[prev in list] [next in list] [prev in thread] [next in thread] 

List:       linux-wireless
Subject:    Re: [PATCH RFC 1/4] brcmfmac: use separate firmware for 43430 revision 4
From:       Stefan Wahren <stefan.wahren () i2se ! com>
Date:       2022-01-31 17:25:11
Message-ID: 17ecdb9e-9b3e-9308-acd6-b1aa7f953002 () i2se ! com
[Download RAW message or body]

Hi Arend,

Am 30.01.22 um 13:56 schrieb Arend Van Spriel:
> I am also a bit confused. The subject mentions revision 4, but your
> log shows 43430/2. Took some digging but found that numerical rev 2
> matches 'b0' and rev 3 matches 'b1'. The BT part is unknown territory
> for me so can not clarify things there.

my patch based on a similar patch [1] by Mikhail Rudenko. I falsely
assumed that the Ampak AP6212 and Raspberry Pi Zero 2 W are using
different wifi chips. But now it i think they are identical. If i grep
the firmware blob for the Raspberry Pi Zero 2 W, i getting this:

# strings firmware.bin | grep 4343
43436b0-roml/sdio-g-pool-p2p-pno-pktfilter-keepalive-aoe-mchan-proptxstatus-ampduhostr \
eorder-lpc-wl11u-fmc-wepso-anqpo-srmem-srfast-11nprop-ob-ndoe-rssiindur-bcol-gtr-mfp-sdiorxenhance-extsae
                
Version: 9.88.4.65 (test) (f149b32@shgit)  (r679549) CRC: 3e0bc4b6 Date:
Fri 2020-10-09 14:45:07 CST Ucode Ver: 1043.20742 FWID: 01-f40f3270

So i will drop this patch from the series and hope the firmware isn't
customized.

Best regards

[1]
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?h=v5.17-rc2&id=f8d6523891cf07468f0bfe21aba8b562a6f75780


> 
> Regards,
> Arend
> 
> On January 30, 2022 12:35:35 PM Stefan Wahren <stefan.wahren@i2se.com>
> wrote:
> 
> > Hi Arend,
> > 
> > Am 24.01.22 um 10:43 schrieb Arend van Spriel:
> > > On 1/22/2022 1:35 PM, Stefan Wahren wrote:
> > > > Am 01.01.22 um 21:26 schrieb Stefan Wahren:
> > > > > A separate firmware is needed, for Broadcom 43430 revision 4. This
> > > > > chip can be found on e.g. certain revisions of Raspberry Pi Zero 2 W.
> > > > > Original firmware file from IC vendor is named
> > > > > 'brcmfmac43436-sdio.bin',
> > > > > but brcmfmac and also btbcm drivers report chip id 43430, so
> > > > > requested
> > > > > firmware file name is 'brcmfmac43430c0-sdio.bin' in line with other
> > > > > 43430 revisions.
> > > > > 
> > > > > Signed-off-by: Stefan Wahren <stefan.wahren@i2se.com>
> > > > > ---
> > > > > 
> > > > > Hi,
> > > > > i'm not sure about all these mappings. All i can say is that the wifi
> > > > > interface of the RPi Zero 2 cames up with this patch.
> > > > gentle ping (yes, i'm aware of the merge window)
> > > 
> > > Sorry, Stefan
> > > 
> > > Should have seen this earlier, but here it is....
> > > 
> > > > > 
> > > > > drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c | 4 +++-
> > > > > 1 file changed, 3 insertions(+), 1 deletion(-)
> > > > > 
> > > > > diff --git a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
> > > > > b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
> > > > > index 8effeb7..c79bd47 100644
> > > > > --- a/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
> > > > > +++ b/drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c
> > > > > @@ -618,6 +618,7 @@ BRCMF_FW_DEF(43430A0, "brcmfmac43430a0-sdio");
> > > > > /* Note the names are not postfixed with a1 for backward
> > > > > compatibility */
> > > > > BRCMF_FW_CLM_DEF(43430A1, "brcmfmac43430-sdio");
> > > > > BRCMF_FW_DEF(43430B0, "brcmfmac43430b0-sdio");
> > > > > +BRCMF_FW_CLM_DEF(43430C0, "brcmfmac43430c0-sdio");
> > > > > BRCMF_FW_CLM_DEF(43455, "brcmfmac43455-sdio");
> > > > > BRCMF_FW_DEF(43456, "brcmfmac43456-sdio");
> > > > > BRCMF_FW_CLM_DEF(4354, "brcmfmac4354-sdio");
> > > > > @@ -649,7 +650,8 @@ static const struct brcmf_firmware_mapping
> > > > > brcmf_sdio_fwnames[] = {
> > > > > BRCMF_FW_ENTRY(BRCM_CC_4339_CHIP_ID, 0xFFFFFFFF, 4339),
> > > > > BRCMF_FW_ENTRY(BRCM_CC_43430_CHIP_ID, 0x00000001, 43430A0),
> > > > > BRCMF_FW_ENTRY(BRCM_CC_43430_CHIP_ID, 0x00000002, 43430A1),
> > > > > -    BRCMF_FW_ENTRY(BRCM_CC_43430_CHIP_ID, 0xFFFFFFFC, 43430B0),
> > > > > +    BRCMF_FW_ENTRY(BRCM_CC_43430_CHIP_ID, 0x00000004, 43430C0),
> > > 
> > > According to the revision mask your firmware seems to be for numerical
> > > revision 2 of this chip...
> > > 
> > > > > +    BRCMF_FW_ENTRY(BRCM_CC_43430_CHIP_ID, 0xFFFFFFF8, 43430B0),
> > > 
> > > and for the b0 the chip revision is 3 (or higher). So the alphanumeric
> > > revision of your chip would be 'a2' instead of 'c0'.
> > 
> > i changed this to a2 for the next version of this series. But before i
> > send them out, please doublecheck the kernel log 5.17-rc1 of my RPi Zero
> > 2 W. I confuses me that the bluetooth part tells about BCM43430B0:
> > 
> > [    9.653199] Bluetooth: hci0: BCM: chip id 115
> > [    9.653749] Bluetooth: hci0: BCM: features 0x0e
> > [    9.655466] Bluetooth: hci0: BCM43430B0
> > [    9.655502] Bluetooth: hci0: BCM43430B0 (002.001.012) build 0000
> > [    9.659512] vc4-drm soc:gpu: bound 3f400000.hvs (ops
> > vc4_drm_unregister [vc4])
> > [    9.667047] Bluetooth: hci0: BCM43430B0 'brcm/BCM43430B0.hcd' Patch
> > [    9.668052] brcmfmac: brcmf_fw_alloc_request: using
> > brcm/brcmfmac43430a2-sdio for chip BCM43430/2
> > [    9.668567] brcmfmac mmc1:0001:1: Direct firmware load for
> > brcm/brcmfmac43430a2-sdio.raspberrypi,model-zero-2-w.bin failed with
> > error -2
> > [    9.673724] vc4-drm soc:gpu: bound 3f400000.hvs (ops
> > vc4_drm_unregister [vc4])
> > [    9.697996] brcmfmac mmc1:0001:1: Direct firmware load for
> > brcm/brcmfmac43430a2-sdio.raspberrypi,model-zero-2-w.txt failed with
> > error -2
> > [    9.816030] brcmfmac: brcmf_fw_alloc_request: using
> > brcm/brcmfmac43430a2-sdio for chip BCM43430/2
> > [    9.840128] brcmfmac: brcmf_c_preinit_dcmds: Firmware: BCM43430/2
> > wl0: Oct  9 2020 14:44:32 version 9.88.4.65 (test) (f149b32@shgit)
> > (r679549) FWID 01-f40f3270
> > [    9.919182] random: crng init done
> > [    9.919200] random: 7 urandom warning(s) missed due to ratelimiting
> > [   10.105524] vc4-drm soc:gpu: bound 3f400000.hvs (ops
> > vc4_drm_unregister [vc4])
> > [   10.118011] vc4-drm soc:gpu: bound 3f902000.hdmi (ops
> > vc4_drm_unregister [vc4])
> > [   10.118190] vc4-drm soc:gpu: bound 3f806000.vec (ops
> > vc4_drm_unregister [vc4])
> > [   10.118364] vc4-drm soc:gpu: bound 3f004000.txp (ops
> > vc4_drm_unregister [vc4])
> > [   10.118503] vc4-drm soc:gpu: bound 3f206000.pixelvalve (ops
> > vc4_drm_unregister [vc4])
> > [   10.118625] vc4-drm soc:gpu: bound 3f207000.pixelvalve (ops
> > vc4_drm_unregister [vc4])
> > [   10.118747] vc4-drm soc:gpu: bound 3f807000.pixelvalve (ops
> > vc4_drm_unregister [vc4])
> > [   10.118847] vc4-drm soc:gpu: bound 3fc00000.v3d (ops
> > vc4_drm_unregister [vc4])
> > [   10.119331] checking generic (1e330000 8ca000) vs hw (0 ffffffff)
> > [   10.119339] fb0: switching to vc4 from simple
> > [   10.122443] Console: switching to colour dummy device 80x30
> > [   10.125147] [drm] Initialized vc4 0.0.0 20140616 for soc:gpu on
> > minor 0
> > [   10.203822] Console: switching to colour frame buffer device 240x75
> > [   10.241067] vc4-drm soc:gpu: [drm] fb0: vc4drmfb frame buffer device
> > [   10.414793] Bluetooth: hci0: BCM4343B0 37.4MHz wlbga_iLNA_iTR
> > [Baseline: 0092]
> > [   10.414822] Bluetooth: hci0: BCM43430B0 (002.001.012) build 0092
> > 
> > Thanks
> 
> 
> 


[prev in list] [next in list] [prev in thread] [next in thread]