2018-03-13 12:14:01

by Arnd Bergmann

[permalink] [raw]
Subject: [PATCH] mailbox: tegra: relax TEGRA_HSP_MBOX Kconfig dependencies

With the addition of the ARCH_TEGRA_194_SOC driver, we get a new Kconfig warning:

warning: (ARCH_TEGRA_186_SOC && ARCH_TEGRA_194_SOC) selects TEGRA_HSP_MBOX which has unmet direct dependencies (MAILBOX && ARCH_TEGRA_186_SOC)

It looks like the dependency is a bit too strict here, allowing the driver to
be built for any Tegra chip avoids the problem.

Fixes: 6f9ed07fde03 ("soc/tegra: Add Tegra194 SoC configuration option")
Signed-off-by: Arnd Bergmann <[email protected]>
---
drivers/mailbox/Kconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/mailbox/Kconfig b/drivers/mailbox/Kconfig
index ba2f1525f4ee..7c24900554a7 100644
--- a/drivers/mailbox/Kconfig
+++ b/drivers/mailbox/Kconfig
@@ -134,7 +134,7 @@ config QCOM_APCS_IPC

config TEGRA_HSP_MBOX
bool "Tegra HSP (Hardware Synchronization Primitives) Driver"
- depends on ARCH_TEGRA_186_SOC
+ depends on ARCH_TEGRA
help
The Tegra HSP driver is used for the interprocessor communication
between different remote processors and host processors on Tegra186
--
2.9.0



2018-03-13 13:38:47

by Thierry Reding

[permalink] [raw]
Subject: Re: [PATCH] mailbox: tegra: relax TEGRA_HSP_MBOX Kconfig dependencies

On Tue, Mar 13, 2018 at 01:11:43PM +0100, Arnd Bergmann wrote:
> With the addition of the ARCH_TEGRA_194_SOC driver, we get a new Kconfig warning:
>
> warning: (ARCH_TEGRA_186_SOC && ARCH_TEGRA_194_SOC) selects TEGRA_HSP_MBOX which has unmet direct dependencies (MAILBOX && ARCH_TEGRA_186_SOC)
>
> It looks like the dependency is a bit too strict here, allowing the driver to
> be built for any Tegra chip avoids the problem.
>
> Fixes: 6f9ed07fde03 ("soc/tegra: Add Tegra194 SoC configuration option")
> Signed-off-by: Arnd Bergmann <[email protected]>
> ---
> drivers/mailbox/Kconfig | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)

The HSP does not exist on any chip prior to Tegra186, which is what the
dependency reflects. However, I agree that there is little point for an
exact list of per-chip dependencies. We don't really support non-multi-
platform kernels anymore, or at least they don't get any testing, so
relaxing the dependency is fine with me.

Acked-by: Thierry Reding <[email protected]>

Thanks for fixing this,
Thierry


Attachments:
(No filename) (1.05 kB)
signature.asc (849.00 B)
Download all attachments