linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: mark.rutland@arm.com (Mark Rutland)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 08/10] arm64: make cpu number a percpu variable
Date: Wed, 19 Oct 2016 20:10:32 +0100	[thread overview]
Message-ID: <1476904234-9511-9-git-send-email-mark.rutland@arm.com> (raw)
In-Reply-To: <1476904234-9511-1-git-send-email-mark.rutland@arm.com>

In the absence of CONFIG_THREAD_INFO_IN_TASK, core code maintains
thread_info::cpu, and low-level architecture code can access this to
build raw_smp_processor_id(). With CONFIG_THREAD_INFO_IN_TASK, core code
maintains task_struct::cpu, which for reasons of hte header soup is not
accessible to low-level arch code.

Instead, we can maintain a percpu variable containing the cpu number.
Non-preemptible code will receive the current CPU's number, while
preemptible code may race (as was previosuly the case), and see a stale
value. Thus, there shouldn't be a change in observable behaviour.

For both the old and new implementation of raw_smp_processor_id(), we
read a syreg into a GPR, add an offset, and load the result. As the
offset is now large, it probably won't be folded into the load, but
otherwise the assembly shouldn't change much.

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: James Morse <james.morse@arm.com>
Cc: Suzuki K Poulose <suzuki.poulose@arm.com>
Cc: Will Deacon <will.deacon@arm.com>
---
 arch/arm64/include/asm/smp.h | 6 +++++-
 arch/arm64/kernel/smp.c      | 5 +++++
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/include/asm/smp.h b/arch/arm64/include/asm/smp.h
index 0226447..f77ac0d 100644
--- a/arch/arm64/include/asm/smp.h
+++ b/arch/arm64/include/asm/smp.h
@@ -29,11 +29,15 @@
 
 #ifndef __ASSEMBLY__
 
+#include <asm/percpu.h>
+
 #include <linux/threads.h>
 #include <linux/cpumask.h>
 #include <linux/thread_info.h>
 
-#define raw_smp_processor_id() (current_thread_info()->cpu)
+DECLARE_PER_CPU_READ_MOSTLY(int, cpu_number);
+
+#define raw_smp_processor_id() (*this_cpu_ptr(&cpu_number))
 
 struct seq_file;
 
diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c
index 7af46bf..2679722 100644
--- a/arch/arm64/kernel/smp.c
+++ b/arch/arm64/kernel/smp.c
@@ -58,6 +58,9 @@
 #define CREATE_TRACE_POINTS
 #include <trace/events/ipi.h>
 
+DEFINE_PER_CPU_READ_MOSTLY(int, cpu_number);
+EXPORT_PER_CPU_SYMBOL(cpu_number);
+
 /*
  * as from 2.5, kernels no longer have an init_tasks structure
  * so we need some other way of telling a new secondary core
@@ -718,6 +721,8 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
 	 */
 	for_each_possible_cpu(cpu) {
 
+		per_cpu(cpu_number, cpu) = cpu;
+
 		if (cpu == smp_processor_id())
 			continue;
 
-- 
1.9.1

  parent reply	other threads:[~2016-10-19 19:10 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-10-19 19:10 [PATCH 00/10] arm64: move thread_info off of the task stack Mark Rutland
2016-10-19 19:10 ` [PATCH 01/10] arm64: thread_info remove stale items Mark Rutland
2016-10-19 19:10 ` [PATCH 02/10] arm64: asm-offsets: remove unused definitions Mark Rutland
2016-10-19 19:10 ` [PATCH 03/10] arm64: factor out current_stack_pointer Mark Rutland
2016-10-19 19:10 ` [PATCH 04/10] arm64: traps: simplify die() and __die() Mark Rutland
2016-10-19 19:10 ` [PATCH 05/10] arm64: prep stack walkers for THREAD_INFO_IN_TASK Mark Rutland
2016-10-19 19:10 ` [PATCH 06/10] arm64: move sp_el0 and tpidr_el1 into cpu_suspend_ctx Mark Rutland
2016-10-19 19:10 ` [PATCH 07/10] arm64: smp: prepare for smp_processor_id() rework Mark Rutland
2016-10-19 19:10 ` Mark Rutland [this message]
2016-10-19 19:10 ` [PATCH 09/10] arm64: assembler: introduce ldr_this_cpu Mark Rutland
2016-10-19 19:10 ` [PATCH 10/10] arm64: split thread_info from task stack Mark Rutland
2016-10-21 14:50   ` James Morse
2016-10-21 15:59     ` Mark Rutland
2016-10-21 17:27       ` Mark Rutland
2016-10-21 16:20     ` Mark Rutland
2016-10-24 17:38 ` [PATCH 00/10] arm64: move thread_info off of the " Laura Abbott
2016-10-24 17:48   ` Mark Rutland
2016-10-24 17:58     ` Laura Abbott
2016-10-24 18:09       ` Mark Rutland
2016-10-24 18:15         ` Mark Rutland
2016-10-24 18:18           ` Kees Cook
2016-10-25 10:05             ` Mark Rutland
2016-10-26  0:46         ` Laura Abbott
2016-10-26  9:55           ` Mark Rutland

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=1476904234-9511-9-git-send-email-mark.rutland@arm.com \
    --to=mark.rutland@arm.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    /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;
as well as URLs for NNTP newsgroup(s).