Re: [PATCH 7/7] bpf: Add tests for new BPF atomic operations

[Yonghong Song <yhs@fb.com>]

On 11/23/20 9:32 AM, Brendan Jackman wrote:
> This relies on the work done by Yonghong Song in
> https://reviews.llvm.org/D72184
> 
> Signed-off-by: Brendan Jackman <jackmanb@google.com>
> ---
>   tools/testing/selftests/bpf/Makefile          |   2 +-
>   .../selftests/bpf/prog_tests/atomics_test.c   | 145 ++++++++++++++++++
>   .../selftests/bpf/progs/atomics_test.c        |  61 ++++++++
>   .../selftests/bpf/verifier/atomic_cmpxchg.c   |  96 ++++++++++++
>   .../selftests/bpf/verifier/atomic_fetch_add.c | 106 +++++++++++++
>   .../selftests/bpf/verifier/atomic_xchg.c      | 113 ++++++++++++++
>   6 files changed, 522 insertions(+), 1 deletion(-)
>   create mode 100644 tools/testing/selftests/bpf/prog_tests/atomics_test.c
>   create mode 100644 tools/testing/selftests/bpf/progs/atomics_test.c
>   create mode 100644 tools/testing/selftests/bpf/verifier/atomic_cmpxchg.c
>   create mode 100644 tools/testing/selftests/bpf/verifier/atomic_fetch_add.c
>   create mode 100644 tools/testing/selftests/bpf/verifier/atomic_xchg.c
> 
> diff --git a/tools/testing/selftests/bpf/Makefile b/tools/testing/selftests/bpf/Makefile
> index 3d5940cd110d..4e28640ca2d8 100644
> --- a/tools/testing/selftests/bpf/Makefile
> +++ b/tools/testing/selftests/bpf/Makefile
> @@ -250,7 +250,7 @@ define CLANG_BPF_BUILD_RULE
>   	$(call msg,CLNG-LLC,$(TRUNNER_BINARY),$2)
>   	$(Q)($(CLANG) $3 -O2 -target bpf -emit-llvm			\
>   		-c $1 -o - || echo "BPF obj compilation failed") | 	\
> -	$(LLC) -mattr=dwarfris -march=bpf -mcpu=v3 $4 -filetype=obj -o $2
> +	$(LLC) -mattr=dwarfris -march=bpf -mcpu=v4 $4 -filetype=obj -o $2

We have an issue here. If we change -mcpu=v4 here, we will force
people to use trunk llvm to run selftests which is not a good idea.

I am wondering whether we can single out progs/atomics_test.c, which 
will be compiled with -mcpu=v4 and run with test_progs.

test_progs-no_alu32 runs tests without alu32. Since -mcpu=v4 implies
alu32, atomic tests should be skipped in test_progs-no-alu32.

In bpf_helpers.h, we already use __clang_major__ macro to compare
to clang version,

#if __clang_major__ >= 8 && defined(__bpf__)
static __always_inline void
bpf_tail_call_static(void *ctx, const void *map, const __u32 slot)
{
         if (!__builtin_constant_p(slot))
                 __bpf_unreachable();
...

I think we could also use __clang_major__ in progs/atomics_test.c
to enable tested intrinsics only if __clang_major__ >= 12? This
way, the same code can be compiled with -mcpu=v2/v3.

Alternatively, you can also use a macro at clang command line like
    clang -mcpu=v4 -DENABLE_ATOMIC ...
    clang -mcpu=v3/v2 ...

progs/atomics_test.c:
    #ifdef ENABLE_ATOMIC
      ... atomic_intrinsics ...
    #endif


>   endef
>   # Similar to CLANG_BPF_BUILD_RULE, but with disabled alu32
>   define CLANG_NOALU32_BPF_BUILD_RULE
> diff --git a/tools/testing/selftests/bpf/prog_tests/atomics_test.c b/tools/testing/selftests/bpf/prog_tests/atomics_test.c
> new file mode 100644
> index 000000000000..a4859d88fc11
> --- /dev/null
> +++ b/tools/testing/selftests/bpf/prog_tests/atomics_test.c
> @@ -0,0 +1,145 @@
> +// SPDX-License-Identifier: GPL-2.0
> +
> +#include <test_progs.h>
> +
> +#include "atomics_test.skel.h"
> +
> +static void test_add(void)
> +{
> +	struct atomics_test *atomics_skel = NULL;
> +	int err, prog_fd;
> +	__u32 duration = 0, retval;
> +
> +	atomics_skel = atomics_test__open_and_load();
> +	if (CHECK(!atomics_skel, "atomics_skel_load", "atomics skeleton failed\n"))
> +		goto cleanup;
> +
> +	err = atomics_test__attach(atomics_skel);
> +	if (CHECK(err, "atomics_attach", "atomics attach failed: %d\n", err))
> +		goto cleanup;
> +
> +	prog_fd = bpf_program__fd(atomics_skel->progs.add);
> +	err = bpf_prog_test_run(prog_fd, 1, NULL, 0,
> +				NULL, NULL, &retval, &duration);
> +	if (CHECK(err || retval, "test_run add",
> +		  "err %d errno %d retval %d duration %d\n",
> +		  err, errno, retval, duration))
> +		goto cleanup;
> +
> +	CHECK(atomics_skel->data->add64_value != 3, "add64_value",
> +	      "64bit atomic add value was not incremented (got %lld want 2)\n",
> +	      atomics_skel->data->add64_value);
> +	CHECK(atomics_skel->bss->add64_result != 1, "add64_result",
> +	      "64bit atomic add bad return value (got %lld want 1)\n",
> +	      atomics_skel->bss->add64_result);
> +
> +	CHECK(atomics_skel->data->add32_value != 3, "add32_value",
> +	      "32bit atomic add value was not incremented (got %d want 2)\n",
> +	      atomics_skel->data->add32_value);
> +	CHECK(atomics_skel->bss->add32_result != 1, "add32_result",
> +	      "32bit atomic add bad return value (got %d want 1)\n",
> +	      atomics_skel->bss->add32_result);
> +
> +	CHECK(atomics_skel->bss->add_stack_value_copy != 3, "add_stack_value",
> +	      "_stackbit atomic add value was not incremented (got %lld want 2)\n",
> +	      atomics_skel->bss->add_stack_value_copy);
> +	CHECK(atomics_skel->bss->add_stack_result != 1, "add_stack_result",
> +	      "_stackbit atomic add bad return value (got %lld want 1)\n",
> +	      atomics_skel->bss->add_stack_result);
> +
> +cleanup:
> +	atomics_test__destroy(atomics_skel);
> +}
> +
[...]
> diff --git a/tools/testing/selftests/bpf/progs/atomics_test.c b/tools/testing/selftests/bpf/progs/atomics_test.c
> new file mode 100644
> index 000000000000..d81f45eb6c45
> --- /dev/null
> +++ b/tools/testing/selftests/bpf/progs/atomics_test.c
> @@ -0,0 +1,61 @@
> +// SPDX-License-Identifier: GPL-2.0
> +#include <linux/bpf.h>
> +#include <bpf/bpf_helpers.h>
> +#include <bpf/bpf_tracing.h>
> +
> +__u64 add64_value = 1;
> +__u64 add64_result;
> +__u32 add32_value = 1;
> +__u32 add32_result;
> +__u64 add_stack_value_copy;
> +__u64 add_stack_result;

To please llvm10, let us initialize all unitialized globals explicitly like
    __u64 add64_result = 0;
    __u32 add32_result = 0;
    ...

llvm11 and above are okay but llvm10 put those uninitialized globals
into COM section (not .bss or .data sections) which BTF did not
handle them.

> +SEC("fentry/bpf_fentry_test1")
> +int BPF_PROG(add, int a)
> +{
> +	__u64 add_stack_value = 1;
> +
> +	add64_result = __sync_fetch_and_add(&add64_value, 2);
> +	add32_result = __sync_fetch_and_add(&add32_value, 2);
> +	add_stack_result = __sync_fetch_and_add(&add_stack_value, 2);
> +	add_stack_value_copy = add_stack_value;
> +
> +	return 0;
> +}
> +
> +__u64 cmpxchg64_value = 1;
> +__u64 cmpxchg64_result_fail;
> +__u64 cmpxchg64_result_succeed;
> +__u32 cmpxchg32_value = 1;
> +__u32 cmpxchg32_result_fail;
> +__u32 cmpxchg32_result_succeed;

same here. explicitly initializing cmpxchg64_result_fail, 
cmpxchg64_result_succeed, cmpxchg32_result_fail, cmpxchg32_result_succeed.

> +SEC("fentry/bpf_fentry_test1")
> +int BPF_PROG(cmpxchg, int a)
> +{
> +	cmpxchg64_result_fail = __sync_val_compare_and_swap(
> +		&cmpxchg64_value, 0, 3);
> +	cmpxchg64_result_succeed = __sync_val_compare_and_swap(
> +		&cmpxchg64_value, 1, 2);
> +
> +	cmpxchg32_result_fail = __sync_val_compare_and_swap(
> +		&cmpxchg32_value, 0, 3);
> +	cmpxchg32_result_succeed = __sync_val_compare_and_swap(
> +		&cmpxchg32_value, 1, 2);
> +
> +	return 0;
> +}
> +
> +__u64 xchg64_value = 1;
> +__u64 xchg64_result;
> +__u32 xchg32_value = 1;
> +__u32 xchg32_result;

explicitly initializing xchg64_result, xchg32_result.

> +SEC("fentry/bpf_fentry_test1")
> +int BPF_PROG(xchg, int a)
> +{
> +	__u64 val64 = 2;
> +	__u32 val32 = 2;
> +
> +	__atomic_exchange(&xchg64_value, &val64, &xchg64_result, __ATOMIC_RELAXED);
> +	__atomic_exchange(&xchg32_value, &val32, &xchg32_result, __ATOMIC_RELAXED);
> +
> +	return 0;
> +}
[...]