[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <d0a6ca2d-250a-47e2-8dda-b4e7d74a7ffb@intel.com>
Date: Fri, 3 Nov 2023 13:23:57 -0700
From: Reinette Chatre <reinette.chatre@...el.com>
To: "Luck, Tony" <tony.luck@...el.com>,
"Moger, Babu" <babu.moger@....com>
CC: "Yu, Fenghua" <fenghua.yu@...el.com>,
Peter Newman <peternewman@...gle.com>,
"x86@...nel.org" <x86@...nel.org>,
Shaopeng Tan <tan.shaopeng@...itsu.com>,
James Morse <james.morse@....com>,
Jamie Iles <quic_jiles@...cinc.com>,
Randy Dunlap <rdunlap@...radead.org>,
"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
"patches@...ts.linux.dev" <patches@...ts.linux.dev>
Subject: Re: [PATCH v2] x86/resctrl: Fix unused variable warning in
cache_alloc_hsw_probe()
On 11/2/2023 3:50 PM, Luck, Tony wrote:
>> This does seem a valid issue and my Ack remains. I'm just puzzled why I do not
>> encounter the same warning.
>
> Reinette,
>
> Some other CONFIG option changing CFLAGS?
>
> Here's what I have for a "make V=1 W=1" for core.o
>
Tony and I went a bit more back-and-forth on this one and it appears that the
relevant code is optimized out in my build while the same did not happen
for him.
I can now trigger the warning by enabling Xen support, with CONFIG_PARAVIRT_XXL
causing rdmsr() to be replaced by paravirt_read_msr(). With the relevant code
no longer optimized out, the warning is triggered.
Thanks Tony!
Reinette
Powered by blists - more mailing lists