[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <20250208162542.fba2bdb0051eeab5344d3f82@linux-foundation.org>
Date: Sat, 8 Feb 2025 16:25:42 -0800
From: Andrew Morton <akpm@...ux-foundation.org>
To: Tamir Duberstein <tamird@...il.com>
Cc: David Gow <davidgow@...gle.com>, Shuah Khan <skhan@...uxfoundation.org>,
Luis Felipe Hernandez <luis.hernandez093@...il.com>, Shuah Khan
<shuah@...nel.org>, linux-kernel@...r.kernel.org,
linux-kselftest@...r.kernel.org, Yu-Chun Lin <eleanor15x@...il.com>
Subject: Re: [PATCH v3 0/2] lib/prime_numbers: convert self-test to KUnit
On Sat, 08 Feb 2025 17:52:28 -0500 Tamir Duberstein <tamird@...il.com> wrote:
> This is a clear example of a unit test.
>
> I tested this using:
>
> $ tools/testing/kunit/kunit.py run --arch arm64 --make_options LLVM=1 math-prime_numbers
>
> On success:
> ; [08:51:41] ============== math-prime_numbers (1 subtest) ==============
> ; [08:51:41] [PASSED] prime_numbers_test
> ; [08:51:41] =============== [PASSED] math-prime_numbers ================
> ; [08:51:41] ============================================================
> ; [08:51:41] Testing complete. Ran 1 tests: passed: 1
>
> On failure:
> ; [08:50:19] ============== math-prime_numbers (1 subtest) ==============
> ; [08:50:19] # prime_numbers_test: ASSERTION FAILED at lib/math/tests/prime_numbers_kunit.c:28
> ; [08:50:19] Expected slow == fast, but
> ; [08:50:19] slow == 0 (0x0)
> ; [08:50:19] fast == 1 (0x1)
> ; [08:50:19] is-prime(2)
> ; [08:50:19] [FAILED] prime_numbers_test
> ; [08:50:19] # module: prime_numbers_kunit
> ; [08:50:19] # math-prime_numbers: primes.{last=61, .sz=64, .primes[]=...x28208a20a08a28ac} = 2-3,5,7,11,13,17,19,23,29,31,37,41,43,47,53,59,61
> ; [08:50:19] =============== [FAILED] math-prime_numbers ================
> ; [08:50:19] ============================================================
> ; [08:50:19] Testing complete. Ran 1 tests: failed: 1
>
I resolved this against "lib/math: add Kunit test suite for gcd()"
(https://lkml.kernel.org/r/20250203075400.3431330-1-eleanor15x@gmail.com)
in the obvious fashion then added this fixup:
From: Andrew Morton <akpm@...ux-foundation.org>
Subject: lib-math-hook-up-tests-makefile-fix
Date: Sat Feb 8 03:33:59 PM PST 2025
don't link gcd_kunit.o twice
Cc: David Gow <davidgow@...gle.com>
Cc: Tamir Duberstein <tamird@...il.com>
Signed-off-by: Andrew Morton <akpm@...ux-foundation.org>
---
lib/math/Makefile | 1 -
1 file changed, 1 deletion(-)
--- a/lib/math/Makefile~lib-math-hook-up-tests-makefile-fix
+++ a/lib/math/Makefile
@@ -8,5 +8,4 @@ obj-$(CONFIG_RATIONAL) += rational.o
obj-$(CONFIG_TEST_DIV64) += test_div64.o
obj-$(CONFIG_TEST_MULDIV64) += test_mul_u64_u64_div_u64.o
obj-$(CONFIG_RATIONAL_KUNIT_TEST) += rational-test.o
-obj-$(CONFIG_GCD_KUNIT_TEST) += tests/gcd_kunit.o
obj-y += tests/
_
and the x86_64 allmodconfig build failed thusly:
lib/math/tests/prime_numbers_kunit.c: In function 'prime_numbers_test':
lib/math/tests/prime_numbers_kunit.c:25:35: error: implicit declaration of function 'slow_is_prime_number'; did you mean 'is_prime_number'? [-Werror=implicit-function-declaration]
25 | const bool slow = slow_is_prime_number(x);
| ^~~~~~~~~~~~~~~~~~~~
| is_prime_number
So, please redo against mm.git's mm-everything branch or its
mm-nonmm-unstable branch and retest carefully.
Powered by blists - more mailing lists