From: Lukasz Majewski <lukma@denx.de>
To: Alistair Francis <alistair23@gmail.com>
Cc: Joseph Myers <joseph@codesourcery.com>,
Zack Weinberg <zackw@panix.com>, Arnd Bergmann <arnd@arndb.de>,
Alistair Francis <alistair.francis@wdc.com>,
GNU C Library <libc-alpha@sourceware.org>,
Adhemerval Zanella <adhemerval.zanella@linaro.org>,
Florian Weimer <fweimer@redhat.com>,
"Carlos O'Donell" <carlos@redhat.com>,
Stepan Golosunov <stepan@golosunov.pp.ru>
Subject: Re: [PATCH v7 0/3] y2038: Linux: Introduce __clock_settime64 function
Date: Fri, 6 Sep 2019 23:55:42 +0200 [thread overview]
Message-ID: <20190906235542.1637b4c0@jawa> (raw)
In-Reply-To: <CAKmqyKMdsVZg_ZS7fLTt23dy2vnZc1z85b_+8heiK-8UiqMx+g@mail.gmail.com>
[-- Attachment #1: Type: text/plain, Size: 7143 bytes --]
Hi Alistair,
> On Fri, Sep 6, 2019 at 7:59 AM Lukasz Majewski <lukma@denx.de> wrote:
> >
> > This patch set introduces the conversion of clock_settime to
> > explicit 64 bit struct __timespec64 arguments. As a result this
> > function is now Y2038 safe.
> >
> > This work is (loosely) based on a previous development/patches:
> > https://libc-alpha.sourceware.narkive.com/zniMOWui/rfc-patch-00-52-make-glibc-y2038-proof#post68
> >
> > Github branch (including the y2038 conversion example):
> > https://github.com/lmajewski/y2038_glibc/commits/glibc__clock_settime-conversion-v7
> >
> > Those patches have been applied on top of master branch:
> > SHA1: a26918cfda4bc4b9dad8aae1496e3ef7cbb63d96
> >
> > Shall be used with provided meta-y2038 for development and testing:
> > https://github.com/lmajewski/meta-y2038
> >
> > I've used guidelines from:
> > https://www.gnu.org/software/libc/manual/html_mono/libc.html
> > "D.2.1 64-bit time symbol handling in the GNU C Library"
> > to convert *clock_settime*.
> >
> > and most notably from:
> > https://sourceware.org/glibc/wiki/Y2038ProofnessDesign#clock_gettime.28.29
> >
>
> Thanks for the patches Lukassz!
>
> With my RV32 port I see this failure when compiling:
>
But I did not change the core code between v5 (to which you refer here):
https://patchwork.ozlabs.org/cover/1155397/
and then:
https://sourceware.org/ml/libc-alpha/2019-05/msg00661.html
and v7:
https://patchwork.ozlabs.org/patch/1159056/
(The only change in v6 was to use 32 bit call to clock_settime syscall
as requested by Arnd to facilitate validation on existing systems.
However, Joseph was against this change).
The notable change (but not in this patch set) was moving clock_settime
symbol solely to glibc (and it has been removed from librt).
> In file included from ../sysdeps/generic/hp-timing.h:23,
> from ../nptl/descr.h:27,
> from ../sysdeps/riscv/nptl/tls.h:41,
> from ../include/errno.h:25,
> from ../sysdeps/unix/sysv/linux/clock_settime.c:18:
> ../include/time.h:129:28: error: conflicting types for
> '__clock_settime' # define __clock_settime64 __clock_settime
> ^~~~~~~~~~~~~~~
> ../sysdeps/unix/sysv/linux/clock_settime.c:25:1: note: in expansion of
> macro '__clock_settime64'
> __clock_settime64 (clockid_t clock_id, const struct __timespec64 *tp)
> ^~~~~~~~~~~~~~~~~
> In file included from <command-line>:
> ../include/time.h:25:20: note: previous declaration of
> '__clock_settime' was here
> libc_hidden_proto (__clock_settime)
> ^~~~~~~~~~~~~~~
> ./../include/libc-symbols.h:598:33: note: in definition of macro
> '__hidden_proto'
> extern thread __typeof (name) name __hidden_proto_hiddenattr
> (attrs); ^~~~
> ./../include/libc-symbols.h:617:44: note: in expansion of macro
> 'hidden_proto' # define libc_hidden_proto(name, attrs...)
> hidden_proto (name, ##attrs) ^~~~~~~~~~~~
> ../include/time.h:25:1: note: in expansion of macro
> 'libc_hidden_proto' libc_hidden_proto (__clock_settime)
> ^~~~~~~~~~~~~~~~~
> ../sysdeps/unix/sysv/linux/clock_settime.c:70:42: error: conflicting
> types for 'clock_settime'
> versioned_symbol (libc, __clock_settime, clock_settime, GLIBC_2_17);
> ^~~~~~~~~~~~~
> ./../include/libc-symbols.h:152:26: note: in definition of macro
> '_weak_alias' extern __typeof (name) aliasname __attribute__ ((weak,
> alias (#name))) \ ^~~~~~~~~
> ../include/shlib-compat.h:74:3: note: in expansion of macro
> 'weak_alias' weak_alias (local, symbol)
> ^~~~~~~~~~
> ../sysdeps/unix/sysv/linux/clock_settime.c:70:1: note: in expansion of
> macro 'versioned_symbol'
> versioned_symbol (libc, __clock_settime, clock_settime, GLIBC_2_17);
> ^~~~~~~~~~~~~~~~
> In file included from ../include/time.h:2,
> from ../sysdeps/generic/hp-timing.h:23,
> from ../nptl/descr.h:27,
> from ../sysdeps/riscv/nptl/tls.h:41,
> from ../include/errno.h:25,
> from ../sysdeps/unix/sysv/linux/clock_settime.c:18:
> ../time/time.h:222:12: note: previous declaration of 'clock_settime'
> was here extern int clock_settime (clockid_t __clock_id, const struct
> timespec *__tp)
>
> Which I can fix with this diff:
>
> diff --git a/include/time.h b/include/time.h
> index 7ed3aa61d1d..28e2722de21 100644
> --- a/include/time.h
> +++ b/include/time.h
> @@ -125,13 +125,9 @@ extern __time64_t __timegm64 (struct tm *__tp)
> __THROW; libc_hidden_proto (__timegm64)
> #endif
>
> -#if __TIMESIZE == 64
What is the value of __TIMESIZE on your port? Is it 32 or 64 ?
Do you also #define __USE_TIME_BITS64 1 ?
> -# define __clock_settime64 __clock_settime
> -#else
> extern int __clock_settime64 (clockid_t clock_id,
> const struct __timespec64 *tp);
> libc_hidden_proto (__clock_settime64)
> -#endif
>
> /* Compute the `struct tm' representation of T,
> offset OFFSET seconds east of UTC,
> diff --git a/sysdeps/unix/sysv/linux/clock_settime.c
> b/sysdeps/unix/sysv/linux/clock_settime.c
> index f5e084238a5..ab033c56ea9 100644
> --- a/sysdeps/unix/sysv/linux/clock_settime.c
> +++ b/sysdeps/unix/sysv/linux/clock_settime.c
> @@ -54,7 +54,6 @@ __clock_settime64 (clockid_t clock_id, const struct
> __timespec64 *tp)
> #endif
> }
>
> -#if __TIMESIZE != 64
When I look on RISC-V patchset:
https://patchwork.ozlabs.org/patch/1155391/
for the __clock_nanosleep for example you follow the same convention.
And you don't need to remove #if __TIMESIZE != 64 for example.
One difference is that the clock_settime64 code will be Y2038 on e.g.
32 bit ARM only when I apply on top of it following patch:
https://github.com/lmajewski/y2038_glibc/commit/c6740c05ea9b224a552847d10402f98da8376994
(it enables redirect for clock_settime and support for -D_TIME_BITS=64
compilation flag).
> int
> __clock_settime (clockid_t clock_id, const struct timespec *tp)
> {
> @@ -63,7 +62,6 @@ __clock_settime (clockid_t clock_id, const struct
> timespec *tp)
> valid_timespec_to_timespec64 (tp, &ts64);
> return __clock_settime64 (clock_id, &ts64);
> }
> -#endif
>
> libc_hidden_def (__clock_settime)
>
> Alistair
>
> >
> >
> > Lukasz Majewski (3):
> > y2038: Introduce internal for glibc struct __timespec64
> > y2038: Provide conversion helpers for struct __timespec64
> > y2038: linux: Provide __clock_settime64 implementation
> >
> > include/time.h | 116
> > ++++++++++++++++++++++++ sysdeps/unix/sysv/linux/clock_settime.c |
> > 38 +++++++- 2 files changed, 150 insertions(+), 4 deletions(-)
> >
> > --
> > 2.20.1
> >
Best regards,
Lukasz Majewski
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: (+49)-8142-66989-59 Fax: (+49)-8142-66989-80 Email: lukma@denx.de
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
next prev parent reply other threads:[~2019-09-06 21:55 UTC|newest]
Thread overview: 24+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-09-06 14:59 [PATCH v7 0/3] y2038: Linux: Introduce __clock_settime64 function Lukasz Majewski
2019-09-06 14:59 ` [PATCH v7 1/3] y2038: Introduce internal for glibc struct __timespec64 Lukasz Majewski
2019-09-06 14:59 ` [PATCH v7 2/3] y2038: Provide conversion helpers for " Lukasz Majewski
2019-09-06 14:59 ` [PATCH v7 3/3] y2038: linux: Provide __clock_settime64 implementation Lukasz Majewski
2019-09-06 16:52 ` [PATCH v7 0/3] y2038: Linux: Introduce __clock_settime64 function Alistair Francis
2019-09-06 21:28 ` Joseph Myers
2019-09-16 22:45 ` Alistair Francis
2019-09-17 0:44 ` Joseph Myers
2019-09-17 10:11 ` Lukasz Majewski
2019-09-17 13:42 ` Joseph Myers
2019-09-17 15:53 ` Lukasz Majewski
2019-09-17 16:51 ` Joseph Myers
2019-09-18 17:03 ` Alistair Francis
2019-09-18 17:25 ` Joseph Myers
2019-09-18 21:37 ` Lukasz Majewski
2019-09-18 21:45 ` Joseph Myers
2019-09-19 21:56 ` Alistair Francis
2019-09-18 21:28 ` Lukasz Majewski
2019-09-18 22:26 ` Alistair Francis
2019-09-19 7:50 ` Lukasz Majewski
2019-09-06 21:55 ` Lukasz Majewski [this message]
2019-09-06 22:01 ` Alistair Francis
2019-09-13 14:36 ` Lukasz Majewski
2019-09-16 21:50 ` Alistair Francis
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
List information: https://www.gnu.org/software/libc/involved.html
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20190906235542.1637b4c0@jawa \
--to=lukma@denx.de \
--cc=adhemerval.zanella@linaro.org \
--cc=alistair.francis@wdc.com \
--cc=alistair23@gmail.com \
--cc=arnd@arndb.de \
--cc=carlos@redhat.com \
--cc=fweimer@redhat.com \
--cc=joseph@codesourcery.com \
--cc=libc-alpha@sourceware.org \
--cc=stepan@golosunov.pp.ru \
--cc=zackw@panix.com \
/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).