diff options
| author | Kyrylo Tkachov <ktkachov@gcc.gnu.org> | 2017-12-19 16:58:22 +0000 |
|---|---|---|
| committer | Kyrylo Tkachov <ktkachov@gcc.gnu.org> | 2017-12-19 16:58:22 +0000 |
| commit | 3a3a8086bed05ffc8d24bb6908e4437bd9e982f7 (patch) | |
| tree | 51e517e0af38cf0dea1b64d47ade9c0235f32191 | |
| parent | 972027748c2aa9f59ef1ab534763b0d6ed37b76d (diff) | |
| download | gcc-3a3a8086bed05ffc8d24bb6908e4437bd9e982f7.zip gcc-3a3a8086bed05ffc8d24bb6908e4437bd9e982f7.tar.gz gcc-3a3a8086bed05ffc8d24bb6908e4437bd9e982f7.tar.bz2 | |
[arm] PR target/82975: Guard against reg_renumber being NULL in arm.h
In this bug we ICE when checking REGNO_OK_FOR_INDEX_P on arm during pre-IRA scheduling.
This is because REGNO_OK_FOR_INDEX_P ends up checking the reg_renumber array.
Before IRA reg_renumber is NULL and thus we segfault.
The fix is to guard the use of reg_renumber in the logic in TEST_REGNO in arm.h.
On aarch64, for example, we also guard against the reg_renumber == NULL case.
This fixes the ICE. I also remove the part of the comment that muses on when reg_renumber
is available as with this patch it should now be safe to use at any point.
Bootstrapped and tested on arm-none-linux-gnueabihf.
PR target/82975
* config/arm/arm.h (TEST_REGNO): Check reg_renumber is set before
accessing it. Adjust comment.
* gcc.dg/pr82975.c: New test.
From-SVN: r255830
| -rw-r--r-- | gcc/ChangeLog | 8 | ||||
| -rw-r--r-- | gcc/config/arm/arm.h | 3 | ||||
| -rw-r--r-- | gcc/testsuite/ChangeLog | 5 | ||||
| -rw-r--r-- | gcc/testsuite/gcc.dg/pr82975.c | 19 |
4 files changed, 33 insertions, 2 deletions
diff --git a/gcc/ChangeLog b/gcc/ChangeLog index 31277b9..76dbd69 100644 --- a/gcc/ChangeLog +++ b/gcc/ChangeLog @@ -1,4 +1,10 @@ -2017-12-18 Jakub Jelinek <jakub@redhat.com> +2017-12-19 Kyrylo Tkachov <kyrylo.tkachov@arm.com> + + PR target/82975 + * config/arm/arm.h (TEST_REGNO): Check reg_renumber is set before + accessing it. Adjust comment. + +2017-12-19 Jakub Jelinek <jakub@redhat.com> PR middle-end/81914 * predict.c (zero_one_minusone): New function. diff --git a/gcc/config/arm/arm.h b/gcc/config/arm/arm.h index ac51412..a482f03 100644 --- a/gcc/config/arm/arm.h +++ b/gcc/config/arm/arm.h @@ -1620,7 +1620,8 @@ enum arm_auto_incmodes has been allocated, which happens in reginfo.c during register allocation. */ #define TEST_REGNO(R, TEST, VALUE) \ - ((R TEST VALUE) || ((unsigned) reg_renumber[R] TEST VALUE)) + ((R TEST VALUE) \ + || (reg_renumber && ((unsigned) reg_renumber[R] TEST VALUE))) /* Don't allow the pc to be used. */ #define ARM_REGNO_OK_FOR_BASE_P(REGNO) \ diff --git a/gcc/testsuite/ChangeLog b/gcc/testsuite/ChangeLog index 200a813..e0eb513 100644 --- a/gcc/testsuite/ChangeLog +++ b/gcc/testsuite/ChangeLog @@ -1,3 +1,8 @@ +2017-12-19 Kyrylo Tkachov <kyrylo.tkachov@arm.com> + + PR target/82975 + * gcc.dg/pr82975.c: New test. + 2017-12-19 Marek Polacek <polacek@redhat.com> PR c++/83489 diff --git a/gcc/testsuite/gcc.dg/pr82975.c b/gcc/testsuite/gcc.dg/pr82975.c new file mode 100644 index 0000000..e6c13bc --- /dev/null +++ b/gcc/testsuite/gcc.dg/pr82975.c @@ -0,0 +1,19 @@ +/* PR target/82975. */ +/* { dg-do compile } */ +/* { dg-options "-mtune=cortex-a57 -fno-sched-pressure -O2" } */ + +typedef __SIZE_TYPE__ size_t; + +struct S1 +{ + char pad1; + char val; + short pad2; +}; + +extern char t[256]; + +void foo (struct S1 a, size_t i) +{ + t[i] = a.val; +} |