bdev/nvme: do not split SGE callbacks on 2MB boundaries

An SGE could be for a payload that is greater than the NVMe
devices MDTS (i.e. 128KB), but that SGE may not be aligned
on a sector-size boundary.  We can safely assume that each
iov is individually physically contiguous - the DPDK
mempools for example guarantee this.

Signed-off-by: Jim Harris <james.r.harris@intel.com>
Change-Id: I8143ed01814c3154d0a06b8bbc548484437c1e88
This commit is contained in:
Jim Harris 2016-12-20 16:10:20 -07:00 committed by Daniel Verkamp
parent df8129fb39
commit 6d4ce17380

@ -791,10 +791,6 @@ queued_reset_sgl(void *ref, uint32_t sgl_offset)
}
}
#define min(a, b) (((a)<(b))?(a):(b))
#define _2MB_OFFSET(ptr) (((uintptr_t)ptr) & (0x200000 - 1))
static int
queued_next_sge(void *ref, void **address, uint32_t *length)
{
@ -814,8 +810,6 @@ queued_next_sge(void *ref, void **address, uint32_t *length)
*length -= bio->iov_offset;
}
*length = min(*length, 0x200000 - _2MB_OFFSET(*address));
bio->iov_offset += *length;
if (bio->iov_offset == iov->iov_len) {
bio->iovpos++;