From: Michael Olbrich <m.olbrich@pengutronix.de>
To: Ian Abbott <abbotti@mev.co.uk>
Cc: ptxdist@pengutronix.de
Subject: Re: [ptxdist] [PATCH] libtahu: Change CFLAGS to use 64-bit system time
Date: Fri, 24 Feb 2023 15:45:13 +0100 [thread overview]
Message-ID: <20230224144513.GV11263@pengutronix.de> (raw)
In-Reply-To: <9af82edb-4f5f-ee85-681f-bbc2d253829f@mev.co.uk>
On Fri, Feb 24, 2023 at 12:03:14PM +0000, Ian Abbott wrote:
> On 23/02/2023 15:42, Michael Olbrich wrote:
> > On Mon, Feb 20, 2023 at 01:41:05PM +0000, Ian Abbott wrote:
> > > libtahu exposes a library function `get_current_timestamp()` to return
> > > the real-time clock as the number of milliseconds since the Unix epoch
> > > as a 64-bit value. It calls `clock_gettime(CLOCK_REALTIME, &ts)` (where
> > > `ts` is a `struct timespec`) to get the system time and does some simple
> > > arithmetic on `ts.tv_sec` and `ts.tv_nsec` to convert the time to a
> > > number of milliseconds.
> > >
> > > Ideally, we want `ts.tv_sec` to have a 64-bit type to avoid Y2038
> > > problems, but on most 32-bit architectures, `ts.tv_sec` will have a
> > > 32-bit type by default. Try and select the 64-bit time interface even
> > > on 32-bit architectures by appending `-D_FILE_OFFSET_BITS=64` and
> > > `-D_TIME_BITS=64` to the `CFLAGS` variable when invoking `make`. This
> > > works for glibc 2.34 or later when the system is running on Linux kernel
> > > 5.1 or later.
> > >
> > > Signed-off-by: Ian Abbott <abbotti@mev.co.uk>
> > > ---
> > > rules/libtahu.make | 5 +++--
> > > 1 file changed, 3 insertions(+), 2 deletions(-)
> > >
> > > diff --git a/rules/libtahu.make b/rules/libtahu.make
> > > index 26c064c6d..3e392216d 100644
> > > --- a/rules/libtahu.make
> > > +++ b/rules/libtahu.make
> > > @@ -37,11 +37,12 @@ LIBTAHU_CONF_TOOL := NO
> > > # Compile
> > > # ----------------------------------------------------------------------------
> > > -#LIBTAHU_MAKE_ENV := $(CROSS_ENV)
> > > -
> > > +LIBTAHU_CPPFLAGS := -Iinclude -D_FILE_OFFSET_BITS=64 -D_TIME_BITS=64
> > > +LIBTAHU_CFLAGS := -O2 -g -g3 -fPIC
> >
> > No need to mess with the existing CFLAGS. You can inject extra CFLAGS
> > though the toolchain wrappers. And for correctness, make it conditional:
> >
> > ifdef PTXCONF_GLOBAL_LARGE_FILE
> > LIBTAHU_CFLAGS := -D_FILE_OFFSET_BITS=64 -D_TIME_BITS=64
> > endif
> >
> > nothing else is needed.
> >
> > Michael
>
> Hi Michael,
>
> Sorry to be a pain. I'm attempting to inject extra CFLAGS through the
> toolchain wrappers, but I'm not sure how to do it. Here is one of my
> attempts that does not work:
>
> #
> ----------------------------------------------------------------------------
> # Prepare
> #
> ----------------------------------------------------------------------------
>
> LIBTAHU_CONF_TOOL := NO
>
> #
> ----------------------------------------------------------------------------
> # Compile
> #
> ----------------------------------------------------------------------------
>
> # -D_TIME_BITS=64 requires -D_FILE_OFFSET_BITS=64
> LIBTAHU_CPPFLAGS := -D_FILE_OFFSET_BITS=64 -D_TIME_BITS=64
> LIBTAHU_CFLAGS := -O2
> LIBTAHU_MAKE_ENV := $(CROSS_ENV)
> # Just build the dynamic library.
> LIBTAHU_MAKE_OPT := \
> $(CROSS_ENV_PROGS) \
> lib/libtahu.so
>
> $(STATEDIR)/libtahu.compile:
> @$(call targetinfo)
> @$(call world/compile, LIBTAHU)
> @$(call touch)
>
> I'm obviously missing something to pass the properly wrapped CC and CPP
> variables (incorporating the extra flags) to 'make'. Could you provide a
> clue how to fix it?
Are you sure it does not work? It will not be visible in the regular make
output. Run PTXdist with '-v' and then check platform-XXX/logfile.
You should find a 'wrapper: ...' line for each compiler run and the extra
flags should be visible here.
Michael
--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
next prev parent reply other threads:[~2023-02-24 14:45 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-02-20 13:41 Ian Abbott
2023-02-23 15:42 ` Michael Olbrich
2023-02-24 9:55 ` Ian Abbott
2023-02-24 14:43 ` Michael Olbrich
2023-02-24 12:03 ` Ian Abbott
2023-02-24 14:45 ` Michael Olbrich [this message]
2023-02-24 16:10 ` Ian Abbott
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=20230224144513.GV11263@pengutronix.de \
--to=m.olbrich@pengutronix.de \
--cc=abbotti@mev.co.uk \
--cc=ptxdist@pengutronix.de \
/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