[v4,0/10] Bluetooth: Add RTL8723BS support

Message

Hans de Goede Aug. 2, 2018, 2:57 p.m. UTC

Hi All,

Here is v4 of my series of patches adding support for RTL8723BS'
bluetooth component through serdev enumeration.

This builds on the great work done on this by Martin Blumenstingl
which makes the btrtl.c code also support firmware+config loading
for the RTL8723BS, as well as Jeremy's Cline work to add ACPI
enumeration support on top of this.

New in v4:
-Add new rtl_dev_err, rtl_dev_info, etc. helper macros and use those
 instead of bt_dev_err, etc
-Add a new patch replacing all existing BT_ERR, bt_dev_err, etc. calls
 in btrtl.c with the new helpers

New in v3:
-Fix build error when CONFIG_RTL_BT is not set, reported by Paul Cercueil

New in v2:
-Dropped devicetree specific patches
 (these can be reposted later by someone who can actually test them)
-Add the option to pass a postfix for the config firmware filename,
 to btrtl_initialize() (new patch). For ACPI serdev enumerated
 devies this will be set to the ACPI HID, so the filename becomes e.g.
 rtl_bt/rtl8732bs_config-OBDA8723.bin
 This is done because the config is board specific and we may need a
 different file for e.g. ARM boards

I've tested this series on several Intel boards with a RTL8723BS chip.

Regards,

Hans

--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Comments

Marcel Holtmann Aug. 3, 2018, 12:11 p.m. UTC | #1

Hi Hans,

> Here is v4 of my series of patches adding support for RTL8723BS'
> bluetooth component through serdev enumeration.
> 
> This builds on the great work done on this by Martin Blumenstingl
> which makes the btrtl.c code also support firmware+config loading
> for the RTL8723BS, as well as Jeremy's Cline work to add ACPI
> enumeration support on top of this.
> 
> New in v4:
> -Add new rtl_dev_err, rtl_dev_info, etc. helper macros and use those
> instead of bt_dev_err, etc
> -Add a new patch replacing all existing BT_ERR, bt_dev_err, etc. calls
> in btrtl.c with the new helpers
> 
> New in v3:
> -Fix build error when CONFIG_RTL_BT is not set, reported by Paul Cercueil
> 
> New in v2:
> -Dropped devicetree specific patches
> (these can be reposted later by someone who can actually test them)
> -Add the option to pass a postfix for the config firmware filename,
> to btrtl_initialize() (new patch). For ACPI serdev enumerated
> devies this will be set to the ACPI HID, so the filename becomes e.g.
> rtl_bt/rtl8732bs_config-OBDA8723.bin
> This is done because the config is board specific and we may need a
> different file for e.g. ARM boards
> 
> I've tested this series on several Intel boards with a RTL8723BS chip.

so I applied all 10 patches to bluetooth-next tree.

However I am bit disappointed that we couldn’t get bt3wire.c fixed to be used for this hardware. Having to hack this into hci_h5.c is a step back in my book. And I know what happens now. Nobody is going to work on it and we keep adding hacks and hacks on top of hci_ldisc.c that was never really designed for this. Maybe the next time around I have to be hardline and just reject unclean driver support.

In addition I am not happy with the -postfix firmware naming support. I think that is dangerous and might get us into a bad spot. Assuming anything about vendor firmware names and policies is bad. And from what I have seen with Realtek drivers, they are not consistent and are really hacking things together to make it just work for a single design. I really hope this is not coming back to haunt us.

Regards

Marcel

--
To unsubscribe from this list: send the line "unsubscribe linux-acpi" in
the body of a message to [e-mail]
More majordomo info at  http://vger.kernel.org/majordomo-info.html