2020-07-17 02:09:33

by YangYuxi

Subject: [PATCH] ebpf: fix parameter naming confusing

Signed-off-by: YangYuxi <[email protected]>
---
kernel/bpf/syscall.c | 10 +++++-----
1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
index 0fd80ac81f70..300ae16baffc 100644
--- a/kernel/bpf/syscall.c
+++ b/kernel/bpf/syscall.c
@@ -1881,13 +1881,13 @@ struct bpf_prog *bpf_prog_inc_not_zero(struct bpf_prog *prog)
EXPORT_SYMBOL_GPL(bpf_prog_inc_not_zero);

bool bpf_prog_get_ok(struct bpf_prog *prog,
- enum bpf_prog_type *attach_type, bool attach_drv)
+ enum bpf_prog_type *prog_type, bool attach_drv)
{
/* not an attachment, just a refcount inc, always allow */
- if (!attach_type)
+ if (!prog_type)
return true;

- if (prog->type != *attach_type)
+ if (prog->type != *prog_type)
return false;
if (bpf_prog_is_dev_bound(prog->aux) && !attach_drv)
return false;
@@ -1895,7 +1895,7 @@ bool bpf_prog_get_ok(struct bpf_prog *prog,
return true;
}

-static struct bpf_prog *__bpf_prog_get(u32 ufd, enum bpf_prog_type *attach_type,
+static struct bpf_prog *__bpf_prog_get(u32 ufd, enum bpf_prog_type *prog_type,
bool attach_drv)
{
struct fd f = fdget(ufd);
@@ -1904,7 +1904,7 @@ static struct bpf_prog *__bpf_prog_get(u32 ufd, enum bpf_prog_type *attach_type,
prog = ____bpf_prog_get(f);
if (IS_ERR(prog))
return prog;
- if (!bpf_prog_get_ok(prog, attach_type, attach_drv)) {
+ if (!bpf_prog_get_ok(prog, prog_type, attach_drv)) {
prog = ERR_PTR(-EINVAL);
goto out;
}
--
1.8.3.1

2020-07-21 19:18:41

by Alexei Starovoitov

Subject: Re: [PATCH] ebpf: fix parameter naming confusing

On Thu, Jul 16, 2020 at 7:08 PM YangYuxi <[email protected]> wrote:
>
> Signed-off-by: YangYuxi <[email protected]>
> ---
> kernel/bpf/syscall.c | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/kernel/bpf/syscall.c b/kernel/bpf/syscall.c
> index 0fd80ac81f70..300ae16baffc 100644
> --- a/kernel/bpf/syscall.c
> +++ b/kernel/bpf/syscall.c
> @@ -1881,13 +1881,13 @@ struct bpf_prog *bpf_prog_inc_not_zero(struct bpf_prog *prog)
> EXPORT_SYMBOL_GPL(bpf_prog_inc_not_zero);
>
> bool bpf_prog_get_ok(struct bpf_prog *prog,
> - enum bpf_prog_type *attach_type, bool attach_drv)
> + enum bpf_prog_type *prog_type, bool attach_drv)
> {
> /* not an attachment, just a refcount inc, always allow */
> - if (!attach_type)
> + if (!prog_type)
> return true;

I think it makes it worse.
Now the comment doesn't match the code.
And attach_drv name also looks out of place.
Technically program type is also an attach type to some degree.
The name could be a bit confusing, but in combination with type:
'enum bpf_prog_type *attach_type'
I think it's pretty clear what these functions are doing.
So I prefer to keep the code as-is.