[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2021-04-29 Thread vries at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 Tom de Vries changed: What|Removed |Added Resolution|--- |FIXED Target Milestone|---

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2020-10-02 Thread burnus at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 --- Comment #11 from Tobias Burnus --- Cross ref: the submitted patch is at https://gcc.gnu.org/pipermail/gcc-patches/2020-October/555352.html

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2020-10-01 Thread vries at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 --- Comment #10 from Tom de Vries --- Tentative patch: ... diff --git a/gcc/omp-expand.c b/gcc/omp-expand.c index 99cb4f9dda4..034de497390 100644 --- a/gcc/omp-expand.c +++ b/gcc/omp-expand.c @@ -6333,6 +6333,8 @@ expand_omp_simd (struct

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2020-10-01 Thread vries at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 --- Comment #9 from Tom de Vries --- I ran into this again, and did another round of minimizing. This time, I added some buffering around where we write, and check the entire buffer afterwards: ... $ cat

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2020-09-25 Thread burnus at gcc dot gnu.org via Gcc-bugs
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 Tobias Burnus changed: What|Removed |Added CC||burnus at gcc dot gnu.org --- Comment

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2017-12-18 Thread vries at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 --- Comment #7 from Tom de Vries --- A point fix could look like: ... diff --git a/gcc/omp-expand.c b/gcc/omp-expand.c index 3990998..2439a89 100644 --- a/gcc/omp-expand.c +++ b/gcc/omp-expand.c @@ -4339,6 +4339,26 @@ expand_omp_for_static_chunk

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2017-12-18 Thread vries at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 --- Comment #6 from Tom de Vries --- This is essentially: ... diff --git a/gcc/omp-expand.c b/gcc/omp-expand.c index 2c28682..3990998 100644 --- a/gcc/omp-expand.c +++ b/gcc/omp-expand.c @@ -4327,6 +4327,18 @@ expand_omp_simd (struct omp_region

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2017-12-18 Thread vries at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 --- Comment #5 from Tom de Vries --- See PR83457 - "Add fhost-simt-vf" for a way to reproduce this bug on x86_64.

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2017-08-09 Thread amonakov at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 Alexander Monakov changed: What|Removed |Added Status|UNCONFIRMED |NEW Last reconfirmed|

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2017-08-09 Thread vries at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 --- Comment #3 from Tom de Vries --- The core loop looks like this: ... $L4: add.u32 %r31,%r42,-1; cvt.u64.u32 %r63,%r31; shl.b64 %r64,%r63,2; add.u64 %r65,%r75,%r64; ld.u32 %r71,[%r65]; add.u32 %r70,%r71,-4; st.u32 [%r65],%r70;

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2017-08-09 Thread vries at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 --- Comment #2 from Tom de Vries --- nvidia card: Quadro M1200 cuda: 7.5 cuda driver: 375.66

[Bug libgomp/81778] libgomp.c/for-5.c failure on nvptx -- illegal memory access

2017-08-09 Thread vries at gcc dot gnu.org
https://gcc.gnu.org/bugzilla/show_bug.cgi?id=81778 --- Comment #1 from Tom de Vries --- Created attachment 41956 --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=41956=edit GOMP_DEBUG=1 log