| Message ID | 20141126155539.GB3850@physik.fu-berlin.de |
|---|---|
| State | New |
| Headers | show |
On 26.11.2014 16:55, Tobias Burnus wrote: > This patch adds a configure check for isl_schedule_constraints_compute_schedule, > which is new in ISL 0.13 - and uses it for conditional compilation. > > The graphite*c patch is based on the one of Jack Howarth, > https://gcc.gnu.org/ml/gcc-patches/2014-11/msg00906.html - without checking > whether the ISL replacements make sense. > > With the patch, I have successfully bootstrapped GCC with ISL 0.12.2 and > ISL 0.14 (both using an in-tree build). [I still have to regtest the two > variants and I also want to do a system-ISL built with ISL 0.12.2] > > Is the patch OK for the trunk? > > Tobias > > PS: I'd be happy if some others could run additional tests. > > PPS: If the patch is in, can someone put ISL 0.14 to infrastructure? Then > one can update contrib/download_prerequisites - such that the graphite > failure is at least fixed for in-tree builds (cf. PR64017, PR62289). That looks good from the graphite site. Cheers, Tobias
This patch bootstraps fine on x86_64-apple-darwin14 against isl
0.14 at r218098.
Jack
On Wed, Nov 26, 2014 at 10:55 AM, Tobias Burnus
<tobias.burnus@physik.fu-berlin.de> wrote:
> This patch adds a configure check for isl_schedule_constraints_compute_schedule,
> which is new in ISL 0.13 - and uses it for conditional compilation.
>
> The graphite*c patch is based on the one of Jack Howarth,
> https://gcc.gnu.org/ml/gcc-patches/2014-11/msg00906.html - without checking
> whether the ISL replacements make sense.
>
> With the patch, I have successfully bootstrapped GCC with ISL 0.12.2 and
> ISL 0.14 (both using an in-tree build). [I still have to regtest the two
> variants and I also want to do a system-ISL built with ISL 0.12.2]
>
> Is the patch OK for the trunk?
>
> Tobias
>
> PS: I'd be happy if some others could run additional tests.
>
> PPS: If the patch is in, can someone put ISL 0.14 to infrastructure? Then
> one can update contrib/download_prerequisites - such that the graphite
> failure is at least fixed for in-tree builds (cf. PR64017, PR62289).
PING for https://gcc.gnu.org/ml/gcc-patches/2014-11/msg03281.html Only the build changes have still to be reviewed; the Graphite part is already approved – as is the loosely related follow-up patch. Tobias Burnus wrote: > This patch adds a configure check for isl_schedule_constraints_compute_schedule, > which is new in ISL 0.13 - and uses it for conditional compilation. > > The graphite*c patch is based on the one of Jack Howarth, > https://gcc.gnu.org/ml/gcc-patches/2014-11/msg00906.html - without checking > whether the ISL replacements make sense. > > With the patch, I have successfully bootstrapped GCC with ISL 0.12.2 and > ISL 0.14 (both using an in-tree build). [I still have to regtest the two > variants and I also want to do a system-ISL built with ISL 0.12.2] > > Is the patch OK for the trunk? > > Tobias
On Sun, Nov 30, 2014 at 2:19 PM, Tobias Burnus <burnus@net-b.de> wrote: > PING for https://gcc.gnu.org/ml/gcc-patches/2014-11/msg03281.html > > Only the build changes have still to be reviewed; the Graphite part is > already approved – as is the loosely related follow-up patch. Ok. Thanks, Richard. > Tobias Burnus wrote: >> >> This patch adds a configure check for >> isl_schedule_constraints_compute_schedule, >> which is new in ISL 0.13 - and uses it for conditional compilation. >> >> The graphite*c patch is based on the one of Jack Howarth, >> https://gcc.gnu.org/ml/gcc-patches/2014-11/msg00906.html - without >> checking >> whether the ISL replacements make sense. >> >> With the patch, I have successfully bootstrapped GCC with ISL 0.12.2 and >> ISL 0.14 (both using an in-tree build). [I still have to regtest the two >> variants and I also want to do a system-ISL built with ISL 0.12.2] >> >> Is the patch OK for the trunk? >> >> Tobias > >
diff --git a/gcc/config.in b/gcc/config.in index 9cc0198..65d5e42 100644 --- a/gcc/config.in +++ b/gcc/config.in @@ -1295,6 +1295,12 @@ #endif +/* Define if isl_schedule_constraints_compute_schedule exists. */ +#ifndef USED_FOR_TARGET +#undef HAVE_ISL_SCHED_CONSTRAINTS_COMPUTE_SCHEDULE +#endif + + /* Define to 1 if you have the `kill' function. */ #ifndef USED_FOR_TARGET #undef HAVE_KILL diff --git a/gcc/configure b/gcc/configure index 4f9d1df..6b46bbb 100755 --- a/gcc/configure +++ b/gcc/configure @@ -28057,6 +28057,47 @@ $as_echo "#define HAVE_isl 1" >>confdefs.h fi +# Check whether isl_schedule_constraints_compute_schedule is available; +# it's new in ISL-0.13. +if test "x${ISLLIBS}" != "x" ; then + saved_CFLAGS="$CFLAGS" + CFLAGS="$CFLAGS $ISLINC" + saved_LIBS="$LIBS" + LIBS="$LIBS $ISLLIBS $GMPLIBS" + + { $as_echo "$as_me:${as_lineno-$LINENO}: checking Checking for isl_schedule_constraints_compute_schedule" >&5 +$as_echo_n "checking Checking for isl_schedule_constraints_compute_schedule... " >&6; } + cat confdefs.h - <<_ACEOF >conftest.$ac_ext +/* end confdefs.h. */ +#include <isl/schedule.h> +int +main () +{ +isl_schedule_constraints_compute_schedule (NULL); + ; + return 0; +} +_ACEOF +if ac_fn_c_try_link "$LINENO"; then : + ac_has_isl_schedule_constraints_compute_schedule=yes +else + ac_has_isl_schedule_constraints_compute_schedule=no +fi +rm -f core conftest.err conftest.$ac_objext \ + conftest$ac_exeext conftest.$ac_ext + { $as_echo "$as_me:${as_lineno-$LINENO}: result: $ac_has_isl_schedule_constraints_compute_schedule" >&5 +$as_echo "$ac_has_isl_schedule_constraints_compute_schedule" >&6; } + + LIBS="$saved_LIBS" + CFLAGS="$saved_CFLAGS" + + if test x"$ac_has_isl_schedule_constraints_compute_schedule" = x"yes"; then + +$as_echo "#define HAVE_ISL_SCHED_CONSTRAINTS_COMPUTE_SCHEDULE 1" >>confdefs.h + + fi +fi + # Check for plugin support # Check whether --enable-plugin was given. if test "${enable_plugin+set}" = set; then :