On 2/11/20 4:14 PM, Alistair Francis wrote: > On Tue, Feb 11, 2020 at 4:14 PM Vineet Gupta <vineetg76@xxxxxxxxx> wrote: >>> diff --git a/sysdeps/unix/sysv/linux/riscv/bits/typesizes.h b/sysdeps/unix/sysv/linux/riscv/bits/typesizes.h >>> new file mode 100644 >>> index 0000000000..0da3bdeb5d >>> --- /dev/null >>> +++ b/sysdeps/unix/sysv/linux/riscv/bits/typesizes.h >> >> I was hoping newer arches could simply use the asm-generic one ? > > We need to specify that RV32 uses a 64-bit time_t. The generic ones > don't do that for 32-bit arches. Right. It pains to see that each new port (despite using asm-generic syscall ABI) will have to make their own copy of something so fundamental as typesizes.h where only a few things will change. The generic file is no longer generic :-( Oh well ! >>> +/* Tell the libc code that off_t and off64_t are actually the same type >>> + for all ABI purposes, even if possibly expressed as different base types >>> + for C type-checking purposes. */ >>> +# define __OFF_T_MATCHES_OFF64_T 1 >> >> This is orthogonal to time_t but since we are on topic of newer ports, how are you >> doing this. The asm-generic uapi defines >> >> typedef __kernel_long_t __kernel_off_t; >> >> and types.h defines >> >> typedef __kernel_off_t off_t; >> >> And I presume long on RV32 is 32-bits > > Can you point me to the code? Last time I looked the kernel used the > 64-bit versions for the syscalls on RV32. You are right. I got confused with the types above: asm-generic syscalls on 32-bit arches use struct stat64 etc which has 64-bit inode. Similarly sys_llseek uses low/high parts for offset. >>> +/* Same for ino_t and ino64_t. */ >>> +# define __INO_T_MATCHES_INO64_T 1 I'm surprised that ARC port doesn't define this in glibc, yet we use the asm-generic syscall interface where this is true. I need to investigate more. Thx, -vineet _______________________________________________ linux-snps-arc mailing list linux-snps-arc@xxxxxxxxxxxxxxxxxxx http://lists.infradead.org/mailman/listinfo/linux-snps-arc
