From: Dave Martin <Dave.Martin@arm.com>
To: x86@kernel.org, linux-kernel@vger.kernel.org
Cc: James Morse <james.morse@arm.com>,
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,
D Scott Phillips OS <scott@os.amperecomputing.com>,
carl@os.amperecomputing.com, lcherian@marvell.com,
bobo.shaobowang@huawei.com,
"Shaopeng Tan (Fujitsu)" <tan.shaopeng@fujitsu.com>,
baolin.wang@linux.alibaba.com,
Jamie Iles <quic_jiles@quicinc.com>,
Xin Hao <xhao@linux.alibaba.com>,
Peter Newman <peternewman@google.com>,
dfustini@baylibre.com, amitsinght@marvell.com,
David Hildenbrand <david@redhat.com>,
Rex Nie <rex.nie@jaguarmicro.com>
Subject: [PATCH v2 04/35] x86/resctrl: Add helper for setting CPU default properties
Date: Fri, 26 Apr 2024 16:08:33 +0100 [thread overview]
Message-ID: <20240426150904.8854-4-Dave.Martin@arm.com> (raw)
In-Reply-To: <20240426150904.8854-1-Dave.Martin@arm.com>
From: James Morse <james.morse@arm.com>
rdtgroup_rmdir_ctrl() and rdtgroup_rmdir_mon() set the per-CPU
pqr_state for CPUs that were part of the rmdir()'d group.
Another architecture might not have a 'pqr_state', its hardware may
need the values in a different format. MPAM's equivalent of RMID values
are not unique, and always need the CLOSID to be provided too.
There is only one caller that modifies a single value,
(rdtgroup_rmdir_mon()). MPAM always needs both CLOSID and RMID
for the hardware value as these are written to the same system
register.
As rdtgroup_rmdir_mon() has the CLOSID on hand, only provide a
helper to set both values. These values are read by
__resctrl_sched_in(), but may be written by a different CPU without
any locking, add READ/WRTE_ONCE() to avoid torn values.
Signed-off-by: James Morse <james.morse@arm.com>
Signed-off-by: Dave Martin <Dave.Martin@arm.com>
---
Changes in v2:
* In rdtgroup_rmdir_mon(), (re)set CPU default closid based on the
parent control group, to avoid the appearance of referencing
something that we're in the process of destroying (even if it
doesn't make a difference because the victim mon group necessarily
has the same closid as the parent control group).
Update comment to match.
No (intentional) functional change.
---
arch/x86/include/asm/resctrl.h | 14 +++++++++++---
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 20 ++++++++++++++------
2 files changed, 25 insertions(+), 9 deletions(-)
diff --git a/arch/x86/include/asm/resctrl.h b/arch/x86/include/asm/resctrl.h
index 12dbd2588ca7..f61382258743 100644
--- a/arch/x86/include/asm/resctrl.h
+++ b/arch/x86/include/asm/resctrl.h
@@ -4,8 +4,9 @@
#ifdef CONFIG_X86_CPU_RESCTRL
-#include <linux/sched.h>
#include <linux/jump_label.h>
+#include <linux/percpu.h>
+#include <linux/sched.h>
/*
* This value can never be a valid CLOSID, and is used when mapping a
@@ -96,8 +97,8 @@ static inline void resctrl_arch_disable_mon(void)
static inline void __resctrl_sched_in(struct task_struct *tsk)
{
struct resctrl_pqr_state *state = this_cpu_ptr(&pqr_state);
- u32 closid = state->default_closid;
- u32 rmid = state->default_rmid;
+ u32 closid = READ_ONCE(state->default_closid);
+ u32 rmid = READ_ONCE(state->default_rmid);
u32 tmp;
/*
@@ -132,6 +133,13 @@ static inline unsigned int resctrl_arch_round_mon_val(unsigned int val)
return val * scale;
}
+static inline void resctrl_arch_set_cpu_default_closid_rmid(int cpu, u32 closid,
+ u32 rmid)
+{
+ WRITE_ONCE(per_cpu(pqr_state.default_closid, cpu), closid);
+ WRITE_ONCE(per_cpu(pqr_state.default_rmid, cpu), rmid);
+}
+
static inline void resctrl_arch_set_closid_rmid(struct task_struct *tsk,
u32 closid, u32 rmid)
{
diff --git a/arch/x86/kernel/cpu/resctrl/rdtgroup.c b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
index 47234562a1dd..98c051ac848d 100644
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -3624,14 +3624,21 @@ static int rdtgroup_mkdir(struct kernfs_node *parent_kn, const char *name,
static int rdtgroup_rmdir_mon(struct rdtgroup *rdtgrp, cpumask_var_t tmpmask)
{
struct rdtgroup *prdtgrp = rdtgrp->mon.parent;
+ u32 closid, rmid;
int cpu;
/* Give any tasks back to the parent group */
rdt_move_group_tasks(rdtgrp, prdtgrp, tmpmask);
- /* Update per cpu rmid of the moved CPUs first */
+ /*
+ * Update per cpu closid/rmid of the moved CPUs first.
+ * Note: the closid will not change, but the arch code still needs it.
+ */
+ closid = prdtgrp->closid;
+ rmid = prdtgrp->mon.rmid;
for_each_cpu(cpu, &rdtgrp->cpu_mask)
- per_cpu(pqr_state.default_rmid, cpu) = prdtgrp->mon.rmid;
+ resctrl_arch_set_cpu_default_closid_rmid(cpu, closid, rmid);
+
/*
* Update the MSR on moved CPUs and CPUs which have moved
* task running on them.
@@ -3664,6 +3671,7 @@ static int rdtgroup_ctrl_remove(struct rdtgroup *rdtgrp)
static int rdtgroup_rmdir_ctrl(struct rdtgroup *rdtgrp, cpumask_var_t tmpmask)
{
+ u32 closid, rmid;
int cpu;
/* Give any tasks back to the default group */
@@ -3674,10 +3682,10 @@ static int rdtgroup_rmdir_ctrl(struct rdtgroup *rdtgrp, cpumask_var_t tmpmask)
&rdtgroup_default.cpu_mask, &rdtgrp->cpu_mask);
/* Update per cpu closid and rmid of the moved CPUs first */
- for_each_cpu(cpu, &rdtgrp->cpu_mask) {
- per_cpu(pqr_state.default_closid, cpu) = rdtgroup_default.closid;
- per_cpu(pqr_state.default_rmid, cpu) = rdtgroup_default.mon.rmid;
- }
+ closid = rdtgroup_default.closid;
+ rmid = rdtgroup_default.mon.rmid;
+ for_each_cpu(cpu, &rdtgrp->cpu_mask)
+ resctrl_arch_set_cpu_default_closid_rmid(cpu, closid, rmid);
/*
* Update the MSR on moved CPUs and CPUs which have moved
--
2.34.1
next prev parent reply other threads:[~2024-04-26 15:09 UTC|newest]
Thread overview: 50+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-04-26 15:05 [PATCH v2 00/35] x86/resctrl: Move the resctrl filesystem code to /fs/resctrl Dave Martin
2024-04-26 15:08 ` [PATCH v2 01/35] x86/resctrl: Fix allocation of cleanest CLOSID on platforms with no monitors Dave Martin
2024-04-26 15:08 ` [PATCH v2 02/35] x86/resctrl: Add a helper to avoid reaching into the arch code resource list Dave Martin
2024-04-26 15:08 ` [PATCH v2 03/35] x86/resctrl: Move ctrlval string parsing policy away from the arch code Dave Martin
2024-04-26 15:08 ` Dave Martin [this message]
2024-04-26 15:08 ` [PATCH v2 05/35] x86/resctrl: Remove rdtgroup from update_cpu_closid_rmid() Dave Martin
2024-04-26 15:08 ` [PATCH v2 06/35] x86/resctrl: Export resctrl fs's init function Dave Martin
2024-05-23 0:48 ` Shaopeng Tan (Fujitsu)
2024-04-26 15:08 ` [PATCH v2 07/35] x86/resctrl: Wrap resctrl_arch_find_domain() around rdt_find_domain() Dave Martin
2024-04-26 15:08 ` [PATCH v2 08/35] x86/resctrl: Move resctrl types to a separate header Dave Martin
2024-04-26 15:08 ` [PATCH v2 09/35] x86/resctrl: Add a resctrl helper to reset all the resources Dave Martin
2024-04-26 15:08 ` [PATCH v2 10/35] x86/resctrl: Move monitor init work to a resctrl init call Dave Martin
2024-04-26 15:08 ` [PATCH v2 11/35] x86/resctrl: Move monitor exit work to a resctrl exit call Dave Martin
2024-04-26 15:08 ` [PATCH v2 12/35] x86/resctrl: Move max_{name,data}_width into resctrl code Dave Martin
2024-04-26 15:08 ` [PATCH v2 13/35] x86/resctrl: Stop using the for_each_*_rdt_resource() walkers Dave Martin
2024-04-26 15:08 ` [PATCH v2 14/35] x86/resctrl: Export the is_mbm_*_enabled() helpers to asm/resctrl.h Dave Martin
2024-04-26 15:08 ` [PATCH v2 15/35] x86/resctrl: Add resctrl_arch_is_evt_configurable() to abstract BMEC Dave Martin
2024-04-26 15:08 ` [PATCH v2 16/35] x86/resctrl: Change mon_event_config_{read,write}() to be arch helpers Dave Martin
2024-04-26 15:08 ` [PATCH v2 17/35] x86/resctrl: Move mbm_cfg_mask to struct rdt_resource Dave Martin
2024-06-14 13:57 ` James Morse
2024-04-26 15:08 ` [PATCH v2 18/35] x86/resctrl: Allow resctrl_arch_mon_event_config_write() to return an error Dave Martin
2024-04-26 15:08 ` [PATCH v2 19/35] x86/resctrl: Add resctrl_arch_ prefix to pseudo lock functions Dave Martin
2024-04-26 15:08 ` [PATCH v2 20/35] x86/resctrl: Allow an architecture to disable pseudo lock Dave Martin
2024-04-26 15:08 ` [PATCH v2 21/35] x86/resctrl: Make prefetch_disable_bits belong to the arch code Dave Martin
2024-04-26 15:08 ` [PATCH v2 22/35] x86/resctrl: Make resctrl_arch_pseudo_lock_fn() take a plr Dave Martin
2024-06-14 13:58 ` James Morse
2024-04-26 15:08 ` [PATCH v2 23/35] x86/resctrl: Move thread_throttle_mode_init() to be managed by resctrl Dave Martin
2024-04-26 15:08 ` [PATCH v2 24/35] x86/resctrl: Move get_config_index() to a header Dave Martin
2024-04-26 15:08 ` [PATCH v2 25/35] x86/resctrl: Claim get_domain_from_cpu() for resctrl Dave Martin
2024-04-26 15:08 ` [PATCH v2 26/35] x86/resctrl: Describe resctrl's bitmap size assumptions Dave Martin
2024-04-26 15:08 ` [PATCH v2 27/35] x86/resctrl: Rename resctrl_sched_in() to begin with "resctrl_arch_" Dave Martin
2024-04-26 15:08 ` [PATCH v2 28/35] x86/resctrl: Drop __init/__exit on assorted symbols Dave Martin
2024-04-26 15:08 ` [PATCH v2 29/35] fs/resctrl: Add boiler plate for external resctrl code Dave Martin
2024-06-14 13:59 ` James Morse
2024-04-26 15:08 ` [PATCH v2 30/35] x86/resctrl: Squelch whitespace anomalies in resctrl core code Dave Martin
2024-04-26 15:09 ` [PATCH v2 31/35] x86/resctrl: Prefer alloc(sizeof(*foo)) idiom in rdt_init_fs_context() Dave Martin
2024-04-26 15:09 ` [PATCH v2 32/35] x86/resctrl: Relax some asm #includes Dave Martin
2024-04-26 15:09 ` [PATCH v2 33/35] x86/resctrl: Move the filesystem bits to headers visible to fs/resctrl Dave Martin
2024-04-26 15:09 ` [PATCH v2 34/35] x86/resctrl: Move the resctrl filesystem code to /fs/resctrl Dave Martin
2024-04-26 15:09 ` [PATCH v2 35/35] [SQUASHME] " Dave Martin
2024-04-29 23:34 ` [PATCH v2 00/35] " Peter Newman
2024-04-30 16:29 ` Dave Martin
2024-05-08 15:41 ` Reinette Chatre
2024-05-14 15:17 ` Dave Martin
2024-05-14 15:56 ` Reinette Chatre
2024-05-23 15:01 ` Dave Martin
2024-05-23 15:07 ` David Hildenbrand
2024-05-23 15:18 ` Dave Martin
2024-05-23 0:39 ` Shaopeng Tan (Fujitsu)
2024-06-14 13:59 ` James Morse
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240426150904.8854-4-Dave.Martin@arm.com \
--to=dave.martin@arm.com \
--cc=Babu.Moger@amd.com \
--cc=amitsinght@marvell.com \
--cc=baolin.wang@linux.alibaba.com \
--cc=bobo.shaobowang@huawei.com \
--cc=bp@alien8.de \
--cc=carl@os.amperecomputing.com \
--cc=david@redhat.com \
--cc=dfustini@baylibre.com \
--cc=fenghua.yu@intel.com \
--cc=hpa@zytor.com \
--cc=james.morse@arm.com \
--cc=lcherian@marvell.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mingo@redhat.com \
--cc=peternewman@google.com \
--cc=quic_jiles@quicinc.com \
--cc=reinette.chatre@intel.com \
--cc=rex.nie@jaguarmicro.com \
--cc=scott@os.amperecomputing.com \
--cc=shameerali.kolothum.thodi@huawei.com \
--cc=tan.shaopeng@fujitsu.com \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
--cc=xhao@linux.alibaba.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox