-----Original Message----- From: Richard Weinberger <richard@xxxxxx> Sent: Thursday, April 20, 2023 5:18 PM To: Christian Loehle <CLoehle@xxxxxxxxxxxxxx> Cc: Biju Das <biju.das.jz@xxxxxxxxxxxxxx>; linux-mmc <linux-mmc@xxxxxxxxxxxxxxx>; linux-renesas-soc <linux-renesas-soc@xxxxxxxxxxxxxxx>; wsa+renesas <wsa+renesas@xxxxxxxxxxxxxxxxxxxx>; ulf hansson <ulf.hansson@xxxxxxxxxx> Subject: Re: Poor write performance to boot area using rcar-gen3-sdhi > > I did further tracing and on Linux I see clearly that tmio_mmc_irq() always much later when writing to the boot area. > > Another idea, since u-boot does not really support interrupts it polls various status bits to detect completion of commands and data transfers. > Crazy idea: Maybe u-boot writes are faster than they should because it polls not always the right bits? > > ...or Linux is too conservative and it matters for the boot area. > > Thanks, > //richard > If u-boot does not wait until the card no longer signals busy, I would expect data loss. There are probably some more differences if you really want to go down that rabbit hole, like different bus modes for uboot and linux. Anyway I don't really know how I could help you out here, if you could provide a trace showing the busy times, that would be a smoking gun for the eMMC vendor. Regards, Christian Hyperstone GmbH | Reichenaustr. 39a | 78467 Konstanz Managing Director: Dr. Jan Peter Berns. Commercial register of local courts: Freiburg HRB381782
