<marex>
Forty-Bot: btw I still don't see it, 4e64cd7763ca5 ("usb: dwc3: fix backwards compat with rockchip devices") refers to 33fb697ec7e5 ("usb: dwc3: Get clocks individually") , which seems unrelated to GFLADJ programming
<marex>
I understand the former part, but not how it is related to GFLADJ programming
<Forty-Bot>
it was a prerequisite to the latter patch for Linux since all the clocks were gotten in bulk
<marex>
ah
<Forty-Bot>
although I suppose you could skip it and just continue using the period property
<Forty-Bot>
I'm not familiar with the current state of the U-Boot driver
<marex>
Forty-Bot: outdated
<Forty-Bot>
I figured
<marex>
I like the positive attitude ;-)
<Forty-Bot>
anyway, I just wanted to point it out because you wouldn't know about it just from looking at the lore link
<marex>
Forty-Bot: yeah, that was inobvious, thanks
<marex>
Forty-Bot: 5114c3ee24875 ("usb: dwc3: Calculate REFCLKPER based on reference clock")
<marex>
this one adds the dependency, now I see it
rvalue- has quit [Remote host closed the connection]
rvalue has joined #u-boot
xypron has quit [Quit: xypron]
xypron has joined #u-boot
mmu_man has joined #u-boot
camus1 has joined #u-boot
camus has quit [Ping timeout: 260 seconds]
camus1 is now known as camus
vagrantc has joined #u-boot
vagrantc has quit [Quit: leaving]
ikarso has joined #u-boot
mmu_man has quit [Ping timeout: 252 seconds]
mmu_man has joined #u-boot
___nick___ has quit [Ping timeout: 246 seconds]
<marex>
Forty-Bot: all right , patches are out, including the clock part
<Forty-Bot>
hm, IMO we shouldn't add support for snps,ref-clock-period-ns if it didn't already exist
<marex>
Forty-Bot: linux does support it though
<Forty-Bot>
yeah, but it's deprecated
<Forty-Bot>
and U-Boot didn't use it before
<Forty-Bot>
no point in adding something which shouldn't be used
<marex>
Forty-Bot: it's a property documented in the bindings for dwc3, so I would argue we can very well easily support it
<marex>
DT review should trap any users who might try to use it
<Forty-Bot>
I don't know about that
<Forty-Bot>
I have seen many crusty boards in U-Boot
<marex>
Forty-Bot: I want to believe the U-Boot and Linux DTs will mostly converge eventually
<marex>
and dtbs_check should already warn on deprecated props, no ?
<Forty-Bot>
do people run that on their U-Boot trees?
<marex>
Forty-Bot: I do run it on Linux DTs and then sync them to U-Boot
<Forty-Bot>
neat
<marex>
it is not always possible tho
<marex>
imx ... still is buggy
<marex>
stm32mp15xx ... damn, clean as a whistle
<Forty-Bot>
I've run it on some upstream trees I've used and gotten screenfuls of errors
<Forty-Bot>
mostly because dtspect insists that every property be documented, and every ordering of compatibles too
<marex>
Forty-Bot: yep, I did send a few outstanding patches for the mp1 in that direction
<marex>
it just takes some effort
<marex>
oh well
<marex>
but the benefit is clear
<Forty-Bot>
personally, I only have so much time to spend cleaning up after vendors...
* marex
holds back on graphic comments
<marex>
:)
prabhakarlad has joined #u-boot
sbach has quit [Read error: Connection reset by peer]
sbach has joined #u-boot
sicelo has quit [Quit: Bye!]
sicelo has joined #u-boot
<hanetzer>
so question. when booting a fit image from u-boot, how does it know which of the included dtbs to use?
<Forty-Bot>
it's specified by the config
<marex>
board_fit_config_name_match
<hanetzer>
hrm. and what bout devices that lack that function defined? i'm not seeing it for, for example, rk3288-veyron-speedy
<marex>
holy s..t ... this DWC3 is real something
<marex>
so like when you send a FS device get descriptor request, the first one, 64 Byte , followed by another get descriptor request in less than 125us, the device fails to respond
<hanetzer>
yeh, their mmc controller is annoying me since they require two clocks but the vendorware version of it uses one clock and there's no documented secondary clocks that they may just not have wired in
<marex>
hanetzer: doesnt socfpga also use that controller with one clock ?
<marex>
hmmm, I guess a USB port with 128 FS keyboards plugged into it is a valid usecase
<Forty-Bot>
hanetzer: they fall back to the default config
<Forty-Bot>
you can also specify a config to bootm manually
ikarso has quit [Quit: Connection closed for inactivity]