Re: [PATCH v2 3/3] tests/tcg/s390x: Test BRASL and BRCL with large negative offsets

[Richard Henderson <richard.henderson@linaro.org>]

On 3/12/22 01:25, Ilya Leoshkevich wrote:
> Add a small test in order to prevent regressions.
> 
> Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
> ---
>   tests/tcg/s390x/Makefile.target        |  1 +
>   tests/tcg/s390x/branch-relative-long.c | 60 ++++++++++++++++++++++++++
>   2 files changed, 61 insertions(+)
>   create mode 100644 tests/tcg/s390x/branch-relative-long.c
> 
> diff --git a/tests/tcg/s390x/Makefile.target b/tests/tcg/s390x/Makefile.target
> index 257c568c58..fd34b130f7 100644
> --- a/tests/tcg/s390x/Makefile.target
> +++ b/tests/tcg/s390x/Makefile.target
> @@ -15,6 +15,7 @@ TESTS+=mvc
>   TESTS+=shift
>   TESTS+=trap
>   TESTS+=signals-s390x
> +TESTS+=branch-relative-long
>   
>   ifneq ($(HAVE_GDB_BIN),)
>   GDB_SCRIPT=$(SRC_PATH)/tests/guest-debug/run-test.py
> diff --git a/tests/tcg/s390x/branch-relative-long.c b/tests/tcg/s390x/branch-relative-long.c
> new file mode 100644
> index 0000000000..c6f3f2db6d
> --- /dev/null
> +++ b/tests/tcg/s390x/branch-relative-long.c
> @@ -0,0 +1,60 @@
> +#include <stddef.h>
> +#include <stdio.h>
> +#include <string.h>
> +#include <sys/mman.h>
> +
> +#define DEFINE_ASM(_name, _code) \
> +    extern const char _name[]; \
> +    extern const char _name ## _end[]; \
> +    asm("    .globl " #_name "\n" \
> +        #_name ":\n" \
> +        "    " _code "\n" \
> +        "    .globl " #_name "_end\n" \
> +        #_name "_end:\n");
> +
> +DEFINE_ASM(br_r14, "br %r14");
> +DEFINE_ASM(brasl_r0, "brasl %r0,.-0x100000000");
> +DEFINE_ASM(brcl_0xf, "brcl 0xf,.-0x100000000");
> +
> +struct test {
> +    const char *code;
> +    const char *code_end;
> +};
> +
> +static const struct test tests[] = {
> +    {
> +        .code = brasl_r0,
> +        .code_end = brasl_r0_end,
> +    },
> +    {
> +        .code = brcl_0xf,
> +        .code_end = brcl_0xf_end,
> +    },
> +};
> +
> +int main(void)
> +{
> +    size_t length = 0x100000006;

This length is a little specific, given the apparent open-endedness of "code_end".  Though 
of course we know that we'll get an entire page allocated for these 6 bytes so there's no 
real problem.

I'm not actually sure what to suggest to improve this though.

For the series:
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>


r~

> +    unsigned char *buf;
> +    void (*code)(void);
> +    size_t i;
> +
> +    buf = mmap(NULL, length, PROT_READ | PROT_WRITE | PROT_EXEC,
> +               MAP_PRIVATE | MAP_ANONYMOUS, -1, 0);
> +    if (buf == MAP_FAILED) {
> +        perror("SKIP: mmap() failed");
> +        return 0;
> +    }
> +
> +    memcpy(buf, br_r14, br_r14_end - br_r14);
> +    for (i = 0; i < sizeof(tests) / sizeof(tests[0]); i++) {
> +        code = (void *)(buf + 0x100000000);
> +        memcpy(code, tests[i].code, tests[i].code_end - tests[i].code);
> +        code();
> +        memset(code, 0, tests[i].code_end - tests[i].code);
> +    }
> +
> +    munmap(buf, length);
> +
> +    return 0;
> +}