lkml.org 
[lkml]   [2022]   [Jul]   [15]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: [PATCH v2 bpf-next 3/5] ftrace: introduce FTRACE_OPS_FL_SHARE_IPMODIFY
Date


> On Jul 15, 2022, at 12:59 PM, Steven Rostedt <rostedt@goodmis.org> wrote:
>
> On Fri, 15 Jul 2022 19:49:00 +0000
> Song Liu <songliubraving@fb.com> wrote:
>
>>>
>>> What about if we release the lock when doing the callback?
>>
>> We can probably unlock ftrace_lock here. But we may break locking order
>> with direct mutex (see below).
>
> You're talking about the multi registering case, right?

We are using the *_ftrace_direct_multi() API here, to be able to specify
ops_func. The direct single API just uses the shared direct_ops.

>
>>
>>>
>>> Then we just need to make sure things are the same after reacquiring the
>>> lock, and if they are different, we release the lock again and do the
>>> callback with the new update. Wash, rinse, repeat, until the state is the
>>> same before and after the callback with locks acquired?
>>
>> Personally, I would like to avoid wash-rinse-repeat here.
>
> But it's common to do. Keeps your hair cleaner that way ;-)
>
>>
>>>
>>> This is a common way to handle callbacks that need to do something that
>>> takes the lock held before doing a callback.
>>>
>>> The reason I say this, is because the more we can keep the accounting
>>> inside of ftrace the better.
>>>
>>> Wouldn't this need to be done anyway if BPF was first and live kernel
>>> patching needed the update? An -EAGAIN would not suffice.
>>
>> prepare_direct_functions_for_ipmodify handles BPF-first-livepatch-later
>> case. The benefit of prepare_direct_functions_for_ipmodify() is that it
>> holds direct_mutex before ftrace_lock, and keeps holding it if necessary.
>> This is enough to make sure we don't need the wash-rinse-repeat.
>>
>> OTOH, if we wait until __ftrace_hash_update_ipmodify(), we already hold
>> ftrace_lock, but not direct_mutex. To make changes to bpf trampoline, we
>> have to unlock ftrace_lock and lock direct_mutex to avoid deadlock.
>> However, this means we will need the wash-rinse-repeat.

What do you think about the prepare_direct_functions_for_ipmodify()
approach? If this is not ideal, maybe we can simplify it so that it only
holds direct_mutex (when necessary). The benefit is that we are sure
direct_mutex is already held in __ftrace_hash_update_ipmodify(). However,
I think it is not safe to unlock ftrace_lock in __ftrace_hash_update_ipmodify().
We can get parallel do_for_each_ftrace_rec(), which is dangerous, no?

>>
>>
>> For livepatch-first-BPF-later case, we can probably handle this in
>> __ftrace_hash_update_ipmodify(), since we hold both direct_mutex and
>> ftrace_lock. We can unlock ftrace_lock and update the BPF trampoline.
>> It is safe against changes to direct ops, because we are still holding
>> direct_mutex. But, is this safe against another IPMODIFY ops? I am not
>> sure yet... Also, this is pretty weird because, we are updating a
>> direct trampoline before we finish registering it for the first time.
>> IOW, we are calling modify_ftrace_direct_multi_nolock for the same
>> trampoline before register_ftrace_direct_multi() returns.
>>
>> The approach in v2 propagates the -EAGAIN to BPF side, so these are two
>> independent calls of register_ftrace_direct_multi(). This does require
>> some protocol between ftrace core and its user, but I still think this
>> is a cleaner approach.
>
> The issue I have with this approach is it couples BPF and ftrace a bit too
> much.
>
> But there is a way with my approach you can still do your approach. That
> is, have ops_func() return zero if everything is fine, and otherwise returns
> a negative value. Then have the register function fail and return whatever
> value that gets returned by the ops_func()
>
> Then have the bpf ops_func() check (does this direct caller handle
> IPMODIFY? if yes, return 0, else return -EAGAIN). Then the registering of
> ftrace fails with your -EAGAIN, and then you can change the direct
> trampoline to handle IPMODIFY and try again. This time when ops_func() is
> called, it sees that the direct trampoline can handle the IPMODIFY and
> returns 0.
>
> Basically, it's a way to still implement my suggestion, but let BPF decide
> to use -EAGAIN to try again. And then BPF and ftrace don't need to have
> these special flags to change the behavior of each other.

I like this one. So there is no protocol about the return value here.

Thanks,
Song

\
 
 \ /
  Last update: 2022-07-15 22:23    [W:0.476 / U:0.076 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site