Re: static_branch_enable() does not work from a __init function?

From: Peter Zijlstra <peterz@infradead.org>
To: Jessica Yu <jeyu@kernel.org>
Cc: Dexuan Cui <decui@microsoft.com>, Ingo Molnar <mingo@kernel.org>,
	Daniel Bristot de Oliveira <bristot@redhat.com>,
	"kvm@vger.kernel.org" <kvm@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Josh Poimboeuf <jpoimboe@redhat.com>
Subject: Re: static_branch_enable() does not work from a __init function?
Date: Wed, 16 Dec 2020 13:47:08 +0100	[thread overview]
Message-ID: <20201216124708.GZ3021@hirez.programming.kicks-ass.net> (raw)
In-Reply-To: <20201216115524.GA13751@linux-8ccs>

On Wed, Dec 16, 2020 at 12:55:25PM +0100, Jessica Yu wrote:
> +++ Peter Zijlstra [16/12/20 10:26 +0100]:
> [snip]
> > > PS, I originally found: in arch/x86/kvm/vmx/vmx.c: vmx_init(), it looks
> > > like the line "static_branch_enable(&enable_evmcs);" does not take effect
> > > in a v5.4-based kernel, but does take effect in the v5.10 kernel in the
> > > same x86-64 virtual machine on Hyper-V, so I made the above test module
> > > to test static_branch_enable(), and found that static_branch_enable() in
> > > the test module does not work with both v5.10 and my v5.4 kernel, if the
> > > __init marker is used.
> 
> Because the jump label code currently does not allow you to update if
> the entry resides in an init section. By marking the module init
> section __init you place it in the .init.text section.
> jump_label_add_module() detects this (by calling within_module_init())
> and marks the entry by calling jump_entry_set_init(). Then you have
> the following sequence of calls (roughly):
> 
> static_branch_enable
>  static_key_enable
>    static_key_enable_cpuslocked
>      jump_label_update
>        jump_label_can_update
>          jump_entry_is_init returns true, so bail out
> 
> Judging from the comment in jump_label_can_update(), this seems to be
> intentional behavior:
> 
> static bool jump_label_can_update(struct jump_entry *entry, bool init)
> {
>        /*
>         * Cannot update code that was in an init text area.
>         */
>        if (!init && jump_entry_is_init(entry))
>                return false;
> 

Only because we're having .init=false, incorrectly. See the other email.