Merge branch 'for-linus' of git://neil.brown.name/md
* 'for-linus' of git://neil.brown.name/md:
async_tx: fix asynchronous raid6 recovery for ddf layouts
async_pq: rename scribble page
async_pq: kill a stray dma_map() call and other cleanups
md/raid6: kill a gcc-4.0.1 'uninitialized variable' warning
raid6/async_tx: handle holes in block list in async_syndrome_val
md/async: don't pass a memory pointer as a page pointer.
md: Fix handling of raid5 array which is being reshaped to fewer devices.
md: fix problems with RAID6 calculations for DDF.
md/raid456: downlevel multicore operations to raid_run_ops
md: drivers/md/unroll.pl replaced with awk analog
md: remove clumsy usage of do_sync_mapping_range from bitmap code
md: raid1/raid10: handle allocation errors during array setup.
md/raid5: initialize conf->device_lock earlier
md/raid1/raid10: add a cond_resched
Revert "md: do not progress the resync process if the stripe was blocked"
diff --git a/crypto/async_tx/async_pq.c b/crypto/async_tx/async_pq.c
index b88db6d..6b5cc4f 100644
--- a/crypto/async_tx/async_pq.c
+++ b/crypto/async_tx/async_pq.c
@@ -26,14 +26,10 @@
#include <linux/async_tx.h>
/**
- * scribble - space to hold throwaway P buffer for synchronous gen_syndrome
+ * pq_scribble_page - space to hold throwaway P or Q buffer for
+ * synchronous gen_syndrome
*/
-static struct page *scribble;
-
-static bool is_raid6_zero_block(struct page *p)
-{
- return p == (void *) raid6_empty_zero_page;
-}
+static struct page *pq_scribble_page;
/* the struct page *blocks[] parameter passed to async_gen_syndrome()
* and async_syndrome_val() contains the 'P' destination address at
@@ -83,7 +79,7 @@
* sources and update the coefficients accordingly
*/
for (i = 0, idx = 0; i < src_cnt; i++) {
- if (is_raid6_zero_block(blocks[i]))
+ if (blocks[i] == NULL)
continue;
dma_src[idx] = dma_map_page(dma->dev, blocks[i], offset, len,
DMA_TO_DEVICE);
@@ -160,9 +156,9 @@
srcs = (void **) blocks;
for (i = 0; i < disks; i++) {
- if (is_raid6_zero_block(blocks[i])) {
+ if (blocks[i] == NULL) {
BUG_ON(i > disks - 3); /* P or Q can't be zero */
- srcs[i] = blocks[i];
+ srcs[i] = (void*)raid6_empty_zero_page;
} else
srcs[i] = page_address(blocks[i]) + offset;
}
@@ -186,10 +182,14 @@
* blocks[disks-1] to NULL. When P or Q is omitted 'len' must be <=
* PAGE_SIZE as a temporary buffer of this size is used in the
* synchronous path. 'disks' always accounts for both destination
- * buffers.
+ * buffers. If any source buffers (blocks[i] where i < disks - 2) are
+ * set to NULL those buffers will be replaced with the raid6_zero_page
+ * in the synchronous path and omitted in the hardware-asynchronous
+ * path.
*
* 'blocks' note: if submit->scribble is NULL then the contents of
- * 'blocks' may be overridden
+ * 'blocks' may be overwritten to perform address conversions
+ * (dma_map_page() or page_address()).
*/
struct dma_async_tx_descriptor *
async_gen_syndrome(struct page **blocks, unsigned int offset, int disks,
@@ -227,11 +227,11 @@
async_tx_quiesce(&submit->depend_tx);
if (!P(blocks, disks)) {
- P(blocks, disks) = scribble;
+ P(blocks, disks) = pq_scribble_page;
BUG_ON(len + offset > PAGE_SIZE);
}
if (!Q(blocks, disks)) {
- Q(blocks, disks) = scribble;
+ Q(blocks, disks) = pq_scribble_page;
BUG_ON(len + offset > PAGE_SIZE);
}
do_sync_gen_syndrome(blocks, offset, disks, len, submit);
@@ -265,8 +265,10 @@
len);
struct dma_device *device = chan ? chan->device : NULL;
struct dma_async_tx_descriptor *tx;
+ unsigned char coefs[disks-2];
enum dma_ctrl_flags dma_flags = submit->cb_fn ? DMA_PREP_INTERRUPT : 0;
dma_addr_t *dma_src = NULL;
+ int src_cnt = 0;
BUG_ON(disks < 4);
@@ -285,22 +287,32 @@
__func__, disks, len);
if (!P(blocks, disks))
dma_flags |= DMA_PREP_PQ_DISABLE_P;
+ else
+ pq[0] = dma_map_page(dev, P(blocks, disks),
+ offset, len,
+ DMA_TO_DEVICE);
if (!Q(blocks, disks))
dma_flags |= DMA_PREP_PQ_DISABLE_Q;
+ else
+ pq[1] = dma_map_page(dev, Q(blocks, disks),
+ offset, len,
+ DMA_TO_DEVICE);
+
if (submit->flags & ASYNC_TX_FENCE)
dma_flags |= DMA_PREP_FENCE;
- for (i = 0; i < disks; i++)
+ for (i = 0; i < disks-2; i++)
if (likely(blocks[i])) {
- BUG_ON(is_raid6_zero_block(blocks[i]));
- dma_src[i] = dma_map_page(dev, blocks[i],
- offset, len,
- DMA_TO_DEVICE);
+ dma_src[src_cnt] = dma_map_page(dev, blocks[i],
+ offset, len,
+ DMA_TO_DEVICE);
+ coefs[src_cnt] = raid6_gfexp[i];
+ src_cnt++;
}
for (;;) {
tx = device->device_prep_dma_pq_val(chan, pq, dma_src,
- disks - 2,
- raid6_gfexp,
+ src_cnt,
+ coefs,
len, pqres,
dma_flags);
if (likely(tx))
@@ -373,9 +385,9 @@
static int __init async_pq_init(void)
{
- scribble = alloc_page(GFP_KERNEL);
+ pq_scribble_page = alloc_page(GFP_KERNEL);
- if (scribble)
+ if (pq_scribble_page)
return 0;
pr_err("%s: failed to allocate required spare page\n", __func__);
@@ -385,7 +397,7 @@
static void __exit async_pq_exit(void)
{
- put_page(scribble);
+ put_page(pq_scribble_page);
}
module_init(async_pq_init);
diff --git a/crypto/async_tx/async_raid6_recov.c b/crypto/async_tx/async_raid6_recov.c
index 6d73dde..943f2ab 100644
--- a/crypto/async_tx/async_raid6_recov.c
+++ b/crypto/async_tx/async_raid6_recov.c
@@ -131,8 +131,8 @@
}
static struct dma_async_tx_descriptor *
-__2data_recov_4(size_t bytes, int faila, int failb, struct page **blocks,
- struct async_submit_ctl *submit)
+__2data_recov_4(int disks, size_t bytes, int faila, int failb,
+ struct page **blocks, struct async_submit_ctl *submit)
{
struct dma_async_tx_descriptor *tx = NULL;
struct page *p, *q, *a, *b;
@@ -143,8 +143,8 @@
void *cb_param = submit->cb_param;
void *scribble = submit->scribble;
- p = blocks[4-2];
- q = blocks[4-1];
+ p = blocks[disks-2];
+ q = blocks[disks-1];
a = blocks[faila];
b = blocks[failb];
@@ -170,8 +170,8 @@
}
static struct dma_async_tx_descriptor *
-__2data_recov_5(size_t bytes, int faila, int failb, struct page **blocks,
- struct async_submit_ctl *submit)
+__2data_recov_5(int disks, size_t bytes, int faila, int failb,
+ struct page **blocks, struct async_submit_ctl *submit)
{
struct dma_async_tx_descriptor *tx = NULL;
struct page *p, *q, *g, *dp, *dq;
@@ -181,21 +181,22 @@
dma_async_tx_callback cb_fn = submit->cb_fn;
void *cb_param = submit->cb_param;
void *scribble = submit->scribble;
- int uninitialized_var(good);
- int i;
+ int good_srcs, good, i;
- for (i = 0; i < 3; i++) {
+ good_srcs = 0;
+ good = -1;
+ for (i = 0; i < disks-2; i++) {
+ if (blocks[i] == NULL)
+ continue;
if (i == faila || i == failb)
continue;
- else {
- good = i;
- break;
- }
+ good = i;
+ good_srcs++;
}
- BUG_ON(i >= 3);
+ BUG_ON(good_srcs > 1);
- p = blocks[5-2];
- q = blocks[5-1];
+ p = blocks[disks-2];
+ q = blocks[disks-1];
g = blocks[good];
/* Compute syndrome with zero for the missing data pages
@@ -263,10 +264,10 @@
* delta p and delta q
*/
dp = blocks[faila];
- blocks[faila] = (void *)raid6_empty_zero_page;
+ blocks[faila] = NULL;
blocks[disks-2] = dp;
dq = blocks[failb];
- blocks[failb] = (void *)raid6_empty_zero_page;
+ blocks[failb] = NULL;
blocks[disks-1] = dq;
init_async_submit(submit, ASYNC_TX_FENCE, tx, NULL, NULL, scribble);
@@ -323,6 +324,8 @@
async_raid6_2data_recov(int disks, size_t bytes, int faila, int failb,
struct page **blocks, struct async_submit_ctl *submit)
{
+ int non_zero_srcs, i;
+
BUG_ON(faila == failb);
if (failb < faila)
swap(faila, failb);
@@ -334,11 +337,13 @@
*/
if (!submit->scribble) {
void **ptrs = (void **) blocks;
- int i;
async_tx_quiesce(&submit->depend_tx);
for (i = 0; i < disks; i++)
- ptrs[i] = page_address(blocks[i]);
+ if (blocks[i] == NULL)
+ ptrs[i] = (void *) raid6_empty_zero_page;
+ else
+ ptrs[i] = page_address(blocks[i]);
raid6_2data_recov(disks, bytes, faila, failb, ptrs);
@@ -347,19 +352,30 @@
return NULL;
}
- switch (disks) {
- case 4:
+ non_zero_srcs = 0;
+ for (i = 0; i < disks-2 && non_zero_srcs < 4; i++)
+ if (blocks[i])
+ non_zero_srcs++;
+ switch (non_zero_srcs) {
+ case 0:
+ case 1:
+ /* There must be at least 2 sources - the failed devices. */
+ BUG();
+
+ case 2:
/* dma devices do not uniformly understand a zero source pq
* operation (in contrast to the synchronous case), so
- * explicitly handle the 4 disk special case
+ * explicitly handle the special case of a 4 disk array with
+ * both data disks missing.
*/
- return __2data_recov_4(bytes, faila, failb, blocks, submit);
- case 5:
+ return __2data_recov_4(disks, bytes, faila, failb, blocks, submit);
+ case 3:
/* dma devices do not uniformly understand a single
* source pq operation (in contrast to the synchronous
- * case), so explicitly handle the 5 disk special case
+ * case), so explicitly handle the special case of a 5 disk
+ * array with 2 of 3 data disks missing.
*/
- return __2data_recov_5(bytes, faila, failb, blocks, submit);
+ return __2data_recov_5(disks, bytes, faila, failb, blocks, submit);
default:
return __2data_recov_n(disks, bytes, faila, failb, blocks, submit);
}
@@ -385,6 +401,7 @@
dma_async_tx_callback cb_fn = submit->cb_fn;
void *cb_param = submit->cb_param;
void *scribble = submit->scribble;
+ int good_srcs, good, i;
struct page *srcs[2];
pr_debug("%s: disks: %d len: %zu\n", __func__, disks, bytes);
@@ -394,11 +411,13 @@
*/
if (!scribble) {
void **ptrs = (void **) blocks;
- int i;
async_tx_quiesce(&submit->depend_tx);
for (i = 0; i < disks; i++)
- ptrs[i] = page_address(blocks[i]);
+ if (blocks[i] == NULL)
+ ptrs[i] = (void*)raid6_empty_zero_page;
+ else
+ ptrs[i] = page_address(blocks[i]);
raid6_datap_recov(disks, bytes, faila, ptrs);
@@ -407,6 +426,20 @@
return NULL;
}
+ good_srcs = 0;
+ good = -1;
+ for (i = 0; i < disks-2; i++) {
+ if (i == faila)
+ continue;
+ if (blocks[i]) {
+ good = i;
+ good_srcs++;
+ if (good_srcs > 1)
+ break;
+ }
+ }
+ BUG_ON(good_srcs == 0);
+
p = blocks[disks-2];
q = blocks[disks-1];
@@ -414,14 +447,13 @@
* Use the dead data page as temporary storage for delta q
*/
dq = blocks[faila];
- blocks[faila] = (void *)raid6_empty_zero_page;
+ blocks[faila] = NULL;
blocks[disks-1] = dq;
- /* in the 4 disk case we only need to perform a single source
- * multiplication
+ /* in the 4-disk case we only need to perform a single source
+ * multiplication with the one good data block.
*/
- if (disks == 4) {
- int good = faila == 0 ? 1 : 0;
+ if (good_srcs == 1) {
struct page *g = blocks[good];
init_async_submit(submit, ASYNC_TX_FENCE, tx, NULL, NULL,
diff --git a/crypto/async_tx/async_xor.c b/crypto/async_tx/async_xor.c
index b459a90..79182dc 100644
--- a/crypto/async_tx/async_xor.c
+++ b/crypto/async_tx/async_xor.c
@@ -44,20 +44,23 @@
void *cb_param_orig = submit->cb_param;
enum async_tx_flags flags_orig = submit->flags;
enum dma_ctrl_flags dma_flags;
- int xor_src_cnt;
+ int xor_src_cnt = 0;
dma_addr_t dma_dest;
/* map the dest bidrectional in case it is re-used as a source */
dma_dest = dma_map_page(dma->dev, dest, offset, len, DMA_BIDIRECTIONAL);
for (i = 0; i < src_cnt; i++) {
/* only map the dest once */
+ if (!src_list[i])
+ continue;
if (unlikely(src_list[i] == dest)) {
- dma_src[i] = dma_dest;
+ dma_src[xor_src_cnt++] = dma_dest;
continue;
}
- dma_src[i] = dma_map_page(dma->dev, src_list[i], offset,
- len, DMA_TO_DEVICE);
+ dma_src[xor_src_cnt++] = dma_map_page(dma->dev, src_list[i], offset,
+ len, DMA_TO_DEVICE);
}
+ src_cnt = xor_src_cnt;
while (src_cnt) {
submit->flags = flags_orig;
@@ -123,7 +126,7 @@
int src_cnt, size_t len, struct async_submit_ctl *submit)
{
int i;
- int xor_src_cnt;
+ int xor_src_cnt = 0;
int src_off = 0;
void *dest_buf;
void **srcs;
@@ -135,8 +138,9 @@
/* convert to buffer pointers */
for (i = 0; i < src_cnt; i++)
- srcs[i] = page_address(src_list[i]) + offset;
-
+ if (src_list[i])
+ srcs[xor_src_cnt++] = page_address(src_list[i]) + offset;
+ src_cnt = xor_src_cnt;
/* set destination address */
dest_buf = page_address(dest) + offset;
diff --git a/drivers/md/Makefile b/drivers/md/Makefile
index 1dc4185..e355e7f 100644
--- a/drivers/md/Makefile
+++ b/drivers/md/Makefile
@@ -46,7 +46,7 @@
obj-$(CONFIG_DM_ZERO) += dm-zero.o
quiet_cmd_unroll = UNROLL $@
- cmd_unroll = $(PERL) $(srctree)/$(src)/unroll.pl $(UNROLL) \
+ cmd_unroll = $(AWK) -f$(srctree)/$(src)/unroll.awk -vN=$(UNROLL) \
< $< > $@ || ( rm -f $@ && exit 1 )
ifeq ($(CONFIG_ALTIVEC),y)
@@ -59,56 +59,56 @@
targets += raid6int1.c
$(obj)/raid6int1.c: UNROLL := 1
-$(obj)/raid6int1.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int1.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int2.c
$(obj)/raid6int2.c: UNROLL := 2
-$(obj)/raid6int2.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int2.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int4.c
$(obj)/raid6int4.c: UNROLL := 4
-$(obj)/raid6int4.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int4.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int8.c
$(obj)/raid6int8.c: UNROLL := 8
-$(obj)/raid6int8.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int8.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int16.c
$(obj)/raid6int16.c: UNROLL := 16
-$(obj)/raid6int16.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int16.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
targets += raid6int32.c
$(obj)/raid6int32.c: UNROLL := 32
-$(obj)/raid6int32.c: $(src)/raid6int.uc $(src)/unroll.pl FORCE
+$(obj)/raid6int32.c: $(src)/raid6int.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
CFLAGS_raid6altivec1.o += $(altivec_flags)
targets += raid6altivec1.c
$(obj)/raid6altivec1.c: UNROLL := 1
-$(obj)/raid6altivec1.c: $(src)/raid6altivec.uc $(src)/unroll.pl FORCE
+$(obj)/raid6altivec1.c: $(src)/raid6altivec.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
CFLAGS_raid6altivec2.o += $(altivec_flags)
targets += raid6altivec2.c
$(obj)/raid6altivec2.c: UNROLL := 2
-$(obj)/raid6altivec2.c: $(src)/raid6altivec.uc $(src)/unroll.pl FORCE
+$(obj)/raid6altivec2.c: $(src)/raid6altivec.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
CFLAGS_raid6altivec4.o += $(altivec_flags)
targets += raid6altivec4.c
$(obj)/raid6altivec4.c: UNROLL := 4
-$(obj)/raid6altivec4.c: $(src)/raid6altivec.uc $(src)/unroll.pl FORCE
+$(obj)/raid6altivec4.c: $(src)/raid6altivec.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
CFLAGS_raid6altivec8.o += $(altivec_flags)
targets += raid6altivec8.c
$(obj)/raid6altivec8.c: UNROLL := 8
-$(obj)/raid6altivec8.c: $(src)/raid6altivec.uc $(src)/unroll.pl FORCE
+$(obj)/raid6altivec8.c: $(src)/raid6altivec.uc $(src)/unroll.awk FORCE
$(call if_changed,unroll)
quiet_cmd_mktable = TABLE $@
diff --git a/drivers/md/bitmap.c b/drivers/md/bitmap.c
index 6986b00..60e2b32 100644
--- a/drivers/md/bitmap.c
+++ b/drivers/md/bitmap.c
@@ -1624,10 +1624,11 @@
bitmap->offset = mddev->bitmap_offset;
if (file) {
get_file(file);
- do_sync_mapping_range(file->f_mapping, 0, LLONG_MAX,
- SYNC_FILE_RANGE_WAIT_BEFORE |
- SYNC_FILE_RANGE_WRITE |
- SYNC_FILE_RANGE_WAIT_AFTER);
+ /* As future accesses to this file will use bmap,
+ * and bypass the page cache, we must sync the file
+ * first.
+ */
+ vfs_fsync(file, file->f_dentry, 1);
}
/* read superblock from bitmap file (this sets bitmap->chunksize) */
err = bitmap_read_sb(bitmap);
diff --git a/drivers/md/md.c b/drivers/md/md.c
index 26ba42a..10eb1fc 100644
--- a/drivers/md/md.c
+++ b/drivers/md/md.c
@@ -2631,7 +2631,7 @@
rdev->desc_nr = i++;
rdev->raid_disk = rdev->desc_nr;
set_bit(In_sync, &rdev->flags);
- } else if (rdev->raid_disk >= mddev->raid_disks) {
+ } else if (rdev->raid_disk >= (mddev->raid_disks - min(0, mddev->delta_disks))) {
rdev->raid_disk = -1;
clear_bit(In_sync, &rdev->flags);
}
diff --git a/drivers/md/raid1.c b/drivers/md/raid1.c
index d1b9bd5..a053423 100644
--- a/drivers/md/raid1.c
+++ b/drivers/md/raid1.c
@@ -64,7 +64,7 @@
/* allocate a r1bio with room for raid_disks entries in the bios array */
r1_bio = kzalloc(size, gfp_flags);
- if (!r1_bio)
+ if (!r1_bio && pi->mddev)
unplug_slaves(pi->mddev);
return r1_bio;
@@ -1683,6 +1683,7 @@
generic_make_request(bio);
}
}
+ cond_resched();
}
if (unplug)
unplug_slaves(mddev);
@@ -1978,13 +1979,14 @@
conf->poolinfo = kmalloc(sizeof(*conf->poolinfo), GFP_KERNEL);
if (!conf->poolinfo)
goto out_no_mem;
- conf->poolinfo->mddev = mddev;
+ conf->poolinfo->mddev = NULL;
conf->poolinfo->raid_disks = mddev->raid_disks;
conf->r1bio_pool = mempool_create(NR_RAID1_BIOS, r1bio_pool_alloc,
r1bio_pool_free,
conf->poolinfo);
if (!conf->r1bio_pool)
goto out_no_mem;
+ conf->poolinfo->mddev = mddev;
spin_lock_init(&conf->device_lock);
mddev->queue->queue_lock = &conf->device_lock;
diff --git a/drivers/md/raid10.c b/drivers/md/raid10.c
index 51c4c5c..c2cb7b8 100644
--- a/drivers/md/raid10.c
+++ b/drivers/md/raid10.c
@@ -68,7 +68,7 @@
/* allocate a r10bio with room for raid_disks entries in the bios array */
r10_bio = kzalloc(size, gfp_flags);
- if (!r10_bio)
+ if (!r10_bio && conf->mddev)
unplug_slaves(conf->mddev);
return r10_bio;
@@ -1632,6 +1632,7 @@
generic_make_request(bio);
}
}
+ cond_resched();
}
if (unplug)
unplug_slaves(mddev);
@@ -2095,7 +2096,6 @@
if (!conf->tmppage)
goto out_free_conf;
- conf->mddev = mddev;
conf->raid_disks = mddev->raid_disks;
conf->near_copies = nc;
conf->far_copies = fc;
@@ -2132,6 +2132,7 @@
goto out_free_conf;
}
+ conf->mddev = mddev;
spin_lock_init(&conf->device_lock);
mddev->queue->queue_lock = &conf->device_lock;
diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
index 9482980..81abefc 100644
--- a/drivers/md/raid5.c
+++ b/drivers/md/raid5.c
@@ -156,13 +156,16 @@
static int raid6_idx_to_slot(int idx, struct stripe_head *sh,
int *count, int syndrome_disks)
{
- int slot;
+ int slot = *count;
+ if (sh->ddf_layout)
+ (*count)++;
if (idx == sh->pd_idx)
return syndrome_disks;
if (idx == sh->qd_idx)
return syndrome_disks + 1;
- slot = (*count)++;
+ if (!sh->ddf_layout)
+ (*count)++;
return slot;
}
@@ -717,7 +720,7 @@
int i;
for (i = 0; i < disks; i++)
- srcs[i] = (void *)raid6_empty_zero_page;
+ srcs[i] = NULL;
count = 0;
i = d0_idx;
@@ -727,9 +730,8 @@
srcs[slot] = sh->dev[i].page;
i = raid6_next_disk(i, disks);
} while (i != d0_idx);
- BUG_ON(count != syndrome_disks);
- return count;
+ return syndrome_disks;
}
static struct dma_async_tx_descriptor *
@@ -814,7 +816,7 @@
* slot number conversion for 'faila' and 'failb'
*/
for (i = 0; i < disks ; i++)
- blocks[i] = (void *)raid6_empty_zero_page;
+ blocks[i] = NULL;
count = 0;
i = d0_idx;
do {
@@ -828,7 +830,6 @@
failb = slot;
i = raid6_next_disk(i, disks);
} while (i != d0_idx);
- BUG_ON(count != syndrome_disks);
BUG_ON(faila == failb);
if (failb < faila)
@@ -845,7 +846,7 @@
init_async_submit(&submit, ASYNC_TX_FENCE, NULL,
ops_complete_compute, sh,
to_addr_conv(sh, percpu));
- return async_gen_syndrome(blocks, 0, count+2,
+ return async_gen_syndrome(blocks, 0, syndrome_disks+2,
STRIPE_SIZE, &submit);
} else {
struct page *dest;
@@ -1139,7 +1140,7 @@
&sh->ops.zero_sum_result, percpu->spare_page, &submit);
}
-static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
+static void __raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
{
int overlap_clear = 0, i, disks = sh->disks;
struct dma_async_tx_descriptor *tx = NULL;
@@ -1204,22 +1205,55 @@
put_cpu();
}
+#ifdef CONFIG_MULTICORE_RAID456
+static void async_run_ops(void *param, async_cookie_t cookie)
+{
+ struct stripe_head *sh = param;
+ unsigned long ops_request = sh->ops.request;
+
+ clear_bit_unlock(STRIPE_OPS_REQ_PENDING, &sh->state);
+ wake_up(&sh->ops.wait_for_ops);
+
+ __raid_run_ops(sh, ops_request);
+ release_stripe(sh);
+}
+
+static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
+{
+ /* since handle_stripe can be called outside of raid5d context
+ * we need to ensure sh->ops.request is de-staged before another
+ * request arrives
+ */
+ wait_event(sh->ops.wait_for_ops,
+ !test_and_set_bit_lock(STRIPE_OPS_REQ_PENDING, &sh->state));
+ sh->ops.request = ops_request;
+
+ atomic_inc(&sh->count);
+ async_schedule(async_run_ops, sh);
+}
+#else
+#define raid_run_ops __raid_run_ops
+#endif
+
static int grow_one_stripe(raid5_conf_t *conf)
{
struct stripe_head *sh;
+ int disks = max(conf->raid_disks, conf->previous_raid_disks);
sh = kmem_cache_alloc(conf->slab_cache, GFP_KERNEL);
if (!sh)
return 0;
- memset(sh, 0, sizeof(*sh) + (conf->raid_disks-1)*sizeof(struct r5dev));
+ memset(sh, 0, sizeof(*sh) + (disks-1)*sizeof(struct r5dev));
sh->raid_conf = conf;
spin_lock_init(&sh->lock);
+ #ifdef CONFIG_MULTICORE_RAID456
+ init_waitqueue_head(&sh->ops.wait_for_ops);
+ #endif
- if (grow_buffers(sh, conf->raid_disks)) {
- shrink_buffers(sh, conf->raid_disks);
+ if (grow_buffers(sh, disks)) {
+ shrink_buffers(sh, disks);
kmem_cache_free(conf->slab_cache, sh);
return 0;
}
- sh->disks = conf->raid_disks;
/* we just created an active stripe so... */
atomic_set(&sh->count, 1);
atomic_inc(&conf->active_stripes);
@@ -1231,7 +1265,7 @@
static int grow_stripes(raid5_conf_t *conf, int num)
{
struct kmem_cache *sc;
- int devs = conf->raid_disks;
+ int devs = max(conf->raid_disks, conf->previous_raid_disks);
sprintf(conf->cache_name[0],
"raid%d-%s", conf->level, mdname(conf->mddev));
@@ -1329,6 +1363,9 @@
nsh->raid_conf = conf;
spin_lock_init(&nsh->lock);
+ #ifdef CONFIG_MULTICORE_RAID456
+ init_waitqueue_head(&nsh->ops.wait_for_ops);
+ #endif
list_add(&nsh->lru, &newstripes);
}
@@ -1899,10 +1936,15 @@
case ALGORITHM_PARITY_N:
break;
case ALGORITHM_ROTATING_N_CONTINUE:
+ /* Like left_symmetric, but P is before Q */
if (sh->pd_idx == 0)
i--; /* P D D D Q */
- else if (i > sh->pd_idx)
- i -= 2; /* D D Q P D */
+ else {
+ /* D D Q P D */
+ if (i < sh->pd_idx)
+ i += raid_disks;
+ i -= (sh->pd_idx + 1);
+ }
break;
case ALGORITHM_LEFT_ASYMMETRIC_6:
case ALGORITHM_RIGHT_ASYMMETRIC_6:
@@ -2896,7 +2938,7 @@
*
*/
-static bool handle_stripe5(struct stripe_head *sh)
+static void handle_stripe5(struct stripe_head *sh)
{
raid5_conf_t *conf = sh->raid_conf;
int disks = sh->disks, i;
@@ -3167,11 +3209,9 @@
ops_run_io(sh, &s);
return_io(return_bi);
-
- return blocked_rdev == NULL;
}
-static bool handle_stripe6(struct stripe_head *sh)
+static void handle_stripe6(struct stripe_head *sh)
{
raid5_conf_t *conf = sh->raid_conf;
int disks = sh->disks;
@@ -3455,17 +3495,14 @@
ops_run_io(sh, &s);
return_io(return_bi);
-
- return blocked_rdev == NULL;
}
-/* returns true if the stripe was handled */
-static bool handle_stripe(struct stripe_head *sh)
+static void handle_stripe(struct stripe_head *sh)
{
if (sh->raid_conf->level == 6)
- return handle_stripe6(sh);
+ handle_stripe6(sh);
else
- return handle_stripe5(sh);
+ handle_stripe5(sh);
}
static void raid5_activate_delayed(raid5_conf_t *conf)
@@ -3503,9 +3540,10 @@
{
raid5_conf_t *conf = mddev->private;
int i;
+ int devs = max(conf->raid_disks, conf->previous_raid_disks);
rcu_read_lock();
- for (i = 0; i < conf->raid_disks; i++) {
+ for (i = 0; i < devs; i++) {
mdk_rdev_t *rdev = rcu_dereference(conf->disks[i].rdev);
if (rdev && !test_bit(Faulty, &rdev->flags) && atomic_read(&rdev->nr_pending)) {
struct request_queue *r_queue = bdev_get_queue(rdev->bdev);
@@ -4277,9 +4315,7 @@
clear_bit(STRIPE_INSYNC, &sh->state);
spin_unlock(&sh->lock);
- /* wait for any blocked device to be handled */
- while (unlikely(!handle_stripe(sh)))
- ;
+ handle_stripe(sh);
release_stripe(sh);
return STRIPE_SECTORS;
@@ -4349,37 +4385,6 @@
return handled;
}
-#ifdef CONFIG_MULTICORE_RAID456
-static void __process_stripe(void *param, async_cookie_t cookie)
-{
- struct stripe_head *sh = param;
-
- handle_stripe(sh);
- release_stripe(sh);
-}
-
-static void process_stripe(struct stripe_head *sh, struct list_head *domain)
-{
- async_schedule_domain(__process_stripe, sh, domain);
-}
-
-static void synchronize_stripe_processing(struct list_head *domain)
-{
- async_synchronize_full_domain(domain);
-}
-#else
-static void process_stripe(struct stripe_head *sh, struct list_head *domain)
-{
- handle_stripe(sh);
- release_stripe(sh);
- cond_resched();
-}
-
-static void synchronize_stripe_processing(struct list_head *domain)
-{
-}
-#endif
-
/*
* This is our raid5 kernel thread.
@@ -4393,7 +4398,6 @@
struct stripe_head *sh;
raid5_conf_t *conf = mddev->private;
int handled;
- LIST_HEAD(raid_domain);
pr_debug("+++ raid5d active\n");
@@ -4430,7 +4434,9 @@
spin_unlock_irq(&conf->device_lock);
handled++;
- process_stripe(sh, &raid_domain);
+ handle_stripe(sh);
+ release_stripe(sh);
+ cond_resched();
spin_lock_irq(&conf->device_lock);
}
@@ -4438,7 +4444,6 @@
spin_unlock_irq(&conf->device_lock);
- synchronize_stripe_processing(&raid_domain);
async_tx_issue_pending_all();
unplug_slaves(mddev);
@@ -4558,13 +4563,9 @@
if (!sectors)
sectors = mddev->dev_sectors;
- if (!raid_disks) {
+ if (!raid_disks)
/* size is defined by the smallest of previous and new size */
- if (conf->raid_disks < conf->previous_raid_disks)
- raid_disks = conf->raid_disks;
- else
- raid_disks = conf->previous_raid_disks;
- }
+ raid_disks = min(conf->raid_disks, conf->previous_raid_disks);
sectors &= ~((sector_t)mddev->chunk_sectors - 1);
sectors &= ~((sector_t)mddev->new_chunk_sectors - 1);
@@ -4665,7 +4666,7 @@
}
per_cpu_ptr(conf->percpu, cpu)->spare_page = spare_page;
}
- scribble = kmalloc(scribble_len(conf->raid_disks), GFP_KERNEL);
+ scribble = kmalloc(conf->scribble_len, GFP_KERNEL);
if (!scribble) {
err = -ENOMEM;
break;
@@ -4686,7 +4687,7 @@
static raid5_conf_t *setup_conf(mddev_t *mddev)
{
raid5_conf_t *conf;
- int raid_disk, memory;
+ int raid_disk, memory, max_disks;
mdk_rdev_t *rdev;
struct disk_info *disk;
@@ -4722,28 +4723,6 @@
conf = kzalloc(sizeof(raid5_conf_t), GFP_KERNEL);
if (conf == NULL)
goto abort;
-
- conf->raid_disks = mddev->raid_disks;
- conf->scribble_len = scribble_len(conf->raid_disks);
- if (mddev->reshape_position == MaxSector)
- conf->previous_raid_disks = mddev->raid_disks;
- else
- conf->previous_raid_disks = mddev->raid_disks - mddev->delta_disks;
-
- conf->disks = kzalloc(conf->raid_disks * sizeof(struct disk_info),
- GFP_KERNEL);
- if (!conf->disks)
- goto abort;
-
- conf->mddev = mddev;
-
- if ((conf->stripe_hashtbl = kzalloc(PAGE_SIZE, GFP_KERNEL)) == NULL)
- goto abort;
-
- conf->level = mddev->new_level;
- if (raid5_alloc_percpu(conf) != 0)
- goto abort;
-
spin_lock_init(&conf->device_lock);
init_waitqueue_head(&conf->wait_for_stripe);
init_waitqueue_head(&conf->wait_for_overlap);
@@ -4757,11 +4736,33 @@
atomic_set(&conf->active_aligned_reads, 0);
conf->bypass_threshold = BYPASS_THRESHOLD;
+ conf->raid_disks = mddev->raid_disks;
+ if (mddev->reshape_position == MaxSector)
+ conf->previous_raid_disks = mddev->raid_disks;
+ else
+ conf->previous_raid_disks = mddev->raid_disks - mddev->delta_disks;
+ max_disks = max(conf->raid_disks, conf->previous_raid_disks);
+ conf->scribble_len = scribble_len(max_disks);
+
+ conf->disks = kzalloc(max_disks * sizeof(struct disk_info),
+ GFP_KERNEL);
+ if (!conf->disks)
+ goto abort;
+
+ conf->mddev = mddev;
+
+ if ((conf->stripe_hashtbl = kzalloc(PAGE_SIZE, GFP_KERNEL)) == NULL)
+ goto abort;
+
+ conf->level = mddev->new_level;
+ if (raid5_alloc_percpu(conf) != 0)
+ goto abort;
+
pr_debug("raid5: run(%s) called.\n", mdname(mddev));
list_for_each_entry(rdev, &mddev->disks, same_set) {
raid_disk = rdev->raid_disk;
- if (raid_disk >= conf->raid_disks
+ if (raid_disk >= max_disks
|| raid_disk < 0)
continue;
disk = conf->disks + raid_disk;
@@ -4793,7 +4794,7 @@
}
memory = conf->max_nr_stripes * (sizeof(struct stripe_head) +
- conf->raid_disks * ((sizeof(struct bio) + PAGE_SIZE))) / 1024;
+ max_disks * ((sizeof(struct bio) + PAGE_SIZE))) / 1024;
if (grow_stripes(conf, conf->max_nr_stripes)) {
printk(KERN_ERR
"raid5: couldn't allocate %dkB for buffers\n", memory);
@@ -4918,7 +4919,8 @@
test_bit(In_sync, &rdev->flags))
working_disks++;
- mddev->degraded = conf->raid_disks - working_disks;
+ mddev->degraded = (max(conf->raid_disks, conf->previous_raid_disks)
+ - working_disks);
if (mddev->degraded > conf->max_degraded) {
printk(KERN_ERR "raid5: not enough operational devices for %s"
diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
index 2390e0e..dd70835 100644
--- a/drivers/md/raid5.h
+++ b/drivers/md/raid5.h
@@ -214,12 +214,20 @@
int disks; /* disks in stripe */
enum check_states check_state;
enum reconstruct_states reconstruct_state;
- /* stripe_operations
+ /**
+ * struct stripe_operations
* @target - STRIPE_OP_COMPUTE_BLK target
+ * @target2 - 2nd compute target in the raid6 case
+ * @zero_sum_result - P and Q verification flags
+ * @request - async service request flags for raid_run_ops
*/
struct stripe_operations {
int target, target2;
enum sum_check_flags zero_sum_result;
+ #ifdef CONFIG_MULTICORE_RAID456
+ unsigned long request;
+ wait_queue_head_t wait_for_ops;
+ #endif
} ops;
struct r5dev {
struct bio req;
@@ -294,6 +302,8 @@
#define STRIPE_FULL_WRITE 13 /* all blocks are set to be overwritten */
#define STRIPE_BIOFILL_RUN 14
#define STRIPE_COMPUTE_RUN 15
+#define STRIPE_OPS_REQ_PENDING 16
+
/*
* Operation request flags
*/
@@ -478,7 +488,7 @@
{
return (layout >= 0 && layout <= 5)
||
- (layout == 8 || layout == 10)
+ (layout >= 8 && layout <= 10)
||
(layout >= 16 && layout <= 20);
}
diff --git a/drivers/md/raid6altivec.uc b/drivers/md/raid6altivec.uc
index 699dfee..2654d5c 100644
--- a/drivers/md/raid6altivec.uc
+++ b/drivers/md/raid6altivec.uc
@@ -15,7 +15,7 @@
*
* $#-way unrolled portable integer math RAID-6 instruction set
*
- * This file is postprocessed using unroll.pl
+ * This file is postprocessed using unroll.awk
*
* <benh> hpa: in process,
* you can just "steal" the vec unit with enable_kernel_altivec() (but
diff --git a/drivers/md/raid6int.uc b/drivers/md/raid6int.uc
index f9bf9cb..d1e276a 100644
--- a/drivers/md/raid6int.uc
+++ b/drivers/md/raid6int.uc
@@ -15,7 +15,7 @@
*
* $#-way unrolled portable integer math RAID-6 instruction set
*
- * This file is postprocessed using unroll.pl
+ * This file is postprocessed using unroll.awk
*/
#include <linux/raid/pq.h>
diff --git a/drivers/md/raid6test/Makefile b/drivers/md/raid6test/Makefile
index 58ffdf4..2874cbe 100644
--- a/drivers/md/raid6test/Makefile
+++ b/drivers/md/raid6test/Makefile
@@ -7,7 +7,7 @@
OPTFLAGS = -O2 # Adjust as desired
CFLAGS = -I.. -I ../../../include -g $(OPTFLAGS)
LD = ld
-PERL = perl
+AWK = awk
AR = ar
RANLIB = ranlib
@@ -35,35 +35,35 @@
raid6test: test.c raid6.a
$(CC) $(CFLAGS) -o raid6test $^
-raid6altivec1.c: raid6altivec.uc ../unroll.pl
- $(PERL) ../unroll.pl 1 < raid6altivec.uc > $@
+raid6altivec1.c: raid6altivec.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=1 < raid6altivec.uc > $@
-raid6altivec2.c: raid6altivec.uc ../unroll.pl
- $(PERL) ../unroll.pl 2 < raid6altivec.uc > $@
+raid6altivec2.c: raid6altivec.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=2 < raid6altivec.uc > $@
-raid6altivec4.c: raid6altivec.uc ../unroll.pl
- $(PERL) ../unroll.pl 4 < raid6altivec.uc > $@
+raid6altivec4.c: raid6altivec.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=4 < raid6altivec.uc > $@
-raid6altivec8.c: raid6altivec.uc ../unroll.pl
- $(PERL) ../unroll.pl 8 < raid6altivec.uc > $@
+raid6altivec8.c: raid6altivec.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=8 < raid6altivec.uc > $@
-raid6int1.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 1 < raid6int.uc > $@
+raid6int1.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=1 < raid6int.uc > $@
-raid6int2.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 2 < raid6int.uc > $@
+raid6int2.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=2 < raid6int.uc > $@
-raid6int4.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 4 < raid6int.uc > $@
+raid6int4.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=4 < raid6int.uc > $@
-raid6int8.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 8 < raid6int.uc > $@
+raid6int8.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=8 < raid6int.uc > $@
-raid6int16.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 16 < raid6int.uc > $@
+raid6int16.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=16 < raid6int.uc > $@
-raid6int32.c: raid6int.uc ../unroll.pl
- $(PERL) ../unroll.pl 32 < raid6int.uc > $@
+raid6int32.c: raid6int.uc ../unroll.awk
+ $(AWK) ../unroll.awk -vN=32 < raid6int.uc > $@
raid6tables.c: mktables
./mktables > raid6tables.c
diff --git a/drivers/md/unroll.awk b/drivers/md/unroll.awk
new file mode 100644
index 0000000..c6aa036
--- /dev/null
+++ b/drivers/md/unroll.awk
@@ -0,0 +1,20 @@
+
+# This filter requires one command line option of form -vN=n
+# where n must be a decimal number.
+#
+# Repeat each input line containing $$ n times, replacing $$ with 0...n-1.
+# Replace each $# with n, and each $* with a single $.
+
+BEGIN {
+ n = N + 0
+}
+{
+ if (/\$\$/) { rep = n } else { rep = 1 }
+ for (i = 0; i < rep; ++i) {
+ tmp = $0
+ gsub(/\$\$/, i, tmp)
+ gsub(/\$\#/, n, tmp)
+ gsub(/\$\*/, "$", tmp)
+ print tmp
+ }
+}
diff --git a/drivers/md/unroll.pl b/drivers/md/unroll.pl
deleted file mode 100644
index 3acc710..0000000
--- a/drivers/md/unroll.pl
+++ /dev/null
@@ -1,24 +0,0 @@
-#!/usr/bin/perl
-#
-# Take a piece of C code and for each line which contains the sequence $$
-# repeat n times with $ replaced by 0...n-1; the sequence $# is replaced
-# by the unrolling factor, and $* with a single $
-#
-
-($n) = @ARGV;
-$n += 0;
-
-while ( defined($line = <STDIN>) ) {
- if ( $line =~ /\$\$/ ) {
- $rep = $n;
- } else {
- $rep = 1;
- }
- for ( $i = 0 ; $i < $rep ; $i++ ) {
- $tmp = $line;
- $tmp =~ s/\$\$/$i/g;
- $tmp =~ s/\$\#/$n/g;
- $tmp =~ s/\$\*/\$/g;
- print $tmp;
- }
-}
