[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <d47235e0-7ecd-483e-82c3-78f13c77df67@amd.com>
Date: Wed, 1 Nov 2023 15:26:23 -0500
From: "Moger, Babu" <babu.moger@....com>
To: Tony Luck <tony.luck@...el.com>, Fenghua Yu <fenghua.yu@...el.com>,
Reinette Chatre <reinette.chatre@...el.com>,
Peter Newman <peternewman@...gle.com>, x86@...nel.org
Cc: 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, patches@...ts.linux.dev
Subject: Re: [PATCH] x86/resctrl: Fix unused variable warning in
cache_alloc_hsw_probe()
Hi Tony,
On 10/31/23 17:05, Tony Luck wrote:
> In a "W=1" build gcc throws a warning:
>
> arch/x86/kernel/cpu/resctrl/core.c: In function ‘cache_alloc_hsw_probe’:
> arch/x86/kernel/cpu/resctrl/core.c:139:16: warning: variable ‘h’ set but not used
>
> Fix by switching from rdmsr() to rdmsrl() using a single u64 argument
> for the MSR value instead of the pair of u32 for the high and low
> halves.
>
> Signed-off-by: Tony Luck <tony.luck@...el.com>
> ---
> This has been annoying me for a while as the only warning from the
> resctrl code when building with W=1.
>
> N.B. compile tested only. I don't have a Haswell system to check this works.
>
> arch/x86/kernel/cpu/resctrl/core.c | 7 ++++---
> 1 file changed, 4 insertions(+), 3 deletions(-)
>
> diff --git a/arch/x86/kernel/cpu/resctrl/core.c b/arch/x86/kernel/cpu/resctrl/core.c
> index 19e0681f0435..4084131d391d 100644
> --- a/arch/x86/kernel/cpu/resctrl/core.c
> +++ b/arch/x86/kernel/cpu/resctrl/core.c
> @@ -136,15 +136,16 @@ static inline void cache_alloc_hsw_probe(void)
> {
> struct rdt_hw_resource *hw_res = &rdt_resources_all[RDT_RESOURCE_L3];
> struct rdt_resource *r = &hw_res->r_resctrl;
> - u32 l, h, max_cbm = BIT_MASK(20) - 1;
> + u32 max_cbm = BIT_MASK(20) - 1;
> + u64 l3_cbm_0;
>
> if (wrmsr_safe(MSR_IA32_L3_CBM_BASE, max_cbm, 0))
> return;
>
> - rdmsr(MSR_IA32_L3_CBM_BASE, l, h);
> + rdmsrl(MSR_IA32_L3_CBM_BASE, l3_cbm_0);
You are writing 32 bit and reading 64 bit. Why don't you change both to 64
bit?
>
> /* If all the bits were set in MSR, return success */
> - if (l != max_cbm)
> + if (l3_cbm_0 != max_cbm)
> return;
>
> hw_res->num_closid = 4;
--
Thanks
Babu Moger
Powered by blists - more mailing lists