[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <6006810f-6419-bc39-7eff-1bd31a572631@linux.vnet.ibm.com>
Date: Fri, 14 Jul 2023 19:55:10 +0530
From: Shrikanth Hegde <sshegde@...ux.vnet.ibm.com>
To: Peter Zijlstra <peterz@...radead.org>
Cc: juri.lelli@...hat.com, dave.hansen@...ux.intel.com,
bsegall@...gle.com, hpa@...or.com, agordeev@...ux.ibm.com,
linux-s390@...r.kernel.org, vincent.guittot@...aro.org,
x86@...nel.org, mingo@...hat.com, mgorman@...e.de,
borntraeger@...ux.ibm.com, gor@...ux.ibm.com, hca@...ux.ibm.com,
npiggin@...il.com, bp@...en8.de, rostedt@...dmis.org,
dietmar.eggemann@....com, linuxppc-dev@...ts.ozlabs.org,
linux-kernel@...r.kernel.org, svens@...ux.ibm.com,
bristot@...hat.com, Thomas Gleixner <tglx@...utronix.de>,
Valentin Schneider <vschneid@...hat.com>
Subject: Re: [RFC][PATCH] sched: Rename DIE domain
On 7/12/23 8:32 PM, Valentin Schneider wrote:
> On 12/07/23 16:10, Peter Zijlstra wrote:
>> Hi
>>
>> Thomas just tripped over the x86 topology setup creating a 'DIE' domain
>> for the package mask :-)
>>
>> Since these names are SCHED_DEBUG only, rename them.
>> I don't think anybody *should* be relying on this, but who knows.
>>
>
> FWIW I don't care much about the actual name.
>
> There are some stray references to DIE in comments - see below. Bit funny
> to see:
> - * - Package (DIE)
> + * - Package (PKG)
>
> With that:
> Acked-by: Valentin Schneider <vschneid@...hat.com>
>
> ---
> diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
> index a80a73909dc2a..190a647534984 100644
> --- a/kernel/sched/fair.c
> +++ b/kernel/sched/fair.c
> @@ -9439,7 +9439,7 @@ static bool sched_use_asym_prio(struct sched_domain *sd, int cpu)
> * can only do it if @group is an SMT group and has exactly on busy CPU. Larger
> * imbalances in the number of CPUS are dealt with in find_busiest_group().
> *
> - * If we are balancing load within an SMT core, or at DIE domain level, always
> + * If we are balancing load within an SMT core, or at PKG domain level, always
> * proceed.
> *
> * Return: true if @env::dst_cpu can do with asym_packing load balance. False
> diff --git a/kernel/sched/topology.c b/kernel/sched/topology.c
> index e9d9cf776b7ab..2cdcfec1d1c89 100644
> --- a/kernel/sched/topology.c
> +++ b/kernel/sched/topology.c
> @@ -1118,7 +1118,7 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu)
> *
> * - Simultaneous multithreading (SMT)
> * - Multi-Core Cache (MC)
> - * - Package (DIE)
> + * - Package (PKG)
> *
> * Where the last one more or less denotes everything up to a NUMA node.
> *
> @@ -1140,13 +1140,13 @@ build_overlap_sched_groups(struct sched_domain *sd, int cpu)
> *
> * CPU 0 1 2 3 4 5 6 7
> *
> - * DIE [ ]
> + * PKG [ ]
> * MC [ ] [ ]
> * SMT [ ] [ ] [ ] [ ]
> *
> * - or -
> *
> - * DIE 0-7 0-7 0-7 0-7 0-7 0-7 0-7 0-7
> + * PKG 0-7 0-7 0-7 0-7 0-7 0-7 0-7 0-7
> * MC 0-3 0-3 0-3 0-3 4-7 4-7 4-7 4-7
> * SMT 0-1 0-1 2-3 2-3 4-5 4-5 6-7 6-7
> *
>
A couple of comments missing the change still? Not sure if it taken care already.
---
arch/powerpc/kernel/smp.c | 2 +-
arch/x86/kernel/smpboot.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/kernel/smp.c b/arch/powerpc/kernel/smp.c
index fbbb695bae3d..9b1853bf6b1d 100644
--- a/arch/powerpc/kernel/smp.c
+++ b/arch/powerpc/kernel/smp.c
@@ -1588,7 +1588,7 @@ static void add_cpu_to_masks(int cpu)
/* Skip all CPUs already part of current CPU core mask */
cpumask_andnot(mask, cpu_online_mask, cpu_core_mask(cpu));
- /* If chip_id is -1; limit the cpu_core_mask to within DIE*/
+ /* If chip_id is -1; limit the cpu_core_mask to within PKG*/
if (chip_id == -1)
cpumask_and(mask, mask, cpu_cpu_mask(cpu));
diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c
index e1aa2cd7734b..3f175e70eb7a 100644
--- a/arch/x86/kernel/smpboot.c
+++ b/arch/x86/kernel/smpboot.c
@@ -647,7 +647,7 @@ static void __init build_sched_topology(void)
};
#endif
/*
- * When there is NUMA topology inside the package skip the DIE domain
+ * When there is NUMA topology inside the package skip the PKG domain
* since the NUMA domains will auto-magically create the right spanning
* domains based on the SLIT.
*/
Powered by blists - more mailing lists