From: Leo Yan <leoy@marvell.com>
To: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Cc: "linux-arm-kernel@lists.infradead.org"
<linux-arm-kernel@lists.infradead.org>,
"linux-pm@vger.kernel.org" <linux-pm@vger.kernel.org>,
Zhou Zhu <zzhu3@marvell.com>, Yu Tang <ytang5@marvell.com>,
Neil Zhang <zhangwm@marvell.com>
Subject: Re: [Question] Verification For arm64: suspend/resume implementation
Date: Mon, 23 Sep 2013 19:11:40 +0800 [thread overview]
Message-ID: <524021EC.2000207@marvell.com> (raw)
In-Reply-To: <20130913144001.GA28531@e102568-lin.cambridge.arm.com>
[-- Attachment #1: Type: text/plain, Size: 2657 bytes --]
On 09/13/2013 10:40 PM, Lorenzo Pieralisi wrote:
> On Fri, Sep 13, 2013 at 03:53:53AM +0100, Leo Yan wrote:
>> hi Lorenzo,
>>
>> I have applied your ARM64's suspend/resume related patches and can built
>> successfully. i want to verify these patches on foundation model
>> firstly, so below are my questions:
>>
>> "Code has been tested on AEM v8 models and a simple CPU idle driver that
>> enables a C-state where CPUs are reset when wfi is hit."
>>
>> 1. Can u help share this simple cpu idle driver?
>
> Yes, I will post a simple skeleton driver, PSCI based (bootwrapper
> implementation), by the end of September.
>
>> 2. On the foundation model, if the core is placed into reset state, then
>> if there have interrupt is routed to the core, the core still cannot be
>> waken up anymore; because the reset bit cannot be released by h/w. So
>> how can let the core return back from the reset state?
>
> Well, I am testing it with the AEM models power controller that is not
> publicly available yet, and _should_ be released with the new version of the
> foundation models.
>
> As a first step, I will write a PSCI suspend implementation that just executes
> wfi and resumes through the reset vector to emulate a power down as a means to
> make the suspend/resume code path usable to everyone.
>
Looking forward the related implementation; After them are ready, i'm
glad have a trying.
At my side, i'm warming up related doc and tried to debug related tear
down opreations according to CA53's TRM; pls see enclosed two patches.
Firstly clarify, these two patches are _ONLY_ for debugging purpose, i
have no plan to commit them for Linux kernel.
0001-cpuidle-add-simple-driver-for-arm64.patch: it's a simple cpuidle
driver;
0002-ARM64-add-cpu-tear-down-function-for-A53-s-power-mod.patch: i tried
to wrote a tear down operations (disable D$/flush L1 cache/Disable SMP
bit, etc);
But i found in the patch 2, if the core execute the instruction "mrs
x0, S3_1_C15_C2_1" to access CPUECTLR_EL1, the kernel will report the
illegal instructions. So just like before i saw the discussion on
mailing list, On ARMv8 we need operate the SMP bit in EL2/EL3/Secure EL1
but not in non-secure EL1.
Here i tried two methods to try to fix this issue, but both of them were
failure:
1. I tried to set the ACTLR_EL2 bit 1 in the boot wrapper code, but when
in the non-secure world's kernel to access CPUECTLR_EL1 it still will
report the panic for illegal instruction;
2. I tried to modify the boot wrapper code to let the kernel stay in
secure world's EL1, but looks like it also failed;
So do u have any suggestion for this failure?
Thx,
Leo Yan
[-- Attachment #2: 0001-cpuidle-add-simple-driver-for-arm64.patch --]
[-- Type: text/x-patch, Size: 5866 bytes --]
>From a1aa5dd0924b8c7493c6fec31c00b2feba7f5ad7 Mon Sep 17 00:00:00 2001
From: Leo Yan <leoy@marvell.com>
Date: Mon, 23 Sep 2013 17:05:42 +0800
Subject: [PATCH 1/2] cpuidle: add simple driver for arm64
Signed-off-by: Leo Yan <leoy@marvell.com>
---
drivers/cpuidle/Kconfig | 6 ++
drivers/cpuidle/Makefile | 1 +
drivers/cpuidle/cpuidle-arm64.c | 180 +++++++++++++++++++++++++++++++++++++++
3 files changed, 187 insertions(+)
create mode 100644 drivers/cpuidle/cpuidle-arm64.c
diff --git a/drivers/cpuidle/Kconfig b/drivers/cpuidle/Kconfig
index c4cc27e..361325e 100644
--- a/drivers/cpuidle/Kconfig
+++ b/drivers/cpuidle/Kconfig
@@ -39,4 +39,10 @@ config CPU_IDLE_CALXEDA
help
Select this to enable cpuidle on Calxeda processors.
+config CPU_IDLE_ARM64
+ bool "CPU Idle Driver for ARM64"
+ depends on ARM64
+ help
+ Select this to enable cpuidle on ARM64 processors.
+
endif
diff --git a/drivers/cpuidle/Makefile b/drivers/cpuidle/Makefile
index 0d8bd55..954b67f 100644
--- a/drivers/cpuidle/Makefile
+++ b/drivers/cpuidle/Makefile
@@ -7,3 +7,4 @@ obj-$(CONFIG_ARCH_NEEDS_CPU_IDLE_COUPLED) += coupled.o
obj-$(CONFIG_CPU_IDLE_CALXEDA) += cpuidle-calxeda.o
obj-$(CONFIG_ARCH_KIRKWOOD) += cpuidle-kirkwood.o
+obj-$(CONFIG_CPU_IDLE_ARM64) += cpuidle-arm64.o
diff --git a/drivers/cpuidle/cpuidle-arm64.c b/drivers/cpuidle/cpuidle-arm64.c
new file mode 100644
index 0000000..7d376db
--- /dev/null
+++ b/drivers/cpuidle/cpuidle-arm64.c
@@ -0,0 +1,180 @@
+/*
+ * ARM64 CPU idle driver.
+ *
+ * Copyright (C) 2012 ARM Ltd.
+ * Author: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/bitmap.h>
+#include <linux/cpuidle.h>
+#include <linux/cpu_pm.h>
+#include <linux/clockchips.h>
+#include <linux/debugfs.h>
+#include <linux/hrtimer.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/tick.h>
+#include <asm/proc-fns.h>
+#include <asm/suspend.h>
+#include <asm/cacheflush.h>
+
+static int arm64_cpuidle_simple_enter(struct cpuidle_device *dev,
+ struct cpuidle_driver *drv, int index)
+{
+ ktime_t time_start, time_end;
+ s64 diff;
+
+ time_start = ktime_get();
+
+ cpu_do_idle();
+
+ time_end = ktime_get();
+
+ local_irq_enable();
+
+ diff = ktime_to_us(ktime_sub(time_end, time_start));
+ if (diff > INT_MAX)
+ diff = INT_MAX;
+
+ dev->last_residency = (int) diff;
+
+ return index;
+}
+
+static int arm64_enter_powerdown(struct cpuidle_device *dev,
+ struct cpuidle_driver *drv, int idx);
+
+static struct cpuidle_state arm64_cpuidle_set[] __initdata = {
+ [0] = {
+ .enter = arm64_cpuidle_simple_enter,
+ .exit_latency = 1,
+ .target_residency = 1,
+ .power_usage = UINT_MAX,
+ .flags = CPUIDLE_FLAG_TIME_VALID,
+ .name = "WFI",
+ .desc = "ARM64 WFI",
+ },
+#if 1
+ [1] = {
+ .enter = arm64_enter_powerdown,
+ .exit_latency = 300,
+ .target_residency = 1000,
+ .flags = CPUIDLE_FLAG_TIME_VALID,
+ .name = "C1",
+ .desc = "ARM64 power down",
+ },
+#endif
+};
+
+struct cpuidle_driver arm64_idle_driver = {
+ .name = "arm64_idle",
+ .owner = THIS_MODULE,
+ .safe_state_index = 0
+};
+
+static DEFINE_PER_CPU(struct cpuidle_device, arm64_idle_dev);
+
+extern void arm64_cpu_tear_down(void);
+typedef void (*phys_reset_t)(unsigned long);
+
+static int notrace arm64_powerdown_finisher(unsigned long arg)
+{
+ phys_reset_t phys_reset;
+
+ setup_mm_for_reboot();
+
+ arm64_cpu_tear_down();
+ wfi();
+
+ phys_reset = (phys_reset_t)(unsigned long)virt_to_phys(cpu_reset);
+ phys_reset(virt_to_phys(cpu_resume));
+
+ /* should never get here */
+ BUG();
+}
+
+/*
+ * arm64_enter_powerdown - Programs CPU to enter the specified state
+ * @dev: cpuidle device
+ * @drv: The target state to be programmed
+ * @idx: state index
+ *
+ * Called from the CPUidle framework to program the device to the
+ * specified target state selected by the governor.
+ */
+static int arm64_enter_powerdown(struct cpuidle_device *dev,
+ struct cpuidle_driver *drv, int idx)
+{
+ struct timespec ts_preidle, ts_postidle, ts_idle;
+ int ret;
+
+ /* Used to keep track of the total time in idle */
+ getnstimeofday(&ts_preidle);
+
+ BUG_ON(!irqs_disabled());
+
+ cpu_pm_enter();
+
+ clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_ENTER, &dev->cpu);
+
+ ret = cpu_suspend((unsigned long) dev, arm64_powerdown_finisher);
+ if (ret)
+ BUG();
+
+ clockevents_notify(CLOCK_EVT_NOTIFY_BROADCAST_EXIT, &dev->cpu);
+
+ cpu_pm_exit();
+
+ getnstimeofday(&ts_postidle);
+ local_irq_enable();
+ ts_idle = timespec_sub(ts_postidle, ts_preidle);
+
+ dev->last_residency = ts_idle.tv_nsec / NSEC_PER_USEC +
+ ts_idle.tv_sec * USEC_PER_SEC;
+ return idx;
+}
+
+/*
+ * arm64_idle_init
+ *
+ * Registers the bl specific cpuidle driver with the cpuidle
+ * framework with the valid set of states.
+ */
+int __init arm64_idle_init(void)
+{
+ struct cpuidle_device *dev;
+ int i, cpu_id;
+ struct cpuidle_driver *drv = &arm64_idle_driver;
+
+ drv->state_count = (sizeof(arm64_cpuidle_set) /
+ sizeof(struct cpuidle_state));
+
+ for (i = 0; i < drv->state_count; i++) {
+ memcpy(&drv->states[i], &arm64_cpuidle_set[i],
+ sizeof(struct cpuidle_state));
+ }
+
+ cpuidle_register_driver(drv);
+
+ for_each_cpu(cpu_id, cpu_online_mask) {
+ pr_err("CPUidle for CPU%d registered\n", cpu_id);
+ dev = &per_cpu(arm64_idle_dev, cpu_id);
+ dev->cpu = cpu_id;
+
+ dev->state_count = drv->state_count;
+
+ if (cpuidle_register_device(dev)) {
+ printk(KERN_ERR "%s: Cpuidle register device failed\n",
+ __func__);
+ return -EIO;
+ }
+ }
+
+ return 0;
+}
+
+device_initcall(arm64_idle_init);
--
1.7.9.5
[-- Attachment #3: 0002-ARM64-add-cpu-tear-down-function-for-A53-s-power-mod.patch --]
[-- Type: text/x-patch, Size: 4314 bytes --]
>From e238ddfc426bd1f8b2ac3b5f396a31150cbe7e5c Mon Sep 17 00:00:00 2001
From: Leo Yan <leoy@marvell.com>
Date: Mon, 23 Sep 2013 17:06:20 +0800
Subject: [PATCH 2/2] ARM64: add cpu tear down function for A53's power mode
Signed-off-by: Leo Yan <leoy@marvell.com>
---
arch/arm64/kernel/Makefile | 2 +-
arch/arm64/kernel/cpu_tear_down.S | 90 +++++++++++++++++++++++++++++++++++++
2 files changed, 91 insertions(+), 1 deletion(-)
create mode 100644 arch/arm64/kernel/cpu_tear_down.S
diff --git a/arch/arm64/kernel/Makefile b/arch/arm64/kernel/Makefile
index 010550a..046fbf0 100644
--- a/arch/arm64/kernel/Makefile
+++ b/arch/arm64/kernel/Makefile
@@ -18,7 +18,7 @@ arm64-obj-$(CONFIG_SMP) += smp.o smp_spin_table.o smp_psci.o
arm64-obj-$(CONFIG_HW_PERF_EVENTS) += perf_event.o
arm64-obj-$(CONFIG_HAVE_HW_BREAKPOINT)+= hw_breakpoint.o
arm64-obj-$(CONFIG_EARLY_PRINTK) += early_printk.o
-arm64-obj-$(CONFIG_ARM_CPU_SUSPEND) += sleep.o suspend.o
+arm64-obj-$(CONFIG_ARM_CPU_SUSPEND) += sleep.o suspend.o cpu_tear_down.o
arm64-obj-$(CONFIG_SUSPEND) += fake_suspend.o
obj-y += $(arm64-obj-y) vdso/
diff --git a/arch/arm64/kernel/cpu_tear_down.S b/arch/arm64/kernel/cpu_tear_down.S
new file mode 100644
index 0000000..9f3d5d0
--- /dev/null
+++ b/arch/arm64/kernel/cpu_tear_down.S
@@ -0,0 +1,90 @@
+/*
+ * arch/arm64/mach-vexpress/sleep.S
+ *
+ * Copyright (c) 2013 Marvell Semiconductor Inc.
+ *
+ * Author: Leo Yan <leoy@marvell.com>
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; either version 2 of the License, or
+ * (at your option) any later version.
+ *
+ * This program is distributed in the hope that it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along
+ * with this program; if not, write to the Free Software Foundation, Inc.,
+ * 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
+ */
+
+#include <linux/linkage.h>
+#include <linux/init.h>
+#include <asm/assembler.h>
+
+/*
+ * Exits SMP coherency.
+ */
+ENTRY(arm64_cpu_tear_down)
+ mov x12, lr
+
+ mrs x0, sctlr_el1
+ bic x0, x0, #1 << 2 // clear SCTLR.C
+ msr sctlr_el1, x0
+ isb
+
+ dsb sy // ensure ordering with previous memory accesses
+ mrs x0, clidr_el1 // read clidr
+ and x3, x0, #0x7000000 // extract loc from clidr
+ lsr x3, x3, #23 // left align loc bit field
+ cbz x3, finished // if loc is 0, then no need to clean
+ mov x10, #0 // start clean at cache level 0
+
+ add x2, x10, x10, lsr #1 // work out 3x current cache level
+ lsr x1, x0, x2 // extract cache type bits from clidr
+ and x1, x1, #7 // mask of the bits for current cache only
+ cmp x1, #2 // see what cache we have at this level
+ b.lt skip // skip if no cache, or just i-cache
+ save_and_disable_irqs x9 // make CSSELR and CCSIDR access atomic
+ msr csselr_el1, x10 // select current cache level in csselr
+ isb // isb to sych the new cssr&csidr
+ mrs x1, ccsidr_el1 // read the new ccsidr
+ restore_irqs x9
+ and x2, x1, #7 // extract the length of the cache lines
+ add x2, x2, #4 // add 4 (line length offset)
+ mov x4, #0x3ff
+ and x4, x4, x1, lsr #3 // find maximum number on the way size
+ clz w5, w4 // find bit position of way size increment
+ mov x7, #0x7fff
+ and x7, x7, x1, lsr #13 // extract max number of the index size
+loop2:
+ mov x9, x4 // create working copy of max way size
+loop3:
+ lsl x6, x9, x5
+ orr x11, x10, x6 // factor way and cache number into x11
+ lsl x6, x7, x2
+ orr x11, x11, x6 // factor index number into x11
+ dc cisw, x11 // clean & invalidate by set/way
+ subs x9, x9, #1 // decrement the way
+ b.ge loop3
+ subs x7, x7, #1 // decrement the index
+ b.ge loop2
+
+skip:
+finished:
+ mov x10, #0 // swith back to cache level 0
+ msr csselr_el1, x10 // select current cache level in csselr
+ dsb sy
+ isb
+
+ mrs x0, S3_1_C15_C2_1
+ bic x0, x0, #0x1 << 6 // disable SMP bit
+ msr S3_1_C15_C2_1, x0
+ dsb sy
+ isb
+
+ ret x12
+ENDPROC(arm64_cpu_tear_down)
+
--
1.7.9.5
next prev parent reply other threads:[~2013-09-23 11:12 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-09-13 2:53 [Question] Verification For arm64: suspend/resume implementation Leo Yan
2013-09-13 14:40 ` Lorenzo Pieralisi
2013-09-23 11:11 ` Leo Yan [this message]
2013-09-23 15:26 ` Achin Gupta
2013-09-24 2:00 ` Leo Yan
2013-09-24 9:02 ` Achin Gupta
2013-09-24 11:49 ` Leo Yan
2013-09-24 13:00 ` Catalin Marinas
2013-09-26 11:12 ` Leo Yan
2013-09-26 16:53 ` Catalin Marinas
2013-09-27 2:53 ` Leo Yan
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=524021EC.2000207@marvell.com \
--to=leoy@marvell.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-pm@vger.kernel.org \
--cc=lorenzo.pieralisi@arm.com \
--cc=ytang5@marvell.com \
--cc=zhangwm@marvell.com \
--cc=zzhu3@marvell.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;
as well as URLs for NNTP newsgroup(s).