On 23/12/25 18:27, Sourabh Jain wrote:

+Cc: PowerPC mailing list.

and added some comments below.

On 22/12/25 12:14, kernel test robot wrote:

quoted

tree: 
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 
master
head:   9448598b22c50c8a5bb77a9103e2d49f134c9578
commit: 38c64dfe0af12778953846df5f259e913275cfe5 kexec: Include 
kernel-end even without crashkernel
date:   6 weeks ago
config: powerpc-randconfig-r111-20251222 
(https://download.01.org/0day-ci/archive/20251222/202512221405.VHPKPjnp-lkp@intel.com/config)
compiler: clang version 22.0.0git 
(https://github.com/llvm/llvm-project 
42b3483ac4987cae1bdb632398e8a3ce2dea6633)
reproduce (this is a W=1 build): 
(https://download.01.org/0day-ci/archive/20251222/202512221405.VHPKPjnp-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 [off-list ref]
| Closes: 
https://lore.kernel.org/oe-kbuild-all/202512221405.VHPKPjnp-lkp@intel.com/ (local)

sparse warnings: (new ones prefixed by >>)

quoted

quoted

arch/powerpc/kexec/core.c:217:20: sparse: sparse: incorrect type in 
assignment (different base types) @@     expected unsigned int 
static [addressable] [toplevel] [usertype] kernel_end @@     got 
restricted __be32 [usertype] @@

    arch/powerpc/kexec/core.c:217:20: sparse:     expected unsigned 
int static [addressable] [toplevel] [usertype] kernel_end
    arch/powerpc/kexec/core.c:217:20: sparse:     got restricted 
__be32 [usertype]

If CONFIG_CRASH_DUMP is enabled (which is not done in config shared by 
kernel test robot) sparse warn
about other assignments in the same file. And these warnings are not 
limited to 32-bit.

../arch/powerpc/kexec/core.c:181:29: sparse: warning: incorrect type 
in assignment (different base types)
../arch/powerpc/kexec/core.c:181:29: sparse:    expected unsigned int 
static [addressable] [toplevel] [usertype] crashk_base
../arch/powerpc/kexec/core.c:181:29: sparse:    got restricted __be32 
[usertype]
../arch/powerpc/kexec/core.c:183:29: sparse: warning: incorrect type 
in assignment (different base types)
../arch/powerpc/kexec/core.c:183:29: sparse:    expected unsigned int 
static [addressable] [toplevel] [usertype] crashk_size
../arch/powerpc/kexec/core.c:183:29: sparse:    got restricted __be32 
[usertype]
../arch/powerpc/kexec/core.c:191:19: sparse: warning: incorrect type 
in assignment (different base types)
../arch/powerpc/kexec/core.c:191:19: sparse:    expected unsigned long 
long static [addressable] [toplevel] mem_limit
../arch/powerpc/kexec/core.c:191:19: sparse:    got restricted __be32 
[usertype]
../arch/powerpc/kexec/core.c:217:20: sparse: warning: incorrect type 
in assignment (different base types)
../arch/powerpc/kexec/core.c:217:20: sparse:    expected unsigned int 
static [addressable] [toplevel] [usertype] kernel_end
../arch/powerpc/kexec/core.c:217:20: sparse:    got restricted __be32 
[usertype]
  CHECK   ../arch/powerpc/kexec/core_32.c

To resolve these warnings:

Should we use __force tpyecast and silence the warnings?
for example:
kernel_end = (__force phys_addr_t) cpu_to_be_ulong(__pa(_end));

Or

Use proper data types __be<32|64> and duplicated the definition and 
assignment? seems like bit too much.

Found a better way to define variables as __be<32|64> without duplication,
so I have posted a fix:
https://lore.kernel.org/all/20251224151257.28672-1-sourabhjain@linux.ibm.com/ (local)

- Sourabh Jain

Or

Do we have better way to handle these warnings? If yes, please share.

- Sourabh Jain

quoted

vim +217 arch/powerpc/kexec/core.c

6f29c3298b1821 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-17  203
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  204  static int __init kexec_setup(void)
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  205  {
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  206      struct device_node *node;
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  207
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  208      node = of_find_node_by_path("/chosen");
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  209      if (!node)
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  210          return -ENOENT;
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  211
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  212      /* remove any stale properties so ours can be 
found */
38c64dfe0af127 arch/powerpc/kexec/core.c           Ben Collins     
2025-04-21  213      of_remove_property(node, of_find_property(node, 
kernel_end_prop.name,
38c64dfe0af127 arch/powerpc/kexec/core.c           Ben Collins     
2025-04-21  214                            NULL));
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  215
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  216      /* information needed by userspace when using 
default_machine_kexec */
ea961a828fe725 arch/powerpc/kernel/machine_kexec.c Anton Blanchard 
2014-01-22 @217      kernel_end = cpu_to_be_ulong(__pa(_end));
79d1c712958f94 arch/powerpc/kernel/machine_kexec.c Nathan Fontenot 
2012-10-02  218      of_add_property(node, &kernel_end_prop);
2e8e4f5b80e101 arch/powerpc/kernel/machine_kexec.c Dale Farnsworth 
2008-12-16  219

:::::: The code at line 217 was first introduced by commit
:::::: ea961a828fe7250e954f086d74d9323c3d44c3e4 powerpc: Fix endian 
issues in kexec and crash dump code

:::::: TO: Anton Blanchard [off-list ref]
:::::: CC: Benjamin Herrenschmidt [off-list ref]