Re: [alobakin:iavf-pp-frag 11/28] net/core/page_pool.c:582:9: sparse: sparse: incorrect type in argument 1 (different address spaces)

Re: [alobakin:iavf-pp-frag 11/28] net/core/page_pool.c:582:9: sparse: sparse: incorrect type in argument 1 (different address spaces)

[Alexander Lobakin]

From: Kernel Test Robot <lkp@intel.com>
Date: Sat, 29 Jul 2023 20:30:10 +0800

> tree:   https://github.com/alobakin/linux iavf-pp-frag
> head:   0eccfd667ed84b425dc0c48eab06b0cb1d5fd8c4
> commit: 67f0169041f76c28d399c2a63947e33a1ae1210d [11/28] page_pool: add a lockdep check for recycling in hardirq
> config: loongarch-randconfig-r072-20230728 (https://download.01.org/0day-ci/archive/20230729/202307292029.cFz0s8Hh-lkp@intel.com/config)
> compiler: loongarch64-linux-gcc (GCC) 12.3.0
> reproduce: (https://download.01.org/0day-ci/archive/20230729/202307292029.cFz0s8Hh-lkp@intel.com/reproduce)
> 
> If you fix the issue in a separate patch/commit (i.e. not just a new version of
> the same patch/commit), kindly add following tags
> | Reported-by: kernel test robot <lkp@intel.com>
> | Closes: https://lore.kernel.org/oe-kbuild-all/202307292029.cFz0s8Hh-lkp@intel.com/

Jakub, could you take a look, is this warning reasonable?

> 
> sparse warnings: (new ones prefixed by >>)
>>> net/core/page_pool.c:582:9: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected void *ptr @@     got unsigned int [noderef] __percpu * @@
>    net/core/page_pool.c:582:9: sparse:     expected void *ptr
>    net/core/page_pool.c:582:9: sparse:     got unsigned int [noderef] __percpu *
>>> net/core/page_pool.c:582:9: sparse: sparse: incorrect type in argument 1 (different address spaces) @@     expected void *ptr @@     got unsigned int [noderef] __percpu * @@
>    net/core/page_pool.c:582:9: sparse:     expected void *ptr
>    net/core/page_pool.c:582:9: sparse:     got unsigned int [noderef] __percpu *
>>> net/core/page_pool.c:582:9: sparse: sparse: incorrect type in argument 1 (different address

[...]

Thanks,
Olek

Re: [alobakin:iavf-pp-frag 11/28] net/core/page_pool.c:582:9: sparse: sparse: incorrect type in argument 1 (different address spaces)

[Jakub Kicinski]

On Tue, 1 Aug 2023 15:46:22 +0200 Alexander Lobakin wrote:
> > If you fix the issue in a separate patch/commit (i.e. not just a new version of
> > the same patch/commit), kindly add following tags
> > | Reported-by: kernel test robot <lkp@intel.com>
> > | Closes: https://lore.kernel.org/oe-kbuild-all/202307292029.cFz0s8Hh-lkp@intel.com/  
> 
> Jakub, could you take a look, is this warning reasonable?

Doesn't look legit to me. I can't repro it on x86 and:

$ head .config
#
# Automatically generated file; DO NOT EDIT.
# Linux/loongarch 6.5.0-rc3 Kernel Configuration
#
CONFIG_CC_VERSION_TEXT="loongarch64-linux-gcc (GCC) 12.3.0"

Adding loongarch to CC, it must be arch specific ?

Re: [alobakin:iavf-pp-frag 11/28] net/core/page_pool.c:582:9: sparse: sparse: incorrect type in argument 1 (different address spaces)

[Alexander Lobakin]

From: Jakub Kicinski <kuba@kernel.org>
Date: Tue, 1 Aug 2023 09:53:21 -0700

> On Tue, 1 Aug 2023 15:46:22 +0200 Alexander Lobakin wrote:
>>> If you fix the issue in a separate patch/commit (i.e. not just a new version of
>>> the same patch/commit), kindly add following tags
>>> | Reported-by: kernel test robot <lkp@intel.com>
>>> | Closes: https://lore.kernel.org/oe-kbuild-all/202307292029.cFz0s8Hh-lkp@intel.com/  
>>
>> Jakub, could you take a look, is this warning reasonable?
> 
> Doesn't look legit to me. I can't repro it on x86 and:
> 
> $ head .config
> #
> # Automatically generated file; DO NOT EDIT.
> # Linux/loongarch 6.5.0-rc3 Kernel Configuration
> #
> CONFIG_CC_VERSION_TEXT="loongarch64-linux-gcc (GCC) 12.3.0"
> 
> Adding loongarch to CC, it must be arch specific ?

Took a look, sounds like something's wrong inside this_cpu_read() on
this architecture (probably with some not-widely-used options enabled),
dunno. Will wait for the architecture folks, there's nothing to fix in
your patch (fully legit and ordinary lockdep expression).

Thanks,
Olek