public inbox for bpf@vger.kernel.org
 help / color / mirror / Atom feed
From: Kui-Feng Lee <sinquersw@gmail.com>
To: Martin KaFai Lau <martin.lau@linux.dev>, thinker.li@gmail.com
Cc: kuifeng@meta.com, bpf@vger.kernel.org, ast@kernel.org,
	song@kernel.org, kernel-team@meta.com, andrii@kernel.org
Subject: Re: [RFC bpf-next v2 1/9] bpf: refactory struct_ops type initialization to a function.
Date: Fri, 15 Sep 2023 18:35:20 -0700	[thread overview]
Message-ID: <b181b09e-9d41-1574-1f12-31f7466c6e4c@gmail.com> (raw)
In-Reply-To: <34a6af4f-ef3d-7e34-0c71-3c76d8f299e2@linux.dev>



On 9/15/23 15:43, Martin KaFai Lau wrote:
> On 9/12/23 11:14 PM, thinker.li@gmail.com wrote:
>> From: Kui-Feng Lee <thinker.li@gmail.com>
>>
>> Move most of code to bpf_struct_ops_init_one() that can be use to
>> initialize new struct_ops types registered dynamically.
> 
> While in RFC, still better to have SOB so that it won't be overlooked in 
> the future.
> 
>> ---
>>   kernel/bpf/bpf_struct_ops.c | 157 +++++++++++++++++++-----------------
>>   1 file changed, 83 insertions(+), 74 deletions(-)
>>
>> diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c
>> index fdc3e8705a3c..1662875e0ebe 100644
>> --- a/kernel/bpf/bpf_struct_ops.c
>> +++ b/kernel/bpf/bpf_struct_ops.c
>> @@ -110,102 +110,111 @@ const struct bpf_prog_ops 
>> bpf_struct_ops_prog_ops = {
>>   static const struct btf_type *module_type;
>> -void bpf_struct_ops_init(struct btf *btf, struct bpf_verifier_log *log)
>> +static void bpf_struct_ops_init_one(struct bpf_struct_ops *st_ops,
>> +                    struct btf *btf,
>> +                    struct bpf_verifier_log *log)
>>   {
>> -    s32 type_id, value_id, module_id;
>>       const struct btf_member *member;
>> -    struct bpf_struct_ops *st_ops;
>>       const struct btf_type *t;
>> +    s32 type_id, value_id;
>>       char value_name[128];
>>       const char *mname;
>> -    u32 i, j;
>> +    int i;
>> -    /* Ensure BTF type is emitted for "struct bpf_struct_ops_##_name" */
>> -#define BPF_STRUCT_OPS_TYPE(_name) BTF_TYPE_EMIT(struct 
>> bpf_struct_ops_##_name);
>> -#include "bpf_struct_ops_types.h"
>> -#undef BPF_STRUCT_OPS_TYPE
>> +    if (strlen(st_ops->name) + VALUE_PREFIX_LEN >=
>> +        sizeof(value_name)) {
>> +        pr_warn("struct_ops name %s is too long\n",
>> +            st_ops->name);
>> +        return;
>> +    }
>> +    sprintf(value_name, "%s%s", VALUE_PREFIX, st_ops->name);
>> -    module_id = btf_find_by_name_kind(btf, "module", BTF_KIND_STRUCT);
>> -    if (module_id < 0) {
>> -        pr_warn("Cannot find struct module in btf_vmlinux\n");
>> +    value_id = btf_find_by_name_kind(btf, value_name,
>> +                     BTF_KIND_STRUCT);
> 
> It needs to do some sanity checks on the value_type since this won't be 
> statically enforced by bpf_struct_ops.c.

Do you mean to check if a value_type has refcnt, state and data field?

> 
> [ ... ]
> 
>> +void bpf_struct_ops_init(struct btf *btf, struct bpf_verifier_log *log)
>> +{
>> +    struct bpf_struct_ops *st_ops;
>> +    s32 module_id;
>> +    u32 i;
>> -            if (__btf_member_bitfield_size(t, member)) {
>> -                pr_warn("bit field member %s in struct %s is not 
>> supported\n",
>> -                    mname, st_ops->name);
>> -                break;
>> -            }
>> +    /* Ensure BTF type is emitted for "struct bpf_struct_ops_##_name" */
>> +#define BPF_STRUCT_OPS_TYPE(_name) BTF_TYPE_EMIT(struct 
>> bpf_struct_ops_##_name);
>> +#include "bpf_struct_ops_types.h"
>> +#undef BPF_STRUCT_OPS_TYPE
> 
> Can this static way of defining struct_ops be removed? bpf_tcp_ca should 
> be able to use the register_bpf_struct_ops() introduced in patch 2.

It sounds good for me.

> 
> For the future subsystem supporting struct_ops, the subsystem could be 
> compiled as a kernel module or as a built-in. register_bpf_struct_ops() 
> should work for both.
> 

  reply	other threads:[~2023-09-16  1:35 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-09-13  6:14 [RFC bpf-next v2 0/9] Registrating struct_ops types from modules thinker.li
2023-09-13  6:14 ` [RFC bpf-next v2 1/9] bpf: refactory struct_ops type initialization to a function thinker.li
2023-09-15 22:43   ` Martin KaFai Lau
2023-09-16  1:35     ` Kui-Feng Lee [this message]
2023-09-13  6:14 ` [RFC bpf-next v2 2/9] bpf: add register and unregister functions for struct_ops thinker.li
2023-09-16  0:05   ` Martin KaFai Lau
2023-09-16  1:14     ` Kui-Feng Lee
2023-09-18 18:47       ` Martin KaFai Lau
2023-09-18 20:40         ` Kui-Feng Lee
2023-09-13  6:14 ` [RFC bpf-next v2 3/9] bpf: attach a module BTF to a bpf_struct_ops thinker.li
2023-09-13  6:14 ` [RFC bpf-next v2 4/9] bpf: use attached BTF to find correct type info of struct_ops progs thinker.li
2023-09-13  6:14 ` [RFC bpf-next v2 5/9] bpf: hold module for bpf_struct_ops_map thinker.li
2023-09-13  6:14 ` [RFC bpf-next v2 6/9] libbpf: Find correct module BTFs for struct_ops maps and progs thinker.li
2023-09-13  6:14 ` [RFC bpf-next v2 7/9] bpf: export btf_ctx_access to modules thinker.li
2023-09-13  6:14 ` [RFC bpf-next v2 8/9] selftests/bpf: test case for register_bpf_struct_ops() thinker.li
2023-09-13  6:14 ` [RFC bpf-next v2 9/9] Comments and debug thinker.li

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b181b09e-9d41-1574-1f12-31f7466c6e4c@gmail.com \
    --to=sinquersw@gmail.com \
    --cc=andrii@kernel.org \
    --cc=ast@kernel.org \
    --cc=bpf@vger.kernel.org \
    --cc=kernel-team@meta.com \
    --cc=kuifeng@meta.com \
    --cc=martin.lau@linux.dev \
    --cc=song@kernel.org \
    --cc=thinker.li@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox