module/accel/ioat: fix bug with preparing a fill command for batch

The accel_fw API taks a uint8_t for fill pattern, the ioat build
command expects a full 64.

Signed-off-by: paul luse <paul.e.luse@intel.com>
Change-Id: I55b33706a9e48fe7efd57a06816e11fdd8b7fa4c
Reviewed-on: https://review.spdk.io/gerrit/c/spdk/spdk/+/3322
Community-CI: Mellanox Build Bot
Reviewed-by: Jim Harris <james.r.harris@intel.com>
Reviewed-by: Ben Walker <benjamin.walker@intel.com>
Tested-by: SPDK CI Jenkins <sys_sgci@intel.com>
This commit is contained in:
paul luse 2020-07-13 18:27:52 -04:00 committed by Jim Harris
parent 0cecfcb19b
commit acee02acc9

View File

@ -304,12 +304,14 @@ ioat_batch_prep_fill(struct spdk_io_channel *ch, struct spdk_accel_batch *batch,
{
struct ioat_io_channel *ioat_ch = spdk_io_channel_get_ctx(ch);
struct ioat_task *ioat_task = (struct ioat_task *)cb_arg;
uint64_t fill_pattern;
ioat_task->cb = cb_fn;
ioat_ch->hw_batch = true;
memset(&fill_pattern, fill, sizeof(uint64_t));
/* Call the IOAT library prep function. */
return spdk_ioat_build_fill(ioat_ch->ioat_ch, ioat_task, ioat_done, dst, fill, nbytes);
return spdk_ioat_build_fill(ioat_ch->ioat_ch, ioat_task, ioat_done, dst, fill_pattern, nbytes);
}
static int