[NFC] Reduce number of run steps in ppc rop-protect test. #139607
Conversation
Test was running both -mcpu=pwrN and -mcpu=powerN compile steps for power N = 8, 9 and 10. Reduce to one run step for each form using power8 for one instance and power 10 for the other.
llvmbot
added
clang
|
@llvm/pr-subscribers-clang-driver Author: Sean Fertile (mandlebug) ChangesTest was running both -mcpu=pwrN and -mcpu=powerN compile steps for power N = 8, 9 and 10. Reduce to one run step for each form using power8 for one instance and power 10 for the other. Full diff: https://github.com/llvm/llvm-project/pull/139607.diff 1 Files Affected:
diff --git a/clang/test/Driver/ppc-mrop-protection-support-check.c b/clang/test/Driver/ppc-mrop-protection-support-check.c
index 50eaef3ed770b..541dc2bf99c3f 100644
--- a/clang/test/Driver/ppc-mrop-protection-support-check.c
+++ b/clang/test/Driver/ppc-mrop-protection-support-check.c
@@ -1,20 +1,10 @@
// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
// RUN: -mcpu=pwr10 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=power10 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=pwr9 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=power9 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=pwr8 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
// RUN: -mcpu=power8 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
// RUN: -mcpu=pwr7 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=NOROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=power7 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=NOROP
#ifdef __ROP_PROTECT__
static_assert(false, "ROP Protect enabled");
|
|
@llvm/pr-subscribers-clang Author: Sean Fertile (mandlebug) ChangesTest was running both -mcpu=pwrN and -mcpu=powerN compile steps for power N = 8, 9 and 10. Reduce to one run step for each form using power8 for one instance and power 10 for the other. Full diff: https://github.com/llvm/llvm-project/pull/139607.diff 1 Files Affected:
diff --git a/clang/test/Driver/ppc-mrop-protection-support-check.c b/clang/test/Driver/ppc-mrop-protection-support-check.c
index 50eaef3ed770b..541dc2bf99c3f 100644
--- a/clang/test/Driver/ppc-mrop-protection-support-check.c
+++ b/clang/test/Driver/ppc-mrop-protection-support-check.c
@@ -1,20 +1,10 @@
// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
// RUN: -mcpu=pwr10 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=power10 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=pwr9 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=power9 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=pwr8 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
// RUN: -mcpu=power8 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP
// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
// RUN: -mcpu=pwr7 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=NOROP
-// RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \
-// RUN: -mcpu=power7 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=NOROP
#ifdef __ROP_PROTECT__
static_assert(false, "ROP Protect enabled");
|
mandlebug
changed the title
| // RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \ | ||
| // RUN: -mcpu=pwr9 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP | ||
| // RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \ | ||
| // RUN: -mcpu=power9 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP |
There was a problem hiding this comment.
Are we removing P9 as it should be covered by both P8/P10? Or did we also mean to keep a P9 line?
There was a problem hiding this comment.
Yeah. We have one test for P8 which is the first CPU version supported, and one test for P10 which is a later CPU that also supports it. Adding a P9 doesn't really add any coverage IMO.
| // RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \ | ||
| // RUN: -mcpu=power9 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP | ||
| // RUN: not %clang -target powerpc64le-unknown-linux-gnu -fsyntax-only \ | ||
| // RUN: -mcpu=pwr8 -mrop-protect %s 2>&1 | FileCheck %s --check-prefix=HASROP |
There was a problem hiding this comment.
I don't feel super strongly either way, but should we keep the pwr version for consistency?
There was a problem hiding this comment.
The P10 run step uses the pwr format. I think one run step for each mcpu format is adequate. Realistically we should have other coverage that ensures that the 2 formats sets up the same environment and this test could use just one format either way, but since we have a run step for a target greater then 8 anyway I thought to test one of each.
4 participants
Test was running both -mcpu=pwrN and -mcpu=powerN compile steps for power N = 8, 9 and 10. Reduce to one run step for each form using power8 for one instance and power 10 for the other.