diff options
| author | FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp> | 2010-08-10 18:03:24 -0700 |
|---|---|---|
| committer | Linus Torvalds <torvalds@linux-foundation.org> | 2010-08-11 08:59:21 -0700 |
| commit | d80e0d96a328cc864a1cb359f545a6ed0c61812d (patch) | |
| tree | 805689907e865c14943b7b77b7b1d64ca8044e01 /drivers/scsi/53c700.c | |
| parent | 7896bfa451b209f73bc8ec14721dcc2c5329a0a9 (diff) | |
| download | linux-d80e0d96a328cc864a1cb359f545a6ed0c61812d.tar.gz | |
scsi: 53c700: remove dma_is_consistent usage
This driver is the only user of dma_is_consistent(). We plan to remove this
API.
The driver uses the API in the following way:
BUG_ON(!dma_is_consistent(hostdata->dev, pScript) && L1_CACHE_BYTES < dma_get_cache_alignment());
The above code tries to see if L1_CACHE_BYTES is greater than
dma_get_cache_alignment() on sysmtes that can not allocate coherent memory
(some old systems can't).
James Bottomley exmplained that this is necesary because the driver packs the
set of mailboxes into a single coherent area and separates the different
usages by a L1 cache stride. So it's fatal if the dma
He also pointed out that we can kill this checking because we don't hit this
BUG_ON on all architectures that actually use the driver.
(akpm: stolen from the scsi tree because
dma-mapping-remove-dma_is_consistent-api.patch needs it)
Signed-off-by: FUJITA Tomonori <fujita.tomonori@lab.ntt.co.jp>
Signed-off-by: James Bottomley <James.Bottomley@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Diffstat (limited to 'drivers/scsi/53c700.c')
| -rw-r--r-- | drivers/scsi/53c700.c | 3 |
1 files changed, 0 insertions, 3 deletions
diff --git a/drivers/scsi/53c700.c b/drivers/scsi/53c700.c index 80dc3ac12cde..89fc1c8af86b 100644 --- a/drivers/scsi/53c700.c +++ b/drivers/scsi/53c700.c @@ -309,9 +309,6 @@ NCR_700_detect(struct scsi_host_template *tpnt, hostdata->msgin = memory + MSGIN_OFFSET; hostdata->msgout = memory + MSGOUT_OFFSET; hostdata->status = memory + STATUS_OFFSET; - /* all of these offsets are L1_CACHE_BYTES separated. It is fatal - * if this isn't sufficient separation to avoid dma flushing issues */ - BUG_ON(!dma_is_consistent(hostdata->dev, pScript) && L1_CACHE_BYTES < dma_get_cache_alignment()); hostdata->slots = (struct NCR_700_command_slot *)(memory + SLOTS_OFFSET); hostdata->dev = dev; |