Re: [kvm-unit-tests PATCH v2] s390x: selftest: Add prefixes to fix report output (was "s390x: selftest: Fix report output")
From: Thomas Huth <hidden>
Date: 2021-06-21 07:23:56
Also in:
kvm
On 10/06/2021 16.19, Janosch Frank wrote:
quoted hunk ↗ jump to hunk
To make our TAP parser (and me) happy we don't want to have two reports with exactly the same wording so I added in two new prefix pushes. Also moving the code inside of the region of a prefix will give us more data when a problem arises. Signed-off-by: Janosch Frank <frankja@linux.ibm.com> Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com> --- s390x/selftest.c | 26 ++++++++++++++++++-------- 1 file changed, 18 insertions(+), 8 deletions(-)diff --git a/s390x/selftest.c b/s390x/selftest.c index b2fe2e7b..0f099ca0 100644 --- a/s390x/selftest.c +++ b/s390x/selftest.c@@ -40,19 +40,28 @@ static void test_pgm_int(void) static void test_malloc(void) { - int *tmp = malloc(sizeof(int)); - int *tmp2 = malloc(sizeof(int)); + int *tmp, *tmp2; + report_prefix_push("malloc"); + + report_prefix_push("ptr_0"); + tmp = malloc(sizeof(int)); + report((uintptr_t)tmp & 0xf000000000000000ul, "allocated memory"); *tmp = 123456789; + mb(); + report(*tmp == 123456789, "wrote allocated memory"); + report_prefix_pop(); + + report_prefix_push("ptr_1"); + tmp2 = malloc(sizeof(int)); + report((uintptr_t)tmp2 & 0xf000000000000000ul, + "allocated memory"); *tmp2 = 123456789; mb(); + report((*tmp2 == 123456789), "wrote allocated memory"); + report_prefix_pop(); - report((uintptr_t)tmp & 0xf000000000000000ul, "malloc: got vaddr"); - report(*tmp == 123456789, "malloc: access works"); - report((uintptr_t)tmp2 & 0xf000000000000000ul, - "malloc: got 2nd vaddr"); - report((*tmp2 == 123456789), "malloc: access works"); - report(tmp != tmp2, "malloc: addresses differ"); + report(tmp != tmp2, "allocated memory addresses differ"); expect_pgm_int(); configure_dat(0);@@ -62,6 +71,7 @@ static void test_malloc(void) free(tmp); free(tmp2); + report_prefix_pop(); } int main(int argc, char**argv)
Reviewed-by: Thomas Huth <redacted>