mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
* [PATCH wireless-next v4 00/10] bcma-ac-rev1: support 4352/4360 rev3 chips
@ 2026-11-09  1:17 Alessio Ferri
  2026-11-09  1:17 ` [PATCH wireless-next v4 01/10] bcma: core: clear only FORCEHT for BCMA_CLKMODE_DYNAMIC Alessio Ferri
                   ` (9 more replies)
  0 siblings, 10 replies; 11+ messages in thread
From: Alessio Ferri @ 2026-11-09  1:17 UTC (permalink / raw)
  To: Rafał Miłecki, Michael Buesch
  Cc: vit9696, ville+git, egc112, linux-wireless, linux-kernel,
	linux-mips, Alessio Ferri, b43-dev

Preparatory work in the bring up of AC family rev1 chips, in particular,
the bcm4352 rev3 and bcm4360 rev3. When public sources were found, the 
same gates applied for operations were ported in this code.

Initialize the PMU, the PCIe2 core and chipcommon of the BCM4352/BCM4360
family on chip based on the open source version of broadcom driver, parse 
SPROM revision 11, read the NVRAM variables set on top of the SPROM, and 
bind bcma-pci-bridge to the BCM4352. Finally, add the missing ids to b43.

Most of the series was tested by Erik Conijin and Ville Takio, both are 
cc-ed, but i don't know if i can put their tested by directly in the 
commit myself or if it is a very suspicious thing to do.

Concerning the risk of applying the series: 
- FORCEHT may change the behaviour of brcmsmac too (can i ask Arend?)
- i do not expect regression from rev11 parsing, especially since no one 
  use it yet.
- The rest of the series has code mostly gated with the same criteria 
  used by public code released by Broadcom, so i guess that it is safe. 
- b43 marks the ac part BROKEN, so there is very little to preserve.

Signed-off-by: Alessio Ferri <alessio.ferri@mythread.it>
---
Changes in v4:
- Readded the missing signed off
- Added pdoffset2g40ma and pdoffset2g40mvalid to the series, work of 
  Ville Takio, mostly.
- Link to v3: https://patch.msgid.link/20261031-bcma-ac-rev1-v3-0-1fbe38e8179b@mythread.it

Changes in v3:
- regated most of the checks on the proper version/flag present
  instead of forcing everything to be rev3 only, as Vitaly suggested
- Added nvram parsing for all params found in a real northstar router
- Removed AvVmid parsing from nvram, as it doesn't seems to be useful
- Added missing pci ids for 4352/4360 variants
- Link to v2: https://patch.msgid.link/20261027-bcma-ac-rev1-v2-0-b69b61c31920@mythread.it

Changes in v2:
- gated chipcommon behaviour on its own revision instead of chip rev as
  pointed out by Vitaly
- Link to v1: https://patch.msgid.link/20261014-bcma-ac-rev1-v1-0-b608e99d3f56@mythread.it

To: Rafał Miłecki <zajec5@gmail.com>
To: Michael Buesch <m@bues.ch>
To: vit9696 <vit9696@protonmail.com>
Cc: linux-wireless@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Cc: linux-mips@vger.kernel.org
Cc: b43-dev@lists.infradead.org

---
Alessio Ferri (10):
      bcma: core: clear only FORCEHT for BCMA_CLKMODE_DYNAMIC
      bcma: pmu: init PLL and resources on BCM4352/BCM4360
      bcma: pcie2: request HT on BCM4352/BCM4360
      bcma: chipcommon: enable asynchronous GPIO interrupts
      bcma: extract SPROM revision 11
      firmware: bcm47xx_sprom: read the SPROM revision 11 fields from NVRAM
      firmware: bcm47xx_sprom: read ledbh4..ledbh15 from NVRAM
      firmware: bcm47xx_sprom: read boardflags3 from NVRAM over the SPROM
      bcma: add the single-band BCM4352/BCM4360 PCI IDs
      wifi: b43: add the 2.4 GHz BCM4352/BCM4360 to the supported bands

 drivers/bcma/core.c                         |   2 +-
 drivers/bcma/driver_chipcommon.c            |   3 +
 drivers/bcma/driver_chipcommon_pmu.c        |  81 +++++++++
 drivers/bcma/driver_pcie2.c                 |   3 +
 drivers/bcma/host_pci.c                     |   8 +-
 drivers/bcma/sprom.c                        | 210 ++++++++++++++++++++++-
 drivers/firmware/broadcom/bcm47xx_sprom.c   | 257 +++++++++++++++++++++++++---
 drivers/net/wireless/broadcom/b43/main.c    |   6 +
 include/linux/bcm47xx_sprom.h               |  15 ++
 include/linux/bcma/bcma_driver_chipcommon.h |  15 ++
 include/linux/ssb/ssb.h                     |  72 ++++++++
 include/linux/ssb/ssb_regs.h                | 103 +++++++++++
 12 files changed, 747 insertions(+), 28 deletions(-)
---
base-commit: 6628e7d4c6b25c8d18f1345e3a20d21829aae7e4
change-id: 20261014-bcma-ac-rev1-da894142790f

Best regards,
--  
Alessio Ferri <alessio.ferri@mythread.it>


^ permalink raw reply	[flat|nested] 11+ messages in thread

end of thread, other threads:[~2026-11-09  1:26 UTC | newest]

Thread overview: 11+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-11-09  1:17 [PATCH wireless-next v4 00/10] bcma-ac-rev1: support 4352/4360 rev3 chips Alessio Ferri
2026-11-09  1:17 ` [PATCH wireless-next v4 01/10] bcma: core: clear only FORCEHT for BCMA_CLKMODE_DYNAMIC Alessio Ferri
2026-11-09  1:17 ` [PATCH wireless-next v4 02/10] bcma: pmu: init PLL and resources on BCM4352/BCM4360 Alessio Ferri
2026-11-09  1:17 ` [PATCH wireless-next v4 03/10] bcma: pcie2: request HT " Alessio Ferri
2026-11-09  1:17 ` [PATCH wireless-next v4 04/10] bcma: chipcommon: enable asynchronous GPIO interrupts Alessio Ferri
2026-11-09  1:17 ` [PATCH wireless-next v4 05/10] bcma: extract SPROM revision 11 Alessio Ferri
2026-11-09  1:17 ` [PATCH wireless-next v4 06/10] firmware: bcm47xx_sprom: read the SPROM revision 11 fields from NVRAM Alessio Ferri
2026-11-09  1:17 ` [PATCH wireless-next v4 07/10] firmware: bcm47xx_sprom: read ledbh4..ledbh15 " Alessio Ferri
2026-11-09  1:17 ` [PATCH wireless-next v4 08/10] firmware: bcm47xx_sprom: read boardflags3 from NVRAM over the SPROM Alessio Ferri
2026-11-09  1:25 ` [PATCH wireless-next v4 09/10] bcma: add the single-band BCM4352/BCM4360 PCI IDs Alessio Ferri
2026-11-09  1:26 ` [PATCH wireless-next v4 10/10] wifi: b43: add the 2.4 GHz BCM4352/BCM4360 to the supported bands Alessio Ferri

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®