This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [RFC] Forcing 64-bits __OFF_T_TYPE and __INO_T_TYPE for new 32-bit architectures?
- From: Arnd Bergmann <arnd at arndb dot de>
- To: libc-alpha at sourceware dot org
- Cc: "Manuel A. Fernandez Montecelo" <manuel dot montezelo at gmail dot com>, Joseph Myers <joseph at codesourcery dot com>, Zack Weinberg <zackw at panix dot com>
- Date: Mon, 06 Jun 2016 23:46:27 +0200
- Subject: Re: [RFC] Forcing 64-bits __OFF_T_TYPE and __INO_T_TYPE for new 32-bit architectures?
- Authentication-results: sourceware.org; auth=none
- References: <20160606185149 dot GA11555 at reva dot itsari dot org> <alpine dot DEB dot 2 dot 20 dot 1606061958200 dot 10972 at digraph dot polyomino dot org dot uk> <20160606205633 dot GA13715 at reva dot itsari dot org>
On Monday, June 6, 2016 9:56:35 PM CEST Manuel A. Fernandez Montecelo wrote:
>
> Also, it is my understanding that this change will be pushed to cover
> existing 32-bit arches like x86 and arm-32 that will need this in the
> near future (similar to -D_FILE_OFFSET_BITS=64, I suppose). It is
> probably better if support for RISC-V is added when there's a clear
> picture on how to proceed.
>
Right, this is the plan for the kernel. I haven't made much progress here
since last year, but we definitely want to do this the same way for all
32-bit architectures, to avoid having more special cases in the kernel.
This means that for every syscall and ioctl command that passes a 32-bit
__kernel_time_t today, we will add a replacement with a 64-bit
__kernel_time64_t. Only after that is in place, we can have new
architectures that leave out the old interfaces based on __kernel_time_t,
similar to how how we no longer use __kernel_off_t today.
Arnd