[committed] Tweak m4/minloc0.m4 indentation to avoid -Wmisleading-indentation warnings (PR libfortran/88807)

Hi! The following patch reindents parts of m4/minloc0.m4 to avoid ../../../libgfortran/generated/minloc0_4_i1.c:138:5: warning: this ‘else’ clause does not guard... [-Wmisleading-indentation] etc. warnings. The indentation of the generated sources is misleading for many of the sources and fixing

[committed][openacc] Add used parameter to TARGET_GOACC_VALIDATE_DIMS

Hi, Add a used parameter to TARGET_GOACC_VALIDATE_DIMS, allowing a target to make decisions in the hook implementation based on whether a dimension is used or not. Committed to trunk. Thanks, - Tom [openacc] Add used parameter to TARGET_GOACC_VALIDATE_DIMS 2019-01-11 Tom de Vries *

[committed][nvptx] Verify dimension limits after applying defaults

Hi, There's a problem in oacc_validate_dims that when f.i. the worker dimension is set using -fopenacc-dim=:32, and the vector_length is set using a "vector_length (128)" clause, the compiler combines, accepts and emits the values, while the combination of the two is invalid. The reason for this

[committed][nvptx] Allow default vl to be overridden in nvptx_goacc_validate_dims_1

Hi, In nvptx_goacc_validate_dims_1, allow oacc_default_dims[DIM_VECTOR] to be overridden, by assigning it to a new variable default_vector_length at the start, and using it at the end. Committed to trunk. Thanks, - Tom [nvptx] Allow default vl to be overridden in nvptx_goacc_validate_dims_1 20

[committed][nvptx] Apply vector-partitionable routines workaround to default vl

Hi, Make "[nvptx] Force vl32 if calling vector-partitionable routines" work as well if vector length is set by modifying PTX_DEFAULT_VECTOR_LENGTH. Committed to trunk. Thanks, - Tom [nvptx] Apply vector-partitionable routines workaround to default vl 2019-01-11 Tom de Vries * confi

Re: [wwwdoc][Patch] Mention Loongson 3a1000 3a2000 3a3000 2k1000 support in gcc9

ping? On Mon, Dec 31, 2018 at 6:27 PM Paul Hua wrote: > > Hi Gerald, > > The attached patch mention Loongson 3a1000 3a2000 3a3000 2k1000 support in > gcc9. > > ok for commit?

Re: [PATCH] PR fortran/61765 -- Avoid ENTRY names in check of repeditive symbols

Hi Steve, This is OK for trunk. Thanks Paul On Sat, 12 Jan 2019 at 04:34, Steve Kargl wrote: > > The attached patch has been tested on x86_64-*-freebsd. There > were no regression. The patch is less then obvious, but simple. > OK to commit? > > 2019-01-11 Steven G. Kargl > > PR f

[wwwdocs] Add __cpp_* feature macros to C++20 entries + other changes that have those in projects/cxx_status.html

Hi! On Fri, Jan 11, 2019 at 04:54:11PM +0100, Jakub Jelinek wrote: > I've noticed we don't have any feature test macros in the table for C++20, > even when a couple of the features have them defined. Here is an updated patch, that in addition to that makes 9 URLs as we now have #cxx in gcc-9/chan

[PATCH, powerpc] Fix speculation barrier and group nop to emit target register names.

Hi, The current implementation of “speculation_barrier” and “group_end_nop” insns emit hard-wired register names which causes tests using them to fail on Darwin, at least, which uses “rNN” instead of “NN”. The patch makes the register names for these insns use the operand output mechanism to s

[C++ PATCH] Add __cpp_guaranteed_copy_elision and __cpp_nontype_template_parameter_auto

Hi! So, from what I can understand, __cpp_guaranteed_copy_elision is a C++17 P0135R1 feature test macro for a feature we claim to support, and __cpp_nontype_template_parameter_auto is a new name for the __cpp_template_auto macro (which doesn't appear anymore in the SD-6 lists, but clang++ keeps it

Re: [wwwdocs] Add __cpp_* feature macros to C++20 entries + other changes that have those in projects/cxx_status.html

On Sat, 12 Jan 2019, Jakub Jelinek wrote: > Ok for wwwdocs (or do you suggest something different for the P0941R2 > imlementation status)? I think that was a question to Jason and Jonathan wrt contents? Markup-wise this looks fine, and adding those links looks like a good idea. Gerald

Re: ISO_Fortran_binding patch

Hi Thomas, Committed as revision 267881. I removed the duplicate include file and added some documentation, as suggested. Many thanks for all the help Paul On Tue, 8 Jan 2019 at 23:19, Thomas Koenig wrote: > > Hi Paul, > > > This is an updated version of the earlier patch. The main addition is

[wwwdocs] svnwrite.html - reduce references to SVN

When migrating from CVS to SVN I already reduced direct references to CVS/SVN, and expecting a migration to GIT in this decade let me take the opportunity to generalize/simplify things a bit further. Committed. (Perhaps we should also start refering to the more general term "commit" vs "check in"

[PATCH, testsuite] Skip new charset tests on Darwin8-10.

Hi, These earlier Darwin versions have “FP_≈” inside a comment in architecture/{ppc,i386}/math.h, which is included by math.h which causes the tests to fail. The intent of the tests (i.e. to ensure that the library itself does not emit non-ascii) is covered by other platforms, including later

[committed] Fix 2 comment typos

Hi! While is't is a contraction of "is it", it doesn't make sense in either of these spots and I believe isn't was meant there instead. Committed as obvious to trunk. 2019-01-12 Jakub Jelinek * tree-ssa-loop-ivopts.c (find_inv_vars): Fix a comment typo. * c-typeck.c (convert

Re: ISO_Fortran_binding patch

On Sat, Jan 12, 2019 at 03:28:02PM +, Paul Richard Thomas wrote: > Hi Thomas, > > Committed as revision 267881. I removed the duplicate include file and > added some documentation, as suggested. > > Many thanks for all the help > Paul, I'm seeing the following failures. Note, I have my un

Re: ISO_Fortran_binding patch

On Sat, Jan 12, 2019 at 09:10:27AM -0800, Steve Kargl wrote: > On Sat, Jan 12, 2019 at 03:28:02PM +, Paul Richard Thomas wrote: > > Hi Thomas, > > > > Committed as revision 267881. I removed the duplicate include file and > > added some documentation, as suggested. > > > > Many thanks for all

Re: [wwwdocs] Add __cpp_* feature macros to C++20 entries + other changes that have those in projects/cxx_status.html

On Sat, Jan 12, 2019 at 04:03:57PM +0100, Gerald Pfeifer wrote: > On Sat, 12 Jan 2019, Jakub Jelinek wrote: > > Ok for wwwdocs (or do you suggest something different for the P0941R2 > > imlementation status)? > > I think that was a question to Jason and Jonathan wrt contents? Yeah. > Markup-wise

Re: ISO_Fortran_binding patch

Hi Steve, Many thanks for the heads up. I had seen similar problems with the the second testcase and I thought that I had fixed them. I will delete them from the tree and will do more work to fix the problem(s). Cheers Paul On Sat, 12 Jan 2019 at 17:17, Steve Kargl wrote: > > On Sat, Jan 12, 2

Set inline-unit-growth to 40

Hello, this patch sets inline-unit-growth to 40. The performance changes are - Firefox, LTO https://treeherder.mozilla.org/perf.html#/compare?originalProject=try&originalRevision=f7bd026e1a931b9a284d1c85c2577a72dd592820&newProject=try&newRevision=74889968abcc688b8d161863566ed273c0401ee4&framewo

Re: ISO_Fortran_binding patch

Done as revision 267884. Thanks again. Paul On Sat, 12 Jan 2019 at 18:29, Paul Richard Thomas wrote: > > Hi Steve, > > Many thanks for the heads up. I had seen similar problems with the the > second testcase and I thought that I had fixed them. I will delete > them from the tree and will do mor

Add split_stack support for GNU/Hurd

Hello, I order to prepare for inclusion of gccgo to GNU/Hurd split stack support has been proposed, and in reality this patch has been applied to Debian since gcc-6. Please commit this patch. Late on updated patches for gccgo in gcc-8 and gcc-9? will be reviewed and committed with the aid of Ian

Re: [PATCH, d] Add README for process contributing to dmd and phobos

On Thu, 10 Jan 2019 at 18:26, Joseph Myers wrote: > > On Thu, 10 Jan 2019, Iain Buclaw wrote: > > > Hi, > > > > Joseph made mention that there isn't a readme documenting where > > changes to d/dmd, libphobos/libdruntime, and libphobos/src should go. > > > > I hope this clears things up. OK for tr

[patch, fortran] Fix the rest of PR 59345

Hello world, this patch fixes the rest of the PR by making sure we do not pack/unpack for function results which are either allocatable or explicit shape arrays. Regression-tested. OK for trunk? Regards Thomas 2019-01-12 Thomas Koenig PR fortran/59345 * trans-array

Re: Add split_stack support for GNU/Hurd

On Sat, Jan 12, 2019 at 11:18 AM Svante Signell wrote: > > I order to prepare for inclusion of gccgo to GNU/Hurd split stack support has > been proposed, and in reality this patch has been applied to Debian since > gcc-6. > > Please commit this patch. Late on updated patches for gccgo in gcc-8 an

[PATCH 3/9] [nvptx] Enable large vectors -- test-cases

Add various test-cases with vector length 128. 2018-12-17 Tom de Vries * testsuite/libgomp.oacc-c-c++-common/vector-length-128-7.c: New test. * testsuite/libgomp.oacc-c-c++-common/vector-length-128-4.c: New test. * testsuite/libgomp.oacc-c-c++-common/vector-length-128-6

[PATCH 1/9] [nvptx] Enable large vectors

Allow vector_length clauses to accept values larger than warp size. Note that this does not enable setting vector_length to values larger than warp size using -fopenacc-dim. 2018-12-17 Tom de Vries * config/nvptx/nvptx.c (nvptx_goacc_validate_dims): Take larger vector lengths

[PATCH 2/9] [nvptx] Update insufficient launch message for variable vector_length

Update message in nvptx libgomp plugin about insufficient resources to launch kernel, to accommodate for the fact the vector_length can now be variable. 19-01-08 Tom de Vries * plugin/plugin-nvptx.c (nvptx_exec): Update insufficient hardware resources diagnostic. --- libgomp/p

[PATCH 0/9] [COVER-LETTER, nvptx] Add support for warp-multiple openacc vector length

I. Current state The current openacc implementation sets vector length to warp-size. There are two aspects that need to be implemented for an openacc implementation to work: communication and synchronization. Synchronization is needed at the end of worker and vector loops. Communication is need

[PATCH 8/9] [nvptx] Enable setting vector length using -fopenacc-dim

Enable setting vector length using -fopenacc-dim, f.i. -fopenacc-dim=::128. 2019-01-12 Tom de Vries * config/nvptx/nvptx.c (nvptx_goacc_validate_dims_1): Alow setting vector length using -fopenacc-dim. * plugin/plugin-nvptx.c (nvptx_exec): Update error message. --- gc

[PATCH 7/9] [nvptx] Add vector_length 64 test-cases

Add some test-cases using vector_length 64. 2019-01-10 Tom de Vries * testsuite/libgomp.oacc-c-c++-common/vector-length-64-1.c: New test. * testsuite/libgomp.oacc-c-c++-common/vector-length-64-2.c: New test. * testsuite/libgomp.oacc-c-c++-common/vector-length-64-3.c: Ne

[PATCH 5/9] [nvptx] Don't emit barriers for empty loops -- test-cases

Add test-cases for PR85381. 2018-12-17 Tom de Vries PR target/85381 * testsuite/libgomp.oacc-c-c++-common/pr85381-5.c: New test. * testsuite/libgomp.oacc-c-c++-common/pr85381.c: New test. --- .../libgomp.oacc-c-c++-common/pr85381-5.c | 24 +

[PATCH 6/9] [nvptx] Force vl32 if calling vector-partitionable routines -- test-cases

Add test-cases for "[nvptx] Force vl32 if calling vector-partitionable routines". 2018-12-17 Tom de Vries PR target/85486 * testsuite/libgomp.oacc-c-c++-common/pr85486-3.c: New test. * testsuite/libgomp.oacc-c-c++-common/pr85486.c: New test. --- .../libgomp.oacc-c-c++-

[PATCH 4/9] [nvptx] Enable large vectors -- reduction testcases

Add various reduction test-cases with vector length 128. 2018-12-17 Tom de Vries * testsuite/libgomp.oacc-c-c++-common/vred2d-128.c: New test. * testsuite/libgomp.oacc-fortran/gemm.f90: New test. * testsuite/libgomp.oacc-c-c++-common/vector-length-128-10.c: New test. --

[PATCH 9/9] [nvptx] Enable setting vector length using -fopenacc-dim -- testcases

Add some test-cases that set vector length using -fopenacc-dim. 2019-01-12 Tom de Vries * testsuite/libgomp.oacc-c-c++-common/pr85486-2.c: New test. * testsuite/libgomp.oacc-c-c++-common/vector-length-128-2.c: New test. * testsuite/libgomp.oacc-c-c++-common/vector-lengt

[patch,libgfortran] PR88776 Namelist read from stdin: loss of data

Hi all, As stated in the PR, the problem turns out to be an ungraceful return after an error. Most namelist errors go through nml_err_ret, The one I am removing did not and in the unique case of UNIT=5 after the error it falls through and hits some code which modifies pointers to the namelist

Re: [patch, fortran] Fix the rest of PR 59345

On Sat, Jan 12, 2019 at 10:09:14PM +0100, Thomas Koenig wrote: > Hello world, > > this patch fixes the rest of the PR by making sure we do not > pack/unpack for function results which are either allocatable > or explicit shape arrays. > > Regression-tested. OK for trunk? > OK. -- Steve

[PATCH] C-family: Replace "may may" with "may" in warning message

gcc/c-family/ * c-warn.c (warn_for_address_or_pointer_of_packed_member): Replace "may may" with "may" in warning message. gcc/c-family/testsuite/ * gcc.dg/pr51628-20.c: Updated. * gcc.dg/pr51628-21.c: Likewise. * gcc.dg/pr51628-25.c: Likewise. --- gcc/c-f

[PATCH] C-amily: Properly strip NOP_EXPR

gcc/c-family/ PR c/51628 * c-warn.c (warn_for_address_or_pointer_of_packed_member): Move NOP_EXPR check to ... (check_and_warn_address_of_packed_member): Here. gcc/testsuite/ PR c/51628 * c-c++-common/pr51628-33.c: New test. --- gcc/c-family/c-war