Re: [patch] Fix PR 84487, large rodata increase in tonto and other programs
Hi Thomas, Thanks for your determination in dealing with this. It has been on my TODO list for a long time but, like you at the outset, I had no idea how to deal with it. OK on the fortran side. Paul On Sat, 13 Apr 2019 at 19:48, Thomas Koenig wrote: > > Hello world, > > the attached patch fixes a 8/9 regression where _def_init, an internal > Fortran variable containing only zeros, was placed into the .rodata > section. This led to a large increase in executable size. > > There should be no impact on other languages because the change to > varasm.c is guarded by lang_GNU_Fortran (). > > Regarding the test case: I did find one other test which checks > for .bss, so I suppose this is safe. > > Regression-tested with a full test (--enable-languages=all and > make -j64 -k check) on POWER9. > > I would like to apply it to both affected branches. > > OK for the general and the Fortran part? > > Regards > > Thomas > > 2019-04-13 Thomas Koenig > > PR fortran/84487 > * trans-decl.c (gfc_get_symbol_decl): Mark _def_init as > artificial. > > 2019-04-13 Thomas Koenig > > PR fortran/84487 > * varasm.c (bss_initializer_p): If we are compiling Fortran, the > decl is artifical and it has a size larger than 255, it can be > put into BSS. > > 2019-04-13 Thomas Koenig > > PR fortran/84487 > * gfortran.dg/def_init_1.f90: New test. > > -- "If you can't explain it simply, you don't understand it well enough" - Albert Einstein
[patch] Fix PR 84487, large rodata increase in tonto and other programs
Hello world,
the attached patch fixes a 8/9 regression where _def_init, an internal
Fortran variable containing only zeros, was placed into the .rodata
section. This led to a large increase in executable size.
There should be no impact on other languages because the change to
varasm.c is guarded by lang_GNU_Fortran ().
Regarding the test case: I did find one other test which checks
for .bss, so I suppose this is safe.
Regression-tested with a full test (--enable-languages=all and
make -j64 -k check) on POWER9.
I would like to apply it to both affected branches.
OK for the general and the Fortran part?
Regards
Thomas
2019-04-13 Thomas Koenig
PR fortran/84487
* trans-decl.c (gfc_get_symbol_decl): Mark _def_init as
artificial.
2019-04-13 Thomas Koenig
PR fortran/84487
* varasm.c (bss_initializer_p): If we are compiling Fortran, the
decl is artifical and it has a size larger than 255, it can be
put into BSS.
2019-04-13 Thomas Koenig
PR fortran/84487
* gfortran.dg/def_init_1.f90: New test.
Index: fortran/trans-decl.c
===
--- fortran/trans-decl.c (Revision 270182)
+++ fortran/trans-decl.c (Arbeitskopie)
@@ -1865,7 +1865,10 @@ gfc_get_symbol_decl (gfc_symbol * sym)
if (sym->attr.vtab
|| (sym->name[0] == '_' && gfc_str_startswith (sym->name, "__def_init")))
-TREE_READONLY (decl) = 1;
+{
+ TREE_READONLY (decl) = 1;
+ DECL_ARTIFICIAL (decl) = 1;
+}
return decl;
}
Index: varasm.c
===
--- varasm.c (Revision 270182)
+++ varasm.c (Arbeitskopie)
@@ -1007,9 +1007,13 @@ decode_reg_name (const char *name)
bool
bss_initializer_p (const_tree decl, bool named)
{
- /* Do not put non-common constants into the .bss section, they belong in
- a readonly section, except when NAMED is true. */
- return ((!TREE_READONLY (decl) || DECL_COMMON (decl) || named)
+ /* Do not put non-common constants into the .bss section, they
+ belong in a readonly section, except when NAMED is true or when
+ we are dealing with an artificial declaration, in the Fortran
+ compiler only, above a certain size. */
+ return ((!TREE_READONLY (decl) || DECL_COMMON (decl) || named
+ || (lang_GNU_Fortran () && DECL_ARTIFICIAL (decl)
+ && (tree_to_uhwi (DECL_SIZE_UNIT (decl)) > 255 )))
&& (DECL_INITIAL (decl) == NULL
/* In LTO we have no errors in program; error_mark_node is used
to mark offlined constructors. */
! { dg-do compile }
! PR 84487 - check that def_init, if it is large,
! is put into .bss.
module TYPES_MODULE
implicit none
type archive_type
character(2**18) :: root_name
end type archive_type
end module TYPES_MODULE
! { dg-final { scan-assembler "\.bss" } }
Re: [libphobos] Work around Solaris ld bug linking __tls_get_addr on 64-bit x86
On Tue, 9 Apr 2019 at 21:36, Rainer Orth wrote: > > Rainer Orth writes: > > > Here's the patch I mentioned in > > > > https://gcc.gnu.org/ml/gcc-patches/2019-01/msg01661.html > > > > to work around an amd64 Solaris ld bug. I'm just posting it for > > reference now: until it's clear if a fix will make it into Solaris 11.5 > > or not, there's no point in applying it yet. > > > > Still, review comments are appreciated. > > With the revised patch for non-dlpi_tls_modid versions of Solaris > > https://gcc.gnu.org/ml/gcc-patches/2019-04/msg00354.html > > __tls_get_addr will always be called and thus this workaround is needed > unconditionally, even should the ld bug be fixed in Solaris 11.5. > > This revision adjusts the patch accordingly and was tested together with > the other one on Solaris 11.[345]/x86. > > Ok for mainline? > OK -- Iain
Re: [PATCH v2, d] PR d/87799 Fix build on windows hosts
On Mon, 25 Mar 2019 at 20:04, Johannes Pfau wrote: > > diff --git a/gcc/d/d-system.h b/gcc/d/d-system.h > index efece15f3bc..6c18b69d4df 100644 > --- a/gcc/d/d-system.h > +++ b/gcc/d/d-system.h > @@ -55,4 +55,10 @@ > #undef tolower > #define tolower(c) TOLOWER(c) > > +/* We do not include direct.h as it conflicts with system.h. */ Maybe a small nit on the comment. /* Forward _mkdir on MinGW to mkdir in system.h. */ Otherwise, OK.
Re: [PATCH] D support for RISC-V
On Tue, 9 Apr 2019 at 11:53, David Abdurachmanov
wrote:
>
> diff --git a/libphobos/libdruntime/rt/sections_elf_shared.d
> b/libphobos/libdruntime/rt/sections_elf_shared.d
> index d4e1ff07699..45c1dcbc7f3 100644
> --- a/libphobos/libdruntime/rt/sections_elf_shared.d
> +++ b/libphobos/libdruntime/rt/sections_elf_shared.d
> @@ -10,6 +10,9 @@
>
> module rt.sections_elf_shared;
>
> +version (RISCV32) version = RISCV_Any;
> +version (RISCV64) version = RISCV_Any;
> +
> version (CRuntime_Glibc) enum SharedELF = true;
> else version (FreeBSD) enum SharedELF = true;
> else version (NetBSD) enum SharedELF = true;
> @@ -671,7 +674,16 @@ version (Shared)
> if (dyn.d_tag == DT_STRTAB)
> {
> version (linux)
> -strtab = cast(const(char)*)dyn.d_un.d_ptr;
> +{
> +// This might change in future glibc releases (after
> 2.29) as dynamic sections
> +// are not required to be read-only on RISC-V. This was
> copy & pasted from MIPS while
> +// upstreaming RISC-V support. Otherwise MIPS is the
> only arch which sets in glibc:
> +// #define DL_RO_DYN_SECTION 1
> +version (RISCV_Any)
> +strtab = cast(const(char)*)(info.dlpi_addr +
> dyn.d_un.d_ptr); // relocate
> +else
> +strtab = cast(const(char)*)dyn.d_un.d_ptr;
> +}
> else version (FreeBSD)
> strtab = cast(const(char)*)(info.dlpi_addr +
> dyn.d_un.d_ptr); // relocate
> else version (NetBSD)
This needs to be rebased to patch gcc/sections/elf_shared.d
> diff --git
> a/libphobos/src/std/experimental/allocator/building_blocks/region.d
> b/libphobos/src/std/experimental/allocator/building_blocks/region.d
> index dfcecce72bd..cafe059a61f 100644
> --- a/libphobos/src/std/experimental/allocator/building_blocks/region.d
> +++ b/libphobos/src/std/experimental/allocator/building_blocks/region.d
> @@ -391,7 +391,8 @@ struct InSituRegion(size_t size, size_t minAlign =
> platformAlignment)
> else version (PPC) enum growDownwards = Yes.growDownwards;
> else version (PPC64) enum growDownwards = Yes.growDownwards;
> else version (MIPS32) enum growDownwards = Yes.growDownwards;
> -else version (MIPS64) enum growDownwards = Yes.growDownwards;
> +else version (RISCV32) enum growDownwards = Yes.growDownwards;
> +else version (RISCV64) enum growDownwards = Yes.growDownwards;
> else version (SPARC) enum growDownwards = Yes.growDownwards;
> else version (SystemZ) enum growDownwards = Yes.growDownwards;
> else static assert(0, "Dunno how the stack grows on this architecture.");
This has been upstreamed and committed in r270296.
> diff --git a/libphobos/src/std/math.d b/libphobos/src/std/math.d
> index e98e746a856..9fe746501a4 100644
> --- a/libphobos/src/std/math.d
> +++ b/libphobos/src/std/math.d
> @@ -162,6 +162,8 @@ version (AArch64) version = ARM_Any;
> version (ARM) version = ARM_Any;
> version (SPARC) version = SPARC_Any;
> version (SPARC64) version = SPARC_Any;
> +version (RISCV32) version = RISCV_Any;
> +version (RISCV64) version = RISCV_Any;
>
> version (D_InlineAsm_X86)
> {
Apart from the asm { } code, same with this as well.
Everything else is OK.
--
Iain
Re: [libphobos] Use sections_elf_shared.d on Solaris 11.5 (PR d/88150)
On Tue, 29 Jan 2019 at 11:25, Rainer Orth wrote: > > I've successfully been using a late prototype of an implementation of > the dlpi_tls_modid field of struct dl_phdr_info on Solaris 11.5 Beta. > This allowed me to get pretty reasonable test results using > sections_elf_shared.d on Solaris. > > This pretty straightforward patch implements this. Only a few points > are worth mentioning: > > * libdruntime/rt/bss_sections.c refers to __bss_start, which only gld > defines. Unfortunately, it's marked weak, so the absence with Solaris > ld went unnoticed at first. Lacking an exact equivalent, I'm using > _edata instead, which is pretty close modulo section alignment. > > * As detailed in the PR, not dlclose()ing the handle in handleForName is > necessary to avoid an assertion failure. > > * I'm removing sections_solaris.d since it wouldn't even compile and > seems pretty useless. > > This patch gave me the following testsuite results on Solaris 11.5/x86 > (a few more minor testsuite fixes were included, too): > > === gdc tests === > > > Running target unix > FAIL: gdc.test/runnable/nulltype.d execution test > FAIL: gdc.test/runnable/nulltype.d -O2 execution test > FAIL: gdc.test/runnable/nulltype.d -O2 -shared-libphobos execution test > FAIL: gdc.test/runnable/nulltype.d -g execution test > FAIL: gdc.test/runnable/nulltype.d -g -O2 execution test > FAIL: gdc.test/runnable/nulltype.d -g -O2 -shared-libphobos execution test > FAIL: gdc.test/runnable/nulltype.d -g -shared-libphobos execution test > FAIL: gdc.test/runnable/nulltype.d -shared-libphobos execution test > > === gdc Summary for unix === > > # of expected passes30785 > # of unexpected failures8 > > Running target unix/-m64 > > === gdc Summary for unix/-m64 === > > # of expected passes30793 > > === gdc Summary === > > # of expected passes61578 > # of unexpected failures8 > /var/gcc/gcc-9.0.1-20190128/11.5-gcc-gas-libphobos/gcc/gdc version 9.0.1 > 20190128 (experimental) [trunk revision 268335] (GCC) > > === libphobos tests === > > > Running target unix > FAIL: libphobos.shared/load.d -shared-libphobos -ldl execution test > FAIL: libphobos.shared/load_13414.d -shared-libphobos -ldl execution test > FAIL: libphobos.shared/finalize.d -shared-libphobos -ldl execution test > FAIL: libphobos.shared/linkD.c lib.so -ldl -pthread execution test > FAIL: libphobos.unittests/druntime/shared/core.sync.mutex > FAIL: libphobos.unittests/druntime/shared/core.time > > === libphobos Summary for unix === > > # of expected passes119 > # of unexpected failures6 > > Running target unix/-m64 > FAIL: libphobos.shared/load.d -shared-libphobos -ldl execution test > FAIL: libphobos.shared/load_13414.d -shared-libphobos -ldl execution test > FAIL: libphobos.shared/finalize.d -shared-libphobos -ldl execution test > FAIL: libphobos.shared/linkD.c lib.so -ldl -pthread execution test > FAIL: libphobos.unittests/druntime/shared/ld.so.1: > FAIL: libphobos.unittests/druntime/shared/unittest: > FAIL: libphobos.unittests/druntime/shared/fatal: > FAIL: libphobos.unittests/druntime/shared/libgdruntime_t.so.0: > FAIL: libphobos.unittests/druntime/shared/open > FAIL: libphobos.unittests/druntime/shared/failed: > FAIL: libphobos.unittests/druntime/shared/No > FAIL: libphobos.unittests/druntime/shared/such > FAIL: libphobos.unittests/druntime/shared/file > FAIL: libphobos.unittests/druntime/shared/or > FAIL: libphobos.unittests/druntime/shared/directory > > === libphobos Summary for unix/-m64 === > > # of expected passes78 > # of unexpected failures15 > > === libphobos Summary === > > # of expected passes197 > # of unexpected failures21 > > The 32-bit nulltype.d failures occur on Linux, too (PR d/87824), and the > 64-bit libphobos.unittests/druntime/shared failures happen because > libgdruntime_t.so.0 is only built for the default multilib. The > libphobos.shared failures clearly bear investigating. > > For the amd64 results, I needed a separate patch to workaround an ld > bug, to be submitted shortly. > > While I did run a sparc-sun-solaris2.11 bootstrap, too, results are > pretty useless due to PR d/88462 (the minfo alignment issue). > Hi, Sorry for the belated response, this has been delayed by myself sorting out moving the sections code out of upstream tree. bss_sections.c is now gone, and the patch should be rebased to target gcc/sections/elf_shared.d. I think only core/sys/solaris/link.d is left to be upstreamed, otherwise this is OK. -- Iain
Re: [libphobos] Work around lack of dlpi_tls_modid before Solaris 11.5
On Tue, 9 Apr 2019 at 21:27, Rainer Orth wrote: > > Rainer Orth writes: > > > Before Solaris 11.5, struct dl_phdr_info lacked the dlpi_tls_modid > > member. While the support might be backported to Solaris 11.4, it > > certainly won't to previous Solaris releases. To work around this, I've > > used the following patch. Again, it's pretty straightforward. > > > > Point of note: > > > > * On Solaris, FreeBSD and NetBSD, dl_phdr_info is always visible in > > , while on Linux one needs to define _GNU_SOURCE. > > AC_USE_SYSTEM_EXTENSION takes care of this, but needs to be called > > early (i.e. not in DRUNTIME_OS_DLPI_TLS_MODID) to avoid an autoconf > > warning: > > > > configure.ac:129: warning: AC_COMPILE_IFELSE was called before > > AC_USE_SYSTEM_EXTENSIONS > > m4/druntime/os.m4:190: DRUNTIME_OS_DLPI_TLS_MODID is expanded from... > > configure.ac:129: the top level > > > > Tested on i386-pc-solaris2.11 (Solaris 11.4) and x86_64-pc-linux-gnu. > > > > Not unexpectedly, there are a couple of regressions compared to the > > Solaris 11.5/x86 results: > > > > +FAIL: gdc.test/runnable/testaa.d execution test > > +FAIL: gdc.test/runnable/testaa.d -fPIC execution test > > +FAIL: gdc.test/runnable/testaa.d -fPIC -shared-libphobos execution test > > +FAIL: gdc.test/runnable/testaa.d -shared-libphobos execution test > > > > 32 and 64-bit > > > > +FAIL: gdc.test/runnable/xtest55.d execution test > > +FAIL: gdc.test/runnable/xtest55.d -shared-libphobos execution test > > > > 64-bit only > > > > +FAIL: libphobos.shared/link_linkdep.d > > -I/vol/gcc/src/hg/trunk/local/libphobos/t > > estsuite/libphobos.shared liblinkdep.so lib.so -shared-libphobos execution > > test > > > > +FAIL: libphobos.shared/load_linkdep.d -shared-libphobos -ldl execution test > > +FAIL: libphobos.shared/load_loaddep.d -shared-libphobos -ldl execution test > > > > +FAIL: libphobos.shared/linkDR.c -shared-libphobos -ldl -pthread execution > > test > > +FAIL: libphobos.shared/host.c -ldl -pthread execution test > > +FAIL: libphobos.shared/loadDR.c -ldl -pthread -g execution test > > > > 32 and 64-bit > > > > +FAIL: libphobos.shared/link.d > > -I/vol/gcc/src/hg/trunk/local/libphobos/testsuite/libphobos.shared lib.so > > -shared-libphobos execution test > > > > 64-bit only > > > > I haven't looked much closer yet, just posting this to see if anything > > along those lines could be acceptable at all. > > Fortunately, Iain just discovered a way better way to work around the > lack of dlpi_tls_modid: dlinfo(RTLD_DI_LINKMAP) returns not only a > pointer to a Link_map * as documented in the man page, but rather a > pointer to a struct Rt_map which also includes a rt_tlsmodid member. > It has been this way since at least Solaris 9 and the Solaris/Illumos > sources ($SRC/cmd/sgs/include/rtld.h) explicitly have a comment around > the start of the structure stateing > > Exposed to rtld_db - don't move, don't delete > > which pretty much guarantees that this can be relied on. > > The only other quirk of note is that before dlpi_tls_modid got added, > tlsmodid started at 0 for the main executable, not 1 as the glibc spec > required and libdruntime assumes. I account for this by adjusting > _tlsMod on dlinfo and undoing the adjustment before passing it to > __tls_get_adddr, the only consumer. > > With the revised patch below, I get clean gdc testresults on Solaris > 11.3 and 11.4 (and still on 11.5 which uses the dlpi_tls_modid path), > and almost identical libphobos testresults on all three versions. > > I hope this is good to go in (the sections_elf_shared.d part via > upstream, the rest directly) now? > > Thanks a lot to Iain for discovering this. > I've finally managed to get around to moving the guts of rt.sections_* to gcc.sections.*, with the patch amended to apply to gcc/sections/elf_shared.d, this is OK. -- Iain
[PATCH, libphobos] Committed move rt.sections modules to gcc.sections
Hi, This patch moved all rt.sections modules to a gcc.sections package. These modules depend on a mixture between how the compiler emits run-time module information, and what functions are exposed by the platform to inquire about loaded global and thread-local data sections. As the upstream implementation is written to work only with how the reference D compiler writes out data, much of what is present does not apply to the GCC D front-end. So it has been moved to a non-upstream location in the source tree, where most of it will be rewritten once each port has been completed. The only tested module sections/elf_shared.d has been cleaned up so that all deprecated declarations have been removed, as well as the brittle module collision checking, which required bss_sections.c. All other ports have been left unchanged apart from a commonizing of attributes. Bootstrapped and regression tested on x86_64-linux-gnu. Committed to trunk as r270341. -- Iain --- libphobos/ChangeLog: 2019-04-13 Iain Buclaw * libdruntime/Makefile.am (DRUNTIME_CSOURCES): Remove bss_sections.c. (DRUNTIME_DSOURCES): Rename rt/sections_* modules to gcc/sections/*. * libdruntime/Makefile.in: Regenerate. * libdruntime/gcc/sections/android.d: New file. * libdruntime/gcc/sections/elf_shared.d: New file. * libdruntime/gcc/sections/osx.d: New file. * libdruntime/gcc/sections/package.d: New file. * libdruntime/gcc/sections/solaris.d: New file. * libdruntime/gcc/sections/win32.d: New file. * libdruntime/gcc/sections/win64.d: New file. * libdruntime/rt/bss_section.c: Remove. * libdruntime/rt/sections.d: Publicly import gcc.sections. * libdruntime/rt/sections_android.d: Remove. * libdruntime/rt/sections_elf_shared.d: Remove. * libdruntime/rt/sections_osx.d: Remove. * libdruntime/rt/sections_solaris.d: Remove. * libdruntime/rt/sections_win32.d: Remove. * libdruntime/rt/sections_win64.d: Remove. --- diff --git a/libphobos/libdruntime/Makefile.am b/libphobos/libdruntime/Makefile.am index abefb435dc7..8e36ce21449 100644 --- a/libphobos/libdruntime/Makefile.am +++ b/libphobos/libdruntime/Makefile.am @@ -122,7 +122,7 @@ DRUNTIME_DSOURCES_GENERATED = gcc/config.d gcc/libbacktrace.d # https://www.gnu.org/software/automake/manual/html_node/Wildcards.html DRUNTIME_SSOURCES = core/threadasm.S -DRUNTIME_CSOURCES = core/stdc/errno_.c rt/bss_section.c +DRUNTIME_CSOURCES = core/stdc/errno_.c DRUNTIME_DSOURCES = core/atomic.d core/attribute.d core/bitop.d \ core/checkedint.d core/cpuid.d core/demangle.d core/exception.d \ @@ -140,14 +140,14 @@ DRUNTIME_DSOURCES = core/atomic.d core/attribute.d core/bitop.d \ core/sync/config.d core/sync/exception.d core/sync/mutex.d \ core/sync/rwmutex.d core/sync/semaphore.d core/thread.d core/time.d \ core/vararg.d gcc/attribute.d gcc/backtrace.d gcc/builtins.d gcc/deh.d \ - gcc/unwind/arm.d gcc/unwind/arm_common.d gcc/unwind/c6x.d \ - gcc/unwind/generic.d gcc/unwind/package.d gcc/unwind/pe.d object.d \ - rt/aApply.d rt/aApplyR.d rt/aaA.d rt/adi.d rt/arrayassign.d \ - rt/arraycast.d rt/arraycat.d rt/cast_.d rt/config.d rt/critical_.d \ - rt/deh.d rt/dmain2.d rt/invariant.d rt/lifetime.d rt/memory.d \ - rt/minfo.d rt/monitor_.d rt/obj.d rt/qsort.d rt/sections.d \ - rt/sections_android.d rt/sections_elf_shared.d rt/sections_osx.d \ - rt/sections_solaris.d rt/sections_win32.d rt/sections_win64.d \ + gcc/sections/android.d gcc/sections/elf_shared.d gcc/sections/osx.d \ + gcc/sections/package.d gcc/sections/solaris.d gcc/sections/win32.d \ + gcc/sections/win64.d gcc/unwind/arm.d gcc/unwind/arm_common.d \ + gcc/unwind/c6x.d gcc/unwind/generic.d gcc/unwind/package.d \ + gcc/unwind/pe.d object.d rt/aApply.d rt/aApplyR.d rt/aaA.d rt/adi.d \ + rt/arrayassign.d rt/arraycast.d rt/arraycat.d rt/cast_.d rt/config.d \ + rt/critical_.d rt/deh.d rt/dmain2.d rt/invariant.d rt/lifetime.d \ + rt/memory.d rt/minfo.d rt/monitor_.d rt/obj.d rt/qsort.d rt/sections.d \ rt/switch_.d rt/tlsgc.d rt/typeinfo/ti_Acdouble.d \ rt/typeinfo/ti_Acfloat.d rt/typeinfo/ti_Acreal.d \ rt/typeinfo/ti_Adouble.d rt/typeinfo/ti_Afloat.d rt/typeinfo/ti_Ag.d \ diff --git a/libphobos/libdruntime/Makefile.in b/libphobos/libdruntime/Makefile.in index 4bfd04c3db5..0a27afac545 100644 --- a/libphobos/libdruntime/Makefile.in +++ b/libphobos/libdruntime/Makefile.in @@ -194,7 +194,10 @@ am__objects_1 = core/atomic.lo core/attribute.lo core/bitop.lo \ core/sync/exception.lo core/sync/mutex.lo core/sync/rwmutex.lo \ core/sync/semaphore.lo core/thread.lo core/time.lo \ core/vararg.lo gcc/attribute.lo gcc/backtrace.lo \ - gcc/builtins.lo gcc/deh.lo gcc/unwind/arm.lo \ + gcc/builtins.lo gcc/deh.lo gcc/sections/android.lo \ + gcc/sections/elf_shared.lo gcc/sections/osx.lo \ + gcc/sections/package.lo gcc/sections/solaris.lo \ + gcc/sections/win32.lo gcc/sections/win64.lo gcc/unwind/arm.lo \
Re: [patch] PR 79842 - i18n: subword translation in "Can't use the same %smodule"
Committed as revision r270338.
Dominique
> Le 11 avr. 2019 à 11:54, Dominique d'Humières a écrit :
>
> Hi all,
>
> I am planning to commit the following patch
>
> --- ../_clean/gcc/fortran/module.c2019-03-21 20:46:46.0 +0100
> +++ gcc/fortran/module.c 2019-04-11 10:28:37.0 +0200
> @@ -7144,8 +7144,12 @@ gfc_use_module (gfc_use_list *module)
> for (p = gfc_state_stack; p; p = p->previous)
> if ((p->state == COMP_MODULE || p->state == COMP_SUBMODULE)
>&& strcmp (p->sym->name, module_name) == 0)
> - gfc_fatal_error ("Cannot USE the same %smodule we're building",
> -p->state == COMP_SUBMODULE ? "sub" : "");
> + {
> + if (p->state == COMP_SUBMODULE)
> + gfc_fatal_error ("Cannot USE a submodule that is currently built");
> + else
> + gfc_fatal_error ("Cannot USE a module that is currently built");
> + }
>
> init_pi_tree ();
> init_true_name_tree ();
>
> While testing it I did not find any coverage of these errors in the test
> suite.
>
> TIA
>
> Dominique
>
[C++ PATCH] PR60531 - Wrong error about unresolved overloaded function
Hi,
For PR60531, GCC wrongly rejects function templates with explicitly
specified template arguments as overloaded. They are resolved by
resolve_nondeduced_context, which is normally called by
cp_default_conversion through decay_conversion, but the latter have
extra effects making them unusable here. Calling the former directly
does work.
Bootstrapped on x86_64-pc-linux-gnu on top of r270264 with
--enable-languages=all; make check shows no regressions.
Does this look okay?
This is my first code contribution to GCC, please let me know if
anything is missing. I have not signed any copyright disclaimer or
copyright assignment; says that is not necessary
for small changes, which I trust this is. If it is needed after all,
please let me know what specifically will be required.
Cheers,
Harald van Dijk
PR c++/60531
* typeck.c (cp_build_binary_op): See if overload can be resolved.
(cp_build_unary_op): Ditto.
* g++.dg/template/operator15.C: New test.
diff --git a/gcc/cp/typeck.c b/gcc/cp/typeck.c
index 03b14024738..e1ffe88ce2c 100644
--- a/gcc/cp/typeck.c
+++ b/gcc/cp/typeck.c
@@ -4384,10 +4384,6 @@ cp_build_binary_op (const op_location_t ,
/* True if both operands have arithmetic type. */
bool arithmetic_types_p;
- /* Apply default conversions. */
- op0 = orig_op0;
- op1 = orig_op1;
-
/* Remember whether we're doing / or %. */
bool doing_div_or_mod = false;
@@ -4397,6 +4393,10 @@ cp_build_binary_op (const op_location_t ,
/* Tree holding instrumentation expression. */
tree instrument_expr = NULL_TREE;
+ /* Apply default conversions. */
+ op0 = resolve_nondeduced_context (orig_op0, complain);
+ op1 = resolve_nondeduced_context (orig_op1, complain);
+
if (code == TRUTH_AND_EXPR || code == TRUTH_ANDIF_EXPR
|| code == TRUTH_OR_EXPR || code == TRUTH_ORIF_EXPR
|| code == TRUTH_XOR_EXPR)
@@ -6204,11 +6204,13 @@ cp_build_unary_op (enum tree_code code, tree xarg, bool
noconvert,
if (!arg || error_operand_p (arg))
return error_mark_node;
+ arg = resolve_nondeduced_context (arg, complain);
+
if ((invalid_op_diag
= targetm.invalid_unary_op ((code == UNARY_PLUS_EXPR
? CONVERT_EXPR
: code),
- TREE_TYPE (xarg
+ TREE_TYPE (arg
{
if (complain & tf_error)
error (invalid_op_diag);
diff --git a/gcc/testsuite/g++.dg/template/operator15.C
b/gcc/testsuite/g++.dg/template/operator15.C
new file mode 100644
index 000..755442266bb
--- /dev/null
+++ b/gcc/testsuite/g++.dg/template/operator15.C
@@ -0,0 +1,6 @@
+// PR c++/60531
+
+template < class T > T foo ();
+
+bool b1 = foo == foo;
+int (*fp1)() = +foo;
