no sooner did I abandon it than two different companies asked me not to.
10 comments:
Patch Set #8, Line 25: keyword to sconfig
Strange line break. […]
it's intentional
Patch Set #8, Line 339: bool "Enable a configurable ramstage."
Please remove the dot at the end.
Done
The minimal PCI scanning will only check those parts that are enabled
in the devicetree.cb. By convention none of those devices should be bridges.
Move this to the description of the option below?
I just blew it away, I did not see a need for this text.
Patch Set #4, Line 1195: dev = dev_find_slot(0, devfn);
Sigh.. dev_find_slot() should be banned. Would following work for you: […]
Done
Patch Set #4, Line 1197: printk(BIOS_WARNING, "%#x is NOT static, skipping it\n", devfn);
We prefer B:D.f notation for PCI devices on console output.
I blew away the print, it's a time waster.
Patch Set #8, Line 1191: dev = pcidev_path_behind(bus, devfn);
I did not see this on first review, but this probably evaluates to same 'dev' as pci_scan_get_dev() […]
yes it should. And it will. What we're showing here is the minimal set of changes to make this work at all, b/c every pass we've taken at this has been bikeshedded into oblivion.
Patch Set #8, Line 1229: *prev = dev->sibling;
It is pci_probe_dev() that filled vendor field. […]
yes we are.
File src/include/device/device.h:
Remove for concise multi-line comments.
Done
File util/sconfig/sconfig.tab.c_shipped:
Patch Set #8, Line 112: #ifndef YY_YY_HOME_RMINNICH_PROJECTS_LINUXBOOT_COREBOOTNERF_GITHUBCOREBOOT_UTIL_SCONFIG_SCONFIG_TAB_H_SHIPPED_INCLUDED
Is this wanted?
this is a generated file. Not much to be done I think.
File util/sconfig/sconfig.tab.h_shipped:
Patch Set #8, Line 33: #ifndef YY_YY_HOME_RMINNICH_PROJECTS_LINUXBOOT_COREBOOTNERF_GITHUBCOREBOOT_UTIL_SCONFIG_SCONFIG_TAB_H_SHIPPED_INCLUDED
Ditto.
ditto :-)
To view, visit change 36221. To unsubscribe, or for help writing mail filters, visit settings.