[PATCH v6 0/5] PCI: add 4x lane support for pci-j721e controllers

Matt Ranostay mranostay at ti.com
Tue Nov 22 21:51:58 PST 2022


On Tue, Nov 15, 2022 at 09:37:35AM -0600, Bjorn Helgaas wrote:
> On Tue, Nov 15, 2022 at 07:03:30AM -0800, Matt Ranostay wrote:
> > Adding of dditional support to Cadence PCIe controller (i.e. pci-j721e.c)
> > for up to 4x lanes, and reworking of driver to define maximum lanes per
> > board configuration.
> > 
> > Changes from v1:
> > * Reworked 'PCI: j721e: Add PCIe 4x lane selection support' to not cause
> >   regressions on 1-2x lane platforms
> > 
> > Changes from v2:
> > * Correct dev_warn format string from %d to %u since lane count is a unsigned
> >   integer
> > * Update CC list
> > 
> > Changes from v3:
> > * Use the max_lanes setting per chip for the mask size required since bootloader
> >   could have set num_lanes to a higher value that the device tree which would leave
> >   in an undefined state
> > * Reorder patches do the previous change to not break bisect
> > * Remove line breaking for dev_warn to allow better grepping and since no strict
> >   80 columns anymore
> > 
> > Changes from v4:
> > * Correct invalid settings for j7200 PCIe RC + EP
> > * Add j784s4 configuration for selection of 4x lanes
> > 
> > Changes from v5:
> > * Dropped 'PCI: j721e: Add warnings on num-lanes misconfiguration' patch from series  
> > * Reworded 'PCI: j721e: Add per platform maximum lane settings' commit message
> > * Added yaml documentation and schema checks for ti,j721e-pci-* lane checking
> > 
> > Matt Ranostay (5):
> >   dt-bindings: PCI: ti,j721e-pci-*: add checks for num-lanes
> >   PCI: j721e: Add per platform maximum lane settings
> >   PCI: j721e: Add PCIe 4x lane selection support
> >   dt-bindings: PCI: ti,j721e-pci-*: add j784s4-pci-* compatible strings
> >   PCI: j721e: add j784s4 PCIe configuration
> 
> Hi Matt,
> 
> Don't repost just for this, but if you have occasion to post this
> again, capitalize this subject line to match the others, i.e.,
> "Add j784s4 configuration".
> 
> Also looks like some commit logs are wrapped at about 65 columns; it's
> nice if they're consistently 75.

Noted... I know this rule, and someone missed wrapping it 75 columns :-/

- Matt



More information about the linux-arm-kernel mailing list