The test currently is a bunch of checks (implemented using BUG_ON()) that can be built into the kernel or as a module.
Convert it to a KUnit test, which can also run in both modes. From a user's perspective, this change adds a CONFIG_KUNIT=y dep and changes the output format of the test [1]. The test itself is the same.
This hopefully makes the test easier to run and more consistent with similar tests in lib/. Since it has no dependencies, it can be run without explicitly setting up a .kunitconfig via $ ./tools/testing/kunit/kunit.py run atomic ... [13:53:44] Starting KUnit Kernel (1/1)... [13:53:44] ============================================================ [13:53:47] =================== atomic (2 subtests) ==================== [13:53:47] [PASSED] test_atomic [13:53:47] [PASSED] test_atomic64 [13:53:47] ===================== [PASSED] atomic ====================== [13:53:47] ============================================================ [13:53:47] Testing complete. Passed: 2, Failed: 0, Crashed: 0, Skipped: 0, Errors: 0 [13:53:47] Elapsed time: 13.902s total, 1.629s configuring, 9.331s building, 2.852s running
It can be run on ARCH=x86_64 (and others) via: $ ./tools/testing/kunit/kunit.py run --arch=x86_64 atomic
The message about which platform the test ran on won't show up in kunit.py, but still gets printed out in dmesg, e.g.
TAP version 14 1..1 # Subtest: atomic 1..2 ok 1 - test_atomic ok 2 - test_atomic64 # atomic: ran on x86-64 platform with CX8 and with SSE # atomic: pass:2 fail:0 skip:0 total:2 # Totals: pass:2 fail:0 skip:0 total:2 ok 1 - atomic
Signed-off-by: Daniel Latypov dlatypov@google.com
[1] https://www.kernel.org/doc/html/latest/dev-tools/ktap.html --- Meta: 1. this patch applies on top of the kunit branch, https://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest.git/?h...
2. checkpatch complains about aligning with parens, but it wants me to indent the `#ifdef CONFIG_X86_64` which seems inappropriate in context.
3. this file doesn't seem to have a clear maintainer, so I assume this conversion is fine to go through the kunit branch. The only observable differences are the new CONFIG_KUNIT=y dep and more standardized (KTAP) output format. --- lib/Kconfig.debug | 4 +- lib/atomic64_test.c | 107 +++++++++++++++++++++----------------------- 2 files changed, 55 insertions(+), 56 deletions(-)
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 075cd25363ac..4cf8d5feda0a 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -2223,7 +2223,9 @@ config PERCPU_TEST If unsure, say N.
config ATOMIC64_SELFTEST - tristate "Perform an atomic64_t self-test" + tristate "Perform an atomic64_t self-test" if !KUNIT_ALL_TESTS + depends on KUNIT + default KUNIT_ALL_TESTS help Enable this option to test the atomic64_t functions at boot or at module load time. diff --git a/lib/atomic64_test.c b/lib/atomic64_test.c index d9d170238165..46cb0130f8d0 100644 --- a/lib/atomic64_test.c +++ b/lib/atomic64_test.c @@ -5,13 +5,9 @@ * Copyright © 2010 Luca Barbieri */
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt +#include <kunit/test.h>
-#include <linux/init.h> -#include <linux/bug.h> -#include <linux/kernel.h> #include <linux/atomic.h> -#include <linux/module.h>
#ifdef CONFIG_X86 #include <asm/cpufeature.h> /* for boot_cpu_has below */ @@ -23,9 +19,7 @@ do { \ r = v0; \ atomic##bit##_##op(val, &v); \ r c_op val; \ - WARN(atomic##bit##_read(&v) != r, "%Lx != %Lx\n", \ - (unsigned long long)atomic##bit##_read(&v), \ - (unsigned long long)r); \ + KUNIT_ASSERT_EQ(test, atomic##bit##_read(&v), r); \ } while (0)
/* @@ -46,8 +40,8 @@ do { \ atomic##bit##_set(&v, v0); \ r = v0; \ r c_op val; \ - BUG_ON(atomic##bit##_##op(val, &v) != r); \ - BUG_ON(atomic##bit##_read(&v) != r); \ + KUNIT_ASSERT_EQ(test, atomic##bit##_##op(val, &v), r); \ + KUNIT_ASSERT_EQ(test, atomic##bit##_read(&v), r); \ } while (0)
#define TEST_FETCH(bit, op, c_op, val) \ @@ -55,8 +49,8 @@ do { \ atomic##bit##_set(&v, v0); \ r = v0; \ r c_op val; \ - BUG_ON(atomic##bit##_##op(val, &v) != v0); \ - BUG_ON(atomic##bit##_read(&v) != r); \ + KUNIT_ASSERT_EQ(test, atomic##bit##_##op(val, &v), v0); \ + KUNIT_ASSERT_EQ(test, atomic##bit##_read(&v), r); \ } while (0)
#define RETURN_FAMILY_TEST(bit, op, c_op, val) \ @@ -72,8 +66,8 @@ do { \ #define TEST_ARGS(bit, op, init, ret, expect, args...) \ do { \ atomic##bit##_set(&v, init); \ - BUG_ON(atomic##bit##_##op(&v, ##args) != ret); \ - BUG_ON(atomic##bit##_read(&v) != expect); \ + KUNIT_ASSERT_EQ(test, atomic##bit##_##op(&v, ##args), ret);\ + KUNIT_ASSERT_EQ(test, atomic##bit##_read(&v), expect); \ } while (0)
#define XCHG_FAMILY_TEST(bit, init, new) \ @@ -101,7 +95,7 @@ do { \ i, (i) - one, (i) - one); \ } while (0)
-static __init void test_atomic(void) +static void test_atomic(struct kunit *test) { int v0 = 0xaaa31337; int v1 = 0xdeadbeef; @@ -144,7 +138,7 @@ static __init void test_atomic(void) }
#define INIT(c) do { atomic64_set(&v, c); r = c; } while (0) -static __init void test_atomic64(void) +static void test_atomic64(struct kunit *test) { long long v0 = 0xaaa31337c001d00dLL; long long v1 = 0xdeadbeefdeafcafeLL; @@ -156,12 +150,12 @@ static __init void test_atomic64(void)
atomic64_t v = ATOMIC64_INIT(v0); long long r = v0; - BUG_ON(v.counter != r); + KUNIT_ASSERT_EQ(test, v.counter, r);
atomic64_set(&v, v1); r = v1; - BUG_ON(v.counter != r); - BUG_ON(atomic64_read(&v) != r); + KUNIT_ASSERT_EQ(test, v.counter, r); + KUNIT_ASSERT_EQ(test, atomic64_read(&v), r);
TEST(64, add, +=, onestwos); TEST(64, add, +=, -one); @@ -190,12 +184,12 @@ static __init void test_atomic64(void) INIT(v0); atomic64_inc(&v); r += one; - BUG_ON(v.counter != r); + KUNIT_ASSERT_EQ(test, v.counter, r);
INIT(v0); atomic64_dec(&v); r -= one; - BUG_ON(v.counter != r); + KUNIT_ASSERT_EQ(test, v.counter, r);
INC_RETURN_FAMILY_TEST(64, v0); DEC_RETURN_FAMILY_TEST(64, v0); @@ -204,73 +198,76 @@ static __init void test_atomic64(void) CMPXCHG_FAMILY_TEST(64, v0, v1, v2);
INIT(v0); - BUG_ON(atomic64_add_unless(&v, one, v0)); - BUG_ON(v.counter != r); + KUNIT_ASSERT_FALSE(test, atomic64_add_unless(&v, one, v0)); + KUNIT_ASSERT_EQ(test, v.counter, r);
INIT(v0); - BUG_ON(!atomic64_add_unless(&v, one, v1)); + KUNIT_ASSERT_TRUE(test, atomic64_add_unless(&v, one, v1)); r += one; - BUG_ON(v.counter != r); + KUNIT_ASSERT_EQ(test, v.counter, r);
INIT(onestwos); - BUG_ON(atomic64_dec_if_positive(&v) != (onestwos - 1)); + KUNIT_ASSERT_EQ(test, atomic64_dec_if_positive(&v), (onestwos - 1)); r -= one; - BUG_ON(v.counter != r); + KUNIT_ASSERT_EQ(test, v.counter, r);
INIT(0); - BUG_ON(atomic64_dec_if_positive(&v) != -one); - BUG_ON(v.counter != r); + KUNIT_ASSERT_EQ(test, atomic64_dec_if_positive(&v), -one); + KUNIT_ASSERT_EQ(test, v.counter, r);
INIT(-one); - BUG_ON(atomic64_dec_if_positive(&v) != (-one - one)); - BUG_ON(v.counter != r); + KUNIT_ASSERT_EQ(test, atomic64_dec_if_positive(&v), (-one - one)); + KUNIT_ASSERT_EQ(test, v.counter, r);
INIT(onestwos); - BUG_ON(!atomic64_inc_not_zero(&v)); + KUNIT_ASSERT_TRUE(test, atomic64_inc_not_zero(&v)); r += one; - BUG_ON(v.counter != r); + KUNIT_ASSERT_EQ(test, v.counter, r);
INIT(0); - BUG_ON(atomic64_inc_not_zero(&v)); - BUG_ON(v.counter != r); + KUNIT_ASSERT_FALSE(test, atomic64_inc_not_zero(&v)); + KUNIT_ASSERT_EQ(test, v.counter, r);
INIT(-one); - BUG_ON(!atomic64_inc_not_zero(&v)); + KUNIT_ASSERT_TRUE(test, atomic64_inc_not_zero(&v)); r += one; - BUG_ON(v.counter != r); + KUNIT_ASSERT_EQ(test, v.counter, r);
/* Confirm the return value fits in an int, even if the value doesn't */ INIT(v3); + r_int = atomic64_inc_not_zero(&v); - BUG_ON(!r_int); + KUNIT_ASSERT_NE(test, r_int, 0); }
-static __init int test_atomics_init(void) -{ - test_atomic(); - test_atomic64(); +static struct kunit_case atomic_test_cases[] = { + KUNIT_CASE(test_atomic), + KUNIT_CASE(test_atomic64), + {}, +};
+static void atomic_suite_exit(struct kunit_suite *suite) +{ #ifdef CONFIG_X86 - pr_info("passed for %s platform %s CX8 and %s SSE\n", + kunit_info(suite, "ran on %s platform %s CX8 and %s SSE\n", #ifdef CONFIG_X86_64 - "x86-64", + "x86-64", #elif defined(CONFIG_X86_CMPXCHG64) - "i586+", + "i586+", #else - "i386+", + "i386+", #endif - boot_cpu_has(X86_FEATURE_CX8) ? "with" : "without", - boot_cpu_has(X86_FEATURE_XMM) ? "with" : "without"); -#else - pr_info("passed\n"); + boot_cpu_has(X86_FEATURE_CX8) ? "with" : "without", + boot_cpu_has(X86_FEATURE_XMM) ? "with" : "without"); #endif - - return 0; }
-static __exit void test_atomics_exit(void) {} +static struct kunit_suite atomic_test_suite = { + .name = "atomic", + .test_cases = atomic_test_cases, + .suite_exit = atomic_suite_exit, +};
-module_init(test_atomics_init); -module_exit(test_atomics_exit); +kunit_test_suites(&atomic_test_suite);
MODULE_LICENSE("GPL");
base-commit: 38289a26e1b8a37755f3e07056ca416c1ee2a2e8
On Tue, May 3, 2022 at 3:23 AM Daniel Latypov dlatypov@google.com wrote:
The test currently is a bunch of checks (implemented using BUG_ON()) that can be built into the kernel or as a module.
Convert it to a KUnit test, which can also run in both modes. From a user's perspective, this change adds a CONFIG_KUNIT=y dep and changes the output format of the test [1]. The test itself is the same.
This hopefully makes the test easier to run and more consistent with similar tests in lib/. Since it has no dependencies, it can be run without explicitly setting up a .kunitconfig via $ ./tools/testing/kunit/kunit.py run atomic ... [13:53:44] Starting KUnit Kernel (1/1)... [13:53:44] ============================================================ [13:53:47] =================== atomic (2 subtests) ==================== [13:53:47] [PASSED] test_atomic [13:53:47] [PASSED] test_atomic64 [13:53:47] ===================== [PASSED] atomic ====================== [13:53:47] ============================================================ [13:53:47] Testing complete. Passed: 2, Failed: 0, Crashed: 0, Skipped: 0, Errors: 0 [13:53:47] Elapsed time: 13.902s total, 1.629s configuring, 9.331s building, 2.852s running
It can be run on ARCH=x86_64 (and others) via: $ ./tools/testing/kunit/kunit.py run --arch=x86_64 atomic
The message about which platform the test ran on won't show up in kunit.py, but still gets printed out in dmesg, e.g.
TAP version 14 1..1 # Subtest: atomic 1..2 ok 1 - test_atomic ok 2 - test_atomic64 # atomic: ran on x86-64 platform with CX8 and with SSE # atomic: pass:2 fail:0 skip:0 total:2 # Totals: pass:2 fail:0 skip:0 total:2 ok 1 - atomic
Signed-off-by: Daniel Latypov dlatypov@google.com
[1] https://www.kernel.org/doc/html/latest/dev-tools/ktap.html
This looks good to me. It's nice to see these tests being standardised.
Meta:
- this patch applies on top of the kunit branch,
https://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest.git/?h...
(To anyone else trying it: make sure you update this first, as the prerequisites only just landed.)
- checkpatch complains about aligning with parens, but it wants me to
indent the `#ifdef CONFIG_X86_64` which seems inappropriate in context.
Huh... checkpatch isn't showing any such warning on my system.
- this file doesn't seem to have a clear maintainer, so I assume this
conversion is fine to go through the kunit branch. The only observable differences are the new CONFIG_KUNIT=y dep and more standardized (KTAP) output format.
I'm happy to vouch for this not breaking anything, even though I'm definitely not an expert on the atomics code.
Reviewed-by: David Gow davidgow@google.com
-- David
lib/Kconfig.debug | 4 +- lib/atomic64_test.c | 107 +++++++++++++++++++++----------------------- 2 files changed, 55 insertions(+), 56 deletions(-)
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug index 075cd25363ac..4cf8d5feda0a 100644 --- a/lib/Kconfig.debug +++ b/lib/Kconfig.debug @@ -2223,7 +2223,9 @@ config PERCPU_TEST If unsure, say N.
config ATOMIC64_SELFTEST
tristate "Perform an atomic64_t self-test"
tristate "Perform an atomic64_t self-test" if !KUNIT_ALL_TESTSdepends on KUNITdefault KUNIT_ALL_TESTS help Enable this option to test the atomic64_t functions at boot or at module load time.diff --git a/lib/atomic64_test.c b/lib/atomic64_test.c index d9d170238165..46cb0130f8d0 100644 --- a/lib/atomic64_test.c +++ b/lib/atomic64_test.c @@ -5,13 +5,9 @@
- Copyright © 2010 Luca Barbieri
*/
-#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt +#include <kunit/test.h>
-#include <linux/init.h> -#include <linux/bug.h> -#include <linux/kernel.h> #include <linux/atomic.h> -#include <linux/module.h>
#ifdef CONFIG_X86 #include <asm/cpufeature.h> /* for boot_cpu_has below */ @@ -23,9 +19,7 @@ do { \ r = v0; \ atomic##bit##_##op(val, &v); \ r c_op val; \
WARN(atomic##bit##_read(&v) != r, "%Lx != %Lx\n", \(unsigned long long)atomic##bit##_read(&v), \(unsigned long long)r); \
KUNIT_ASSERT_EQ(test, atomic##bit##_read(&v), r); \} while (0)
/* @@ -46,8 +40,8 @@ do { \ atomic##bit##_set(&v, v0); \ r = v0; \ r c_op val; \
BUG_ON(atomic##bit##_##op(val, &v) != r); \BUG_ON(atomic##bit##_read(&v) != r); \
KUNIT_ASSERT_EQ(test, atomic##bit##_##op(val, &v), r); \KUNIT_ASSERT_EQ(test, atomic##bit##_read(&v), r); \} while (0)
#define TEST_FETCH(bit, op, c_op, val) \ @@ -55,8 +49,8 @@ do { \ atomic##bit##_set(&v, v0); \ r = v0; \ r c_op val; \
BUG_ON(atomic##bit##_##op(val, &v) != v0); \BUG_ON(atomic##bit##_read(&v) != r); \
KUNIT_ASSERT_EQ(test, atomic##bit##_##op(val, &v), v0); \KUNIT_ASSERT_EQ(test, atomic##bit##_read(&v), r); \} while (0)
#define RETURN_FAMILY_TEST(bit, op, c_op, val) \ @@ -72,8 +66,8 @@ do { \ #define TEST_ARGS(bit, op, init, ret, expect, args...) \ do { \ atomic##bit##_set(&v, init); \
BUG_ON(atomic##bit##_##op(&v, ##args) != ret); \BUG_ON(atomic##bit##_read(&v) != expect); \
KUNIT_ASSERT_EQ(test, atomic##bit##_##op(&v, ##args), ret);\KUNIT_ASSERT_EQ(test, atomic##bit##_read(&v), expect); \} while (0)
#define XCHG_FAMILY_TEST(bit, init, new) \ @@ -101,7 +95,7 @@ do { \ i, (i) - one, (i) - one); \ } while (0)
-static __init void test_atomic(void) +static void test_atomic(struct kunit *test) { int v0 = 0xaaa31337; int v1 = 0xdeadbeef; @@ -144,7 +138,7 @@ static __init void test_atomic(void) }
#define INIT(c) do { atomic64_set(&v, c); r = c; } while (0) -static __init void test_atomic64(void) +static void test_atomic64(struct kunit *test) { long long v0 = 0xaaa31337c001d00dLL; long long v1 = 0xdeadbeefdeafcafeLL; @@ -156,12 +150,12 @@ static __init void test_atomic64(void)
atomic64_t v = ATOMIC64_INIT(v0); long long r = v0;
BUG_ON(v.counter != r);
KUNIT_ASSERT_EQ(test, v.counter, r); atomic64_set(&v, v1); r = v1;
BUG_ON(v.counter != r);BUG_ON(atomic64_read(&v) != r);
KUNIT_ASSERT_EQ(test, v.counter, r);KUNIT_ASSERT_EQ(test, atomic64_read(&v), r); TEST(64, add, +=, onestwos); TEST(64, add, +=, -one);@@ -190,12 +184,12 @@ static __init void test_atomic64(void) INIT(v0); atomic64_inc(&v); r += one;
BUG_ON(v.counter != r);
KUNIT_ASSERT_EQ(test, v.counter, r); INIT(v0); atomic64_dec(&v); r -= one;
BUG_ON(v.counter != r);
KUNIT_ASSERT_EQ(test, v.counter, r); INC_RETURN_FAMILY_TEST(64, v0); DEC_RETURN_FAMILY_TEST(64, v0);@@ -204,73 +198,76 @@ static __init void test_atomic64(void) CMPXCHG_FAMILY_TEST(64, v0, v1, v2);
INIT(v0);
BUG_ON(atomic64_add_unless(&v, one, v0));BUG_ON(v.counter != r);
KUNIT_ASSERT_FALSE(test, atomic64_add_unless(&v, one, v0));KUNIT_ASSERT_EQ(test, v.counter, r); INIT(v0);
BUG_ON(!atomic64_add_unless(&v, one, v1));
KUNIT_ASSERT_TRUE(test, atomic64_add_unless(&v, one, v1)); r += one;
BUG_ON(v.counter != r);
KUNIT_ASSERT_EQ(test, v.counter, r); INIT(onestwos);
BUG_ON(atomic64_dec_if_positive(&v) != (onestwos - 1));
KUNIT_ASSERT_EQ(test, atomic64_dec_if_positive(&v), (onestwos - 1)); r -= one;
BUG_ON(v.counter != r);
KUNIT_ASSERT_EQ(test, v.counter, r); INIT(0);
BUG_ON(atomic64_dec_if_positive(&v) != -one);BUG_ON(v.counter != r);
KUNIT_ASSERT_EQ(test, atomic64_dec_if_positive(&v), -one);KUNIT_ASSERT_EQ(test, v.counter, r); INIT(-one);
BUG_ON(atomic64_dec_if_positive(&v) != (-one - one));BUG_ON(v.counter != r);
KUNIT_ASSERT_EQ(test, atomic64_dec_if_positive(&v), (-one - one));KUNIT_ASSERT_EQ(test, v.counter, r); INIT(onestwos);
BUG_ON(!atomic64_inc_not_zero(&v));
KUNIT_ASSERT_TRUE(test, atomic64_inc_not_zero(&v)); r += one;
BUG_ON(v.counter != r);
KUNIT_ASSERT_EQ(test, v.counter, r); INIT(0);
BUG_ON(atomic64_inc_not_zero(&v));BUG_ON(v.counter != r);
KUNIT_ASSERT_FALSE(test, atomic64_inc_not_zero(&v));KUNIT_ASSERT_EQ(test, v.counter, r); INIT(-one);
BUG_ON(!atomic64_inc_not_zero(&v));
KUNIT_ASSERT_TRUE(test, atomic64_inc_not_zero(&v)); r += one;
BUG_ON(v.counter != r);
KUNIT_ASSERT_EQ(test, v.counter, r); /* Confirm the return value fits in an int, even if the value doesn't */ INIT(v3);r_int = atomic64_inc_not_zero(&v);
BUG_ON(!r_int);
KUNIT_ASSERT_NE(test, r_int, 0);}
-static __init int test_atomics_init(void) -{
test_atomic();test_atomic64();+static struct kunit_case atomic_test_cases[] = {
KUNIT_CASE(test_atomic),KUNIT_CASE(test_atomic64),{},+};
+static void atomic_suite_exit(struct kunit_suite *suite) +{ #ifdef CONFIG_X86
pr_info("passed for %s platform %s CX8 and %s SSE\n",
kunit_info(suite, "ran on %s platform %s CX8 and %s SSE\n",#ifdef CONFIG_X86_64
"x86-64",
"x86-64",#elif defined(CONFIG_X86_CMPXCHG64)
"i586+",
"i586+",#else
"i386+",
"i386+",#endif
boot_cpu_has(X86_FEATURE_CX8) ? "with" : "without",boot_cpu_has(X86_FEATURE_XMM) ? "with" : "without");-#else
pr_info("passed\n");
boot_cpu_has(X86_FEATURE_CX8) ? "with" : "without",boot_cpu_has(X86_FEATURE_XMM) ? "with" : "without");#endif
return 0;}
-static __exit void test_atomics_exit(void) {} +static struct kunit_suite atomic_test_suite = {
.name = "atomic",.test_cases = atomic_test_cases,.suite_exit = atomic_suite_exit,+};
-module_init(test_atomics_init); -module_exit(test_atomics_exit); +kunit_test_suites(&atomic_test_suite);
MODULE_LICENSE("GPL");
base-commit: 38289a26e1b8a37755f3e07056ca416c1ee2a2e8
2.36.0.464.gb9c8b46e94-goog
On Mon, May 2, 2022 at 3:23 PM Daniel Latypov dlatypov@google.com wrote:
The test currently is a bunch of checks (implemented using BUG_ON()) that can be built into the kernel or as a module.
Convert it to a KUnit test, which can also run in both modes. From a user's perspective, this change adds a CONFIG_KUNIT=y dep and changes the output format of the test [1]. The test itself is the same.
This hopefully makes the test easier to run and more consistent with similar tests in lib/. Since it has no dependencies, it can be run without explicitly setting up a .kunitconfig via $ ./tools/testing/kunit/kunit.py run atomic ... [13:53:44] Starting KUnit Kernel (1/1)... [13:53:44] ============================================================ [13:53:47] =================== atomic (2 subtests) ==================== [13:53:47] [PASSED] test_atomic [13:53:47] [PASSED] test_atomic64 [13:53:47] ===================== [PASSED] atomic ====================== [13:53:47] ============================================================ [13:53:47] Testing complete. Passed: 2, Failed: 0, Crashed: 0, Skipped: 0, Errors: 0 [13:53:47] Elapsed time: 13.902s total, 1.629s configuring, 9.331s building, 2.852s running
It can be run on ARCH=x86_64 (and others) via: $ ./tools/testing/kunit/kunit.py run --arch=x86_64 atomic
The message about which platform the test ran on won't show up in kunit.py, but still gets printed out in dmesg, e.g.
TAP version 14 1..1 # Subtest: atomic 1..2 ok 1 - test_atomic ok 2 - test_atomic64 # atomic: ran on x86-64 platform with CX8 and with SSE # atomic: pass:2 fail:0 skip:0 total:2 # Totals: pass:2 fail:0 skip:0 total:2 ok 1 - atomic
Signed-off-by: Daniel Latypov dlatypov@google.com
I am also not an expert, but it looks good to me.
Reviewed-by: Brendan Higgins brendanhiggins@google.com
Daniel Latypov dlatypov@google.com writes:
The test currently is a bunch of checks (implemented using BUG_ON()) that can be built into the kernel or as a module.
Convert it to a KUnit test, which can also run in both modes. From a user's perspective, this change adds a CONFIG_KUNIT=y dep and changes the output format of the test [1]. The test itself is the same.
...
I don't know why I got Cc'ed on this :), but I gave it a quick test anyway.
Seems to work fine on a Power9. I also flipped some of the conditionals to make sure failure is detected correctly.
Tested-by: Michael Ellerman mpe@ellerman.id.au (powerpc)
Meta:
- this patch applies on top of the kunit branch,
https://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest.git/?h...
- checkpatch complains about aligning with parens, but it wants me to
indent the `#ifdef CONFIG_X86_64` which seems inappropriate in context.
- this file doesn't seem to have a clear maintainer, so I assume this
conversion is fine to go through the kunit branch.
I think you want to at least Cc the atomic folks:
ATOMIC INFRASTRUCTURE M: Will Deacon will@kernel.org M: Peter Zijlstra peterz@infradead.org R: Boqun Feng boqun.feng@gmail.com R: Mark Rutland mark.rutland@arm.com L: linux-kernel@vger.kernel.org S: Maintained
cheers
On Thu, May 12, 2022 at 8:06 PM Michael Ellerman mpe@ellerman.id.au wrote:
Daniel Latypov dlatypov@google.com writes:
The test currently is a bunch of checks (implemented using BUG_ON()) that can be built into the kernel or as a module.
Convert it to a KUnit test, which can also run in both modes. From a user's perspective, this change adds a CONFIG_KUNIT=y dep and changes the output format of the test [1]. The test itself is the same.
...
I don't know why I got Cc'ed on this :), but I gave it a quick test anyway.
Seems to work fine on a Power9. I also flipped some of the conditionals to make sure failure is detected correctly.
Tested-by: Michael Ellerman mpe@ellerman.id.au (powerpc)
Thanks!
You were just the last person to have made a real change to this file [1]. The people signing off no commits seemed inconsistent and get_maintainers didn't give anything back.
commit ffba19ccae8d98beb0a17345a0b1ee9e415b23b8 Author: Michael Ellerman mpe@ellerman.id.au Date: Fri Jul 14 14:49:41 2017 -0700
Meta:
- this patch applies on top of the kunit branch,
https://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest.git/?h...
- checkpatch complains about aligning with parens, but it wants me to
indent the `#ifdef CONFIG_X86_64` which seems inappropriate in context.
- this file doesn't seem to have a clear maintainer, so I assume this
conversion is fine to go through the kunit branch.
I think you want to at least Cc the atomic folks:
ATOMIC INFRASTRUCTURE M: Will Deacon will@kernel.org M: Peter Zijlstra peterz@infradead.org R: Boqun Feng boqun.feng@gmail.com R: Mark Rutland mark.rutland@arm.com L: linux-kernel@vger.kernel.org S: Maintained
Ah, thanks for the pointer.
I'll see if I can add F: lib/atomic64_test.c to that entry so this files owners get tracked properly.
Daniel
linux-kselftest-mirror@lists.linaro.org
-
Brendan Higgins -
Daniel Latypov -
David Gow -
Michael Ellerman