It's possible that the data cache for the buffer still holds data to be flushed to memory, since the buffer was probably used as stack before. Thus we need to make sure to flush it also on reads, since it's possible that the cache is automatically flused to memory after the NVMe DMA transfer happened, thus overwriting the NVMe transfer's data. Also add a missing dcache flush for the prp list.
Signed-off-by: Patrick Wildt patrick@blueri.se --- drivers/nvme/nvme.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
diff --git a/drivers/nvme/nvme.c b/drivers/nvme/nvme.c index dee92b613d..f915817aaa 100644 --- a/drivers/nvme/nvme.c +++ b/drivers/nvme/nvme.c @@ -123,6 +123,9 @@ static int nvme_setup_prps(struct nvme_dev *dev, u64 *prp2, } *prp2 = (ulong)dev->prp_pool;
+ flush_dcache_range((ulong)dev->prp_pool, (ulong)dev->prp_pool + + dev->prp_entry_num * sizeof(u64)); + return 0; }
@@ -717,9 +720,8 @@ static ulong nvme_blk_rw(struct udevice *udev, lbaint_t blknr, u16 lbas = 1 << (dev->max_transfer_shift - ns->lba_shift); u64 total_lbas = blkcnt;
- if (!read) - flush_dcache_range((unsigned long)buffer, - (unsigned long)buffer + total_len); + flush_dcache_range((unsigned long)buffer, + (unsigned long)buffer + total_len);
c.rw.opcode = read ? nvme_cmd_read : nvme_cmd_write; c.rw.flags = 0;