(Resend, as I forgot to switch to plain-text for linux-usb@vger rules).
Uhh... I think this is a bad idea. The 120K limit was chosen based on
seeing a lot of devices which broke if you tried to transfer more. At
least, that's my memory. Otherwise, it's a really goofy number.
I certainly wouldn't mind you increasing it in the case of a USB 3.x
device, but not globally.
Matt
On Thu, Oct 29, 2015 at 1:21 PM, Felipe Balbi <balbi@xxxxxx> wrote:
> Instead of using only 120K for transfer size, let's
> increase it to 1024K. This has a very noticeable
> impact on USB3 storage devices.
>
> The following measurements were done with v4.2 on
> the host side and latest linux-next available at the
> time (20151022). Following are results with default
> and modified max_sectors. Each number is the average
> of 100 measurements with each of the two settings.
>
> Pre 244.87
> Post 322.10
>
> That's a 24% increase in throughput by just allowing
> more sectors to transferred in a single
> request. Considering there are other SCSI devices
> using 2048 (and sometimes more) max_sectors, this
> shouldn't bring any surprises to anybody other than
> the extra memory used.
>
> Signed-off-by: Felipe Balbi <balbi@xxxxxx>
> ---
> drivers/usb/storage/scsiglue.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/usb/storage/scsiglue.c b/drivers/usb/storage/scsiglue.c
> index dba51362d2e2..36450111e8b8 100644
> --- a/drivers/usb/storage/scsiglue.c
> +++ b/drivers/usb/storage/scsiglue.c
> @@ -565,8 +565,8 @@ static const struct scsi_host_template usb_stor_host_template = {
> /* lots of sg segments can be handled */
> .sg_tablesize = SCSI_MAX_SG_CHAIN_SEGMENTS,
>
> - /* limit the total size of a transfer to 120 KB */
> - .max_sectors = 240,
> + /* limit the total size of a transfer to 1024 KB */
> + .max_sectors = 2048
>
> /* merge commands... this seems to help performance, but
> * periodically someone should test to see which setting is more
> --
> 2.6.2
>
--
Matthew Dharm
Maintainer, USB Mass Storage driver for Linux
--
To unsubscribe from this list: send the line "unsubscribe linux-usb" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
