From: Dave Marchevsky <davemarchevsky@fb.com>
To: Yonghong Song <yhs@fb.com>, bpf@vger.kernel.org
Cc: Alexei Starovoitov <ast@kernel.org>,
Daniel Borkmann <daniel@iogearbox.net>,
Andrii Nakryiko <andrii@kernel.org>,
Kernel Team <kernel-team@fb.com>, Tejun Heo <tj@kernel.org>
Subject: Re: [RFC PATCH bpf-next 00/11] bpf: Introduce rbtree map
Date: Wed, 10 Aug 2022 13:54:30 -0400 [thread overview]
Message-ID: <3a5a3c88-98e9-bc0e-c5b3-4c6bf0f49620@fb.com> (raw)
In-Reply-To: <e7d884ee-e0d3-02b2-c3d4-3c7bac8f13fc@fb.com>
On 7/28/22 3:04 AM, Yonghong Song wrote:
>
>
> On 7/22/22 11:34 AM, Dave Marchevsky wrote:
>> Introduce bpf_rbtree map data structure. As the name implies, rbtree map
>> allows bpf programs to use red-black trees similarly to kernel code.
>> Programs interact with rbtree maps in a much more open-coded way than
>> more classic map implementations. Some example code to demonstrate:
>>
>> node = bpf_rbtree_alloc_node(&rbtree, sizeof(struct node_data));
>> if (!node)
>> return 0;
>>
>> node->one = calls;
>> node->two = 6;
>> bpf_rbtree_lock(bpf_rbtree_get_lock(&rbtree));
>
> Can we just do
> bpf_rbtree_lock(&rbtree)
> bpf_rbtree_unlock(&rbtree)
> ? Looks like the only places bpf_rbtree_get_lock() used are
> as arguments of bpf_rbtree_lock/unlock or bpf_spin_lock/unlock?
>
Summarizing our VC convo: the intent here is to have the lock live separately
from the tree, meaning it'd be separately initialized and passed into the tree
instead of current state where it's a bpf_rbtree field.
If the tree still keeps a pointer to the lock - ideally passed in when rbtree
map is instantiated - it'll be possible to move to a cleaner API as you
describe. The very explicit way it's done in this series is not the end state
I'd like either.
>>
>> ret = (struct node_data *)bpf_rbtree_add(&rbtree, node, less);
>> if (!ret) {
>> bpf_rbtree_free_node(&rbtree, node);
>> goto unlock_ret;
>> }
>>
>> unlock_ret:
>> bpf_rbtree_unlock(bpf_rbtree_get_lock(&rbtree));
>> return 0;
>>
>>
>> This series is in a heavy RFC state, with some added verifier semantics
>> needing improvement before they can be considered safe. I am sending
>> early to gather feedback on approach:
>>
>> * Does the API seem reasonable and might it be useful for others?
>>
>> * Do new verifier semantics added in this series make logical sense?
>> Are there any glaring safety holes aside from those called out in
>> individual patches?
>>
>> Please see individual patches for more in-depth explanation. A quick
>> summary of patches follows:
>>
>>
>> Patches 1-3 extend verifier and BTF searching logic in minor ways to
>> prepare for rbtree implementation patch.
>> bpf: Pull repeated reg access bounds check into helper fn
>> bpf: Add verifier support for custom callback return range
>> bpf: Add rb_node_off to bpf_map
>>
>>
>> Patch 4 adds basic rbtree map implementation.
>> bpf: Add rbtree map
>>
>> Note that 'complete' implementation requires concepts and changes
>> introduced in further patches in the series. The series is currently
>> arranged in this way to ease RFC review.
>>
>>
>> Patches 5-7 add a spinlock to the rbtree map, with some differing
>> semantics from existing verifier spinlock handling.
>> bpf: Add bpf_spin_lock member to rbtree
>> bpf: Add bpf_rbtree_{lock,unlock} helpers
>> bpf: Enforce spinlock hold for bpf_rbtree_{add,remove,find}
>>
>> Notably, rbtree's bpf_spin_lock must be held while manipulating the tree
>> via helpers, while existing spinlock verifier logic prevents any helper
>> calls while lock is held. In current state this is worked around by not
>> having the verifier treat rbtree's lock specially in any way. This
>> needs to be improved before leaving RFC state as it's unsafe.
>>
> [...]
next prev parent reply other threads:[~2022-08-10 17:54 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-07-22 18:34 [RFC PATCH bpf-next 00/11] bpf: Introduce rbtree map Dave Marchevsky
2022-07-22 18:34 ` [RFC PATCH bpf-next 01/11] bpf: Pull repeated reg access bounds check into helper fn Dave Marchevsky
2022-07-22 18:34 ` [RFC PATCH bpf-next 02/11] bpf: Add verifier support for custom callback return range Dave Marchevsky
2022-07-22 18:34 ` [RFC PATCH bpf-next 03/11] bpf: Add rb_node_off to bpf_map Dave Marchevsky
2022-08-01 22:19 ` Alexei Starovoitov
2022-07-22 18:34 ` [RFC PATCH bpf-next 04/11] bpf: Add rbtree map Dave Marchevsky
2022-08-01 21:49 ` Alexei Starovoitov
2022-07-22 18:34 ` [RFC PATCH bpf-next 05/11] bpf: Add bpf_spin_lock member to rbtree Dave Marchevsky
2022-08-01 22:17 ` Alexei Starovoitov
2022-08-02 13:59 ` Kumar Kartikeya Dwivedi
2022-08-02 15:30 ` Alexei Starovoitov
2022-08-10 21:46 ` Kumar Kartikeya Dwivedi
2022-08-10 22:06 ` Alexei Starovoitov
2022-08-10 23:16 ` Kumar Kartikeya Dwivedi
2022-08-15 5:33 ` Yonghong Song
2022-08-15 5:37 ` Kumar Kartikeya Dwivedi
2022-07-22 18:34 ` [RFC PATCH bpf-next 06/11] bpf: Add bpf_rbtree_{lock,unlock} helpers Dave Marchevsky
2022-08-01 21:58 ` Alexei Starovoitov
2022-07-22 18:34 ` [RFC PATCH bpf-next 07/11] bpf: Enforce spinlock hold for bpf_rbtree_{add,remove,find} Dave Marchevsky
2022-07-22 18:34 ` [RFC PATCH bpf-next 08/11] bpf: Add OBJ_NON_OWNING_REF type flag Dave Marchevsky
2022-08-01 22:41 ` Alexei Starovoitov
2022-07-22 18:34 ` [RFC PATCH bpf-next 09/11] bpf: Add CONDITIONAL_RELEASE " Dave Marchevsky
2022-08-01 22:23 ` Alexei Starovoitov
2022-07-22 18:34 ` [RFC PATCH bpf-next 10/11] bpf: Introduce PTR_ITER and PTR_ITER_END type flags Dave Marchevsky
2022-07-29 16:31 ` Tejun Heo
2022-08-01 22:44 ` Alexei Starovoitov
2022-08-02 13:05 ` Kumar Kartikeya Dwivedi
2022-08-02 15:10 ` Alexei Starovoitov
2022-08-10 17:56 ` Dave Marchevsky
2022-07-22 18:34 ` [RFC PATCH bpf-next 11/11] selftests/bpf: Add rbtree map tests Dave Marchevsky
2022-07-28 7:18 ` Yonghong Song
2022-08-10 17:48 ` Dave Marchevsky
2022-07-28 7:04 ` [RFC PATCH bpf-next 00/11] bpf: Introduce rbtree map Yonghong Song
2022-08-10 17:54 ` Dave Marchevsky [this message]
2022-08-01 21:27 ` Alexei Starovoitov
2022-08-10 18:11 ` Dave Marchevsky
2022-08-02 22:02 ` Andrii Nakryiko
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=3a5a3c88-98e9-bc0e-c5b3-4c6bf0f49620@fb.com \
--to=davemarchevsky@fb.com \
--cc=andrii@kernel.org \
--cc=ast@kernel.org \
--cc=bpf@vger.kernel.org \
--cc=daniel@iogearbox.net \
--cc=kernel-team@fb.com \
--cc=tj@kernel.org \
--cc=yhs@fb.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;
as well as URLs for NNTP newsgroup(s).