Use config.py in all new bignum tests

This previous test is correct, as it is using make. Switch to using
config.py for robustness and consistency.

Signed-off-by: Janos Follath <janos.follath@arm.com>
This commit is contained in:
Janos Follath 2023-07-31 10:07:57 +01:00
parent 82823b2fe8
commit f3135af159

View File

@ -4216,7 +4216,8 @@ component_test_have_int32_cmake_new_bignum () {
scripts/config.py unset MBEDTLS_PADLOCK_C
scripts/config.py unset MBEDTLS_AESCE_C
scripts/config.py set MBEDTLS_TEST_HOOKS
make CC=gcc CFLAGS="$ASAN_CFLAGS -Werror -Wall -Wextra -DMBEDTLS_HAVE_INT32 -DMBEDTLS_ECP_WITH_MPI_UINT" LDFLAGS="$ASAN_CFLAGS"
scripts/config.py -f "$CONFIG_NEW_BIGNUM_H" set MBEDTLS_ECP_WITH_MPI_UINT
make CC=gcc CFLAGS="$ASAN_CFLAGS -Werror -Wall -Wextra -DMBEDTLS_HAVE_INT32" LDFLAGS="$ASAN_CFLAGS"
msg "test: gcc, force 32-bit bignum limbs, new bignum interface, test hooks (ASan build)"
make test