Re: [PATCH v4 3/3] rcu: Use _full() API to debug synchronize_rcu()

[Joel Fernandes <joelagnelf@nvidia.com>]

On 3/3/2025 12:30 PM, Joel Fernandes wrote:
> 
> 
> On 3/3/2025 12:07 PM, Boqun Feng wrote:
>> On Mon, Mar 03, 2025 at 12:00:40PM -0500, Joel Fernandes wrote:
>> [...]
>>>
>>> I see the original patch "rcu: Fix get_state_synchronize_rcu_full() GP-start
>>> detection" is not yet on -next. Once we are convinced about the fix, do we
>>> want to squash the fix into this patch and have Boqun take it?
>>>
>>
>> Which "-next" are you talking about? The original patch and the fix is
>> already in next-20250303 of linux-next:
>>
>> 	https://web.git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git/log/?h=next-20250303&qt=range&q=153fc45000e0058435ec0609258fb16e7ea257d2
> I see it now during manual inspection, but I'm confused why git cherry tells me
> otherwise.
> 
> I tried the following command and it shows the patch in question in the first
> line of output. Basically the question that the command asks is "What is in
> Paul's dev branch that is not in RCU tree's -next branch".  This question is
> asked for the obvious raisins.
> So I am obviously missing something in the command. Thoughts?
> 
> (rcugit is the RCU tree, and paul/dev is Paul's dev branch)
> 
> git cherry --abbrev -v rcugit/next paul/dev | grep "^+" | cut -d' ' -f2,3-
> 
> 012f47f0f806 rcu: Fix get_state_synchronize_rcu_full() GP-start detection
Ah, its because Paul's dev branch has an older version of the patch I think
(without the fix being discussed here) :-D

So that explains it...

thanks,

 - Joel