Re: [PATCH v2 bpf-next 1/6] bpf: add a __btf_get_by_fd helper

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Tue, Nov 26, 2024 at 8:30 AM Anton Protopopov <aspsk@xxxxxxxxxxxxx> wrote:
>
> On 24/11/25 05:31PM, Alexei Starovoitov wrote:
> > On Tue, Nov 19, 2024 at 2:17 AM Anton Protopopov <aspsk@xxxxxxxxxxxxx> wrote:
> > >
> > > Add a new helper to get a pointer to a struct btf from a file
> > > descriptor which doesn't increase a refcount.
> > >
> > > Signed-off-by: Anton Protopopov <aspsk@xxxxxxxxxxxxx>
> > > ---
> > >  include/linux/btf.h | 13 +++++++++++++
> > >  kernel/bpf/btf.c    | 13 ++++---------
> > >  2 files changed, 17 insertions(+), 9 deletions(-)
> > >
> > > diff --git a/include/linux/btf.h b/include/linux/btf.h
> > > index 4214e76c9168..050051a578a8 100644
> > > --- a/include/linux/btf.h
> > > +++ b/include/linux/btf.h
> > > @@ -4,6 +4,7 @@
> > >  #ifndef _LINUX_BTF_H
> > >  #define _LINUX_BTF_H 1
> > >
> > > +#include <linux/file.h>
> > >  #include <linux/types.h>
> > >  #include <linux/bpfptr.h>
> > >  #include <linux/bsearch.h>
> > > @@ -143,6 +144,18 @@ void btf_get(struct btf *btf);
> > >  void btf_put(struct btf *btf);
> > >  const struct btf_header *btf_header(const struct btf *btf);
> > >  int btf_new_fd(const union bpf_attr *attr, bpfptr_t uattr, u32 uattr_sz);
> > > +
> > > +static inline struct btf *__btf_get_by_fd(struct fd f)
> > > +{
> > > +       if (fd_empty(f))
> > > +               return ERR_PTR(-EBADF);
> > > +
> > > +       if (unlikely(fd_file(f)->f_op != &btf_fops))
> > > +               return ERR_PTR(-EINVAL);
> > > +
> > > +       return fd_file(f)->private_data;
> > > +}
> >
> > Maybe let's call it __btf_get() and place it next to __bpf_map_get() ?
> > So names and function bodies are directly comparable?
>
> I named it so because the corresponding helper which is taking a ref is named
> btf_get_by_fd(). And btf_get() is actually increasing a refcnt. In the bpf_map
> case naming is a bit different (and also not super-consistent,
> bpf_map_inc/bpf_map_put to +- refcnt).

I see. __btf_get_by_fd() is fine then.
Only place it next to __bpf_map_get() and add a comment that
double underscore helpers don't inc refcnt while correspoing bpf_map_get()
and btf_get_by_fd().

> Do you want me to make names more
> consistent, globally?

No. let's avoid the churn.





[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux