On 10/16/24 11:02 AM, Mathieu Poirier wrote:
On Wed, Oct 16, 2024 at 10:37:35AM -0500, Andrew Davis wrote:
On 10/16/24 10:26 AM, Mathieu Poirier wrote:
On Mon, Oct 14, 2024 at 09:56:11AM -0500, Andrew Davis wrote:
On 10/7/24 8:23 AM, Arnd Bergmann wrote:
From: Arnd Bergmann <arnd@xxxxxxxx>
The k3-m4 remoteproc driver was merged with incorrect dependencies.
Despite multiple people trying to fix this, the version 6.12-rc2
remains broken and causes a build failure with CONFIG_TI_SCI_PROTOCOL=m
when the driver is built-in.
arm-linux-gnueabi-ld: drivers/remoteproc/ti_k3_m4_remoteproc.o: in function `k3_m4_rproc_probe':
ti_k3_m4_remoteproc.c:(.text.k3_m4_rproc_probe+0x76): undefined reference to `devm_ti_sci_get_by_phandle'
Fix the dependency again to make it work in all configurations.
The 'select OMAP2PLUS_MBOX' no longer matches what the other drivers
dependencies. The link failure can be avoided with a simple 'depends
do, so turn that into the same 'depends' to ensure we get no circular
on TI_SCI_PROTOCOL', but the extra COMPILE_TEST alternative is what
we use elsehwere. On the other hand, building for OMAP2PLUS makes
no sense since the hardware only exists on K3.
Fixes: ebcf9008a895 ("remoteproc: k3-m4: Add a remoteproc driver for M4F subsystem")
Fixes: ba0c0cb56f22 ("remoteproc: k3-m4: use the proper dependencies")
Fixes: 54595f2807d2 ("mailbox, remoteproc: omap2+: fix compile testing")
Signed-off-by: Arnd Bergmann <arnd@xxxxxxxx>
---
drivers/remoteproc/Kconfig | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/remoteproc/Kconfig b/drivers/remoteproc/Kconfig
index 955e4e38477e..62f8548fb46a 100644
--- a/drivers/remoteproc/Kconfig
+++ b/drivers/remoteproc/Kconfig
@@ -341,9 +341,9 @@ config TI_K3_DSP_REMOTEPROC
config TI_K3_M4_REMOTEPROC
tristate "TI K3 M4 remoteproc support"
- depends on ARCH_OMAP2PLUS || ARCH_K3
- select MAILBOX
- select OMAP2PLUS_MBOX
+ depends on ARCH_K3 || COMPILE_TEST
+ depends on TI_SCI_PROTOCOL || (COMPILE_TEST && TI_SCI_PROTOCOL=n)
This line is odd. IMHO "COMPILE_TEST" should only be added to ARCH_*
dependencies, as often only one ARCH can be selected which prevents
compile testing drivers with various multiple architecture deps in
one compile test.
Normal dependencies, on the other hand, can simply be enabled if one
wants to compile test its dependent drivers. In this case, TI_SCI_PROTOCOL
cannot be enabled as it has a dependency up the chain that doesn't
allow selecting when not on a TI platform. We can fix that as I posted
here[0]. With that fix in, this line can be simply become:
depends on TI_SCI_PROTOCOL
From the above and the follow-on conversation with Nishanth, should I understand
you are working on a patchset to address this issue? If not I will apply Arnd's
patch. People are sending different fix [1] - the issue needs to be addressed
well before the end of the cycle.
This is a bit complex as it touches multiple subsystems. You should take Arnd's
patch. This will fix this for M4, I will then add a follow on patch doing the same
for the other two remoteproc drivers (DSP and R5).
When my series here[0] goes in I will then send a final patch removing the depends
TI_SCI_PROTOCOL=n oddness from all 3 drivers. This final step does not need to happen
this cycle though, only the first two steps are needed to prevent any further compile
test issues.
I have applied Arnd's patch.
Follow up series sent: https://lore.kernel.org/lkml/20241016164141.93401-1-afd@xxxxxx/
Andrew
Andrew
[0] https://lore.kernel.org/lkml/20241015213322.2649011-1-afd@xxxxxx/
[1]. https://lore.kernel.org/linux-arm-kernel/20241016013922.1392290-1-zengheng4@xxxxxxxxxx/T/
Andrew
[0] https://lore.kernel.org/lkml/20241014144821.15094-1-afd@xxxxxx/
+ depends on OMAP2PLUS_MBOX
help
Say m here to support TI's M4 remote processor subsystems
on various TI K3 family of SoCs through the remote processor
