fork of 9front i guess
Find a file
kemal c5c79d61e6 etheriwl: fixes for 7260 ac
2021-08-14 17:50 GMT, kemal <kemalinanc8@gmail.com>:
> 1- as driver reads 8 bytes from nvm instead of 6 so fw doesn't
> spit us an ADVANCED_SYSASSERT, it was reading 2 more
> extra bytes. apparently those 2 extra bytes were put to
> the first 2 bytes of our buffer, so we got to skip that.

some more thoughts on this, i think as 0x15*2 is not multiple
of 8, fw rounds the offset to 0x14*2. i have touched to code
to read data from 0x14*2 then ignore the first 2 bytes, just
so it's not confusing. if this causes mac to be read wrong again,
report.

also, some more changes:

1. set the fwname at iwlpci, just to align the behavior with 8000+.
this is a cosmetic change.

2. i have discovered that on device boot/reset/shutdown functions,
our driver slept way much more than it should. the reason for that is,
driver used the function delay() on places where it needs to use
microdelay() instead. i have modified the code to use microdelay().
wpi likely needs similar changes too. i hope that this does not
break the code.

3. zzz a bit more on tx/rx scheduler shutdowns and niclock.

4. openbsd's iwm and linux apparently does not check if ownership
was obtained anymore in their handover functions. instead they
just loop until the hw is ready. aligned the behavior.
see linux commit: 289e5501c3141191dd830957f1d764d3dc14a54f

5. don't take antenna masks from nvm. it's apparently empty
in some cards from 7k family. we will rely on what the fw file gives
us.

6. when the calibration is completed, wakeup the proc that runs
postboot. otherwise that thing sleeps for like 2 whole seconds
even if calibration completed earlier.

i honestly don't think any of these changes will fix 7260 not
being able to get calibration results, but i don't see anything
wrong at all in postboot7000 at this point. i will just hope
these changes somehow make it get calibration results.

NOTE: latest patch on the 9front ml, posted Mon, 14 Feb 2022 15:26:55 +0300
(non functional as of yet)
2021-08-21 22:23:12 +00:00
386 ape: sync flaot.h macros with u.h 2021-02-08 15:45:11 -08:00
68000 add signed fixed size integer typedefs 2018-05-12 19:19:52 +02:00
68020 ape: sync flaot.h macros with u.h 2021-02-08 15:45:11 -08:00
acme win: fix window recreation command 2020-09-16 12:41:24 +02:00
adm/timezone timezones: change DST timezone string for South Australia 2019-11-20 11:49:16 +10:30
amd64 libc: change usize to 64-bit for amd64 and arm64, make memory(2) functions use usize 2021-07-25 15:54:22 +00:00
arm ape: sync flaot.h macros with u.h 2021-02-08 15:45:11 -08:00
arm64 libc: change usize to 64-bit for amd64 and arm64, make memory(2) functions use usize 2021-07-25 15:54:22 +00:00
lib /lib/theo: You obviously don't know what you are doing, and I think you don't deserve help. 2022-01-23 03:08:35 +00:00
mips ape: sync flaot.h macros with u.h 2021-02-08 15:45:11 -08:00
power ape: sync flaot.h macros with u.h 2021-02-08 15:45:11 -08:00
power64 add signed fixed size integer typedefs 2018-05-12 19:19:52 +02:00
rc nusbrc: fix FTDI detection 2022-02-06 00:11:49 +00:00
sparc ape: sync flaot.h macros with u.h 2021-02-08 15:45:11 -08:00
sparc64 ape: sync flaot.h macros with u.h 2021-02-08 15:45:11 -08:00
spim add signed fixed size integer typedefs 2018-05-12 19:19:52 +02:00
sys etheriwl: fixes for 7260 ac 2021-08-21 22:23:12 +00:00