Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 23.128.96.18 as permitted sender) client-ip=23.128.96.18;
Date:   Wed, 11 Aug 2021 17:32:03 +0100
From:   Jamie Iles <jamie@nuviainc.com>
To:     James Morse <james.morse@arm.com>
Cc:     x86@kernel.org, linux-kernel@vger.kernel.org,
        Fenghua Yu <fenghua.yu@intel.com>,
        Reinette Chatre <reinette.chatre@intel.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
        H Peter Anvin <hpa@zytor.com>,
        Babu Moger <Babu.Moger@amd.com>,
        shameerali.kolothum.thodi@huawei.com,
        Jamie Iles <jamie@nuviainc.com>,
        D Scott Phillips OS <scott@os.amperecomputing.com>,
        lcherian@marvell.com, bobo.shaobowang@huawei.com
Subject: Re: [PATCH v1 05/20] x86/resctrl: Create mba_sc configuration in the
 rdt_domain
Message-ID: <YRP7g5/T2kJ9BypS@hazel>
References: <20210729223610.29373-1-james.morse@arm.com>
 <20210729223610.29373-6-james.morse@arm.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20210729223610.29373-6-james.morse@arm.com>
Precedence: bulk

Hi James,

On Thu, Jul 29, 2021 at 10:35:55PM +0000, James Morse wrote:
> To support resctrl's MBA software controller, the architecture must provide
> a second configuration array to hold the mbps_val from user-space.
> 
> This complicates the interface between the architecture code.
> 
> Make the filesystem parts of resctrl create an array for the mba_sc
> values when the struct resctrl_schema is created. The software controller
> can be changed to use this, allowing the architecture code to only
> consider the values configured in hardware.
> 
> Signed-off-by: James Morse <james.morse@arm.com>
> ---
>  arch/x86/kernel/cpu/resctrl/internal.h |  1 -
>  arch/x86/kernel/cpu/resctrl/rdtgroup.c | 69 ++++++++++++++++++++++++++
>  include/linux/resctrl.h                | 13 +++++
>  3 files changed, 82 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kernel/cpu/resctrl/internal.h b/arch/x86/kernel/cpu/resctrl/internal.h
> index e12b55f815bf..a7e2cbce29d5 100644
> --- a/arch/x86/kernel/cpu/resctrl/internal.h
> +++ b/arch/x86/kernel/cpu/resctrl/internal.h
> @@ -36,7 +36,6 @@
>  #define MBM_OVERFLOW_INTERVAL		1000
>  #define MAX_MBA_BW			100u
>  #define MBA_IS_LINEAR			0x4
> -#define MBA_MAX_MBPS			U32_MAX
>  #define MAX_MBA_BW_AMD			0x800
>  #define MBM_CNTR_WIDTH_OFFSET_AMD	20
>  
> diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
> index cf0db0b7a5d0..185f9bb992d1 100644
> --- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
> +++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
> @@ -2030,6 +2030,60 @@ static int mkdir_mondata_all(struct kernfs_node *parent_kn,
>  			     struct rdtgroup *prgrp,
>  			     struct kernfs_node **mon_data_kn);
>  
> +static int mba_sc_domain_allocate(struct rdt_resource *res,
> +				  struct rdt_domain *d)
> +{
> +	u32 num_closid = closid_free_map_len;
> +	int cpu = cpumask_any(&d->cpu_mask);
> +	int i;
> +
> +	d->mba_sc = kcalloc_node(num_closid, sizeof(*d->mba_sc),
> +				 GFP_KERNEL, cpu_to_node(cpu));
> +	if (!d->mba_sc)
> +		return -ENOMEM;

If a CPU was hotplugged before resctrl is mounted then isn't it possible 
for this to already be allocated?  I might be misunderstanding the flows 
here though...

> +	for (i = 0; i < num_closid; i++)
> +		d->mba_sc[i].mbps_val = MBA_MAX_MBPS;
> +
> +	return 0;
> +}
> +
> +static int mba_sc_allocate(struct rdt_resource *r)
> +{
> +	struct rdt_domain *d;
> +	int ret;
> +
> +	lockdep_assert_cpus_held();
> +
> +	if (!is_mba_sc(r))
> +		return 0;
> +
> +	list_for_each_entry(d, &r->domains, list) {
> +		ret = mba_sc_domain_allocate(r, d);
> +		if (ret)
> +			break;
> +	}
> +
> +	return ret;
> +}
> +
> +static void mba_sc_domain_destroy(struct rdt_resource *r,
> +				  struct rdt_domain *d)
> +{
> +	kfree(d->mba_sc);
> +	d->mba_sc = NULL;
> +}
> +
> +static void mba_sc_destroy(struct rdt_resource *r)
> +{
> +	struct rdt_domain *d;
> +
> +	lockdep_assert_cpus_held();
> +
> +	list_for_each_entry(d, &r->domains, list)
> +		mba_sc_domain_destroy(r, d);
> +}
> +
>  static int rdt_enable_ctx(struct rdt_fs_context *ctx)
>  {
>  	int ret = 0;
> @@ -2117,17 +2171,27 @@ static int schemata_list_create(void)
>  
>  		if (ret)
>  			break;
> +
> +		ret = mba_sc_allocate(r);
> +		if (ret)
> +			break;
>  	}
>  
>  	return ret;
>  }
>  
> +/*
> + * During rdt_kill_sb(), the mba_sc state is reset before
> + * destroy_schemata_list() is called: unconditionally try to free the
> + * array.
> + */
>  static void schemata_list_destroy(void)
>  {
>  	struct resctrl_schema *s, *tmp;
>  
>  	list_for_each_entry_safe(s, tmp, &resctrl_schema_all, list) {
>  		list_del(&s->list);
> +		mba_sc_destroy(s->res);
>  		kfree(s);
>  	}
>  }
> @@ -3255,6 +3319,8 @@ void resctrl_offline_domain(struct rdt_resource *r, struct rdt_domain *d)
>  		__check_limbo(d, true);
>  		cancel_delayed_work(&d->cqm_limbo);
>  	}
> +	if (static_branch_unlikely(&rdt_enable_key) && is_mba_sc(r))
> +		mba_sc_domain_destroy(r, d);
>  	bitmap_free(d->rmid_busy_llc);
>  	kfree(d->mbm_total);
>  	kfree(d->mbm_local);
> @@ -3287,6 +3353,9 @@ static int domain_setup_mon_state(struct rdt_resource *r, struct rdt_domain *d)
>  		}
>  	}
>  
> +	if (is_mba_sc(r))
> +		mba_sc_domain_allocate(r, d);

This looks to be missing an error check.

Thanks,

Jamie