On Tue, Jul 11, 2017 at 3:07 PM, kbuild test robot lkp@intel.com wrote:
make ARCH=i386
Yeah, either this code shouldn't have been built on 32-bit arch at all, or be portable.
arch/x86/pci/fixup.c: In function 'pci_amd_enable_64bit_bar':
arch/x86/pci/fixup.c:674:15: warning: large integer implicitly truncated to unsigned type [-Woverflow]
res->start = 0x100000000ull; ^~~~~~~~~~~~~~arch/x86/pci/fixup.c:675:13: warning: large integer implicitly truncated to unsigned type [-Woverflow] res->end = 0xfd00000000ull - 1; ^~~~~~~~~~~~~~~
arch/x86/pci/fixup.c:686:22: warning: right shift count >= width of type [-Wshift-count-overflow]
I suppose explicit casting will help.
high = ((res->start >> 40) & AMD_141b_MMIO_HIGH_MMIOBASE_MASK) | ^~arch/x86/pci/fixup.c:687:21: warning: right shift count >= width of type [-Wshift-count-overflow] ((((res->end + 1) >> 40) << AMD_141b_MMIO_HIGH_MMIOLIMIT_SHIFT) ^~
These can be done via upper_32_bits() + shift.