This is the mail archive of the
libc-alpha@sourceware.org
mailing list for the glibc project.
Re: [patches] Re: [PATCH v4 04/17] Add documentation for __riscv_flush_icache
- From: Joseph Myers <joseph at codesourcery dot com>
- To: Palmer Dabbelt <palmer at dabbelt dot com>
- Cc: <libc-alpha at sourceware dot org>, <patches at groups dot riscv dot org>
- Date: Mon, 22 Jan 2018 21:58:00 +0000
- Subject: Re: [patches] Re: [PATCH v4 04/17] Add documentation for __riscv_flush_icache
- Authentication-results: sourceware.org; auth=none
- References: <mhng-03b563f6-b38d-4368-ad62-7f078a050cc9@palmer-si-x1c4>
On Mon, 22 Jan 2018, Palmer Dabbelt wrote:
> -@deftypefun {void} __riscv_flush_icache(void *start, void *end, unsigned long
> int flags)
> +@deftypefun {void} __riscv_flush_icache(void *@var{__start}, void
> *@var{__end},
> + unsigned long int @var{__flags})
No __ on parameter names in the manual, only in the installed header. And
you can't split the @deftypefun line like that. (Texinfo supports a
syntax with @ at end of line when such an @def* line is split, but I don't
think glibc's own scripts processing .texi files handle that.)
> @safety{@prelim{}@mtsafe{}@assafe{}@acsafe{}}
> -Enforce ordering between stores and instruction cache fetches. The flags
> -variable is used to determine if this applies just to the local thread or to
> -all threads in the system.
> +Enforce ordering between stores and instruction cache fetches. The range of
> +addresses over which ordering is enforced is specified by @var{__start} and
> +@var{__end}. The @var{__flags} variable controls how the ordering is
> enforced,
As I said before I think this should be called an argument not a variable.
> +with @code{SYS_RISCV_FLUSH_ICACHE_LOCAL} specifying that the ordering is
> +enforced only with respect to the local instruction stream (as opposed to all
> +instruction streams).
And what other values are valid and what do they mean? Is 0 the only
other alternative value, meaning ordering over all cores? Are other
values reserved for future use? You seem to describe only one value,
SYS_RISCV_FLUSH_ICACHE_LOCAL, without saying anything about any other
values such as 0.
--
Joseph S. Myers
joseph@codesourcery.com