On Mon, Dec 27, 2021 at 7:25 AM Hengqi Chen <hengqi.chen@xxxxxxxxx> wrote:
>
>
>
> On 2021/12/26 4:37 AM, grantseltzer wrote:
> > From: Grant Seltzer <grantseltzer@xxxxxxxxx>
> >
> > This adds documentation for:
> >
> > - bpf_map_delete_batch()
> > - bpf_map_lookup_batch()
> > - bpf_map_lookup_and_delete_batch()
> > - bpf_map_update_batch()
> >
> > Signed-off-by: Grant Seltzer <grantseltzer@xxxxxxxxx>
> > ---
> > tools/lib/bpf/bpf.c | 4 +-
> > tools/lib/bpf/bpf.h | 112 +++++++++++++++++++++++++++++++++++++++++++-
> > 2 files changed, 112 insertions(+), 4 deletions(-)
> >
> > diff --git a/tools/lib/bpf/bpf.c b/tools/lib/bpf/bpf.c
> > index 9b64eed2b003..25f3d6f85fe5 100644
> > --- a/tools/lib/bpf/bpf.c
> > +++ b/tools/lib/bpf/bpf.c
> > @@ -691,7 +691,7 @@ static int bpf_map_batch_common(int cmd, int fd, void *in_batch,
> > return libbpf_err_errno(ret);
> > }
> >
> > -int bpf_map_delete_batch(int fd, void *keys, __u32 *count,
> > +int bpf_map_delete_batch(int fd, const void *keys, __u32 *count,
>
> Maybe you should drop these const qualifier changes.
Can you help me understand the benefit of using the const qualifier in
this context? I added it at Andrii's suggestion without proper
understanding. I understand that it will properly convey that the keys
or values aren't output parameters like in other batch operation
functions, I don't think it would change where the underlying data is
stored, just the pointer variable.
Is it worth it to have seperate 'common' functions for the sake of
having a const qualifier?
>
> All batch operations use `bpf_map_batch_common`, which has the following signature:
>
> static int bpf_map_batch_common(int cmd, int fd, void *in_batch,
> void *out_batch, void *keys, void *values,
> __u32 *count,
> const struct bpf_map_batch_opts *opts)
>
> Adding these const qualifiers causes the following error:
>
> bpf.c:698:15: error: passing argument 5 of ‘bpf_map_batch_common’ discards
> ‘const’ qualifier from pointer target type [-Werror=discarded-qualifiers]
>
> > const struct bpf_map_batch_opts *opts)
> > {
> > return bpf_map_batch_common(BPF_MAP_DELETE_BATCH, fd, NULL,
> > @@ -715,7 +715,7 @@ int bpf_map_lookup_and_delete_batch(int fd, void *in_batch, void *out_batch,
> > count, opts);
> > }
> >
> > -int bpf_map_update_batch(int fd, void *keys, void *values, __u32 *count,
> > +int bpf_map_update_batch(int fd, const void *keys, const void *values, __u32 *count,
> > const struct bpf_map_batch_opts *opts)
> > {
> > return bpf_map_batch_common(BPF_MAP_UPDATE_BATCH, fd, NULL, NULL,
> > diff --git a/tools/lib/bpf/bpf.h b/tools/lib/bpf/bpf.h
> > index 00619f64a040..01011747f127 100644
> > --- a/tools/lib/bpf/bpf.h
> > +++ b/tools/lib/bpf/bpf.h
> > @@ -254,20 +254,128 @@ struct bpf_map_batch_opts {
> > };
> > #define bpf_map_batch_opts__last_field flags
> >
> > -LIBBPF_API int bpf_map_delete_batch(int fd, void *keys,
> > +
> > +/**
> > + * @brief **bpf_map_delete_batch()** allows for batch deletion of multiple
> > + * elements in a BPF map.
> > + *
> > + * @param fd BPF map file descriptor
> > + * @param keys pointer to an array of *count* keys
> > + * @param count number of elements in the map to sequentially delete
> > + * @param opts options for configuring the way the batch deletion works
> > + * @return 0, on success; negative error code, otherwise (errno is also set to
> > + * the error code)
> > + */
> > +LIBBPF_API int bpf_map_delete_batch(int fd, const void *keys,
> > __u32 *count,
> > const struct bpf_map_batch_opts *opts);
> > +
> > +/**
> > + * @brief **bpf_map_lookup_batch()** allows for batch lookup of BPF map elements.
> > + *
> > + * The parameter *in_batch* is the address of the first element in the batch to read.
> > + * *out_batch* is an output parameter that should be passed as *in_batch* to subsequent
> > + * calls to **bpf_map_lookup_batch()**. NULL can be passed for *in_batch* to indicate
> > + * that the batched lookup starts from the beginning of the map.
> > + *
> > + * The *keys* and *values* are output parameters which must point to memory large enough to
> > + * hold *count* items based on the key and value size of the map *map_fd*. The *keys*
> > + * buffer must be of *key_size* * *count*. The *values* buffer must be of
> > + * *value_size* * *count*.
> > + *
> > + * @param fd BPF map file descriptor
> > + * @param in_batch address of the first element in batch to read, can pass NULL to
> > + * indicate that the batched lookup starts from the beginning of the map.
> > + * @param out_batch output parameter that should be passed to next call as *in_batch*
> > + * @param keys pointer to an array large enough for *count* keys
> > + * @param values pointer to an array large enough for *count* values
> > + * @param count number of elements in the map to read in batch. If ENOENT is
> > + * returned, count will be set as the number of elements that were read before
> > + * running out of entries in the map
> > + * @param opts options for configuring the way the batch lookup works
> > + * @return 0, on success; negative error code, otherwise (errno is also set to
> > + * the error code)
> > + */
> > LIBBPF_API int bpf_map_lookup_batch(int fd, void *in_batch, void *out_batch,
> > void *keys, void *values, __u32 *count,
> > const struct bpf_map_batch_opts *opts);
> > +
> > +/**
> > + * @brief **bpf_map_lookup_and_delete_batch()** allows for batch lookup and deletion
> > + * of BPF map elements where each element is deleted after being retrieved.
> > + *
> > + * Note that *count* is an input and output parameter, where on output it
> > + * represents how many elements were successfully deleted. Also note that if
> > + * **EFAULT** is returned up to *count* elements may have been deleted without
> > + * being returned via the *keys* and *values* output parameters. If **ENOENT**
> > + * is returned then *count* will be set to the number of elements that were read
> > + * before running out of entries in the map.
> > + *
> > + * @param fd BPF map file descriptor
> > + * @param in_batch address of the first element in batch to read, can pass NULL to
> > + * get address of the first element in *out_batch*
> > + * @param out_batch output parameter that should be passed to next call as *in_batch*
> > + * @param keys pointer to an array of *count* keys
> > + * @param values pointer to an array large enough for *count* values
> > + * @param count input and output parameter; on input it's the number of elements
> > + * in the map to read and delete in batch; on output it represents number of elements
> > + * that were successfully read and deleted
> > + * If ENOENT is returned, count will be set as the number of elements that were
> > + * read before running out of entries in the map
> > + * @param opts options for configuring the way the batch lookup and delete works
> > + * @return 0, on success; negative error code, otherwise (errno is also set to
> > + * the error code)
> > + */
> > LIBBPF_API int bpf_map_lookup_and_delete_batch(int fd, void *in_batch,
> > void *out_batch, void *keys,
> > void *values, __u32 *count,
> > const struct bpf_map_batch_opts *opts);
> > -LIBBPF_API int bpf_map_update_batch(int fd, void *keys, void *values,
> > +
> > +/**
> > + * @brief **bpf_map_update_batch()** updates multiple elements in a map
> > + * by specifying keys and their corresponding values.
> > + *
> > + * The *keys* and *values* parameters must point to memory large enough
> > + * to hold *count* items based on the key and value size of the map.
> > + *
> > + * The *opts* parameter can be used to control how *bpf_map_update_batch()*
> > + * should handle keys that either do or do not already exist in the map.
> > + * In particular the *flags* parameter of *bpf_map_batch_opts* can be
> > + * one of the following:
> > + *
> > + * Note that *count* is an input and output parameter, where on output it
> > + * represents how many elements were successfully updated. Also note that if
> > + * **EFAULT** then *count* should not be trusted to be correct.
> > + *
> > + * **BPF_ANY**
> > + * Create new elements or update existing.
> > + *
> > + * **BPF_NOEXIST**
> > + * Create new elements only if they do not exist.
> > + *
> > + * **BPF_EXIST**
> > + * Update existing elements.
> > + *
> > + * **BPF_F_LOCK**
> > + * Update spin_lock-ed map elements. This must be
> > + * specified if the map value contains a spinlock.
> > + *
> > + * @param fd BPF map file descriptor
> > + * @param keys pointer to an array of *count* keys
> > + * @param values pointer to an array of *count* values
> > + * @param count input and output parameter; on input it's the number of elements
> > + * in the map to update in batch; on output it represents the number of elements
> > + * that were successfully updated. If EFAULT is returned, *count* should not
> > + * be trusted to be correct.
> > + * @param opts options for configuring the way the batch update works
> > + * @return 0, on success; negative error code, otherwise (errno is also set to
> > + * the error code)
> > + */
> > +LIBBPF_API int bpf_map_update_batch(int fd, const void *keys, const void *values,
> > __u32 *count,
> > const struct bpf_map_batch_opts *opts);
> >
> > +
> > LIBBPF_API int bpf_obj_pin(int fd, const char *pathname);
> > LIBBPF_API int bpf_obj_get(const char *pathname);
> >
