From: Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
To: "CHIGOT, CLEMENT" <clement.chigot@atos.net>
Cc: libstdc++ <libstdc++@gcc.gnu.org>,
David Edelsohn <dje.gcc@gmail.com>,
Jonathan Wakely <jwakely@redhat.com>,
David Edelsohn via Gcc-patches <gcc-patches@gcc.gnu.org>
Subject: Re: [PATCH] libstdc++: implement locale support for AIX
Date: Fri, 22 Jan 2021 12:04:22 +0100 [thread overview]
Message-ID: <ydd7do58c7d.fsf@CeBiTec.Uni-Bielefeld.DE> (raw)
In-Reply-To: <PA4PR02MB6686A1AE6FA0C0EA009BF81DEAA00@PA4PR02MB6686.eurprd02.prod.outlook.com> (CLEMENT CHIGOT's message of "Fri, 22 Jan 2021 09:57:08 +0000")
Hi Clement,
>> > 3) POSIX 2017 and non-POSIX functions
>> > Many of the *_l functions being used in GNU or dragonfly models aren't
>> > POSIX 2008, but mainly POSIX 2017 or like strtof_l not POSIX at all.
>> > However, there are really useful in the code, thus I've made a double
>> > implementation based on "#ifdef HAVE_". Is it ok for you ? It's not really
>> > POSIX 2008 but more POSIX 2008 with 2017 compatibility.
>> > For the configure, I didn't find any better way to check each syscall, as
>> > they all depend on different includes. Tell me if you have a better idea.
>>
>> First a general observation: there are two groups of functions you're
>> testing for:
>>
>> * Pure BSD additions, not available in either POSIX.1, ISO C, or glibc:
>>
>> localeconv_l
>> mbstowcs_l
>> strtod_l
>> strtof_l
>> strtold_l
>> wcsftime_l
>>
>> * Part of XPG7:
>>
>> iswctype_l
>> strcoll_l
>> strftime_l
>> strxfrm_l
>> towlower_l
>> towupper_l
>> wcscoll_l
>> wcsxfrm_l
>> wctype_l
>>
>> My suggestion would be not to have configure tests _GLIBCXX_HAVE_<FUNC>
>> for any of the second group at all: this is ieee_1003.1-2008, after all,
>> so if some OS selects that clocale variant, it better implement all of
>> those. If really need be, one could a configure check for those and
>> error out if any is missing. This makes the code way more readable than
>> trying to handle some hypothetical partial implementation.
>
> In this case, it would be better to call it ieee_1003.1-2017 but I agree
why? I've just double-checked the OpenGroup pages: all of the functions
listed as XPG7 above were part of IEEE 1003.1-2008, just some of them
have Technical Corrigenda applied. IIUC IEEE 1003.1-2017 is just a
revision of the -2008 standard, not a new issue (XPG8 or something).
> it would be better to avoid all these #ifdef.
> Some are still needed as for example only the last version of AIX have
> strftime_l.
Then that version doesn't conform to XPG7 and shouldn't use that clocale
variant. Until we have a clearer understanding of the variation here,
I'd argue that only P1003.1-2008 conforming OSes should use
ieee_1003.1-2008, rather than creating an impenetrable maze of #ifdefs
for all sorts of partial implementations.
>> As for the BSD group, I suggest to have one representative configure
>> test (for localeconv_l perhaps) and then use an appropriate name for the
>> group as a whole. Again, this will most likely be an all-or-nothing
>> thing.
>
> I'm not sure this is really all-or-nothing for these. Maybe strtof_l and cie
> can be grouped by. But the 3 others are really different. Linux have wcsftime_l
> but not the others. AIX avec none. BSD have all.
TBH, I don't care about Linux here: it will continue to use the gnu
variant anyway. Besides, since the patch will have to work on targets
without wcsftime_l and the other BSD functions, I don't see any harm in
not using one non-standard one of them although it's present.
>> Besides, your configure tests are way too complicated: just use
>> AC_CHECK_FUNCS doing a link test and be done with it.
>
> Sadly, you can't pass includes to AC_CHECK_FUNCS. That's why I had to do
> that. I've made a first version with AC_CHECK_DECLS which allows extra
> headers, but it didn't work too. I might know why though.
Why would you need to? AC_CHECK_FUNCS just perform a link test to check
if the functions are present in libc; no need to have a declaration
present.
>> In a similar vein, configure.ac already has
>> AC_CHECK_HEADERS([xlocale.h]). Rather than hardcoding the existance of
>> the header based on the configure triple, just use the existing
>> HAVE_XLOCALE_H. This ways, things will simply fall into place for
>> e.g. NetBSD, OpenBSD and possibly others.
>
> Right, I'll make the change. Thanks !
>
>> > 4) ctype_configure_char.cc
>> > I've some troubles knowing what is supposed to be implemented on this file.
>> > I don't really understand the part with setlocale which appears in many
>> > os. When I'm adding it, some tests start failing, some start working...
>> > Moreover, on Linux, if I understand correctly, there is some optimizations
>> > based on classic_table(), _M_toupper and _M_tolower. Could you confirm
>> > that it's only useful on Linux ?
>>
>> I don't know myself. However, when trying the first version of your
>> patch (augmented to compile on Solaris), the corresponding change to the
>> solaris file made no difference in test results.
>
> I might have found the correct code since yesterday's mail. The problem seems
> to come from _M_c_locale_ctype initialization. With locale support, it must be
> _S_clone_c_locale(__cloc), without it, it must be the default locale which
> ends up
> being "C". I might push a newer patch this afternoon, with the correct code.
Nice; I'll certainly give it a whirl!
>> > Feel free to try in on other OS. But I've made modifications only for AIX and
>> > Linux, as I can test the other ones.
>>
>> While reading through the patch, I saw that in two places you still use
>> __DragonFly__ || __FreeBSD__ tests. For one, it's hard to tell what
>> feature they are really about, besides they will require fiddling with
>> e.g. for other BSDs. Please use a descriptive macro which says which
>> difference this is about.
>
> Right, because I don't know how to handle them (and I've forgotten to ask
> for it...).
> The first is for typedef __c_locale. It seems to be int* instead of locale_t.
> Could you confirm that this is wanted and mandatory ?
> The second is in about some functions in ctype_members.cc which are
> defined in config/os/../ctype_inlines.h for FreeBSD and Dragonfly. Someone
> has to confirm that it can be merged with the new code, or if this is mandatory.
That someone would most likely be Jonathan ;-)
>> That said, I gave the new patch a try on Solaris 11.4. To get it to
>> compile, I had to apply two changes that I'd mentioned (without an actual
>> patch) when commenting on the first patch:
>>
>> * The C99 fields of struct lconv need _LCONV_C99 to be visible for
>> C++11.
>>
>> * Some ctype macros need __bitmapsize = 15, as the generic clocale
>> implementation uses.
>
> If I'm not mistaking, POSIX is only defining 11 bit for ctype. If we want
> some optimizations we can have a define of bitmasksize or we can simply
> fill the whole mask by setting bitmasksize=15 as in generic.
> I don't know what's best.
AFAIK the constants used in <ctype.h> (or <iso/ctype_iso.h> on Solaris)
to describe the different character classes are just an implementation
detail of the respective OS; certainly none of them are listed on the
OpenGroup page for XPG7 <ctype.h>. So OSes are free to implement this
any way they wish, and the generic variant (and Solaris) show that 11
bits are not enough on some.
Rainer
--
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University
next prev parent reply other threads:[~2021-01-22 11:04 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <PA4PR02MB6686075C6C254E583B72BC2AEAAB0@PA4PR02MB6686.eurprd02.prod.outlook.com>
[not found] ` <CAGWvny=XpcWGnyb=MWg5ziYSND7O1AnQ6-NAX811p1b5urH0YA@mail.gmail.com>
2021-01-11 15:35 ` Rainer Orth
2021-01-11 15:40 ` Jonathan Wakely
2021-01-11 15:56 ` CHIGOT, CLEMENT
2021-01-11 22:20 ` David Edelsohn
2021-01-12 15:14 ` CHIGOT, CLEMENT
2021-01-12 15:23 ` CHIGOT, CLEMENT
2021-01-12 15:25 ` Jonathan Wakely
2021-01-12 15:40 ` CHIGOT, CLEMENT
2021-01-12 15:44 ` David Edelsohn
2021-01-12 17:34 ` Jonathan Wakely
2021-01-12 15:52 ` Rainer Orth
2021-01-12 17:41 ` Rainer Orth
2021-01-12 17:44 ` David Edelsohn
2021-01-12 19:58 ` Rainer Orth
2021-01-13 11:57 ` Rainer Orth
2021-01-13 12:23 ` CHIGOT, CLEMENT
2021-01-13 12:31 ` Rainer Orth
2021-01-13 12:41 ` CHIGOT, CLEMENT
2021-01-13 12:47 ` Rainer Orth
2021-01-21 12:48 ` CHIGOT, CLEMENT
2021-01-21 16:36 ` Rainer Orth
2021-01-22 9:57 ` CHIGOT, CLEMENT
2021-01-22 11:04 ` Rainer Orth [this message]
2021-01-22 11:29 ` Jonathan Wakely
2021-01-22 11:54 ` Rainer Orth
2021-01-22 12:23 ` CHIGOT, CLEMENT
2021-01-27 12:52 ` CHIGOT, CLEMENT
2021-01-27 14:26 ` Rainer Orth
2021-01-27 14:44 ` CHIGOT, CLEMENT
2021-01-28 10:09 ` CHIGOT, CLEMENT
2021-05-17 9:17 ` CHIGOT, CLEMENT
2021-06-08 6:59 ` CHIGOT, CLEMENT
2021-06-09 14:50 ` Rainer Orth
2021-07-21 12:00 ` CHIGOT, CLEMENT
2021-07-21 13:04 ` Rainer Orth
2021-07-22 12:09 ` CHIGOT, CLEMENT
2021-07-22 12:19 ` Rainer Orth
2021-07-30 14:02 ` CHIGOT, CLEMENT
2022-03-16 9:57 ` CHIGOT, CLEMENT
2021-01-22 11:12 ` Jonathan Wakely
2021-01-22 11:02 ` Jonathan Wakely
2021-01-12 16:00 ` Rainer Orth
[not found] ` <PA4PR02MB6686C2022E2B42D82DC9F269EAAB0@PA4PR02MB6686.eurprd02.prod.outlook.com>
2021-01-11 15:38 ` Rainer Orth
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=ydd7do58c7d.fsf@CeBiTec.Uni-Bielefeld.DE \
--to=ro@cebitec.uni-bielefeld.de \
--cc=clement.chigot@atos.net \
--cc=dje.gcc@gmail.com \
--cc=gcc-patches@gcc.gnu.org \
--cc=jwakely@redhat.com \
--cc=libstdc++@gcc.gnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).