mbox series

[0/2] hw/arm: Unconditionally map MMIO-based USB host controllers

Message ID 20240115165615.78323-1-philmd@linaro.org
Headers show
Series hw/arm: Unconditionally map MMIO-based USB host controllers | expand

Message

Philippe Mathieu-Daudé Jan. 15, 2024, 4:56 p.m. UTC 
When a chipset contain a USB controller, we can not simply
remove it. We could disable it, but that requires more changes
this series isn't aiming for. For more context:
https://lore.kernel.org/qemu-devel/56fde49f-7dc6-4f8e-9bbf-0336a20a9ebf@roeck-us.net/

Philippe Mathieu-Daudé (2):
  hw/arm/allwinner-a10: Unconditionally map the USB Host controllers
  hw/arm/nseries: Unconditionally map the TUSB6010 USB Host controller

 hw/arm/allwinner-a10.c | 49 +++++++++++++++++-------------------------
 hw/arm/nseries.c       |  4 +---
 2 files changed, 21 insertions(+), 32 deletions(-)

Comments

Peter Maydell Jan. 15, 2024, 5 p.m. UTC | #1 
On Mon, 15 Jan 2024 at 16:56, Philippe Mathieu-Daudé <philmd@linaro.org> wrote:
>
> When a chipset contain a USB controller, we can not simply
> remove it. We could disable it, but that requires more changes
> this series isn't aiming for. For more context:
> https://lore.kernel.org/qemu-devel/56fde49f-7dc6-4f8e-9bbf-0336a20a9ebf@roeck-us.net/
>
> Philippe Mathieu-Daudé (2):
>   hw/arm/allwinner-a10: Unconditionally map the USB Host controllers
>   hw/arm/nseries: Unconditionally map the TUSB6010 USB Host controller

Yes, I think this makes sense. It is technically a migration
compat break for the affected machines if the user was
passing the '-usb none' option, but I think we're OK with
that; worth mentioning in the commit messages I guess.

-- PMM
Peter Maydell Jan. 19, 2024, 4:51 p.m. UTC | #2 
On Mon, 15 Jan 2024 at 16:56, Philippe Mathieu-Daudé <philmd@linaro.org> wrote:
>
> When a chipset contain a USB controller, we can not simply
> remove it. We could disable it, but that requires more changes
> this series isn't aiming for. For more context:
> https://lore.kernel.org/qemu-devel/56fde49f-7dc6-4f8e-9bbf-0336a20a9ebf@roeck-us.net/
>
> Philippe Mathieu-Daudé (2):
>   hw/arm/allwinner-a10: Unconditionally map the USB Host controllers
>   hw/arm/nseries: Unconditionally map the TUSB6010 USB Host controller
>



Applied to target-arm.next, thanks.

-- PMM
Philippe Mathieu-Daudé Jan. 19, 2024, 10:06 p.m. UTC | #3 
On 19/1/24 17:51, Peter Maydell wrote:
> On Mon, 15 Jan 2024 at 16:56, Philippe Mathieu-Daudé <philmd@linaro.org> wrote:
>>
>> When a chipset contain a USB controller, we can not simply
>> remove it. We could disable it, but that requires more changes
>> this series isn't aiming for. For more context:
>> https://lore.kernel.org/qemu-devel/56fde49f-7dc6-4f8e-9bbf-0336a20a9ebf@roeck-us.net/
>>
>> Philippe Mathieu-Daudé (2):
>>    hw/arm/allwinner-a10: Unconditionally map the USB Host controllers
>>    hw/arm/nseries: Unconditionally map the TUSB6010 USB Host controller
>>
> 
> 
> 
> Applied to target-arm.next, thanks.

Thanks! I posted a v2 with the requested changes (mention
migration compat break):
https://lore.kernel.org/qemu-devel/20240119215106.45776-1-philmd@linaro.org/