Hi,
On Fri, Aug 7, 2020 at 1:51 AM kernel test robot <[email protected]> wrote:
>
> Hi Sergio,
>
> First bad commit (maybe != root cause):
>
> tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git
> master
> head: 7b4ea9456dd3f73238408126ab00f1d906963d81
> commit: 3b2fa0c92686562ac0b8cf00c0326a45814f8e18 MIPS: ralink: enable PCI
> support only if driver for mt7621 SoC is selected
> date: 9 months ago
> config: mips-randconfig-r005-20200807 (attached as .config)
> compiler: mipsel-linux-gcc (GCC) 9.3.0
> reproduce (this is a W=1 build):
> wget
> https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O
> ~/bin/make.cross
> chmod +x ~/bin/make.cross
> git checkout 3b2fa0c92686562ac0b8cf00c0326a45814f8e18
> # save the attached .config to linux build tree
> COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-9.3.0 make.cross
> ARCH=mips
>
> If you fix the issue, kindly add following tag as appropriate
> Reported-by: kernel test robot <[email protected]>
>
> All errors (new ones prefixed by >>):
>
> >> drivers/staging/mt7621-pci/pci-mt7621.c:189:11: error:
> >> 'pci_generic_config_read' undeclared here (not in a function)
> 189 | .read = pci_generic_config_read,
> | ^~~~~~~~~~~~~~~~~~~~~~~
> >> drivers/staging/mt7621-pci/pci-mt7621.c:190:12: error:
> >> 'pci_generic_config_write' undeclared here (not in a function)
> 190 | .write = pci_generic_config_write,
> | ^~~~~~~~~~~~~~~~~~~~~~~~
[snip]
PCI is not selected if CONFIG_SOC_MT7621 is not set which is the case
for the attached kernel config... There was a time when the driver
itself was depending on PCI (which for me seems to be the correct
thing to do) but
it was removed creating a regression for gnubee and ralink with pci
based boards. This was done in 'c4d48cf5e2f0 ("MIPS: ralink:
deactivate PCI support for SOC_MT7621")'. To try to fix this issue and
being able again to properly compile the driver I sent the patch is
mentioned here by the test robot in where HAVE_PCI was selected only
in PCI_MT7621 SOC was selected... I was told in the mips kernel list
that this way was not the best thing to do to fix the issue but the
patch was accepted. I also asked about the original compile issue to
the guy who removed first the 'depends PCI' stuff with no answer at
all... So, what should I do to properly fix this?
Thanks in advance for your time.
Best regards,
Sergio Paracuellos
