This is a note to let you know that I've just added the patch titled
x86/resctrl: Apply READ_ONCE/WRITE_ONCE to task_struct.{rmid,closid}
to the 5.4-stable tree which can be found at:
http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
The filename of the patch is:
x86-resctrl-apply-read_once-write_once-to-task_struct.-rmid-closid.patch
and it can be found in the queue-5.4 subdirectory.
If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@xxxxxxxxxxxxxxx> know about it.
>From 6d3b47ddffed70006cf4ba360eef61e9ce097d8f Mon Sep 17 00:00:00 2001
From: Valentin Schneider <valentin.schneider@xxxxxxx>
Date: Thu, 17 Dec 2020 14:31:21 -0800
Subject: x86/resctrl: Apply READ_ONCE/WRITE_ONCE to task_struct.{rmid,closid}
From: Valentin Schneider <valentin.schneider@xxxxxxx>
commit 6d3b47ddffed70006cf4ba360eef61e9ce097d8f upstream.
A CPU's current task can have its {closid, rmid} fields read locally
while they are being concurrently written to from another CPU.
This can happen anytime __resctrl_sched_in() races with either
__rdtgroup_move_task() or rdt_move_group_tasks().
Prevent load / store tearing for those accesses by giving them the
READ_ONCE() / WRITE_ONCE() treatment.
Signed-off-by: Valentin Schneider <valentin.schneider@xxxxxxx>
Signed-off-by: Reinette Chatre <reinette.chatre@xxxxxxxxx>
Signed-off-by: Borislav Petkov <bp@xxxxxxx>
Link: https://lkml.kernel.org/r/9921fda88ad81afb9885b517fbe864a2bc7c35a9.1608243147.git.reinette.chatre@xxxxxxxxx
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
arch/x86/include/asm/resctrl_sched.h | 11 +++++++----
arch/x86/kernel/cpu/resctrl/rdtgroup.c | 10 +++++-----
2 files changed, 12 insertions(+), 9 deletions(-)
--- a/arch/x86/include/asm/resctrl_sched.h
+++ b/arch/x86/include/asm/resctrl_sched.h
@@ -56,19 +56,22 @@ static void __resctrl_sched_in(void)
struct resctrl_pqr_state *state = this_cpu_ptr(&pqr_state);
u32 closid = state->default_closid;
u32 rmid = state->default_rmid;
+ u32 tmp;
/*
* If this task has a closid/rmid assigned, use it.
* Else use the closid/rmid assigned to this cpu.
*/
if (static_branch_likely(&rdt_alloc_enable_key)) {
- if (current->closid)
- closid = current->closid;
+ tmp = READ_ONCE(current->closid);
+ if (tmp)
+ closid = tmp;
}
if (static_branch_likely(&rdt_mon_enable_key)) {
- if (current->rmid)
- rmid = current->rmid;
+ tmp = READ_ONCE(current->rmid);
+ if (tmp)
+ rmid = tmp;
}
if (closid != state->cur_closid || rmid != state->cur_rmid) {
--- a/arch/x86/kernel/cpu/resctrl/rdtgroup.c
+++ b/arch/x86/kernel/cpu/resctrl/rdtgroup.c
@@ -563,11 +563,11 @@ static int __rdtgroup_move_task(struct t
*/
if (rdtgrp->type == RDTCTRL_GROUP) {
- tsk->closid = rdtgrp->closid;
- tsk->rmid = rdtgrp->mon.rmid;
+ WRITE_ONCE(tsk->closid, rdtgrp->closid);
+ WRITE_ONCE(tsk->rmid, rdtgrp->mon.rmid);
} else if (rdtgrp->type == RDTMON_GROUP) {
if (rdtgrp->mon.parent->closid == tsk->closid) {
- tsk->rmid = rdtgrp->mon.rmid;
+ WRITE_ONCE(tsk->rmid, rdtgrp->mon.rmid);
} else {
rdt_last_cmd_puts("Can't move task to different control group\n");
return -EINVAL;
@@ -2177,8 +2177,8 @@ static void rdt_move_group_tasks(struct
for_each_process_thread(p, t) {
if (!from || is_closid_match(t, from) ||
is_rmid_match(t, from)) {
- t->closid = to->closid;
- t->rmid = to->mon.rmid;
+ WRITE_ONCE(t->closid, to->closid);
+ WRITE_ONCE(t->rmid, to->mon.rmid);
/*
* Order the closid/rmid stores above before the loads
Patches currently in stable-queue which might be from valentin.schneider@xxxxxxx are
queue-5.4/x86-resctrl-apply-read_once-write_once-to-task_struct.-rmid-closid.patch
