The following commit has been merged in the linux branch: commit 5dd33c9a4c29015f6d87568d33521c98931a387e Author: NeilBrown neilb@suse.de Date: Fri Oct 16 16:40:25 2009 +1100

md/async: don't pass a memory pointer as a page pointer.

md/raid6 passes a list of 'struct page *' to the async_tx routines, which then either DMA map them for offload, or take the page_address for CPU based calculations.

For RAID6 we sometime leave 'blanks' in the list of pages. For CPU based calcs, we want to treat theses as a page of zeros. For offloaded calculations, we simply don't pass a page to the hardware.

Currently the 'blanks' are encoded as a pointer to raid6_empty_zero_page. This is a 4096 byte memory region, not a 'struct page'. This is mostly handled correctly but is rather ugly.

So change the code to pass and expect a NULL pointer for the blanks. When taking page_address of a page, we need to check for a NULL and in that case use raid6_empty_zero_page.

Signed-off-by: NeilBrown neilb@suse.de

diff --git a/crypto/async_tx/async_pq.c b/crypto/async_tx/async_pq.c index b88db6d..9ab1ce4 100644 --- a/crypto/async_tx/async_pq.c +++ b/crypto/async_tx/async_pq.c @@ -30,11 +30,6 @@ */ static struct page *scribble;

-static bool is_raid6_zero_block(struct page *p) -{ - return p == (void *) raid6_empty_zero_page; -} - /* the struct page *blocks[] parameter passed to async_gen_syndrome() * and async_syndrome_val() contains the 'P' destination address at * blocks[disks-2] and the 'Q' destination address at blocks[disks-1] @@ -83,7 +78,7 @@ do_async_gen_syndrome(struct dma_chan *chan, struct page **blocks, * 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 +155,9 @@ do_sync_gen_syndrome(struct page **blocks, unsigned int offset, int disks, 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; } @@ -290,12 +285,10 @@ async_syndrome_val(struct page **blocks, unsigned int offset, int disks, if (submit->flags & ASYNC_TX_FENCE) dma_flags |= DMA_PREP_FENCE; for (i = 0; i < disks; i++) - if (likely(blocks[i])) { - BUG_ON(is_raid6_zero_block(blocks[i])); + if (likely(blocks[i])) dma_src[i] = dma_map_page(dev, blocks[i], offset, len, DMA_TO_DEVICE); - }

for (;;) { tx = device->device_prep_dma_pq_val(chan, pq, dma_src, diff --git a/crypto/async_tx/async_raid6_recov.c b/crypto/async_tx/async_raid6_recov.c index 6d73dde..8e30b6e 100644 --- a/crypto/async_tx/async_raid6_recov.c +++ b/crypto/async_tx/async_raid6_recov.c @@ -263,10 +263,10 @@ __2data_recov_n(int disks, size_t bytes, int faila, int failb, * 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); @@ -338,7 +338,10 @@ async_raid6_2data_recov(int disks, size_t bytes, int faila, int failb,

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);

@@ -398,7 +401,10 @@ async_raid6_datap_recov(int disks, size_t bytes, int faila,

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);

@@ -414,7 +420,7 @@ async_raid6_datap_recov(int disks, size_t bytes, int faila, * 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 diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c index c4366c9..dcd9e65 100644 --- a/drivers/md/raid5.c +++ b/drivers/md/raid5.c @@ -720,7 +720,7 @@ static int set_syndrome_sources(struct page **srcs, struct stripe_head *sh) int i;

for (i = 0; i < disks; i++) - srcs[i] = (void *)raid6_empty_zero_page; + srcs[i] = NULL;

count = 0; i = d0_idx; @@ -816,7 +816,7 @@ ops_run_compute6_2(struct stripe_head *sh, struct raid5_percpu *percpu) * 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 {