[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-Id: <D032M8W5K5M5.1ZBHEIJFFNTCI@kernel.org>
Date: Mon, 25 Mar 2024 21:16:55 +0200
From: "Jarkko Sakkinen" <jarkko@...nel.org>
To: "Jarkko Sakkinen" <jarkko@...nel.org>, "Masami Hiramatsu"
<mhiramat@...nel.org>
Cc: <linux-riscv@...ts.infradead.org>, "Paul Walmsley"
<paul.walmsley@...ive.com>, "Palmer Dabbelt" <palmer@...belt.com>, "Albert
Ou" <aou@...s.berkeley.edu>, <linux-kernel@...r.kernel.org>, "Naveen N .
Rao" <naveen.n.rao@...ux.ibm.com>, "Anil S Keshavamurthy"
<anil.s.keshavamurthy@...el.com>, "David S . Miller" <davem@...emloft.net>,
<linux-trace-kernel@...r.kernel.org>, "Calvin Owens"
<jcalvinowens@...il.com>
Subject: Re: [PATCH v2] arch/riscv: Enable kprobes when CONFIG_MODULES=n
On Mon Mar 25, 2024 at 9:11 PM EET, Jarkko Sakkinen wrote:
> On Mon Mar 25, 2024 at 8:37 PM EET, Jarkko Sakkinen wrote:
> > > You also should consider using IS_ENABLED(CONFIG_MODULE) in the code to
> > > avoid using #ifdefs.
>
> Hmm... I need make a couple of remarks but open for feedback ofc.
>
> First, trace_kprobe_module_exist depends on find_module()
>
> Second, there is a notifier callback that heavily binds to the module
> subsystem.
>
> In both cases using IS_ENABLED would emit a lot of compilation errors.
Also I think adding 'gfp' makes sense exactly at the point as it has
a use case, i.e. two call sites with differing flags. It makes sense
but should be IMHO added exactly at that time.
Leaving it from my patch set does not do any measurable harm but please
correct if I'm missing something.
BR, Jarkko
Powered by blists - more mailing lists