOMEM;
 
-		_bios = bioset_create(16, 16, 4);
+		_bios = bioset_create(16, 16);
 		if (!_bios) {
 			mempool_destroy(_io_pool);
 			_io_pool = NULL;
diff --git a/drivers/md/dm.c b/drivers/md/dm.c
index 3668b170ea68..11a98df298ec 100644
--- a/drivers/md/dm.c
+++ b/drivers/md/dm.c
@@ -1012,7 +1012,7 @@ static struct mapped_device *alloc_dev(int minor)
 	if (!md->tio_pool)
 		goto bad3;
 
-	md->bs = bioset_create(16, 16, 4);
+	md->bs = bioset_create(16, 16);
 	if (!md->bs)
 		goto bad_no_bioset;
 
diff --git a/drivers/md/linear.c b/drivers/md/linear.c
index c625ddb8833d..d5ecd2d53046 100644
--- a/drivers/md/linear.c
+++ b/drivers/md/linear.c
@@ -188,7 +188,7 @@ static linear_conf_t *linear_conf(mddev_t *mddev, int raid_disks)
 	for (i=0; i < cnt-1 ; i++) {
 		sector_t sz = 0;
 		int j;
-		for (j=i; i<cnt-1 && sz < min_spacing ; j++)
+		for (j = i; j < cnt - 1 && sz < min_spacing; j++)
 			sz += conf->disks[j].size;
 		if (sz >= min_spacing && sz < conf->hash_spacing)
 			conf->hash_spacing = sz;
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 05febfd9f071..509171ca7fa8 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -1296,27 +1296,17 @@ static struct super_type super_types[] = {
 		.sync_super	= super_1_sync,
 	},
 };
-	
-static mdk_rdev_t * match_dev_unit(mddev_t *mddev, mdk_rdev_t *dev)
-{
-	struct list_head *tmp;
-	mdk_rdev_t *rdev;
-
-	ITERATE_RDEV(mddev,rdev,tmp)
-		if (rdev->bdev->bd_contains == dev->bdev->bd_contains)
-			return rdev;
-
-	return NULL;
-}
 
 static int match_mddev_units(mddev_t *mddev1, mddev_t *mddev2)
 {
-	struct list_head *tmp;
-	mdk_rdev_t *rdev;
+	struct list_head *tmp, *tmp2;
+	mdk_rdev_t *rdev, *rdev2;
 
 	ITERATE_RDEV(mddev1,rdev,tmp)
-		if (match_dev_unit(mddev2, rdev))
-			return 1;
+		ITERATE_RDEV(mddev2, rdev2, tmp2)
+			if (rdev->bdev->bd_contains ==
+			    rdev2->bdev->bd_contains)
+				return 1;
 
 	return 0;
 }
@@ -1325,10 +1315,10 @@ static LIST_HEAD(pending_raid_disks);
 
 static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 {
-	mdk_rdev_t *same_pdev;
-	char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];
+	char b[BDEVNAME_SIZE];
 	struct kobject *ko;
 	char *s;
+	int err;
 
 	if (rdev->mddev) {
 		MD_BUG();
@@ -1342,14 +1332,6 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 		else
 			mddev->size = rdev->size;
 	}
-	same_pdev = match_dev_unit(mddev, rdev);
-	if (same_pdev)
-		printk(KERN_WARNING
-			"%s: WARNING: %s appears to be on the same physical"
-	 		" disk as %s. True\n     protection against single-disk"
-			" failure might be compromised.\n",
-			mdname(mddev), bdevname(rdev->bdev,b),
-			bdevname(same_pdev->bdev,b2));
 
 	/* Verify rdev->desc_nr is unique.
 	 * If it is -1, assign a free number, else
@@ -1371,20 +1353,35 @@ static int bind_rdev_to_array(mdk_rdev_t * rdev, mddev_t * mddev)
 	while ( (s=strchr(rdev->kobj.k_name, '/')) != NULL)
 		*s = '!';
 			
-	list_add(&rdev->same_set, &mddev->disks);
 	rdev->mddev = mddev;
 	printk(KERN_INFO "md: bind<%s>\n", b);
 
 	rdev->kobj.parent = &mddev->kobj;
-	kobject_add(&rdev->kobj);
+	if ((err = kobject_add(&rdev->kobj)))
+		goto fail;
 
 	if (rdev->bdev->bd_part)
 		ko = &rdev->bdev->bd_part->kobj;
 	else
 		ko = &rdev->bdev->bd_disk->kobj;
-	sysfs_create_link(&rdev->kobj, ko, "block");
+	if ((err = sysfs_create_link(&rdev->kobj, ko, "block"))) {
+		kobject_del(&rdev->kobj);
+		goto fail;
+	}
+	list_add(&rdev->same_set, &mddev->disks);
 	bd_claim_by_disk(rdev->bdev, rdev, mddev->gendisk);
 	return 0;
+
+ fail:
+	printk(KERN_WARNING "md: failed to register dev-%s for %s\n",
+	       b, mdname(mddev));
+	return err;
+}
+
+static void delayed_delete(struct work_struct *ws)
+{
+	mdk_rdev_t *rdev = container_of(ws, mdk_rdev_t, del_work);
+	kobject_del(&rdev->kobj);
 }
 
 static void unbind_rdev_from_array(mdk_rdev_t * rdev)
@@ -1399,7 +1396,12 @@ static void unbind_rdev_from_array(mdk_rdev_t * rdev)
 	printk(KERN_INFO "md: unbind<%s>\n", bdevname(rdev->bdev,b));
 	rdev->mddev = NULL;
 	sysfs_remove_link(&rdev->kobj, "block");
-	kobject_del(&rdev->kobj);
+
+	/* We need to delay this, otherwise we can deadlock when
+	 * writing to 'remove' to "dev/state"
+	 */
+	INIT_WORK(&rdev->del_work, delayed_delete);
+	schedule_work(&rdev->del_work);
 }
 
 /*
@@ -2985,7 +2987,9 @@ static struct kobject *md_probe(dev_t dev, int *part, void *data)
 	mddev->kobj.k_name = NULL;
 	snprintf(mddev->kobj.name, KOBJ_NAME_LEN, "%s", "md");
 	mddev->kobj.ktype = &md_ktype;
-	kobject_register(&mddev->kobj);
+	if (kobject_register(&mddev->kobj))
+		printk(KERN_WARNING "md: cannot register %s/md - name in use\n",
+		       disk->disk_name);
 	return NULL;
 }
 
@@ -3109,6 +3113,36 @@ static int do_md_run(mddev_t * mddev)
 		return -EINVAL;
 	}
 
+	if (pers->sync_request) {
+		/* Warn if this is a potentially silly
+		 * configuration.
+		 */
+		char b[BDEVNAME_SIZE], b2[BDEVNAME_SIZE];
+		mdk_rdev_t *rdev2;
+		struct list_head *tmp2;
+		int warned = 0;
+		ITERATE_RDEV(mddev, rdev, tmp) {
+			ITERATE_RDEV(mddev, rdev2, tmp2) {
+				if (rdev < rdev2 &&
+				    rdev->bdev->bd_contains ==
+				    rdev2->bdev->bd_contains) {
+					printk(KERN_WARNING
+					       "%s: WARNING: %s appears to be"
+					       " on the same physical disk as"
+					       " %s.\n",
+					       mdname(mddev),
+					       bdevname(rdev->bdev,b),
+					       bdevname(rdev2->bdev,b2));
+					warned = 1;
+				}
+			}
+		}
+		if (warned)
+			printk(KERN_WARNING
+			       "True protection against single-disk"
+			       " failure might be compromised.\n");
+	}
+
 	mddev->recovery = 0;
 	mddev->resync_max_sectors = mddev->size << 1; /* may be over-ridden by personality */
 	mddev->barriers_work = 1;
@@ -3133,9 +3167,12 @@ static int do_md_run(mddev_t * mddev)
 		bitmap_destroy(mddev);
 		return err;
 	}
-	if (mddev->pers->sync_request)
-		sysfs_create_group(&mddev->kobj, &md_redundancy_group);
-	else if (mddev->ro == 2) /* auto-readonly not meaningful */
+	if (mddev->pers->sync_request) {
+		if (sysfs_create_group(&mddev->kobj, &md_redundancy_group))
+			printk(KERN_WARNING
+			       "md: cannot register extra attributes for %s\n",
+			       mdname(mddev));
+	} else if (mddev->ro == 2) /* auto-readonly not meaningful */
 		mddev->ro = 0;
 
  	atomic_set(&mddev->writes_pending,0);
@@ -3149,7 +3186,9 @@ static int do_md_run(mddev_t * mddev)
 		if (rdev->raid_disk >= 0) {
 			char nm[20];
 			sprintf(nm, "rd%d", rdev->raid_disk);
-			sysfs_create_link(&mddev->kobj, &rdev->kobj, nm);
+			if (sysfs_create_link(&mddev->kobj, &rdev->kobj, nm))
+				printk("md: cannot register %s for %s\n",
+				       nm, mdname(mddev));
 		}
 	
 	set_bit(MD_RECOVERY_NEEDED, &mddev->recovery);
@@ -3311,6 +3350,10 @@ static int do_md_stop(mddev_t * mddev, int mode)
 				set_disk_ro(disk, 0);
 			blk_queue_make_request(mddev->queue, md_fail_request);
 			mddev->pers->stop(mddev);
+			mddev->queue->merge_bvec_fn = NULL;
+			mddev->queue->unplug_fn = NULL;
+			mddev->queue->issue_flush_fn = NULL;
+			mddev->queue->backing_dev_info.congested_fn = NULL;
 			if (mddev->pers->sync_request)
 				sysfs_remove_group(&mddev->kobj, &md_redundancy_group);
 
@@ -3357,6 +3400,9 @@ static int do_md_stop(mddev_t * mddev, int mode)
 				sysfs_remove_link(&mddev->kobj, nm);
 			}
 
+		/* make sure all delayed_delete calls have finished */
+		flush_scheduled_work();
+
 		export_array(mddev);
 
 		mddev->array_size = 0;
@@ -5343,6 +5389,48 @@ void md_do_sync(mddev_t *mddev)
 EXPORT_SYMBOL_GPL(md_do_sync);
 
 
+static int remove_and_add_spares(mddev_t *mddev)
+{
+	mdk_rdev_t *rdev;
+	struct list_head *rtmp;
+	int spares = 0;
+
+	ITERATE_RDEV(mddev,rdev,rtmp)
+		if (rdev->raid_disk >= 0 &&
+		    (test_bit(Faulty, &rdev->flags) ||
+		     ! test_bit(In_sync, &rdev->flags)) &&
+		    atomic_read(&rdev->nr_pending)==0) {
+			if (mddev->pers->hot_remove_disk(
+				    mddev, rdev->raid_disk)==0) {
+				char nm[20];
+				sprintf(nm,"rd%d", rdev->raid_disk);
+				sysfs_remove_link(&mddev->kobj, nm);
+				rdev->raid_disk = -1;
+			}
+		}
+
+	if (mddev->degraded) {
+		ITERATE_RDEV(mddev,rdev,rtmp)
+			if (rdev->raid_disk < 0
+			    && !test_bit(Faulty, &rdev->flags)) {
+				rdev->recovery_offset = 0;
+				if (mddev->pers->hot_add_disk(mddev,rdev)) {
+					char nm[20];
+					sprintf(nm, "rd%d", rdev->raid_disk);
+					if (sysfs_create_link(&mddev->kobj,
+							      &rdev->kobj, nm))
+						printk(KERN_WARNING
+						       "md: cannot register "
+						       "%s for %s\n",
+						       nm, mdname(mddev));
+					spares++;
+					md_new_event(mddev);
+				} else
+					break;
+			}
+	}
+	return spares;
+}
 /*
  * This routine is regularly called by all per-raid-array threads to
  * deal with generic issues like resync and super-block update.
@@ -5397,7 +5485,7 @@ void md_check_recovery(mddev_t *mddev)
 		return;
 
 	if (mddev_trylock(mddev)) {
-		int spares =0;
+		int spares = 0;
 
 		spin_lock_irq(&mddev->write_lock);
 		if (mddev->safemode && !atomic_read(&mddev->writes_pending) &&
@@ -5460,35 +5548,13 @@ void md_check_recovery(mddev_t *mddev)
 		 * Spare are also removed and re-added, to allow
 		 * the personality to fail the re-add.
 		 */
-		ITERATE_RDEV(mddev,rdev,rtmp)
-			if (rdev->raid_disk >= 0 &&
-			    (test_bit(Faulty, &rdev->flags) || ! test_bit(In_sync, &rdev->flags)) &&
-			    atomic_read(&rdev->nr_pending)==0) {
-				if (mddev->pers->hot_remove_disk(mddev, rdev->raid_disk)==0) {
-					char nm[20];
-					sprintf(nm,"rd%d", rdev->raid_disk);
-					sysfs_remove_link(&mddev->kobj, nm);
-					rdev->raid_disk = -1;
-				}
-			}
-
-		if (mddev->degraded) {
-			ITERATE_RDEV(mddev,rdev,rtmp)
-				if (rdev->raid_disk < 0
-				    && !test_bit(Faulty, &rdev->flags)) {
-					rdev->recovery_offset = 0;
-					if (mddev->pers->hot_add_disk(mddev,rdev)) {
-						char nm[20];
-						sprintf(nm, "rd%d", rdev->raid_disk);
-						sysfs_create_link(&mddev->kobj, &rdev->kobj, nm);
-						spares++;
-						md_new_event(mddev);
-					} else
-						break;
-				}
-		}
 
-		if (spares) {
+		if (mddev->reshape_position != MaxSector) {
+			if (mddev->pers->check_reshape(mddev) != 0)
+				/* Cannot proceed */
+				goto unlock;
+			set_bit(MD_RECOVERY_RESHAPE, &mddev->recovery);
+		} else if ((spares = remove_and_add_spares(mddev))) {
 			clear_bit(MD_RECOVERY_SYNC, &mddev->recovery);
 			clear_bit(MD_RECOVERY_CHECK, &mddev->recovery);
 		} else if (mddev->recovery_cp < MaxSector) {
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index a9401c017e35..82249a69014f 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -429,7 +429,7 @@ static sector_t raid10_find_virt(conf_t *conf, sector_t sector, int dev)
 		if (dev < 0)
 			dev += conf->raid_disks;
 	} else {
-		while (sector > conf->stride) {
+		while (sector >= conf->stride) {
 			sector -= conf->stride;
 			if (dev < conf->near_copies)
 				dev += conf->raid_disks - conf->near_copies;
@@ -1801,6 +1801,7 @@ static sector_t sync_request(mddev_t *mddev, sector_t sector_nr, int *skipped, i
 						for (k=0; k<conf->copies; k++)
 							if (r10_bio->devs[k].devnum == i)
 								break;
+						BUG_ON(k == conf->copies);
 						bio = r10_bio->devs[1].bio;
 						bio->bi_next = biolist;
 						biolist = bio;
@@ -2021,19 +2022,30 @@ static int run(mddev_t *mddev)
 	if (!conf->tmppage)
 		goto out_free_conf;
 
+	conf->mddev = mddev;
+	conf->raid_disks = mddev->raid_disks;
 	conf->near_copies = nc;
 	conf->far_copies = fc;
 	conf->copies = nc*fc;
 	conf->far_offset = fo;
 	conf->chunk_mask = (sector_t)(mddev->chunk_size>>9)-1;
 	conf->chunk_shift = ffz(~mddev->chunk_size) - 9;
+	size = mddev->size >> (conf->chunk_shift-1);
+	sector_div(size, fc);
+	size = size * conf->raid_disks;
+	sector_div(size, nc);
+	/* 'size' is now the number of chunks in the array */
+	/* calculate "used chunks per device" in 'stride' */
+	stride = size * conf->copies;
+	sector_div(stride, conf->raid_disks);
+	mddev->size = stride  << (conf->chunk_shift-1);
+
 	if (fo)
-		conf->stride = 1 << conf->chunk_shift;
-	else {
-		stride = mddev->size >> (conf->chunk_shift-1);
+		stride = 1;
+	else
 		sector_div(stride, fc);
-		conf->stride = stride << conf->chunk_shift;
-	}
+	conf->stride = stride << conf->chunk_shift;
+
 	conf->r10bio_pool = mempool_create(NR_RAID10_BIOS, r10bio_pool_alloc,
 						r10bio_pool_free, conf);
 	if (!conf->r10bio_pool) {
@@ -2063,8 +2075,6 @@ static int run(mddev_t *mddev)
 
 		disk->head_position = 0;
 	}
-	conf->raid_disks = mddev->raid_disks;
-	conf->mddev = mddev;
 	spin_lock_init(&conf->device_lock);
 	INIT_LIST_HEAD(&conf->retry_list);
 
@@ -2106,16 +2116,8 @@ static int run(mddev_t *mddev)
 	/*
 	 * Ok, everything is just fine now
 	 */
-	if (conf->far_offset) {
-		size = mddev->size >> (conf->chunk_shift-1);
-		size *= conf->raid_disks;
-		size <<= conf->chunk_shift;
-		sector_div(size, conf->far_copies);
-	} else
-		size = conf->stride * conf->raid_disks;
-	sector_div(size, conf->near_copies);
-	mddev->array_size = size/2;
-	mddev->resync_max_sectors = size;
+	mddev->array_size = size << (conf->chunk_shift-1);
+	mddev->resync_max_sectors = size << conf->chunk_shift;
 
 	mddev->queue->unplug_fn = raid10_unplug;
 	mddev->queue->issue_flush_fn = raid10_issue_flush;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 11c3d7bfa797..8d59914f2057 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -1050,7 +1050,7 @@ static void compute_parity5(struct stripe_head *sh, int method)
 static void compute_parity6(struct stripe_head *sh, int method)
 {
 	raid6_conf_t *conf = sh->raid_conf;
-	int i, pd_idx = sh->pd_idx, qd_idx, d0_idx, disks = conf->raid_disks, count;
+	int i, pd_idx = sh->pd_idx, qd_idx, d0_idx, disks = sh->disks, count;
 	struct bio *chosen;
 	/**** FIX THIS: This could be very bad if disks is close to 256 ****/
 	void *ptrs[disks];
@@ -1131,8 +1131,7 @@ static void compute_parity6(struct stripe_head *sh, int method)
 /* Compute one missing block */
 static void compute_block_1(struct stripe_head *sh, int dd_idx, int nozero)
 {
-	raid6_conf_t *conf = sh->raid_conf;
-	int i, count, disks = conf->raid_disks;
+	int i, count, disks = sh->disks;
 	void *ptr[MAX_XOR_BLOCKS], *p;
 	int pd_idx = sh->pd_idx;
 	int qd_idx = raid6_next_disk(pd_idx, disks);
@@ -1170,8 +1169,7 @@ static void compute_block_1(struct stripe_head *sh, int dd_idx, int nozero)
 /* Compute two missing blocks */
 static void compute_block_2(struct stripe_head *sh, int dd_idx1, int dd_idx2)
 {
-	raid6_conf_t *conf = sh->raid_conf;
-	int i, count, disks = conf->raid_disks;
+	int i, count, disks = sh->disks;
 	int pd_idx = sh->pd_idx;
 	int qd_idx = raid6_next_disk(pd_idx, disks);
 	int d0_idx = raid6_next_disk(qd_idx, disks);
@@ -1887,11 +1885,11 @@ static void handle_stripe5(struct stripe_head *sh)
 static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
 {
 	raid6_conf_t *conf = sh->raid_conf;
-	int disks = conf->raid_disks;
+	int disks = sh->disks;
 	struct bio *return_bi= NULL;
 	struct bio *bi;
 	int i;
-	int syncing;
+	int syncing, expanding, expanded;
 	int locked=0, uptodate=0, to_read=0, to_write=0, failed=0, written=0;
 	int non_overwrite = 0;
 	int failed_num[2] = {0, 0};
@@ -1909,6 +1907,8 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
 	clear_bit(STRIPE_DELAYED, &sh->state);
 
 	syncing = test_bit(STRIPE_SYNCING, &sh->state);
+	expanding = test_bit(STRIPE_EXPAND_SOURCE, &sh->state);
+	expanded = test_bit(STRIPE_EXPAND_READY, &sh->state);
 	/* Now to look around and see what can be done */
 
 	rcu_read_lock();
@@ -2114,13 +2114,15 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
 	 * parity, or to satisfy requests
 	 * or to load a block that is being partially written.
 	 */
-	if (to_read || non_overwrite || (to_write && failed) || (syncing && (uptodate < disks))) {
+	if (to_read || non_overwrite || (to_write && failed) ||
+	    (syncing && (uptodate < disks)) || expanding) {
 		for (i=disks; i--;) {
 			dev = &sh->dev[i];
 			if (!test_bit(R5_LOCKED, &dev->flags) && !test_bit(R5_UPTODATE, &dev->flags) &&
 			    (dev->toread ||
 			     (dev->towrite && !test_bit(R5_OVERWRITE, &dev->flags)) ||
 			     syncing ||
+			     expanding ||
 			     (failed >= 1 && (sh->dev[failed_num[0]].toread || to_write)) ||
 			     (failed >= 2 && (sh->dev[failed_num[1]].toread || to_write))
 				    )
@@ -2355,6 +2357,79 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
 				}
 			}
 		}
+
+	if (expanded && test_bit(STRIPE_EXPANDING, &sh->state)) {
+		/* Need to write out all blocks after computing P&Q */
+		sh->disks = conf->raid_disks;
+		sh->pd_idx = stripe_to_pdidx(sh->sector, conf,
+					     conf->raid_disks);
+		compute_parity6(sh, RECONSTRUCT_WRITE);
+		for (i = conf->raid_disks ; i-- ;  ) {
+			set_bit(R5_LOCKED, &sh->dev[i].flags);
+			locked++;
+			set_bit(R5_Wantwrite, &sh->dev[i].flags);
+		}
+		clear_bit(STRIPE_EXPANDING, &sh->state);
+	} else if (expanded) {
+		clear_bit(STRIPE_EXPAND_READY, &sh->state);
+		atomic_dec(&conf->reshape_stripes);
+		wake_up(&conf->wait_for_overlap);
+		md_done_sync(conf->mddev, STRIPE_SECTORS, 1);
+	}
+
+	if (expanding && locked == 0) {
+		/* We have read all the blocks in this stripe and now we need to
+		 * copy some of them into a target stripe for expand.
+		 */
+		clear_bit(STRIPE_EXPAND_SOURCE, &sh->state);
+		for (i = 0; i < sh->disks ; i++)
+			if (i != pd_idx && i != qd_idx) {
+				int dd_idx2, pd_idx2, j;
+				struct stripe_head *sh2;
+
+				sector_t bn = compute_blocknr(sh, i);
+				sector_t s = raid5_compute_sector(
+					bn, conf->raid_disks,
+					conf->raid_disks - conf->max_degraded,
+					&dd_idx2, &pd_idx2, conf);
+				sh2 = get_active_stripe(conf, s,
+							conf->raid_disks,
+						       pd_idx2, 1);
+				if (sh2 == NULL)
+					/* so for only the early blocks of
+					 * this stripe have been requests.
+					 * When later blocks get requests, we
+					 * will try again
+					 */
+					continue;
+				if (!test_bit(STRIPE_EXPANDING, &sh2->state) ||
+				    test_bit(R5_Expanded,
+					     &sh2->dev[dd_idx2].flags)) {
+					/* must have already done this block */
+					release_stripe(sh2);
+					continue;
+				}
+				memcpy(page_address(sh2->dev[dd_idx2].page),
+				       page_address(sh->dev[i].page),
+				       STRIPE_SIZE);
+				set_bit(R5_Expanded, &sh2->dev[dd_idx2].flags);
+				set_bit(R5_UPTODATE, &sh2->dev[dd_idx2].flags);
+				for (j = 0 ; j < conf->raid_disks ; j++)
+					if (j != sh2->pd_idx &&
+					    j != raid6_next_disk(sh2->pd_idx,
+							   sh2->disks) &&
+					    !test_bit(R5_Expanded,
+						      &sh2->dev[j].flags))
+						break;
+				if (j == conf->raid_disks) {
+					set_bit(STRIPE_EXPAND_READY,
+						&sh2->state);
+					set_bit(STRIPE_HANDLE, &sh2->state);
+				}
+				release_stripe(sh2);
+			}
+	}
+
 	spin_unlock(&sh->lock);
 
 	while ((bi=return_bi)) {
@@ -2395,7 +2470,7 @@ static void handle_stripe6(struct stripe_head *sh, struct page *tmp_page)
 		rcu_read_unlock();
 
 		if (rdev) {
-			if (syncing)
+			if (syncing || expanding || expanded)
 				md_sync_acct(rdev->bdev, STRIPE_SECTORS);
 
 			bi->bi_bdev = rdev->bdev;
@@ -2915,8 +2990,9 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
 	struct stripe_head *sh;
 	int pd_idx;
 	sector_t first_sector, last_sector;
-	int raid_disks;
-	int data_disks;
+	int raid_disks = conf->previous_raid_disks;
+	int data_disks = raid_disks - conf->max_degraded;
+	int new_data_disks = conf->raid_disks - conf->max_degraded;
 	int i;
 	int dd_idx;
 	sector_t writepos, safepos, gap;
@@ -2925,7 +3001,7 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
 	    conf->expand_progress != 0) {
 		/* restarting in the middle, skip the initial sectors */
 		sector_nr = conf->expand_progress;
-		sector_div(sector_nr, conf->raid_disks-1);
+		sector_div(sector_nr, new_data_disks);
 		*skipped = 1;
 		return sector_nr;
 	}
@@ -2939,14 +3015,14 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
 	 * to after where expand_lo old_maps to
 	 */
 	writepos = conf->expand_progress +
-		conf->chunk_size/512*(conf->raid_disks-1);
-	sector_div(writepos, conf->raid_disks-1);
+		conf->chunk_size/512*(new_data_disks);
+	sector_div(writepos, new_data_disks);
 	safepos = conf->expand_lo;
-	sector_div(safepos, conf->previous_raid_disks-1);
+	sector_div(safepos, data_disks);
 	gap = conf->expand_progress - conf->expand_lo;
 
 	if (writepos >= safepos ||
-	    gap > (conf->raid_disks-1)*3000*2 /*3Meg*/) {
+	    gap > (new_data_disks)*3000*2 /*3Meg*/) {
 		/* Cannot proceed until we've updated the superblock... */
 		wait_event(conf->wait_for_overlap,
 			   atomic_read(&conf->reshape_stripes)==0);
@@ -2976,6 +3052,9 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
 			sector_t s;
 			if (j == sh->pd_idx)
 				continue;
+			if (conf->level == 6 &&
+			    j == raid6_next_disk(sh->pd_idx, sh->disks))
+				continue;
 			s = compute_blocknr(sh, j);
 			if (s < (mddev->array_size<<1)) {
 				skipped = 1;
@@ -2992,28 +3071,27 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
 		release_stripe(sh);
 	}
 	spin_lock_irq(&conf->device_lock);
-	conf->expand_progress = (sector_nr + i)*(conf->raid_disks-1);
+	conf->expand_progress = (sector_nr + i) * new_data_disks;
 	spin_unlock_irq(&conf->device_lock);
 	/* Ok, those stripe are ready. We can start scheduling
 	 * reads on the source stripes.
 	 * The source stripes are determined by mapping the first and last
 	 * block on the destination stripes.
 	 */
-	raid_disks = conf->previous_raid_disks;
-	data_disks = raid_disks - 1;
 	first_sector =
-		raid5_compute_sector(sector_nr*(conf->raid_disks-1),
+		raid5_compute_sector(sector_nr*(new_data_disks),
 				     raid_disks, data_disks,
 				     &dd_idx, &pd_idx, conf);
 	last_sector =
 		raid5_compute_sector((sector_nr+conf->chunk_size/512)
-				     *(conf->raid_disks-1) -1,
+				     *(new_data_disks) -1,
 				     raid_disks, data_disks,
 				     &dd_idx, &pd_idx, conf);
 	if (last_sector >= (mddev->size<<1))
 		last_sector = (mddev->size<<1)-1;
 	while (first_sector <= last_sector) {
-		pd_idx = stripe_to_pdidx(first_sector, conf, conf->previous_raid_disks);
+		pd_idx = stripe_to_pdidx(first_sector, conf,
+					 conf->previous_raid_disks);
 		sh = get_active_stripe(conf, first_sector,
 				       conf->previous_raid_disks, pd_idx, 0);
 		set_bit(STRIPE_EXPAND_SOURCE, &sh->state);
@@ -3348,35 +3426,44 @@ static int run(mddev_t *mddev)
 		 */
 		sector_t here_new, here_old;
 		int old_disks;
+		int max_degraded = (mddev->level == 5 ? 1 : 2);
 
 		if (mddev->new_level != mddev->level ||
 		    mddev->new_layout != mddev->layout ||
 		    mddev->new_chunk != mddev->chunk_size) {
-			printk(KERN_ERR "raid5: %s: unsupported reshape required - aborting.\n",
+			printk(KERN_ERR "raid5: %s: unsupported reshape "
+			       "required - aborting.\n",
 			       mdname(mddev));
 			return -EINVAL;
 		}
 		if (mddev->delta_disks <= 0) {
-			printk(KERN_ERR "raid5: %s: unsupported reshape (reduce disks) required - aborting.\n",
+			printk(KERN_ERR "raid5: %s: unsupported reshape "
+			       "(reduce disks) required - aborting.\n",
 			       mdname(mddev));
 			return -EINVAL;
 		}
 		old_disks = mddev->raid_disks - mddev->delta_disks;
 		/* reshape_position must be on a new-stripe boundary, and one
-		 * further up in new geometry must map after here in old geometry.
+		 * further up in new geometry must map after here in old
+		 * geometry.
 		 */
 		here_new = mddev->reshape_position;
-		if (sector_div(here_new, (mddev->chunk_size>>9)*(mddev->raid_disks-1))) {
-			printk(KERN_ERR "raid5: reshape_position not on a stripe boundary\n");
+		if (sector_div(here_new, (mddev->chunk_size>>9)*
+			       (mddev->raid_disks - max_degraded))) {
+			printk(KERN_ERR "raid5: reshape_position not "
+			       "on a stripe boundary\n");
 			return -EINVAL;
 		}
 		/* here_new is the stripe we will write to */
 		here_old = mddev->reshape_position;
-		sector_div(here_old, (mddev->chunk_size>>9)*(old_disks-1));
-		/* here_old is the first stripe that we might need to read from */
+		sector_div(here_old, (mddev->chunk_size>>9)*
+			   (old_disks-max_degraded));
+		/* here_old is the first stripe that we might need to read
+		 * from */
 		if (here_new >= here_old) {
 			/* Reading from the same stripe as writing to - bad */
-			printk(KERN_ERR "raid5: reshape_position too early for auto-recovery - aborting.\n");
+			printk(KERN_ERR "raid5: reshape_position too early for "
+			       "auto-recovery - aborting.\n");
 			return -EINVAL;
 		}
 		printk(KERN_INFO "raid5: reshape will continue\n");
@@ -3555,12 +3642,15 @@ static int run(mddev_t *mddev)
 	}
 
 	/* Ok, everything is just fine now */
-	sysfs_create_group(&mddev->kobj, &raid5_attrs_group);
+	if (sysfs_create_group(&mddev->kobj, &raid5_attrs_group))
+		printk(KERN_WARNING
+		       "raid5: failed to create sysfs attributes for %s\n",
+		       mdname(mddev));
 
 	mddev->queue->unplug_fn = raid5_unplug_device;
 	mddev->queue->issue_flush_fn = raid5_issue_flush;
-	mddev->queue->backing_dev_info.congested_fn = raid5_congested;
 	mddev->queue->backing_dev_info.congested_data = mddev;
+	mddev->queue->backing_dev_info.congested_fn = raid5_congested;
 
 	mddev->array_size =  mddev->size * (conf->previous_raid_disks -
 					    conf->max_degraded);
@@ -3591,6 +3681,7 @@ static int stop(mddev_t *mddev)
 	mddev->thread = NULL;
 	shrink_stripes(conf);
 	kfree(conf->stripe_hashtbl);
+	mddev->queue->backing_dev_info.congested_fn = NULL;
 	blk_sync_queue(mddev->queue); /* the unplug fn references 'conf'*/
 	sysfs_remove_group(&mddev->kobj, &raid5_attrs_group);
 	kfree(conf->disks);
@@ -3814,6 +3905,8 @@ static int raid5_check_reshape(mddev_t *mddev)
 	if (err)
 		return err;
 
+	if (mddev->degraded > conf->max_degraded)
+		return -EINVAL;
 	/* looks like we might be able to manage this */
 	return 0;
 }
@@ -3827,8 +3920,7 @@ static int raid5_start_reshape(mddev_t *mddev)
 	int added_devices = 0;
 	unsigned long flags;
 
-	if (mddev->degraded ||
-	    test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
+	if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery))
 		return -EBUSY;
 
 	ITERATE_RDEV(mddev, rdev, rtmp)
@@ -3836,7 +3928,7 @@ static int raid5_start_reshape(mddev_t *mddev)
 		    !test_bit(Faulty, &rdev->flags))
 			spares++;
 
-	if (spares < mddev->delta_disks-1)
+	if (spares - mddev->degraded < mddev->delta_disks - conf->max_degraded)
 		/* Not enough devices even to make a degraded array
 		 * of that size
 		 */
@@ -3862,7 +3954,12 @@ static int raid5_start_reshape(mddev_t *mddev)
 				added_devices++;
 				rdev->recovery_offset = 0;
 				sprintf(nm, "rd%d", rdev->raid_disk);
-				sysfs_create_link(&mddev->kobj, &rdev->kobj, nm);
+				if (sysfs_create_link(&mddev->kobj,
+						      &rdev->kobj, nm))
+					printk(KERN_WARNING
+					       "raid5: failed to create "
+					       " link %s for %s\n",
+					       nm, mdname(mddev));
 			} else
 				break;
 		}
@@ -3899,7 +3996,8 @@ static void end_reshape(raid5_conf_t *conf)
 	struct block_device *bdev;
 
 	if (!test_bit(MD_RECOVERY_INTR, &conf->mddev->recovery)) {
-		conf->mddev->array_size = conf->mddev->size * (conf->raid_disks-1);
+		conf->mddev->array_size = conf->mddev->size *
+			(conf->raid_disks - conf->max_degraded);
 		set_capacity(conf->mddev->gendisk, conf->mddev->array_size << 1);
 		conf->mddev->changed = 1;
 
@@ -3972,6 +4070,10 @@ static struct mdk_personality raid6_personality =
 	.spare_active	= raid5_spare_active,
 	.sync_request	= sync_request,
 	.resize		= raid5_resize,
+#ifdef CONFIG_MD_RAID5_RESHAPE
+	.check_reshape	= raid5_check_reshape,
+	.start_reshape  = raid5_start_reshape,
+#endif
 	.quiesce	= raid5_quiesce,
 };
 static struct mdk_personality raid5_personality =
@@ -4011,6 +4113,10 @@ static struct mdk_personality raid4_personality =
 	.spare_active	= raid5_spare_active,
 	.sync_request	= sync_request,
 	.resize		= raid5_resize,
+#ifdef CONFIG_MD_RAID5_RESHAPE
+	.check_reshape	= raid5_check_reshape,
+	.start_reshape  = raid5_start_reshape,
+#endif
 	.quiesce	= raid5_quiesce,
 };
 
diff --git a/drivers/md/raid6mmx.c b/drivers/md/raid6mmx.c
index 359157aaf9e0..6181a5a3365a 100644
--- a/drivers/md/raid6mmx.c
+++ b/drivers/md/raid6mmx.c
@@ -30,14 +30,8 @@ const struct raid6_mmx_constants {
 
 static int raid6_have_mmx(void)
 {
-#ifdef __KERNEL__
 	/* Not really "boot_cpu" but "all_cpus" */
 	return boot_cpu_has(X86_FEATURE_MMX);
-#else
-	/* User space test code */
-	u32 features = cpuid_features();
-	return ( (features & (1<<23)) == (1<<23) );
-#endif
 }
 
 /*
@@ -48,13 +42,12 @@ static void raid6_mmx1_gen_syndrome(int disks, size_t bytes, void **ptrs)
 	u8 **dptr = (u8 **)ptrs;
 	u8 *p, *q;
 	int d, z, z0;
-	raid6_mmx_save_t sa;
 
 	z0 = disks - 3;		/* Highest data disk */
 	p = dptr[z0+1];		/* XOR parity */
 	q = dptr[z0+2];		/* RS syndrome */
 
-	raid6_before_mmx(&sa);
+	kernel_fpu_begin();
 
 	asm volatile("movq %0,%%mm0" : : "m" (raid6_mmx_constants.x1d));
 	asm volatile("pxor %mm5,%mm5");	/* Zero temp */
@@ -78,7 +71,7 @@ static void raid6_mmx1_gen_syndrome(int disks, size_t bytes, void **ptrs)
 		asm volatile("pxor %mm4,%mm4");
 	}
 
-	raid6_after_mmx(&sa);
+	kernel_fpu_end();
 }
 
 const struct raid6_calls raid6_mmxx1 = {
@@ -96,13 +89,12 @@ static void raid6_mmx2_gen_syndrome(int disks, size_t bytes, void **ptrs)
 	u8 **dptr = (u8 **)ptrs;
 	u8 *p, *q;
 	int d, z, z0;
-	raid6_mmx_save_t sa;
 
 	z0 = disks - 3;		/* Highest data disk */
 	p = dptr[z0+1];		/* XOR parity */
 	q = dptr[z0+2];		/* RS syndrome */
 
-	raid6_before_mmx(&sa);
+	kernel_fpu_begin();
 
 	asm volatile("movq %0,%%mm0" : : "m" (raid6_mmx_constants.x1d));
 	asm volatile("pxor %mm5,%mm5");	/* Zero temp */
@@ -137,7 +129,7 @@ static void raid6_mmx2_gen_syndrome(int disks, size_t bytes, void **ptrs)
 		asm volatile("movq %%mm6,%0" : "=m" (q[d+8]));
 	}
 
-	raid6_after_mmx(&sa);
+	kernel_fpu_end();
 }
 
 const struct raid6_calls raid6_mmxx2 = {
diff --git a/drivers/md/raid6sse1.c b/drivers/md/raid6sse1.c
index f7e7859f71aa..f0a1ba8f40ba 100644
--- a/drivers/md/raid6sse1.c
+++ b/drivers/md/raid6sse1.c
@@ -33,16 +33,10 @@ extern const struct raid6_mmx_constants {
 
 static int raid6_have_sse1_or_mmxext(void)
 {
-#ifdef __KERNEL__
 	/* Not really boot_cpu but "all_cpus" */
 	return boot_cpu_has(X86_FEATURE_MMX) &&
 		(boot_cpu_has(X86_FEATURE_XMM) ||
 		 boot_cpu_has(X86_FEATURE_MMXEXT));
-#else
-	/* User space test code - this incorrectly breaks on some Athlons */
-	u32 features = cpuid_features();
-	return ( (features & (5<<23)) == (5<<23) );
-#endif
 }
 
 /*
@@ -53,14 +47,12 @@ static void raid6_sse11_gen_syndrome(int disks, size_t bytes, void **ptrs)
 	u8 **dptr = (u8 **)ptrs;
 	u8 *p, *q;
 	int d, z, z0;
-	raid6_mmx_save_t sa;
 
 	z0 = disks - 3;		/* Highest data disk */
 	p = dptr[z0+1];		/* XOR parity */
 	q = dptr[z0+2];		/* RS syndrome */
 
-	/* This is really MMX code, not SSE */
-	raid6_before_mmx(&sa);
+	kernel_fpu_begin();
 
 	asm volatile("movq %0,%%mm0" : : "m" (raid6_mmx_constants.x1d));
 	asm volatile("pxor %mm5,%mm5");	/* Zero temp */
@@ -94,8 +86,8 @@ static void raid6_sse11_gen_syndrome(int disks, size_t bytes, void **ptrs)
 		asm volatile("movntq %%mm4,%0" : "=m" (q[d]));
 	}
 
-	raid6_after_mmx(&sa);
 	asm volatile("sfence" : : : "memory");
+	kernel_fpu_end();
 }
 
 const struct raid6_calls raid6_sse1x1 = {
@@ -113,13 +105,12 @@ static void raid6_sse12_gen_syndrome(int disks, size_t bytes, void **ptrs)
 	u8 **dptr = (u8 **)ptrs;
 	u8 *p, *q;
 	int d, z, z0;
-	raid6_mmx_save_t sa;
 
 	z0 = disks - 3;		/* Highest data disk */
 	p = dptr[z0+1];		/* XOR parity */
 	q = dptr[z0+2];		/* RS syndrome */
 
-	raid6_before_mmx(&sa);
+	kernel_fpu_begin();
 
 	asm volatile("movq %0,%%mm0" : : "m" (raid6_mmx_constants.x1d));
 	asm volatile("pxor %mm5,%mm5");	/* Zero temp */
@@ -157,8 +148,8 @@ static void raid6_sse12_gen_syndrome(int disks, size_t bytes, void **ptrs)
 		asm volatile("movntq %%mm6,%0" : "=m" (q[d+8]));
 	}
 
-	raid6_after_mmx(&sa);
 	asm volatile("sfence" : :: "memory");
+	kernel_fpu_end();
 }
 
 const struct raid6_calls raid6_sse1x2 = {
diff --git a/drivers/md/raid6sse2.c b/drivers/md/raid6sse2.c
index b3aa7fe0877e..0f019762a7c3 100644
--- a/drivers/md/raid6sse2.c
+++ b/drivers/md/raid6sse2.c
@@ -30,17 +30,11 @@ static const struct raid6_sse_constants {
 
 static int raid6_have_sse2(void)
 {
-#ifdef __KERNEL__
 	/* Not really boot_cpu but "all_cpus" */
 	return boot_cpu_has(X86_FEATURE_MMX) &&
 		boot_cpu_has(X86_FEATURE_FXSR) &&
 		boot_cpu_has(X86_FEATURE_XMM) &&
 		boot_cpu_has(X86_FEATURE_XMM2);
-#else
-	/* User space test code */
-	u32 features = cpuid_features();
-	return ( (features & (15<<23)) == (15<<23) );
-#endif
 }
 
 /*
@@ -51,13 +45,12 @@ static void raid6_sse21_gen_syndrome(int disks, size_t bytes, void **ptrs)
 	u8 **dptr = (u8 **)ptrs;
 	u8 *p, *q;
 	int d, z, z0;
-	raid6_sse_save_t sa;
 
 	z0 = disks - 3;		/* Highest data disk */
 	p = dptr[z0+1];		/* XOR parity */
 	q = dptr[z0+2];		/* RS syndrome */
 
-	raid6_before_sse2(&sa);
+	kernel_fpu_begin();
 
 	asm volatile("movdqa %0,%%xmm0" : : "m" (raid6_sse_constants.x1d[0]));
 	asm volatile("pxor %xmm5,%xmm5");	/* Zero temp */
@@ -93,8 +86,8 @@ static void raid6_sse21_gen_syndrome(int disks, size_t bytes, void **ptrs)
 		asm volatile("pxor %xmm4,%xmm4");
 	}
 
-	raid6_after_sse2(&sa);
 	asm volatile("sfence" : : : "memory");
+	kernel_fpu_end();
 }
 
 const struct raid6_calls raid6_sse2x1 = {
@@ -112,13 +105,12 @@ static void raid6_sse22_gen_syndrome(int disks, size_t bytes, void **ptrs)
 	u8 **dptr = (u8 **)ptrs;
 	u8 *p, *q;
 	int d, z, z0;
-	raid6_sse_save_t sa;
 
 	z0 = disks - 3;		/* Highest data disk */
 	p = dptr[z0+1];		/* XOR parity */
 	q = dptr[z0+2];		/* RS syndrome */
 
-	raid6_before_sse2(&sa);
+	kernel_fpu_begin();
 
 	asm volatile("movdqa %0,%%xmm0" : : "m" (raid6_sse_constants.x1d[0]));
 	asm volatile("pxor %xmm5,%xmm5"); /* Zero temp */
@@ -156,8 +148,8 @@ static void raid6_sse22_gen_syndrome(int disks, size_t bytes, void **ptrs)
 		asm volatile("movntdq %%xmm6,%0" : "=m" (q[d+16]));
 	}
 
-	raid6_after_sse2(&sa);
 	asm volatile("sfence" : : : "memory");
+	kernel_fpu_end();
 }
 
 const struct raid6_calls raid6_sse2x2 = {
@@ -179,13 +171,12 @@ static void raid6_sse24_gen_syndrome(int disks, size_t bytes, void **ptrs)
 	u8 **dptr = (u8 **)ptrs;
 	u8 *p, *q;
 	int d, z, z0;
-	raid6_sse16_save_t sa;
 
 	z0 = disks - 3;		/* Highest data disk */
 	p = dptr[z0+1];		/* XOR parity */
 	q = dptr[z0+2];		/* RS syndrome */
 
-	raid6_before_sse16(&sa);
+	kernel_fpu_begin();
 
 	asm volatile("movdqa %0,%%xmm0" :: "m" (raid6_sse_constants.x1d[0]));
 	asm volatile("pxor %xmm2,%xmm2");	/* P[0] */
@@ -256,8 +247,9 @@ static void raid6_sse24_gen_syndrome(int disks, size_t bytes, void **ptrs)
 		asm volatile("movntdq %%xmm14,%0" : "=m" (q[d+48]));
 		asm volatile("pxor %xmm14,%xmm14");
 	}
+
 	asm volatile("sfence" : : : "memory");
-	raid6_after_sse16(&sa);
+	kernel_fpu_end();
 }
 
 const struct raid6_calls raid6_sse2x4 = {
diff --git a/drivers/md/raid6x86.h b/drivers/md/raid6x86.h
index 4cf20534fe44..9111950414ff 100644
--- a/drivers/md/raid6x86.h
+++ b/drivers/md/raid6x86.h
@@ -21,224 +21,40 @@
 
 #if defined(__i386__) || defined(__x86_64__)
 
-#ifdef __x86_64__
-
-typedef struct {
-	unsigned int fsave[27];
-	unsigned long cr0;
-} raid6_mmx_save_t __attribute__((aligned(16)));
-
-/* N.B.: For SSE we only save %xmm0-%xmm7 even for x86-64, since
-   the code doesn't know about the additional x86-64 registers */
-typedef struct {
-	unsigned int sarea[8*4+2];
-	unsigned long cr0;
-} raid6_sse_save_t __attribute__((aligned(16)));
-
-/* This is for x86-64-specific code which uses all 16 XMM registers */
-typedef struct {
-	unsigned int sarea[16*4+2];
-	unsigned long cr0;
-} raid6_sse16_save_t __attribute__((aligned(16)));
-
-/* On x86-64 the stack *SHOULD* be 16-byte aligned, but currently this
-   is buggy in the kernel and it's only 8-byte aligned in places, so
-   we need to do this anyway.  Sigh. */
-#define SAREA(x) ((unsigned int *)((((unsigned long)&(x)->sarea)+15) & ~15))
-
-#else /* __i386__ */
-
-typedef struct {
-	unsigned int fsave[27];
-	unsigned long cr0;
-} raid6_mmx_save_t;
-
-/* On i386, the stack is only 8-byte aligned, but SSE requires 16-byte
-   alignment.  The +3 is so we have the slack space to manually align
-   a properly-sized area correctly.  */
-typedef struct {
-	unsigned int sarea[8*4+3];
-	unsigned long cr0;
-} raid6_sse_save_t;
-
-/* Find the 16-byte aligned save area */
-#define SAREA(x) ((unsigned int *)((((unsigned long)&(x)->sarea)+15) & ~15))
-
-#endif
-
 #ifdef __KERNEL__ /* Real code */
 
-/* Note: %cr0 is 32 bits on i386 and 64 bits on x86-64 */
-
-static inline unsigned long raid6_get_fpu(void)
-{
-	unsigned long cr0;
-
-	preempt_disable();
-	asm volatile("mov %%cr0,%0 ; clts" : "=r" (cr0));
-	return cr0;
-}
-
-static inline void raid6_put_fpu(unsigned long cr0)
-{
-	asm volatile("mov %0,%%cr0" : : "r" (cr0));
-	preempt_enable();
-}
+#include <asm/i387.h>
 
 #else /* Dummy code for user space testing */
 
-static inline unsigned long raid6_get_fpu(void)
-{
-	return 0xf00ba6;
-}
-
-static inline void raid6_put_fpu(unsigned long cr0)
-{
-	(void)cr0;
-}
-
-#endif
-
-static inline void raid6_before_mmx(raid6_mmx_save_t *s)
-{
-	s->cr0 = raid6_get_fpu();
-	asm volatile("fsave %0 ; fwait" : "=m" (s->fsave[0]));
-}
-
-static inline void raid6_after_mmx(raid6_mmx_save_t *s)
-{
-	asm volatile("frstor %0" : : "m" (s->fsave[0]));
-	raid6_put_fpu(s->cr0);
-}
-
-static inline void raid6_before_sse(raid6_sse_save_t *s)
-{
-	unsigned int *rsa = SAREA(s);
-
-	s->cr0 = raid6_get_fpu();
-
-	asm volatile("movaps %%xmm0,%0" : "=m" (rsa[0]));
-	asm volatile("movaps %%xmm1,%0" : "=m" (rsa[4]));
-	asm volatile("movaps %%xmm2,%0" : "=m" (rsa[8]));
-	asm volatile("movaps %%xmm3,%0" : "=m" (rsa[12]));
-	asm volatile("movaps %%xmm4,%0" : "=m" (rsa[16]));
-	asm volatile("movaps %%xmm5,%0" : "=m" (rsa[20]));
-	asm volatile("movaps %%xmm6,%0" : "=m" (rsa[24]));
-	asm volatile("movaps %%xmm7,%0" : "=m" (rsa[28]));
-}
-
-static inline void raid6_after_sse(raid6_sse_save_t *s)
-{
-	unsigned int *rsa = SAREA(s);
-
-	asm volatile("movaps %0,%%xmm0" : : "m" (rsa[0]));
-	asm volatile("movaps %0,%%xmm1" : : "m" (rsa[4]));
-	asm volatile("movaps %0,%%xmm2" : : "m" (rsa[8]));
-	asm volatile("movaps %0,%%xmm3" : : "m" (rsa[12]));
-	asm volatile("movaps %0,%%xmm4" : : "m" (rsa[16]));
-	asm volatile("movaps %0,%%xmm5" : : "m" (rsa[20]));
-	asm volatile("movaps %0,%%xmm6" : : "m" (rsa[24]));
-	asm volatile("movaps %0,%%xmm7" : : "m" (rsa[28]));
-
-	raid6_put_fpu(s->cr0);
-}
-
-static inline void raid6_before_sse2(raid6_sse_save_t *s)
+static inline void kernel_fpu_begin(void)
 {
-	unsigned int *rsa = SAREA(s);
-
-	s->cr0 = raid6_get_fpu();
-
-	asm volatile("movdqa %%xmm0,%0" : "=m" (rsa[0]));
-	asm volatile("movdqa %%xmm1,%0" : "=m" (rsa[4]));
-	asm volatile("movdqa %%xmm2,%0" : "=m" (rsa[8]));
-	asm volatile("movdqa %%xmm3,%0" : "=m" (rsa[12]));
-	asm volatile("movdqa %%xmm4,%0" : "=m" (rsa[16]));
-	asm volatile("movdqa %%xmm5,%0" : "=m" (rsa[20]));
-	asm volatile("movdqa %%xmm6,%0" : "=m" (rsa[24]));
-	asm volatile("movdqa %%xmm7,%0" : "=m" (rsa[28]));
 }
 
-static inline void raid6_after_sse2(raid6_sse_save_t *s)
+static inline void kernel_fpu_end(void)
 {
-	unsigned int *rsa = SAREA(s);
-
-	asm volatile("movdqa %0,%%xmm0" : : "m" (rsa[0]));
-	asm volatile("movdqa %0,%%xmm1" : : "m" (rsa[4]));
-	asm volatile("movdqa %0,%%xmm2" : : "m" (rsa[8]));
-	asm volatile("movdqa %0,%%xmm3" : : "m" (rsa[12]));
-	asm volatile("movdqa %0,%%xmm4" : : "m" (rsa[16]));
-	asm volatile("movdqa %0,%%xmm5" : : "m" (rsa[20]));
-	asm volatile("movdqa %0,%%xmm6" : : "m" (rsa[24]));
-	asm volatile("movdqa %0,%%xmm7" : : "m" (rsa[28]));
-
-	raid6_put_fpu(s->cr0);
 }
 
-#ifdef __x86_64__
-
-static inline void raid6_before_sse16(raid6_sse16_save_t *s)
-{
-	unsigned int *rsa = SAREA(s);
-
-	s->cr0 = raid6_get_fpu();
+#define X86_FEATURE_MMX		(0*32+23) /* Multimedia Extensions */
+#define X86_FEATURE_FXSR	(0*32+24) /* FXSAVE and FXRSTOR instructions
+					   * (fast save and restore) */
+#define X86_FEATURE_XMM		(0*32+25) /* Streaming SIMD Extensions */
+#define X86_FEATURE_XMM2	(0*32+26) /* Streaming SIMD Extensions-2 */
+#define X86_FEATURE_MMXEXT	(1*32+22) /* AMD MMX extensions */
 
-	asm volatile("movdqa %%xmm0,%0" : "=m" (rsa[0]));
-	asm volatile("movdqa %%xmm1,%0" : "=m" (rsa[4]));
-	asm volatile("movdqa %%xmm2,%0" : "=m" (rsa[8]));
-	asm volatile("movdqa %%xmm3,%0" : "=m" (rsa[12]));
-	asm volatile("movdqa %%xmm4,%0" : "=m" (rsa[16]));
-	asm volatile("movdqa %%xmm5,%0" : "=m" (rsa[20]));
-	asm volatile("movdqa %%xmm6,%0" : "=m" (rsa[24]));
-	asm volatile("movdqa %%xmm7,%0" : "=m" (rsa[28]));
-	asm volatile("movdqa %%xmm8,%0" : "=m" (rsa[32]));
-	asm volatile("movdqa %%xmm9,%0" : "=m" (rsa[36]));
-	asm volatile("movdqa %%xmm10,%0" : "=m" (rsa[40]));
-	asm volatile("movdqa %%xmm11,%0" : "=m" (rsa[44]));
-	asm volatile("movdqa %%xmm12,%0" : "=m" (rsa[48]));
-	asm volatile("movdqa %%xmm13,%0" : "=m" (rsa[52]));
-	asm volatile("movdqa %%xmm14,%0" : "=m" (rsa[56]));
-	asm volatile("movdqa %%xmm15,%0" : "=m" (rsa[60]));
-}
-
-static inline void raid6_after_sse16(raid6_sse16_save_t *s)
+/* Should work well enough on modern CPUs for testing */
+static inline int boot_cpu_has(int flag)
 {
-	unsigned int *rsa = SAREA(s);
+	u32 eax = (flag >> 5) ? 0x80000001 : 1;
+	u32 edx;
 
-	asm volatile("movdqa %0,%%xmm0" : : "m" (rsa[0]));
-	asm volatile("movdqa %0,%%xmm1" : : "m" (rsa[4]));
-	asm volatile("movdqa %0,%%xmm2" : : "m" (rsa[8]));
-	asm volatile("movdqa %0,%%xmm3" : : "m" (rsa[12]));
-	asm volatile("movdqa %0,%%xmm4" : : "m" (rsa[16]));
-	asm volatile("movdqa %0,%%xmm5" : : "m" (rsa[20]));
-	asm volatile("movdqa %0,%%xmm6" : : "m" (rsa[24]));
-	asm volatile("movdqa %0,%%xmm7" : : "m" (rsa[28]));
-	asm volatile("movdqa %0,%%xmm8" : : "m" (rsa[32]));
-	asm volatile("movdqa %0,%%xmm9" : : "m" (rsa[36]));
-	asm volatile("movdqa %0,%%xmm10" : : "m" (rsa[40]));
-	asm volatile("movdqa %0,%%xmm11" : : "m" (rsa[44]));
-	asm volatile("movdqa %0,%%xmm12" : : "m" (rsa[48]));
-	asm volatile("movdqa %0,%%xmm13" : : "m" (rsa[52]));
-	asm volatile("movdqa %0,%%xmm14" : : "m" (rsa[56]));
-	asm volatile("movdqa %0,%%xmm15" : : "m" (rsa[60]));
+	asm volatile("cpuid"
+		     : "+a" (eax), "=d" (edx)
+		     : : "ecx", "ebx");
 
-	raid6_put_fpu(s->cr0);
+	return (edx >> (flag & 31)) & 1;
 }
 
-#endif /* __x86_64__ */
-
-/* User space test hack */
-#ifndef __KERNEL__
-static inline int cpuid_features(void)
-{
-	u32 eax = 1;
-	u32 ebx, ecx, edx;
-
-	asm volatile("cpuid" :
-		     "+a" (eax), "=b" (ebx), "=c" (ecx), "=d" (edx));
-
-	return edx;
-}
 #endif /* ndef __KERNEL__ */
 
 #endif