libbpf: make remark about zero-initializing bpf_*_info structs
PATCH v3 1/1 libbpf: make remark about zero-initializing bpf_*_info structs
In some situations, if you fail to zero-initialize the bpf_{prog,map,btf,link}_info structs supplied to the set of LIBBPF helpers bpf_{prog,map,btf,link}_get_info_by_fd(), you can expect the helper to return an error. This can possibly leave people in a situation where they're scratching their heads for an unnnecessary amount of time. Make an explicit remark about the requirement of zero-initializing the supplied bpf_{prog,map,btf,link}_info structs for the respective LIBBPF helpers.
Internally, LIBBPF helpers bpf_{prog,map,btf,link}_get_info_by_fd() call into bpf_obj_get_info_by_fd() where the bpf(2) BPF_OBJ_GET_INFO_BY_FD command is used. This specific command is effectively backed by restrictions enforced by the bpf_check_uarg_tail_zero() helper. This function ensures that if the size of the supplied bpf_{prog,map,btf,link}_info structs are larger than what the kernel can handle, trailing bits are zeroed. This can be a problem when compiling against UAPI headers that don't necessarily match the sizes of the same underlying types known to the kernel.
---
v2 to v3:
* Modified the comment wording a little to make it less misleading. Specifically, noting that the supplied bpf_{prog,map,btf,link}_info structs should be zero-initialized or initialized as expected. In some cases, subsequent invocations to bpf_{prog,map,btf,link}_get_info_by_fd() helpers don't necessarily require the bpf_{prog,map,btf,link}_info struct to be zero-initialized, but rather that it just has been properly initialized at some point.
tools/lib/bpf/bpf.h | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-)
tools/lib/bpf/bpf.h +17 -5
@@ -500,7 +500,10 @@ LIBBPF_API int bpf_obj_get_info_by_fd(int bpf_fd, void *info, __u32 *info_len);* program corresponding to *prog_fd*. * * Populates up to *info_len* bytes of *info* and updates *info_len* with the -* actual number of bytes written to *info*. +* actual number of bytes written to *info*. Note that *info* should be +* zero-initialized or initialized as expected by the requested *info* +* type. Failing to (zero-)initialize *info* under certain circumstances can +* result in this helper returning an error. * * @param prog_fd BPF program file descriptor * @param info pointer to **struct bpf_prog_info** that will be populated with
@@ -517,7 +520,10 @@ LIBBPF_API int bpf_prog_get_info_by_fd(int prog_fd, struct bpf_prog_info *info,* map corresponding to *map_fd*. * * Populates up to *info_len* bytes of *info* and updates *info_len* with the -* actual number of bytes written to *info*. +* actual number of bytes written to *info*. Note that *info* should be +* zero-initialized or initialized as expected by the requested *info* +* type. Failing to (zero-)initialize *info* under certain circumstances can +* result in this helper returning an error. * * @param map_fd BPF map file descriptor * @param info pointer to **struct bpf_map_info** that will be populated with
@@ -530,11 +536,14 @@ LIBBPF_API int bpf_prog_get_info_by_fd(int prog_fd, struct bpf_prog_info *info, LIBBPF_API int bpf_map_get_info_by_fd(int map_fd, struct bpf_map_info *info, __u32 *info_len);/** -* @brief **bpf_btf_get_info_by_fd()** obtains information about the +* @brief **bpf_btf_get_info_by_fd()** obtains information about the * BTF object corresponding to *btf_fd*. * * Populates up to *info_len* bytes of *info* and updates *info_len* with the -* actual number of bytes written to *info*. +* actual number of bytes written to *info*. Note that *info* should be +* zero-initialized or initialized as expected by the requested *info* +* type. Failing to (zero-)initialize *info* under certain circumstances can +* result in this helper returning an error. * * @param btf_fd BTF object file descriptor * @param info pointer to **struct bpf_btf_info** that will be populated with
@@ -551,7 +560,10 @@ LIBBPF_API int bpf_btf_get_info_by_fd(int btf_fd, struct bpf_btf_info *info, __u* link corresponding to *link_fd*. * * Populates up to *info_len* bytes of *info* and updates *info_len* with the -* actual number of bytes written to *info*. +* actual number of bytes written to *info*. Note that *info* should be +* zero-initialized or initialized as expected by the requested *info* +* type. Failing to (zero-)initialize *info* under certain circumstances can +* result in this helper returning an error. * * @param link_fd BPF link file descriptor * @param info pointer to **struct bpf_link_info** that will be populated with
Signature · 3 lines
2.43.0.687.g38aa6559b0-goog /M
Quoted 24 lines · Matt Bobrowski
In some situations, if you fail to zero-initialize the bpf_{prog,map,btf,link}_info structs supplied to the set of LIBBPF helpers bpf_{prog,map,btf,link}_get_info_by_fd(), you can expect the helper to return an error. This can possibly leave people in a situation where they're scratching their heads for an unnnecessary amount of time. Make an explicit remark about the requirement of zero-initializing the supplied bpf_{prog,map,btf,link}_info structs for the respective LIBBPF helpers.
Internally, LIBBPF helpers bpf_{prog,map,btf,link}_get_info_by_fd() call into
bpf_obj_get_info_by_fd()where the bpf(2)BPF_OBJ_GET_INFO_BY_FDcommand is used. This specific command is effectively backed by restrictions enforced by thebpf_check_uarg_tail_zero()helper. This function ensures that if the size of the supplied bpf_{prog,map,btf,link}_info structs are larger than what the kernel can handle, trailing bits are zeroed. This can be a problem when compiling against UAPI headers that don't necessarily match the sizes of the same underlying types known to the kernel.Signed-off-by Matt Bobrowski <[email protected]>---
v2 to v3:
Quoted 8 lines
* Modified the comment wording a little to make it less misleading. Specifically, noting that the supplied bpf_{prog,map,btf,link}_info structs should be zero-initialized or initialized as expected. In some cases, subsequent invocations to bpf_{prog,map,btf,link}_get_info_by_fd() helpers don't necessarily require the bpf_{prog,map,btf,link}_info struct to be zero-initialized, but rather that it just has been properly initialized at some point.
thanks,
jirka
Quoted 65 lines
tools/lib/bpf/bpf.h | 22 +++++++++++++++++----- 1 file changed, 17 insertions(+), 5 deletions(-)tools/lib/bpf/bpf.h +17
-5@@ -500,7 +500,10 @@ LIBBPF_API int bpf_obj_get_info_by_fd(int bpf_fd, void *info, __u32 *info_len);* program corresponding to *prog_fd*. * * Populates up to *info_len* bytes of *info* and updates *info_len* with the -* actual number of bytes written to *info*. +* actual number of bytes written to *info*. Note that *info* should be +* zero-initialized or initialized as expected by the requested *info* +* type. Failing to (zero-)initialize *info* under certain circumstances can +* result in this helper returning an error. * * @param prog_fd BPF program file descriptor * @param info pointer to **struct bpf_prog_info** that will be populated with @@ -517,7 +520,10 @@ LIBBPF_API int bpf_prog_get_info_by_fd(int prog_fd, struct bpf_prog_info *info,* map corresponding to *map_fd*. * * Populates up to *info_len* bytes of *info* and updates *info_len* with the -* actual number of bytes written to *info*. +* actual number of bytes written to *info*. Note that *info* should be +* zero-initialized or initialized as expected by the requested *info* +* type. Failing to (zero-)initialize *info* under certain circumstances can +* result in this helper returning an error. * * @param map_fd BPF map file descriptor * @param info pointer to **struct bpf_map_info** that will be populated with @@ -530,11 +536,14 @@ LIBBPF_API int bpf_prog_get_info_by_fd(int prog_fd, struct bpf_prog_info *info, LIBBPF_API int bpf_map_get_info_by_fd(int map_fd, struct bpf_map_info *info, __u32 *info_len);/** -* @brief **bpf_btf_get_info_by_fd()** obtains information about the +* @brief **bpf_btf_get_info_by_fd()** obtains information about the * BTF object corresponding to *btf_fd*. * * Populates up to *info_len* bytes of *info* and updates *info_len* with the -* actual number of bytes written to *info*. +* actual number of bytes written to *info*. Note that *info* should be +* zero-initialized or initialized as expected by the requested *info* +* type. Failing to (zero-)initialize *info* under certain circumstances can +* result in this helper returning an error. * * @param btf_fd BTF object file descriptor * @param info pointer to **struct bpf_btf_info** that will be populated with @@ -551,7 +560,10 @@ LIBBPF_API int bpf_btf_get_info_by_fd(int btf_fd, struct bpf_btf_info *info, __u* link corresponding to *link_fd*. * * Populates up to *info_len* bytes of *info* and updates *info_len* with the -* actual number of bytes written to *info*. +* actual number of bytes written to *info*. Note that *info* should be +* zero-initialized or initialized as expected by the requested *info* +* type. Failing to (zero-)initialize *info* under certain circumstances can +* result in this helper returning an error. * * @param link_fd BPF link file descriptor * @param info pointer to **struct bpf_link_info** that will be populated with Signature · 3 lines
2.43.0.687.g38aa6559b0-goog /M
This patch was applied to bpf/bpf-next.git (master) by Andrii Nakryiko <[email protected]>:
Hello:
This patch was applied to bpf/bpf-next.git (master) by Andrii Nakryiko <[email protected]>:
Quoted 10 lines · On Wed, 14 Feb 2024 09:14:23 +0000 you wrote:
In some situations, if you fail to zero-initialize the bpf_{prog,map,btf,link}_info structs supplied to the set of LIBBPF helpers bpf_{prog,map,btf,link}_get_info_by_fd(), you can expect the helper to return an error. This can possibly leave people in a situation where they're scratching their heads for an unnnecessary amount of time. Make an explicit remark about the requirement of zero-initializing the supplied bpf_{prog,map,btf,link}_info structs for the respective LIBBPF helpers.
[...]
Here is the summary with links:
- [v3,bpf-next] libbpf: make remark about zero-initializing bpf_*_info structs
https://git.kernel.org/bpf/bpf-next/c/1159d2785220
You are awesome, thank you!
Signature · 2 lines
Deet-doot-dot, I am a bot. https://korg.docs.kernel.org/patchwork/pwbot.html