[bpf-next,v4,3/4] libbpf: Add libbpf_probe_bpf_kfunc API

Commit Message

Tao Chen Feb. 6, 2025, 5:15 a.m. UTC

Similarly to libbpf_probe_bpf_helper, the libbpf_probe_bpf_kfunc
used to test the availability of the different eBPF kfuncs on the
current system.

Signed-off-by: Tao Chen <[e-mail]>
---
 tools/lib/bpf/libbpf.h        | 18 +++++++++++-
 tools/lib/bpf/libbpf.map      |  1 +
 tools/lib/bpf/libbpf_probes.c | 55 +++++++++++++++++++++++++++++++++++
 3 files changed, 73 insertions(+), 1 deletion(-)

Comments

Eduard Zingerman Feb. 7, 2025, 10:35 p.m. UTC | #1

On Thu, 2025-02-06 at 13:15 +0800, Tao Chen wrote:

[...]

>  LIBBPF_API int libbpf_probe_bpf_helper(enum bpf_prog_type prog_type,
>  				       enum bpf_func_id helper_id, const void *opts);
> -
> +/**
> + * @brief **libbpf_probe_bpf_kfunc()** detects if host kernel supports the
> + * use of a given BPF kfunc from specified BPF program type.
> + * @param prog_type BPF program type used to check the support of BPF kfunc
> + * @param kfunc_id The btf ID of BPF kfunc to check support for
> + * @param btf_fd The module BTF FD, if kfunc is defined in kernel module,
> + * btf_fd is used to point to module's BTF, 0 means kfunc defined in vmlinux.

Regarding '0' as special value:
in general FD is considered invalid only if it's negative, 0 is a valid FD.
Andrii, I remember there was a lengthy discussion about FD==0 and BPF,
but I don't remember the conclusion.

> + * @param opts reserved for future extensibility, should be NULL
> + * @return 1, if given combination of program type and kfunc is supported; 0,
> + * if the combination is not supported; negative error code if feature
> + * detection for provided input arguments failed or can't be performed
> + *
> + * Make sure the process has required set of CAP_* permissions (or runs as
> + * root) when performing feature checking.
> + */
> +LIBBPF_API int libbpf_probe_bpf_kfunc(enum bpf_prog_type prog_type,
> +				      int kfunc_id, int btf_fd, const void *opts);
>  /**
>   * @brief **libbpf_num_possible_cpus()** is a helper function to get the
>   * number of possible CPUs that the host kernel supports and expects.
> diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
> index a8b2936a1646..e93fae101efd 100644
> --- a/tools/lib/bpf/libbpf.map
> +++ b/tools/lib/bpf/libbpf.map
> @@ -436,4 +436,5 @@ LIBBPF_1.6.0 {
>  		bpf_linker__add_buf;
>  		bpf_linker__add_fd;
>  		bpf_linker__new_fd;
> +		libbpf_probe_bpf_kfunc;

This is now in conflict with bpf-next.

>  } LIBBPF_1.5.0;
> diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
> index e142130cb83c..c7f2b2dfbcf1 100644
> --- a/tools/lib/bpf/libbpf_probes.c
> +++ b/tools/lib/bpf/libbpf_probes.c
> @@ -433,6 +433,61 @@ static bool can_probe_prog_type(enum bpf_prog_type prog_type)
>  	return true;
>  }
>  
> +int libbpf_probe_bpf_kfunc(enum bpf_prog_type prog_type, int kfunc_id, int btf_fd,
> +			   const void *opts)
> +{
> +	struct bpf_insn insns[] = {
> +		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, btf_fd, kfunc_id),
> +		BPF_EXIT_INSN(),
> +	};
> +	const size_t insn_cnt = ARRAY_SIZE(insns);
> +	char buf[4096];
> +	int *fd_array = NULL;
> +	size_t fd_array_cnt = 0, fd_array_cap = fd_array_cnt;
> +	int ret;
> +
> +	if (opts)
> +		return libbpf_err(-EINVAL);
> +
> +	if (!can_probe_prog_type(prog_type))
> +		return -EOPNOTSUPP;
> +
> +	if (btf_fd) {
> +		ret = libbpf_ensure_mem((void **)&fd_array, &fd_array_cap,
> +					sizeof(int), fd_array_cnt + btf_fd);

Please take a look at the tools/testing/selftests/bpf/prog_tests/fd_array.c,
e.g. test case check_fd_array_cnt__fd_array_ok(). The offset field of the
call instruction does not have to be an fd (as it only has 16 bits),
instead it's an offset inside the fd_array.
Here it would be sufficient to allocate a small array on stack.

> +		if (ret)
> +			return ret;
> +
> +		/* In kernel, obtain the btf fd by means of the offset of
> +		 * the fd_array, and the offset is the btf fd.
> +		 */
> +		fd_array[btf_fd] = btf_fd;
> +	}

[...]

Tao Chen Feb. 8, 2025, 3:37 p.m. UTC | #2

在 2025/2/8 06:35, Eduard Zingerman 写道:
> On Thu, 2025-02-06 at 13:15 +0800, Tao Chen wrote:
> 
> [...]
> 
>>   LIBBPF_API int libbpf_probe_bpf_helper(enum bpf_prog_type prog_type,
>>   				       enum bpf_func_id helper_id, const void *opts);
>> -
>> +/**
>> + * @brief **libbpf_probe_bpf_kfunc()** detects if host kernel supports the
>> + * use of a given BPF kfunc from specified BPF program type.
>> + * @param prog_type BPF program type used to check the support of BPF kfunc
>> + * @param kfunc_id The btf ID of BPF kfunc to check support for
>> + * @param btf_fd The module BTF FD, if kfunc is defined in kernel module,
>> + * btf_fd is used to point to module's BTF, 0 means kfunc defined in vmlinux.
> 
> Regarding '0' as special value:
> in general FD is considered invalid only if it's negative, 0 is a valid FD.
> Andrii, I remember there was a lengthy discussion about FD==0 and BPF,
> but I don't remember the conclusion.
> 
>> + * @param opts reserved for future extensibility, should be NULL
>> + * @return 1, if given combination of program type and kfunc is supported; 0,
>> + * if the combination is not supported; negative error code if feature
>> + * detection for provided input arguments failed or can't be performed
>> + *
>> + * Make sure the process has required set of CAP_* permissions (or runs as
>> + * root) when performing feature checking.
>> + */
>> +LIBBPF_API int libbpf_probe_bpf_kfunc(enum bpf_prog_type prog_type,
>> +				      int kfunc_id, int btf_fd, const void *opts);
>>   /**
>>    * @brief **libbpf_num_possible_cpus()** is a helper function to get the
>>    * number of possible CPUs that the host kernel supports and expects.
>> diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
>> index a8b2936a1646..e93fae101efd 100644
>> --- a/tools/lib/bpf/libbpf.map
>> +++ b/tools/lib/bpf/libbpf.map
>> @@ -436,4 +436,5 @@ LIBBPF_1.6.0 {
>>   		bpf_linker__add_buf;
>>   		bpf_linker__add_fd;
>>   		bpf_linker__new_fd;
>> +		libbpf_probe_bpf_kfunc;
> 
> This is now in conflict with bpf-next.
> 

My bad, i will rebase the repo.

>>   } LIBBPF_1.5.0;
>> diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
>> index e142130cb83c..c7f2b2dfbcf1 100644
>> --- a/tools/lib/bpf/libbpf_probes.c
>> +++ b/tools/lib/bpf/libbpf_probes.c
>> @@ -433,6 +433,61 @@ static bool can_probe_prog_type(enum bpf_prog_type prog_type)
>>   	return true;
>>   }
>>   
>> +int libbpf_probe_bpf_kfunc(enum bpf_prog_type prog_type, int kfunc_id, int btf_fd,
>> +			   const void *opts)
>> +{
>> +	struct bpf_insn insns[] = {
>> +		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, btf_fd, kfunc_id),
>> +		BPF_EXIT_INSN(),
>> +	};
>> +	const size_t insn_cnt = ARRAY_SIZE(insns);
>> +	char buf[4096];
>> +	int *fd_array = NULL;
>> +	size_t fd_array_cnt = 0, fd_array_cap = fd_array_cnt;
>> +	int ret;
>> +
>> +	if (opts)
>> +		return libbpf_err(-EINVAL);
>> +
>> +	if (!can_probe_prog_type(prog_type))
>> +		return -EOPNOTSUPP;
>> +
>> +	if (btf_fd) {
>> +		ret = libbpf_ensure_mem((void **)&fd_array, &fd_array_cap,
>> +					sizeof(int), fd_array_cnt + btf_fd);
> 
> Please take a look at the tools/testing/selftests/bpf/prog_tests/fd_array.c,
> e.g. test case check_fd_array_cnt__fd_array_ok(). The offset field of the
> call instruction does not have to be an fd (as it only has 16 bits),
> instead it's an offset inside the fd_array.
> Here it would be sufficient to allocate a small array on stack.
> 

Good idea，thanks for your guidance，I'll make the modifications in the 
next version.

>> +		if (ret)
>> +			return ret;
>> +
>> +		/* In kernel, obtain the btf fd by means of the offset of
>> +		 * the fd_array, and the offset is the btf fd.
>> +		 */
>> +		fd_array[btf_fd] = btf_fd;
>> +	}
> 
> [...]
>

Tao Chen Feb. 9, 2025, 6:56 a.m. UTC | #3

在 2025/2/8 06:35, Eduard Zingerman 写道:
> On Thu, 2025-02-06 at 13:15 +0800, Tao Chen wrote:
> 
> [...]
> 
>>   LIBBPF_API int libbpf_probe_bpf_helper(enum bpf_prog_type prog_type,
>>   				       enum bpf_func_id helper_id, const void *opts);
>> -
>> +/**
>> + * @brief **libbpf_probe_bpf_kfunc()** detects if host kernel supports the
>> + * use of a given BPF kfunc from specified BPF program type.
>> + * @param prog_type BPF program type used to check the support of BPF kfunc
>> + * @param kfunc_id The btf ID of BPF kfunc to check support for
>> + * @param btf_fd The module BTF FD, if kfunc is defined in kernel module,
>> + * btf_fd is used to point to module's BTF, 0 means kfunc defined in vmlinux.
> 
> Regarding '0' as special value:
> in general FD is considered invalid only if it's negative, 0 is a valid FD.
> Andrii, I remember there was a lengthy discussion about FD==0 and BPF,
> but I don't remember the conclusion.
> 

Hi Eduard,
As you said, so what about "-1 means kfunc defined in vmlinux", -1 just 
used to distinguish whether it is vmlinux，then processing in 
libbpf_probe_bpf_kfunc like:

offset = 0;
// vmlinux btf offset default is 0
insn.off = offset;
if (btf_fd >= 0) {
	offset = 1;
	insn.off = offset;
	fd_array[offset] = btf_fd;
}
What do you think?

>> + * @param opts reserved for future extensibility, should be NULL
>> + * @return 1, if given combination of program type and kfunc is supported; 0,
>> + * if the combination is not supported; negative error code if feature
>> + * detection for provided input arguments failed or can't be performed
>> + *
>> + * Make sure the process has required set of CAP_* permissions (or runs as
>> + * root) when performing feature checking.
>> + */
>> +LIBBPF_API int libbpf_probe_bpf_kfunc(enum bpf_prog_type prog_type,
>> +				      int kfunc_id, int btf_fd, const void *opts);
>>   /**
>>    * @brief **libbpf_num_possible_cpus()** is a helper function to get the
>>    * number of possible CPUs that the host kernel supports and expects.
>> diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
>> index a8b2936a1646..e93fae101efd 100644
>> --- a/tools/lib/bpf/libbpf.map
>> +++ b/tools/lib/bpf/libbpf.map
>> @@ -436,4 +436,5 @@ LIBBPF_1.6.0 {
>>   		bpf_linker__add_buf;
>>   		bpf_linker__add_fd;
>>   		bpf_linker__new_fd;
>> +		libbpf_probe_bpf_kfunc;
> 
> This is now in conflict with bpf-next.
> 
>>   } LIBBPF_1.5.0;
>> diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
>> index e142130cb83c..c7f2b2dfbcf1 100644
>> --- a/tools/lib/bpf/libbpf_probes.c
>> +++ b/tools/lib/bpf/libbpf_probes.c
>> @@ -433,6 +433,61 @@ static bool can_probe_prog_type(enum bpf_prog_type prog_type)
>>   	return true;
>>   }
>>   
>> +int libbpf_probe_bpf_kfunc(enum bpf_prog_type prog_type, int kfunc_id, int btf_fd,
>> +			   const void *opts)
>> +{
>> +	struct bpf_insn insns[] = {
>> +		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, btf_fd, kfunc_id),
>> +		BPF_EXIT_INSN(),
>> +	};
>> +	const size_t insn_cnt = ARRAY_SIZE(insns);
>> +	char buf[4096];
>> +	int *fd_array = NULL;
>> +	size_t fd_array_cnt = 0, fd_array_cap = fd_array_cnt;
>> +	int ret;
>> +
>> +	if (opts)
>> +		return libbpf_err(-EINVAL);
>> +
>> +	if (!can_probe_prog_type(prog_type))
>> +		return -EOPNOTSUPP;
>> +
>> +	if (btf_fd) {
>> +		ret = libbpf_ensure_mem((void **)&fd_array, &fd_array_cap,
>> +					sizeof(int), fd_array_cnt + btf_fd);
> 
> Please take a look at the tools/testing/selftests/bpf/prog_tests/fd_array.c,
> e.g. test case check_fd_array_cnt__fd_array_ok(). The offset field of the
> call instruction does not have to be an fd (as it only has 16 bits),
> instead it's an offset inside the fd_array.
> Here it would be sufficient to allocate a small array on stack.
> 
>> +		if (ret)
>> +			return ret;
>> +
>> +		/* In kernel, obtain the btf fd by means of the offset of
>> +		 * the fd_array, and the offset is the btf fd.
>> +		 */
>> +		fd_array[btf_fd] = btf_fd;
>> +	}
> 
> [...]
>

Patch

diff --git a/tools/lib/bpf/libbpf.h b/tools/lib/bpf/libbpf.h
index 3020ee45303a..596b27f58c58 100644
--- a/tools/lib/bpf/libbpf.h
+++ b/tools/lib/bpf/libbpf.h
@@ -1680,7 +1680,23 @@  LIBBPF_API int libbpf_probe_bpf_map_type(enum bpf_map_type map_type, const void
  */
 LIBBPF_API int libbpf_probe_bpf_helper(enum bpf_prog_type prog_type,
 				       enum bpf_func_id helper_id, const void *opts);
-
+/**
+ * @brief **libbpf_probe_bpf_kfunc()** detects if host kernel supports the
+ * use of a given BPF kfunc from specified BPF program type.
+ * @param prog_type BPF program type used to check the support of BPF kfunc
+ * @param kfunc_id The btf ID of BPF kfunc to check support for
+ * @param btf_fd The module BTF FD, if kfunc is defined in kernel module,
+ * btf_fd is used to point to module's BTF, 0 means kfunc defined in vmlinux.
+ * @param opts reserved for future extensibility, should be NULL
+ * @return 1, if given combination of program type and kfunc is supported; 0,
+ * if the combination is not supported; negative error code if feature
+ * detection for provided input arguments failed or can't be performed
+ *
+ * Make sure the process has required set of CAP_* permissions (or runs as
+ * root) when performing feature checking.
+ */
+LIBBPF_API int libbpf_probe_bpf_kfunc(enum bpf_prog_type prog_type,
+				      int kfunc_id, int btf_fd, const void *opts);
 /**
  * @brief **libbpf_num_possible_cpus()** is a helper function to get the
  * number of possible CPUs that the host kernel supports and expects.
diff --git a/tools/lib/bpf/libbpf.map b/tools/lib/bpf/libbpf.map
index a8b2936a1646..e93fae101efd 100644
--- a/tools/lib/bpf/libbpf.map
+++ b/tools/lib/bpf/libbpf.map
@@ -436,4 +436,5 @@  LIBBPF_1.6.0 {
 		bpf_linker__add_buf;
 		bpf_linker__add_fd;
 		bpf_linker__new_fd;
+		libbpf_probe_bpf_kfunc;
 } LIBBPF_1.5.0;
diff --git a/tools/lib/bpf/libbpf_probes.c b/tools/lib/bpf/libbpf_probes.c
index e142130cb83c..c7f2b2dfbcf1 100644
--- a/tools/lib/bpf/libbpf_probes.c
+++ b/tools/lib/bpf/libbpf_probes.c
@@ -433,6 +433,61 @@  static bool can_probe_prog_type(enum bpf_prog_type prog_type)
 	return true;
 }
 
+int libbpf_probe_bpf_kfunc(enum bpf_prog_type prog_type, int kfunc_id, int btf_fd,
+			   const void *opts)
+{
+	struct bpf_insn insns[] = {
+		BPF_RAW_INSN(BPF_JMP | BPF_CALL, 0, BPF_PSEUDO_KFUNC_CALL, btf_fd, kfunc_id),
+		BPF_EXIT_INSN(),
+	};
+	const size_t insn_cnt = ARRAY_SIZE(insns);
+	char buf[4096];
+	int *fd_array = NULL;
+	size_t fd_array_cnt = 0, fd_array_cap = fd_array_cnt;
+	int ret;
+
+	if (opts)
+		return libbpf_err(-EINVAL);
+
+	if (!can_probe_prog_type(prog_type))
+		return -EOPNOTSUPP;
+
+	if (btf_fd) {
+		ret = libbpf_ensure_mem((void **)&fd_array, &fd_array_cap,
+					sizeof(int), fd_array_cnt + btf_fd);
+		if (ret)
+			return ret;
+
+		/* In kernel, obtain the btf fd by means of the offset of
+		 * the fd_array, and the offset is the btf fd.
+		 */
+		fd_array[btf_fd] = btf_fd;
+	}
+
+	buf[0] = '\0';
+	ret = probe_prog_load(prog_type, insns, insn_cnt, fd_array,
+			      fd_array_cnt, buf, sizeof(buf));
+	if (ret < 0) {
+		free(fd_array);
+		return libbpf_err(ret);
+	}
+
+	free(fd_array);
+	/* If BPF verifier recognizes BPF kfunc but it's not supported for
+	 * given BPF program type, it will emit "calling kernel function
+	 * bpf_cpumask_create is not allowed", if the kfunc id is invalid,
+	 * it will emit "kernel btf_id 4294967295 is not a function". If btf fd
+	 * invalid in module btf, it will emit "invalid module BTF fd specified" or
+	 * "negative offset disallowed for kernel module function call"
+	 */
+	if (ret == 0 && (strstr(buf, "not allowed") || strstr(buf, "not a function") ||
+			(strstr(buf, "invalid module BTF fd")) ||
+			(strstr(buf, "negative offset disallowed"))))
+		return 0;
+
+	return 1; /* assume supported */
+}
+
 int libbpf_probe_bpf_helper(enum bpf_prog_type prog_type, enum bpf_func_id helper_id,
 			    const void *opts)
 {