Hello,
Nicholas Piggin <npig...@gmail.com> writes: > Christophe Leroy's on May 16, 2019 2:47 pm: >> >> >> Le 16/05/2019 à 04:04, Nicholas Piggin a écrit: >>> Radix boot looks like this: >>> >>> ----------------------------------------------------- >>> phys_mem_size = 0x200000000 >>> dcache_bsize = 0x80 >>> icache_bsize = 0x80 >>> cpu_features = 0x0000c06f8f5fb1a7 >>> possible = 0x0000fbffcf5fb1a7 >>> always = 0x00000003800081a1 >>> cpu_user_features = 0xdc0065c2 0xaee00000 >>> mmu_features = 0xbc006041 >>> firmware_features = 0x0000000010000000 >>> hash-mmu: ppc64_pft_size = 0x0 >>> hash-mmu: kernel vmalloc start = 0xc008000000000000 >>> hash-mmu: kernel IO start = 0xc00a000000000000 >>> hash-mmu: kernel vmemmap start = 0xc00c000000000000 >>> ----------------------------------------------------- >>> >>> Fix: >>> >>> ----------------------------------------------------- >>> phys_mem_size = 0x200000000 >>> dcache_bsize = 0x80 >>> icache_bsize = 0x80 >>> cpu_features = 0x0000c06f8f5fb1a7 >>> possible = 0x0000fbffcf5fb1a7 >>> always = 0x00000003800081a1 >>> cpu_user_features = 0xdc0065c2 0xaee00000 >>> mmu_features = 0xbc006041 >>> firmware_features = 0x0000000010000000 >>> vmalloc start = 0xc008000000000000 >>> IO start = 0xc00a000000000000 >>> vmemmap start = 0xc00c000000000000 >>> ----------------------------------------------------- >>> >>> Signed-off-by: Nicholas Piggin <npig...@gmail.com> >> >> I fear your change defeats most of the purpose of commit >> https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/commit/?h=next-20190515&id=e4dccf9092ab48a6f902003b9558c0e45d0e849a > > I think it's still a significant improvement without introducing > the regression :) > >> As far as I understand, the main issue is the "hash-mmu:" prefix ? >> That's due to the following define in top of book3s64/hash_utils.c: >> >> #define pr_fmt(fmt) "hash-mmu: " fmt >> >> Could we simply undef it just before print_system_hash_info() ? > > Little bit fragile I think. > >> Or move print_system_hash_info() in another book3s64 specific file which >> doesn't set pr_fmt ? > > print_system_info() would be okay for me and allow getting rid of > that PPC64 config. Although it also needs to go in a file without > pr_fmt I guess that's not so hard. Or it could use printk(KERN_INFO "...") instead of pr_info(), which I think makes sense if the pr_fmt prefix is undesired. -- Thiago Jung Bauermann IBM Linux Technology Center
