Hi Simon,
looking at this patch some more I have another question, see below.
-----"Simon Glass" sjg@chromium.org schrieb: -----
This function needs to be used by sandbox for tests. Move it into the generic directory.
Signed-off-by: Simon Glass sjg@chromium.org Reviewed-by: Bin Meng bmeng.cn@gmail.com
[snip]
@@ -81,3 +82,30 @@ static int dm_test_acpi_create_dmar(struct unit_test_state *uts) return 0; } DM_TEST(dm_test_acpi_create_dmar, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
+/* Test acpi_fill_header() */ +static int dm_test_acpi_fill_header(struct unit_test_state *uts) +{
- struct acpi_table_header hdr;
- /* Make sure these 5 fields are not changed */
- hdr.length = 0x11;
- hdr.revision = 0x22;
- hdr.checksum = 0x33;
- hdr.aslc_revision = 0x44;
- acpi_fill_header(&hdr, "ABCD");
- ut_assertok(memcmp("ABCD", hdr.signature, sizeof(hdr.signature)));
Could ut_asserteq_mem() be used here? The output would be a little more verbose.
With ut_assertok(memcmp()): test/dm/acpi.c:104, dm_test_acpi_fill_header(): 0 == memcmp("ABCD", hdr.signature, sizeof(hdr.signature)): Expected 0x0 (0), got 0x13 (19)
With ut_assertmem_eq(): test/dm/acpi.c:103, dm_test_acpi_fill_header(): "ABCD" = hdr.signature: Expected "41424344", got "41424331"
- ut_asserteq(0x11, hdr.length);
- ut_asserteq(0x22, hdr.revision);
- ut_asserteq(0x33, hdr.checksum);
- ut_assertok(memcmp(OEM_ID, hdr.oem_id, sizeof(hdr.oem_id)));
- ut_assertok(memcmp(OEM_TABLE_ID, hdr.oem_table_id,
sizeof(hdr.oem_table_id)));- ut_asserteq(U_BOOT_BUILD_DATE, hdr.oem_revision);
- ut_assertok(memcmp(ASLC_ID, hdr.aslc_id, sizeof(hdr.aslc_id)));
- ut_asserteq(0x44, hdr.aslc_revision);
- return 0;
+}
+DM_TEST(dm_test_acpi_fill_header, DM_TESTF_SCAN_PDATA | DM_TESTF_SCAN_FDT);
2.25.1.481.gfbce0eb801-goog
regards, Wolfgang