RE: [PATCH 1/9] hw/block/nvme: remove pointless rw indirection

[Dmitry Fomichev]

> -----Original Message-----
> From: Keith Busch <kbusch@kernel.org>
> Sent: Wednesday, September 30, 2020 6:04 PM
> To: qemu-block@nongnu.org; qemu-devel@nongnu.org; Klaus Jensen
> <k.jensen@samsung.com>
> Cc: Niklas Cassel <Niklas.Cassel@wdc.com>; Dmitry Fomichev
> <Dmitry.Fomichev@wdc.com>; Kevin Wolf <kwolf@redhat.com>; Philippe
> Mathieu-Daudé <philmd@redhat.com>; Keith Busch <kbusch@kernel.org>
> Subject: [PATCH 1/9] hw/block/nvme: remove pointless rw indirection
> 
> The code switches on the opcode to invoke a function specific to that
> opcode. There's no point in consolidating back to a common function that
> just switches on that same opcode without any actual common code.
> Restore the opcode specific behavior without going back through another
> level of switches.
> 
> Signed-off-by: Keith Busch <kbusch@kernel.org>
> ---
>  hw/block/nvme.c | 91 ++++++++++++++++---------------------------------
>  1 file changed, 29 insertions(+), 62 deletions(-)
> 
> diff --git a/hw/block/nvme.c b/hw/block/nvme.c
> index da8344f196..db52ea0db9 100644
> --- a/hw/block/nvme.c
> +++ b/hw/block/nvme.c
> @@ -927,68 +927,12 @@ static void nvme_rw_cb(void *opaque, int ret)
>      nvme_enqueue_req_completion(nvme_cq(req), req);
>  }
> 
> -static uint16_t nvme_do_aio(BlockBackend *blk, int64_t offset, size_t len,
> -                            NvmeRequest *req)
> -{
> -    BlockAcctCookie *acct = &req->acct;
> -    BlockAcctStats *stats = blk_get_stats(blk);
> -
> -    bool is_write = false;
> -
> -    trace_pci_nvme_do_aio(nvme_cid(req), req->cmd.opcode,
> -                          nvme_io_opc_str(req->cmd.opcode), blk_name(blk),
> -                          offset, len);
> -
> -    switch (req->cmd.opcode) {
> -    case NVME_CMD_FLUSH:
> -        block_acct_start(stats, acct, 0, BLOCK_ACCT_FLUSH);
> -        req->aiocb = blk_aio_flush(blk, nvme_rw_cb, req);
> -        break;
> -
> -    case NVME_CMD_WRITE_ZEROES:
> -        block_acct_start(stats, acct, len, BLOCK_ACCT_WRITE);
> -        req->aiocb = blk_aio_pwrite_zeroes(blk, offset, len,
> -                                           BDRV_REQ_MAY_UNMAP, nvme_rw_cb,
> -                                           req);
> -        break;
> -
> -    case NVME_CMD_WRITE:
> -        is_write = true;
> -
> -        /* fallthrough */
> -
> -    case NVME_CMD_READ:
> -        block_acct_start(stats, acct, len,
> -                         is_write ? BLOCK_ACCT_WRITE : BLOCK_ACCT_READ);
> -
> -        if (req->qsg.sg) {
> -            if (is_write) {
> -                req->aiocb = dma_blk_write(blk, &req->qsg, offset,
> -                                           BDRV_SECTOR_SIZE, nvme_rw_cb, 
> req);
> -            } else {
> -                req->aiocb = dma_blk_read(blk, &req->qsg, offset,
> -                                          BDRV_SECTOR_SIZE, nvme_rw_cb, req);
> -            }
> -        } else {
> -            if (is_write) {
> -                req->aiocb = blk_aio_pwritev(blk, offset, &req->iov, 0,
> -                                             nvme_rw_cb, req);
> -            } else {
> -                req->aiocb = blk_aio_preadv(blk, offset, &req->iov, 0,
> -                                            nvme_rw_cb, req);
> -            }
> -        }
> -
> -        break;
> -    }
> -
> -    return NVME_NO_COMPLETE;
> -}
> -
>  static uint16_t nvme_flush(NvmeCtrl *n, NvmeRequest *req)
>  {
> -    NvmeNamespace *ns = req->ns;
> -    return nvme_do_aio(ns->blkconf.blk, 0, 0, req);
> +    block_acct_start(blk_get_stats(n->conf.blk), &req->acct, 0,

This should be
block_acct_start(blk_get_stats(req->ns->blkconf.blk), &req->acct, 0,

> +                     BLOCK_ACCT_FLUSH);
> +    req->aiocb = blk_aio_flush(n->conf.blk, nvme_rw_cb, req);

and here
req->aiocb = blk_aio_flush(req->ns->blkconf.blk, nvme_rw_cb, req);

> +    return NVME_NO_COMPLETE;
>  }
> 
>  static uint16_t nvme_write_zeroes(NvmeCtrl *n, NvmeRequest *req)
> @@ -1009,7 +953,11 @@ static uint16_t nvme_write_zeroes(NvmeCtrl *n,
> NvmeRequest *req)
>          return status;
>      }
> 
> -    return nvme_do_aio(ns->blkconf.blk, offset, count, req);
> +    block_acct_start(blk_get_stats(n->conf.blk), &req->acct, 0,

 block_acct_start(blk_get_stats(ns->blkconf.blk), &req->acct, 0,

> +                     BLOCK_ACCT_WRITE);
> +    req->aiocb = blk_aio_pwrite_zeroes(n->conf.blk, offset, count,

Here as well,
 req->aiocb = blk_aio_pwrite_zeroes(ns->blkconf.blk, offset, count,

> +                                       BDRV_REQ_MAY_UNMAP, nvme_rw_cb, req);
> +    return NVME_NO_COMPLETE;
>  }
> 
>  static uint16_t nvme_rw(NvmeCtrl *n, NvmeRequest *req)
> @@ -1023,6 +971,7 @@ static uint16_t nvme_rw(NvmeCtrl *n,
> NvmeRequest *req)
>      uint64_t data_offset = nvme_l2b(ns, slba);
>      enum BlockAcctType acct = req->cmd.opcode == NVME_CMD_WRITE ?
>          BLOCK_ACCT_WRITE : BLOCK_ACCT_READ;
> +    BlockBackend *blk = ns->blkconf.blk;
>      uint16_t status;
> 
>      trace_pci_nvme_rw(nvme_cid(req), nvme_io_opc_str(rw->opcode),
> @@ -1045,7 +994,25 @@ static uint16_t nvme_rw(NvmeCtrl *n,
> NvmeRequest *req)
>          goto invalid;
>      }
> 
> -    return nvme_do_aio(ns->blkconf.blk, data_offset, data_size, req);
> +    block_acct_start(blk_get_stats(blk), &req->acct, data_size, acct);
> +    if (req->qsg.sg) {
> +        if (acct == BLOCK_ACCT_WRITE) {
> +            req->aiocb = dma_blk_write(blk, &req->qsg, data_offset,
> +                                       BDRV_SECTOR_SIZE, nvme_rw_cb, req);
> +        } else {
> +            req->aiocb = dma_blk_read(blk, &req->qsg, data_offset,
> +                                      BDRV_SECTOR_SIZE, nvme_rw_cb, req);
> +        }
> +    } else {
> +        if (acct == BLOCK_ACCT_WRITE) {
> +            req->aiocb = blk_aio_pwritev(blk, data_offset, &req->iov, 0,
> +                                         nvme_rw_cb, req);
> +        } else {
> +            req->aiocb = blk_aio_preadv(blk, data_offset, &req->iov, 0,
> +                                        nvme_rw_cb, req);
> +        }
> +    }
> +    return NVME_NO_COMPLETE;
> 
>  invalid:
>      block_acct_invalid(blk_get_stats(ns->blkconf.blk), acct);
> --
> 2.24.1