On Wednesday, September 24, 2014 at 05:31:26 AM, Stephen Warren wrote:
On 09/21/2014 07:13 AM, Marek Vasut wrote:
From: Oleksandr Tymoshenko gonzo@bluezbox.com
This is the USB host controller used on the Altera SoCFPGA and Raspbery Pi.
This code has three checkpatch warnings, but to make sure it stays at least readable and clear, these are not fixed. These bugs are in the USB request handling combinatorial logic, so any abstracting of those is out of question.
Tested on DENX MCV (Altera SoCFPGA 5CSFXC6C6U23C8N) and RPi B+ (BCM2835).
Patches 1, 3, Tested-by: Stephen Warren swarren@nvidia.com
That said, I needed "ARM: rpi_b: query internal MAC address from firmware" to get the MAC address set up correctly. I obviously used that when testing. I guess I should send that patch:-)
Patch 3, Acked-by: Stephen Warren swarren@nvidia.com
Signed-off-by: Oleksandr Tymoshenko gonzo@bluezbox.com Signed-off-by: Stephen Warren swarren@wwwdotorg.org
Given this patch has quite a few changes relative to what Oleksandr or I sent, it might be nice to include some notes in the commit description re: what you changed.
[Marek, fiddled with the frob ...]
For the most part, it's only based on what Oleksander sent. You can probably make a diff between versions, but it will look like a complete rewrite there.
Signed-off-by: Marek Vasut marex@denx.de
README | 3 + drivers/usb/host/Makefile | 3 + drivers/usb/host/dwc2.c | 952 ++++++++++++++++++++++++++++++++++++++++++++++ drivers/usb/host/dwc2.h | 784 ++++++++++++++++++++++++++++++++++++++
In the version Oleksandr sent, there was some header that defined the HW registers that had some problematic Synopsis license. Did dwc2.h come from the Linux kernel? If that is the case, aren't we supposed to say which kernel commit and path it came from?
Yes, the problematic file is dropped completely. The dwc2.h was written from scratch.
Best regards, Marek Vasut