On Thu, 8 Nov 2018 13:29:54 -0600 Adam Ford <aford173@xxxxxxxxx> wrote: > On Thu, Nov 8, 2018 at 12:41 PM Andreas Kemnade <andreas@xxxxxxxxxxxx> wrote: > > > > On Thu, 8 Nov 2018 09:06:37 -0800 > > Tony Lindgren <tony@xxxxxxxxxxx> wrote: > > > > > * Andreas Kemnade <andreas@xxxxxxxxxxxx> [181106 21:37]: > > > > Hi Tony, > > > > > > > > On Tue, 6 Nov 2018 12:39:14 -0800 > > > > Tony Lindgren <tony@xxxxxxxxxxx> wrote: > > > > > > > > > * Andreas Kemnade <andreas@xxxxxxxxxxxx> [181106 20:24]: > > > > > > Mainline kernel gives only about 100KByte/s, with sdio irqs I get > > > > > > > 1000KByte/s. > > > > > > > I have been following this thread which had me run some experiments of > my own using a DM3730 and a WiFi chip with the SDIO interface. Can > you tell me what you did to measure this performance difference? I'd > like to run some tests on my hardware to see if the same applies to my > boards too. > well, there are tools like iperf. Or you can simply run nc -lp some_wild_port </dev/zero on one side and nc -n ip_of_other_device some_wild_port | buffer -z 128k >/dev/null on the other. Or you can simply run ssh ip_of_other_device "cat /dev/zero" | buffer -z 128k >/dev/null Note: the last ones are not the most precise ones, so you have to at least wait until values stabilize. The faster you type your password the less you have to wait. Also check the mmc debugfs files to see whether you are using 4bit or 1bit, polled or sdio-irq mode. Regards, Andreas
Attachment:
pgpJwrqyFuN02.pgp
Description: OpenPGP digital signature
