Hi, Lee Jones <lee.jones@xxxxxxxxxx> writes: >> In Linux, the DWC3 core exists as its own independent platform device. >> Thus when describing relationships in Device Tree, the current default >> boot configuration table option, the DWC3 core often resides as a child >> of the platform specific node. Both of which are given their own >> address space descriptions and the drivers can be mostly agnostic to >> each other. >> >> However, other Operating Systems have taken a more monolithic approach, >> which is evident in the configuration ACPI tables for the Qualcomm >> Snapdragon SDM850, where all DWC3 (core and platform) components are >> described under a single IO memory region. >> >> To ensure successful booting using the supplied ACPI tables, we need to >> devise a way to chop up the address regions provided and subsequently >> register the DWC3 core with the resultant information, which is >> precisely what this patch aims to achieve. >> >> Signed-off-by: Lee Jones <lee.jones@xxxxxxxxxx> >> Reviewed-by: Bjorn Andersson <bjorn.andersson@xxxxxxxxxx> >> --- >> drivers/usb/dwc3/Kconfig | 2 +- >> drivers/usb/dwc3/dwc3-qcom.c | 206 ++++++++++++++++++++++++++++++----- >> 2 files changed, 179 insertions(+), 29 deletions(-) > > I'm starting to get a little twitchy about these patches now. Due to > the release cadence of the larger Linux distros, it's pretty important > that these changes land in v5.3. Without them, it is impossible to > install Linux on some pretty high profile emerging platforms. > > It's already -rc5 and I'm concerned that we're going to miss the > merge-window. Would you be kind enough to review these patches > please? The Pinctrl and I2C parts of the set have already been > merged. I don't seem to have this series in my inbox. This is the only email I have in this series. -- balbi
Attachment:
signature.asc
Description: PGP signature
