On 11/18/2024 8:43 PM, Johan Hovold wrote:
> On Sat, Nov 16, 2024 at 07:49:23AM -0800, Zijun Hu wrote:
>> For WCN6855, board ID specific NVM needs to be downloaded once board ID
>> is available, but the default NVM is always downloaded currently, and
>> the wrong NVM causes poor RF performance which effects user experience.
>>
>> Fix by downloading board ID specific NVM if board ID is available.
>>
>> Cc: Bjorn Andersson <bjorande@xxxxxxxxxxx>
>> Cc: Aiqun Yu (Maria) <quic_aiquny@xxxxxxxxxxx>
>> Cc: Cheng Jiang <quic_chejiang@xxxxxxxxxxx>
>> Cc: Johan Hovold <johan@xxxxxxxxxx>
>> Cc: Jens Glathe <jens.glathe@xxxxxxxxxxxxxxxxxxxxxx>
>> Cc: Steev Klimaszewski <steev@xxxxxxxx>
>> Cc: Paul Menzel <pmenzel@xxxxxxxxxxxxx>
>
> Nit: These Cc tags should typically not be here in the commit message,
> and should at least not be needed for people who git-send-email will
> already include because of Tested-by and Reviewed-by tags.
>
> If they help with your workflow then perhaps you can just put them below
> the cut-off (---) line.
>
thank you for pointing out this and sharing good suggestions
will follow these suggestions for further patches.
>> Fixes: 095327fede00 ("Bluetooth: hci_qca: Add support for QTI Bluetooth chip wcn6855")
>> Cc: stable@xxxxxxxxxxxxxxx # 6.4
>> Reviewed-by: Johan Hovold <johan+linaro@xxxxxxxxxx>
>
> When making non-trivial changes, like the addition of the fallback NVM
> feature in v2, you should probably have dropped any previous Reviewed-by
> tags.
>
make sense. will notice these aspects for further patches.
> The fallback handling looks good to me though (and also works as
> expected).
>
so, is it okay to make this patch still keep tags given by you ?
>> Tested-by: Johan Hovold <johan+linaro@xxxxxxxxxx>
>> Tested-by: Steev Klimaszewski <steev@xxxxxxxx>
>> Tested-by: Jens Glathe <jens.glathe@xxxxxxxxxxxxxxxxxxxxxx>
>> Signed-off-by: Zijun Hu <quic_zijuhu@xxxxxxxxxxx>
>
>> Changes in v2:
>> - Correct subject and commit message
>> - Temporarily add nvm fallback logic to speed up backport.
>> — Add fix/stable tags as suggested by Luiz and Johan
>> - Link to v1: https://lore.kernel.org/r/20241113-x13s_wcn6855_fix-v1-1-15af0aa2549c@xxxxxxxxxxx
>
>> +download_nvm:
>> err = qca_download_firmware(hdev, &config, soc_type, rom_ver);
>> if (err < 0) {
>> bt_dev_err(hdev, "QCA Failed to download NVM (%d)", err);
>> + if (err == -ENOENT && boardid != 0 &&
>> + soc_type == QCA_WCN6855) {
>> + boardid = 0;
>> + qca_get_hsp_nvm_name_generic(&config, ver,
>> + rom_ver, boardid);
>> + bt_dev_warn(hdev, "QCA fallback to default NVM");
>> + goto download_nvm;
>> + }
>> return err;
>
> If you think it's ok for people to continue using the wrong (default)
> NVM file for a while still until their distros ship the board-specific
> ones, then this looks good to me and should ease the transition:
>
yes. i think it is okay now.
> [ 6.125626] Bluetooth: hci0: QCA Downloading qca/hpnv21g.b8c
> [ 6.126730] bluetooth hci0: Direct firmware load for qca/hpnv21g.b8c failed with error -2
> [ 6.126826] Bluetooth: hci0: QCA Failed to request file: qca/hpnv21g.b8c (-2)
> [ 6.126894] Bluetooth: hci0: QCA Failed to download NVM (-2)
> [ 6.126951] Bluetooth: hci0: QCA fallback to default NVM
> [ 6.127003] Bluetooth: hci0: QCA Downloading qca/hpnv21g.bin
> [ 6.309322] Bluetooth: hci0: QCA setup on UART is completed
>
> Johan
