* [Bug] ARM: mxs: STI: console can't wake up from freeze
From: Stefan Wahren @ 2016-11-05 11:07 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20161101092304.GM1041@n2100.armlinux.org.uk>
Hi,
[add Rui]
> Russell King - ARM Linux <linux@armlinux.org.uk> hat am 1. November 2016 um
> 10:23 geschrieben:
>
>
> On Mon, Oct 31, 2016 at 08:54:33PM +0100, Stefan Wahren wrote:
> > [ 366.696043] INFO: task ext4lazyinit:70 blocked for more than 120 seconds.
> > [ 366.703046] Not tainted 4.9.0-rc1 #7
> > [ 366.707188] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables
> > this
> > message.
> > [ 366.715161] ext4lazyinit D c05aa6ac 0 70 2 0x00000000
>
> This looks like a very different problem - I guess one for ext4 people.
>
i investigated the issue more further. This trace wasn't representative, because
the stacktrace is different in most cases. The "endless loop" is located in
freeze_enter(). So i added some debug messages:
-------------------------------->8------------------------------------
diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c
index 1e7f5da..079c08d 100644
--- a/kernel/power/suspend.c
+++ b/kernel/power/suspend.c
@@ -67,17 +67,20 @@ static void freeze_enter(void)
spin_unlock_irq(&suspend_freeze_lock);
get_online_cpus();
+ pr_info("PM: calling cpuidle_resume()\n");
cpuidle_resume();
/* Push all the CPUs into the idle loop. */
+ pr_info("PM: calling wake_up_all_idle_cpus()\n");
wake_up_all_idle_cpus();
- pr_debug("PM: suspend-to-idle\n");
+ pr_info("PM: suspend-to-idle\n");
/* Make the current CPU wait so it can enter the idle loop too. */
wait_event(suspend_freeze_wait_head,
suspend_freeze_state == FREEZE_STATE_WAKE);
- pr_debug("PM: resume from suspend-to-idle\n");
+ pr_info("PM: resume from suspend-to-idle\n");
cpuidle_pause();
+ pr_info("PM: called cpuidle_pause()\n");
put_online_cpus();
spin_lock_irq(&suspend_freeze_lock);
@@ -91,6 +94,8 @@ void freeze_wake(void)
{
unsigned long flags;
+ pr_info("PM: freeze_wake()\n");
+
spin_lock_irqsave(&suspend_freeze_lock, flags);
if (suspend_freeze_state > FREEZE_STATE_NONE) {
suspend_freeze_state = FREEZE_STATE_WAKE;
-------------------------------->8------------------------------------
and repeated the test cases for freeze which shows none the representative
stacktraces:
1. cmdline contains no_console_suspend
* echo freeze > /sys/power/state
...
[ 139.371308] PM: suspend of devices complete after 1342.721 msecs
[ 139.385203] PM: late suspend of devices complete after 7.668 msecs
[ 139.399428] PM: noirq suspend of devices complete after 7.936 msecs
[ 139.406639] PM: calling cpuidle_resume()
[ 139.410619] PM: calling wake_up_all_idle_cpus()
[ 139.415339] PM: suspend-to-idle
>>> no reaction to input via Debug UART
[ 366.683570] INFO: task bash:373 blocked for more than 120 seconds.
[ 366.689814] Not tainted 4.9.0-rc1-dirty #14
[ 366.694705] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this
message.
[ 366.702685] bash D c05aa6ec 0 373 275 0x00000000
[ 366.709227] [<c05aa6ec>] (__schedule) from [<c05aaff8>] (schedule+0x3c/0xbc)
[ 366.716495] [<c05aaff8>] (schedule) from [<c005b588>]
(suspend_devices_and_enter+0x888/0xa78)
[ 366.725228] [<c005b588>] (suspend_devices_and_enter) from [<c005bea4>]
(pm_suspend+0x72c/0x81c)
[ 366.734150] [<c005bea4>] (pm_suspend) from [<c005a008>]
(state_store+0x80/0xcc)
[ 366.741554] [<c005a008>] (state_store) from [<c02f0270>]
(kobj_attr_store+0x18/0x1c)
[ 366.749515] [<c02f0270>] (kobj_attr_store) from [<c01911e8>]
(sysfs_kf_write+0x48/0x4c)
[ 366.757735] [<c01911e8>] (sysfs_kf_write) from [<c0190308>]
(kernfs_fop_write+0xfc/0x1d0)
[ 366.766130] [<c0190308>] (kernfs_fop_write) from [<c011f578>]
(__vfs_write+0x2c/0x124)
[ 366.774255] [<c011f578>] (__vfs_write) from [<c011f724>]
(vfs_write+0xb4/0x1a4)
[ 366.781640] [<c011f724>] (vfs_write) from [<c011f8e8>] (SyS_write+0x44/0x88)
[ 366.788890] [<c011f8e8>] (SyS_write) from [<c000a2c0>]
(ret_fast_syscall+0x0/0x1c)
[ 366.796627]
[ 366.796627] Showing all locks held in the system:
[ 366.803011] 2 locks held by khungtaskd/10:
[ 366.807149] #0: [ 366.808931] (
rcu_read_lock[ 366.811784] ){......}
, at: [ 366.814813] [<c0093a40>] watchdog+0xb4/0x61c
[ 366.819128] #1: [ 366.820902] (
tasklist_lock[ 366.823876] ){.+.+..}
, at: [ 366.826765] [<c0051dbc>] debug_show_all_locks+0x28/0x1bc
[ 366.832151] 4 locks held by bash/373:
[ 366.835973] #0: [ 366.837765] (
sb_writers[ 366.840365] #4
){.+.+.+}[ 366.842987] , at:
[ 366.845079] [<c011f804>] vfs_write+0x194/0x1a4
[ 366.849551] #1: [ 366.851324] (
&of->mutex[ 366.854058] ){+.+.+.}
, at: [ 366.856944] [<c01902cc>] kernfs_fop_write+0xc0/0x1d0
[ 366.861941] #2: [ 366.863839] (
s_active[ 366.866288] #43
){.+.+.+}[ 366.868877] , at:
[ 366.870938] [<c01902d4>] kernfs_fop_write+0xc8/0x1d0
[ 366.876053] #3: [ 366.877843] (
pm_mutex[ 366.880272] ){+.+.+.}
, at: [ 366.883266] [<c005b808>] pm_suspend+0x90/0x81c
[ 366.887757]
[ 366.889268] =============================================
[ 366.889268]
2. cmdline contains no_console_suspend
* echo enabled > /sys/class/tty/ttyAMA0/power/wakeup
* echo freeze > /sys/power/state
the same as in 1.
3. cmdline doesn't contains no_console_suspend
* echo enabled > /sys/class/tty/ttyAMA0/power/wakeup
* echo freeze > /sys/power/state
[ 161.093187] PM: Syncing filesystems ... [ 161.734413] done.
[ 161.793242] Freezing user space processes ... (elapsed 0.008 seconds) done.
[ 161.809797] Freezing remaining freezable tasks ... (elapsed 0.004 seconds)
done.
[ 161.821953] Suspending console(s) (use no_console_suspend to debug)
>>>> no reaction to Debug UART
I expected that in all 3 cases freeze_wake() will be called. Why not?
Here my config again:
CONFIG_PM_SLEEP=y
CONFIG_SUSPEND=y
CONFIG_SUSPEND_FREEZER=y
CONFIG_PM=y
CONFIG_CPU_IDLE is not set
^ permalink raw reply related
* Low network throughput on i.MX28
From: Stefan Wahren @ 2016-11-05 11:33 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478299359.26659.5.camel@embedded.rocks>
Hi J?rg,
> J?rg Krause <joerg.krause@embedded.rocks> hat am 4. November 2016 um 23:42
> geschrieben:
>
>
> Hi Stefan,
>
> sorry, I forget the link in the previous mail.
>
> On Fri, 2016-11-04 at 20:30 +0100, Stefan Wahren wrote:
> > Hi J?rg,
> >
> > > J?rg Krause <joerg.krause@embedded.rocks> hat am 4. November 2016
> > > um 19:44
> > > geschrieben:
> > >
> > >
> > > Hi Shawn,
> > >
> > > On Wed, 2016-11-02 at 09:24 +0100, Stefan Wahren wrote:
> > > > Am 02.11.2016 um 09:14 schrieb J?rg Krause:
> > > > > On Sat, 2016-10-29 at 11:08 +0200, Stefan Wahren wrote:
> > > > > > > J?rg Krause <joerg.krause@embedded.rocks> hat am 29.
> > > > > > > Oktober
> > > > > > > 2016
> > > > > > > um 01:07
> > > > > > > geschrieben:
> > > > > > >
> > > > > > >
> > > > > > > You mentioned [1] an optimization in the Freescale vendor
> > > > > > > Linux
> > > > > > > kernel
> > > > > > > [2]. I would really like to see this optimization in the
> > > > > > > mainline
> > > > > > > kernel.
> > > > > > >
> > > > > > > Did you ever tried to port this code from Freescale to
> > > > > > > mainline?
> > > > > >
> > > > > > Yes, i tried once but i was frustrated soon because of the
> > > > > > lot of
> > > > > > required
> > > > > > changes and resulting issues.
> > > > >
> > > > > I got the PIO mode working for the mxs-mmc driver. For this I
> > > > > ported
> > > > > the PIO code from the vendor kernel and removed the usage of
> > > > > the
> > > > > DMA
> > > > > engine entirely.
> > > >
> > > > Good job
> > > >
> > > > >
> > > > > Testing network bandwidth with iperf, I get about ~10Mbit/sec
> > > > > with
> > > > > PIO
> > > > > mode compared to ~6.5Mbit/sec with DMA mode for UDP and about
> > > > > ~6.5Mbit/sec compared to ~4.5Mbit/sec with DMA mode for TCP.
> > > >
> > > > And how about MMC / sd card performance?
> > >
> > > I noticed poor performance with the i.MX28 MMC and/or DMA driver
> > > using
> > > the mainline kernel compared to the vendor Freescale kernel 2.6.35.
> > > I've seen that hou have added the drivers to mainline some years
> > > ago.
> > >
> > > My custom i.MX28 board has a wifi chip attached to the SSP2
> > > interface.
> > > Comparing the bandwith with iperf I get >20Mbits/sec on the vendor
> > > kernel and <5Mbits/sec on the mainline kernel.
> >
> > there is one thing about the clock handling. I noticed that the
> > Vendor Kernel
> > round up the clock frequency and the Mainline Kernel round down the
> > clock
> > frequency [1]. So don't trust the clock ratings from DT / board code.
> > Better
> > verify the register settings or check it with an osci.
> >
> > [1] - http://www.spinics.net/lists/linux-mmc/msg09132.html
>
> I checked the clock rate setting by reading the register 0x80014070
> (HW_SSP2_TIMING). CLOCK_DIVIDE is 0x2 and CLOCK_RATE is 0x0. As SSP CLK
> is 96MHz this makes a clock rate of 48MHz.
>
> There was a discussion on the mailing list [1] about that tasklets
> might be slow.
>
> [1] http://lists.infradead.org/pipermail/linux-arm-kernel/2011-February
> /043395.html
if i unterstand it right the tasklet is not the problem, but the design of the
MXS DMA driver. Please refer to the chapter "General Design Notes" to the
documentation of the DMA provider [2].
I think the MXS DMA driver is affected. Maybe you should ask Vinod Koul about
this.
[2] - https://www.kernel.org/doc/Documentation/dmaengine/provider.txt
>
> J?rg
^ permalink raw reply
* [Bug] ARM: mxs: STI: console can't wake up from freeze
From: Zhang Rui @ 2016-11-05 11:39 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <322177156.158733.9867e3e7-5710-4844-a098-6f44bd852a6d.open-xchange@email.1und1.de>
On Sat, 2016-11-05 at 12:07 +0100, Stefan Wahren wrote:
> Hi,
>
> [add Rui]
>
> >
> > Russell King - ARM Linux <linux@armlinux.org.uk> hat am 1. November
> > 2016 um
> > 10:23 geschrieben:
> >
> >
> > On Mon, Oct 31, 2016 at 08:54:33PM +0100, Stefan Wahren wrote:
> > >
> > > [??366.696043] INFO: task ext4lazyinit:70 blocked for more than
> > > 120 seconds.
> > > [??366.703046]???????Not tainted 4.9.0-rc1 #7
> > > [??366.707188] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
> > > disables
> > > this
> > > message.
> > > [??366.715161] ext4lazyinit????D c05aa6ac?????0????70??????2
> > > 0x00000000
> > This looks like a very different problem - I guess one for ext4
> > people.
> >
> i investigated the issue more further. This trace wasn't
> representative, because
> the stacktrace is different in most cases. The "endless loop" is
> located in
> freeze_enter(). So i added some debug messages:
>
is this a regression?
> -------------------------------->8-----------------------------------
> -
> diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c
> index 1e7f5da..079c08d 100644
> --- a/kernel/power/suspend.c
> +++ b/kernel/power/suspend.c
> @@ -67,17 +67,20 @@ static void freeze_enter(void)
> ? spin_unlock_irq(&suspend_freeze_lock);
> ?
> ? get_online_cpus();
> + pr_info("PM: calling cpuidle_resume()\n");
> ? cpuidle_resume();
> ?
> ? /* Push all the CPUs into the idle loop. */
> + pr_info("PM: calling wake_up_all_idle_cpus()\n");
> ? wake_up_all_idle_cpus();
> - pr_debug("PM: suspend-to-idle\n");
> + pr_info("PM: suspend-to-idle\n");
> ? /* Make the current CPU wait so it can enter the idle loop
> too. */
> ? wait_event(suspend_freeze_wait_head,
> ? ???suspend_freeze_state == FREEZE_STATE_WAKE);
> - pr_debug("PM: resume from suspend-to-idle\n");
> + pr_info("PM: resume from suspend-to-idle\n");
> ?
> ? cpuidle_pause();
> + pr_info("PM: called cpuidle_pause()\n");
> ? put_online_cpus();
> ?
> ? spin_lock_irq(&suspend_freeze_lock);
> @@ -91,6 +94,8 @@ void freeze_wake(void)
> ?{
> ? unsigned long flags;
> ?
> + pr_info("PM: freeze_wake()\n");
> +
> ? spin_lock_irqsave(&suspend_freeze_lock, flags);
> ? if (suspend_freeze_state > FREEZE_STATE_NONE) {
> ? suspend_freeze_state = FREEZE_STATE_WAKE;
>
> -------------------------------->8-----------------------------------
> -
>
> and repeated the test cases for freeze which shows none the
> representative
> stacktraces:
>
> 1. cmdline contains no_console_suspend
>
> * echo freeze > /sys/power/state
>
> ...
> [??139.371308] PM: suspend of devices complete after 1342.721 msecs
> [??139.385203] PM: late suspend of devices complete after 7.668 msecs
> [??139.399428] PM: noirq suspend of devices complete after 7.936
> msecs
> [??139.406639] PM: calling cpuidle_resume()
> [??139.410619] PM: calling wake_up_all_idle_cpus()
> [??139.415339] PM: suspend-to-idle
>
> >
> > >
> > > >
> > > > no reaction to input via Debug UART
> [??366.683570] INFO: task bash:373 blocked for more than 120 seconds.
> [??366.689814]???????Not tainted 4.9.0-rc1-dirty #14
> [??366.694705] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
> disables this
> message.
> [??366.702685] bash????????????D c05aa6ec?????0???373????275
> 0x00000000
> [??366.709227] [<c05aa6ec>] (__schedule) from [<c05aaff8>]
> (schedule+0x3c/0xbc)
> [??366.716495] [<c05aaff8>] (schedule) from [<c005b588>]
> (suspend_devices_and_enter+0x888/0xa78)
> [??366.725228] [<c005b588>] (suspend_devices_and_enter) from
> [<c005bea4>]
> (pm_suspend+0x72c/0x81c)
> [??366.734150] [<c005bea4>] (pm_suspend) from [<c005a008>]
> (state_store+0x80/0xcc)
> [??366.741554] [<c005a008>] (state_store) from [<c02f0270>]
> (kobj_attr_store+0x18/0x1c)
> [??366.749515] [<c02f0270>] (kobj_attr_store) from [<c01911e8>]
> (sysfs_kf_write+0x48/0x4c)
> [??366.757735] [<c01911e8>] (sysfs_kf_write) from [<c0190308>]
> (kernfs_fop_write+0xfc/0x1d0)
> [??366.766130] [<c0190308>] (kernfs_fop_write) from [<c011f578>]
> (__vfs_write+0x2c/0x124)
> [??366.774255] [<c011f578>] (__vfs_write) from [<c011f724>]
> (vfs_write+0xb4/0x1a4)
> [??366.781640] [<c011f724>] (vfs_write) from [<c011f8e8>]
> (SyS_write+0x44/0x88)
> [??366.788890] [<c011f8e8>] (SyS_write) from [<c000a2c0>]
> (ret_fast_syscall+0x0/0x1c)
> [??366.796627]
> [??366.796627] Showing all locks held in the system:
> [??366.803011] 2 locks held by khungtaskd/10:
> [??366.807149]??#0: [??366.808931]??(
> rcu_read_lock[??366.811784] ){......}
> , at: [??366.814813] [<c0093a40>] watchdog+0xb4/0x61c
> [??366.819128]??#1: [??366.820902]??(
> tasklist_lock[??366.823876] ){.+.+..}
> , at: [??366.826765] [<c0051dbc>] debug_show_all_locks+0x28/0x1bc
> [??366.832151] 4 locks held by bash/373:
> [??366.835973]??#0: [??366.837765]??(
> sb_writers[??366.840365] #4
> ){.+.+.+}[??366.842987] , at:
> [??366.845079] [<c011f804>] vfs_write+0x194/0x1a4
> [??366.849551]??#1: [??366.851324]??(
> &of->mutex[??366.854058] ){+.+.+.}
> , at: [??366.856944] [<c01902cc>] kernfs_fop_write+0xc0/0x1d0
> [??366.861941]??#2: [??366.863839]??(
> s_active[??366.866288] #43
> ){.+.+.+}[??366.868877] , at:
> [??366.870938] [<c01902d4>] kernfs_fop_write+0xc8/0x1d0
> [??366.876053]??#3: [??366.877843]??(
> pm_mutex[??366.880272] ){+.+.+.}
> , at: [??366.883266] [<c005b808>] pm_suspend+0x90/0x81c
> [??366.887757]
> [??366.889268] =============================================
> [??366.889268]
>
> 2. cmdline contains no_console_suspend
>
> * echo enabled > /sys/class/tty/ttyAMA0/power/wakeup
> * echo freeze > /sys/power/state
>
> the same as in 1.
>
> 3. cmdline doesn't contains no_console_suspend
>
> * echo enabled > /sys/class/tty/ttyAMA0/power/wakeup
> * echo freeze > /sys/power/state
>
> [??161.093187] PM: Syncing filesystems ... [??161.734413] done.
> [??161.793242] Freezing user space processes ... (elapsed 0.008
> seconds) done.
> [??161.809797] Freezing remaining freezable tasks ... (elapsed 0.004
> seconds)
> done.
> [??161.821953] Suspending console(s) (use no_console_suspend to
> debug)
>
> >?
> > > > > no reaction to Debug UART
Then the system does not have any response? or the system freezes and
wakes up as expected?
> I expected that in all 3 cases freeze_wake() will be called. Why not?
>
> Here my config again:
>
> CONFIG_PM_SLEEP=y
> CONFIG_SUSPEND=y
> CONFIG_SUSPEND_FREEZER=y
> CONFIG_PM=y
> CONFIG_CPU_IDLE is not set
hmmm, why not have CONFIG_CPU_IDLE set?
thanks,
rui
^ permalink raw reply
* [PATCH v8 0/3] ARM, arm64: renesas: Enable UHS-I SDR-104
From: Wolfram Sang @ 2016-11-05 11:45 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478185645-25556-1-git-send-email-horms+renesas@verge.net.au>
> To aid review the following git branches are provided:
> * This patchset:
> https:://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git topic/sdr104-integration-v8
> * This patch-set and above dependency
> https:://git.kernel.org/pub/scm/linux/kernel/git/horms/renesas.git topic/sdr104-v8
driver-v8 is there, but those two are not. Forgot to push out? :)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20161105/84af31ec/attachment.sig>
^ permalink raw reply
* [PATCH] ARM: dts: imx6: Add imx-weim parameters to dtsi's
From: Shawn Guo @ 2016-11-05 11:47 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478044305-5666-1-git-send-email-stillcompiling@gmail.com>
On Tue, Nov 01, 2016 at 04:51:45PM -0700, Joshua Clayton wrote:
> imx-weim should always set address-cells to 2,
> and size_cells to 1.
> On imx6, fsl,weim-cs-gpr will always be &gpr
>
> Set these common parameters in the dtsi file,
> rather than in a downstream dts.
>
> Signed-off-by: Joshua Clayton <stillcompiling@gmail.com>
Applied, thanks.
^ permalink raw reply
* [Bug] ARM: mxs: STI: console can't wake up from freeze
From: Stefan Wahren @ 2016-11-05 12:01 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478345972.2206.15.camel@intel.com>
Hi Rui,
> Zhang Rui <rui.zhang@intel.com> hat am 5. November 2016 um 12:39 geschrieben:
>
>
> On Sat, 2016-11-05 at 12:07 +0100, Stefan Wahren wrote:
> > Hi,
> >
> > [add Rui]
> >
> > >
> > > Russell King - ARM Linux <linux@armlinux.org.uk> hat am 1. November
> > > 2016 um
> > > 10:23 geschrieben:
> > >
> > >
> > > On Mon, Oct 31, 2016 at 08:54:33PM +0100, Stefan Wahren wrote:
> > > >
> > > > [??366.696043] INFO: task ext4lazyinit:70 blocked for more than
> > > > 120 seconds.
> > > > [??366.703046]???????Not tainted 4.9.0-rc1 #7
> > > > [??366.707188] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
> > > > disables
> > > > this
> > > > message.
> > > > [??366.715161] ext4lazyinit????D c05aa6ac?????0????70??????2
> > > > 0x00000000
> > > This looks like a very different problem - I guess one for ext4
> > > people.
> > >
> > i investigated the issue more further. This trace wasn't
> > representative, because
> > the stacktrace is different in most cases. The "endless loop" is
> > located in
> > freeze_enter(). So i added some debug messages:
> >
> is this a regression?
i've never seen this working on MXS platform, but maybe i hit a corner case. The
oldest kernel that i tested was 3.18 and this issue was reproducible. Should i
test an older version?
> >
> > and repeated the test cases for freeze which shows none the
> > representative
> > stacktraces:
> >
> > 1. cmdline contains no_console_suspend
> >
> > * echo freeze > /sys/power/state
> >
> > ...
> > [??139.371308] PM: suspend of devices complete after 1342.721 msecs
> > [??139.385203] PM: late suspend of devices complete after 7.668 msecs
> > [??139.399428] PM: noirq suspend of devices complete after 7.936
> > msecs
> > [??139.406639] PM: calling cpuidle_resume()
> > [??139.410619] PM: calling wake_up_all_idle_cpus()
> > [??139.415339] PM: suspend-to-idle
> >
> > >
> > > >
> > > > >
> > > > > no reaction to input via Debug UART
> > [??366.683570] INFO: task bash:373 blocked for more than 120 seconds.
> > [??366.689814]???????Not tainted 4.9.0-rc1-dirty #14
> > [??366.694705] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs"
> > disables this
> > message.
> > [??366.702685] bash????????????D c05aa6ec?????0???373????275
> > 0x00000000
> > [??366.709227] [<c05aa6ec>] (__schedule) from [<c05aaff8>]
> > (schedule+0x3c/0xbc)
> > [??366.716495] [<c05aaff8>] (schedule) from [<c005b588>]
> > (suspend_devices_and_enter+0x888/0xa78)
> > [??366.725228] [<c005b588>] (suspend_devices_and_enter) from
> > [<c005bea4>]
> > (pm_suspend+0x72c/0x81c)
> > [??366.734150] [<c005bea4>] (pm_suspend) from [<c005a008>]
> > (state_store+0x80/0xcc)
> > [??366.741554] [<c005a008>] (state_store) from [<c02f0270>]
> > (kobj_attr_store+0x18/0x1c)
> > [??366.749515] [<c02f0270>] (kobj_attr_store) from [<c01911e8>]
> > (sysfs_kf_write+0x48/0x4c)
> > [??366.757735] [<c01911e8>] (sysfs_kf_write) from [<c0190308>]
> > (kernfs_fop_write+0xfc/0x1d0)
> > [??366.766130] [<c0190308>] (kernfs_fop_write) from [<c011f578>]
> > (__vfs_write+0x2c/0x124)
> > [??366.774255] [<c011f578>] (__vfs_write) from [<c011f724>]
> > (vfs_write+0xb4/0x1a4)
> > [??366.781640] [<c011f724>] (vfs_write) from [<c011f8e8>]
> > (SyS_write+0x44/0x88)
> > [??366.788890] [<c011f8e8>] (SyS_write) from [<c000a2c0>]
> > (ret_fast_syscall+0x0/0x1c)
> > [??366.796627]
> > [??366.796627] Showing all locks held in the system:
> > [??366.803011] 2 locks held by khungtaskd/10:
> > [??366.807149]??#0: [??366.808931]??(
> > rcu_read_lock[??366.811784] ){......}
> > , at: [??366.814813] [<c0093a40>] watchdog+0xb4/0x61c
> > [??366.819128]??#1: [??366.820902]??(
> > tasklist_lock[??366.823876] ){.+.+..}
> > , at: [??366.826765] [<c0051dbc>] debug_show_all_locks+0x28/0x1bc
> > [??366.832151] 4 locks held by bash/373:
> > [??366.835973]??#0: [??366.837765]??(
> > sb_writers[??366.840365] #4
> > ){.+.+.+}[??366.842987] , at:
> > [??366.845079] [<c011f804>] vfs_write+0x194/0x1a4
> > [??366.849551]??#1: [??366.851324]??(
> > &of->mutex[??366.854058] ){+.+.+.}
> > , at: [??366.856944] [<c01902cc>] kernfs_fop_write+0xc0/0x1d0
> > [??366.861941]??#2: [??366.863839]??(
> > s_active[??366.866288] #43
> > ){.+.+.+}[??366.868877] , at:
> > [??366.870938] [<c01902d4>] kernfs_fop_write+0xc8/0x1d0
> > [??366.876053]??#3: [??366.877843]??(
> > pm_mutex[??366.880272] ){+.+.+.}
> > , at: [??366.883266] [<c005b808>] pm_suspend+0x90/0x81c
> > [??366.887757]
> > [??366.889268] =============================================
> > [??366.889268]
> >
> > 2. cmdline contains no_console_suspend
> >
> > * echo enabled > /sys/class/tty/ttyAMA0/power/wakeup
> > * echo freeze > /sys/power/state
> >
> > the same as in 1.
> >
> > 3. cmdline doesn't contains no_console_suspend
> >
> > * echo enabled > /sys/class/tty/ttyAMA0/power/wakeup
> > * echo freeze > /sys/power/state
> >
> > [??161.093187] PM: Syncing filesystems ... [??161.734413] done.
> > [??161.793242] Freezing user space processes ... (elapsed 0.008
> > seconds) done.
> > [??161.809797] Freezing remaining freezable tasks ... (elapsed 0.004
> > seconds)
> > done.
> > [??161.821953] Suspending console(s) (use no_console_suspend to
> > debug)
> >
> > >?
> > > > > > no reaction to Debug UART
>
> Then the system does not have any response? or the system freezes and
> wakes up as expected?
The system doesn't react to any input from Debug UART and after 2 minutes i got
the following stacktrace about hung task. But i didn't get any change to come
back to the console. Interestingly basic PM debugging tests doesn't show this
issue. Also "echo mem > /sys/power/state" doesn't cause this issue.
>
> > I expected that in all 3 cases freeze_wake() will be called. Why not?
> >
> > Here my config again:
> >
> > CONFIG_PM_SLEEP=y
> > CONFIG_SUSPEND=y
> > CONFIG_SUSPEND_FREEZER=y
> > CONFIG_PM=y
> > CONFIG_CPU_IDLE is not set
>
> hmmm, why not have CONFIG_CPU_IDLE set?
I'm using mxs_defconfig which doesn't select the ARM CPU idle. Is this
necessary?
>
> thanks,
> rui
^ permalink raw reply
* [GIT PULL] i.MX fixes for 4.9
From: Fabio Estevam @ 2016-11-05 12:02 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <CAOMZO5D9uBBwWV8UJOLj-2iwWO6sReB5BAtzdi1q6OqbtvcSOQ@mail.gmail.com>
Hi Olof,
On Mon, Oct 31, 2016 at 9:29 PM, Fabio Estevam <festevam@gmail.com> wrote:
> I think this will be fixed with this pcie-designware patch:
> http://www.spinics.net/lists/linux-pci/msg55244.html
Bjorn sent this fix to Linus, so we should be fine now.
>
>> Wandboard, has a peculiar MMC error. First I thought it was just the
>> card going bad, but now I'm not so sure since multi_v7 boots just fine
>> on the same hardware:
>
> Yes, I have been noticing this mmc error reported in your autobooter
> and I also thought it was some kind of a bad SD card.
>
> This particular error I have never been able to reproduce. Also
> checked kernelci and do not see this mmc error there.
>
> Interesting that it happens only with imx_v6_v7_defconfig and not with
> multi_v7_defconfig.
Today wandboard does not show the mmc error with imx_v6_v7_defconfig:
http://arm-soc.lixom.net/bootlogs/mainline/v4.9-rc3-376-gfb415f2/wandboard-arm-imx_v6_v7_defconfig.html
Looks like the mmc error come and go?
^ permalink raw reply
* [PATCH v7 0/2] Introducing Exynos ChipId driver
From: Pankaj Dubey @ 2016-11-05 12:03 UTC (permalink / raw)
To: linux-arm-kernel
Once again I am attempting this quite old patch series to introduce
Exynos Chipid driver.
Each Exynos SoC has ChipID block which can give information about SoC's
product Id and revision number.
This patch series introduces Exynos Chipid SoC driver. At the same time
it reduces dependency of mach-exynos files from plat-samsung, by removing
soc_is_exynosMMMM and samsung_rev API. Instead of it now we can use
soc_device_match API proposed by Arnd and getting discussed in thread [1].
Until now we are using static mapping of Exynos Chipid and using this static
mapping to know about SoC name and revision via soc_is_exynosMMMM macro. This
is quite cumbersome and every time new ARMv7 based Exynos SoC supports lands in
mainline a bunch of such new macros needs to be added. Quite long back during
support of Exynos5260 it has been discussed to solve this problem.
To solve this issue this patchset replaces use of soc_is_exynosMMMM by either
of_machine_is_compatible or soc_device_match depending upon usecase.
This patchset depends on [2], and I have prepared it on top of my other patchset [3]
(under-review) which introduces SCU device node for Exynos
I have tested this patch series on Exynos4210 based Origen board for normal SMP
boot.
[1]: https://patchwork.kernel.org/patch/9361389/
[2]: http://www.mail-archive.com/linux-kernel at vger.kernel.org/msg1261739.html
[3]: http://www.spinics.net/lists/arm-kernel/msg540498.html
Revision 6 and it's discussion can be found here:
- https://lkml.org/lkml/2016/5/25/101
Revision 5 and it's discussion can be found here:
- http://lists.infradead.org/pipermail/linux-arm-kernel/2014-December/310046.html
Revision 4 and it's discussion can be found here:
- https://lkml.org/lkml/2014/12/3/115
Chances since v6:
- Removed platform driver from chipid, instead use early_init to register soc_device
- Removed export functions from exynos chipid driver
- Replace soc_is_exynosMMMM via either of_machine_is_compatible or soc_device_match in
files in arm/mach-exynos folder
- This patchset depends on the following patch series by Geert Uytterhoeven. This series
includes patch which introduces soc_device_match.
http://www.mail-archive.com/linux-kernel at vger.kernel.org/msg1261739.html
- Rebased on latest krzk/for-next branch and retested.
Change since v5:
- Addressed Rob's review comments.
- Rebased on latest krzk/for-next branch and retested.
Changes since v4:
- Removed custom sysfs entries as they were not providing any new information
as pointed out by Arnd.
- Removed functions exporting product_id and revision, instead we will export
exynos_chipid_info structure. It will be helpfull when we need to provide more
fields of chipid outside of chipid, as commented by Yadwinder
- Converted all funcions as __init.
Change since v3:
- This patch set contains 5/6 and 6/6 patch from v3 series.
- Made EXYNOS_CHIPID config option non-user selectable,
as suggested by Tomasz Figa.
- Made uniform macro for EXYNOS4/5_SOC_MASK as EXYNOS_SOC_MASK as
suggested by Tomasz Figa.
- Made local variables static in chipid driver.
- Added existing SoC's product id's.
- Added platform driver support.
Changes since v2:
- Reorganized patches as suggested by Tomasz Figa.
- Addressed review comments of Tomasz Figa in i2c-s3c2410.c file.
Pankaj Dubey (2):
soc: samsung: add exynos chipid driver support
ARM: EXYNOS: refactoring of mach-exynos to enable chipid driver
arch/arm/mach-exynos/Kconfig | 1 +
arch/arm/mach-exynos/common.h | 92 -----------------
arch/arm/mach-exynos/exynos.c | 31 ------
arch/arm/mach-exynos/firmware.c | 10 +-
arch/arm/mach-exynos/include/mach/map.h | 21 ----
arch/arm/mach-exynos/platsmp.c | 22 ++--
arch/arm/mach-exynos/pm.c | 41 +++++---
arch/arm/plat-samsung/cpu.c | 14 ---
arch/arm/plat-samsung/include/plat/cpu.h | 2 -
arch/arm/plat-samsung/include/plat/map-s5p.h | 2 -
drivers/soc/samsung/Kconfig | 5 +
drivers/soc/samsung/Makefile | 1 +
drivers/soc/samsung/exynos-chipid.c | 148 +++++++++++++++++++++++++++
13 files changed, 201 insertions(+), 189 deletions(-)
delete mode 100644 arch/arm/mach-exynos/include/mach/map.h
create mode 100644 drivers/soc/samsung/exynos-chipid.c
--
2.7.4
^ permalink raw reply
* [PATCH v7 1/2] soc: samsung: add exynos chipid driver support
From: Pankaj Dubey @ 2016-11-05 12:03 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478347427-28409-1-git-send-email-pankaj.dubey@samsung.com>
Exynos SoCs have Chipid, for identification of product IDs
and SoC revisions. This patch intends to provide initialization
code for all these functionalities, at the same time it provides some
sysfs entries for accessing these information to user-space.
This driver uses existing binding for exynos-chipid.
CC: Grant Likely <grant.likely@linaro.org>
CC: Rob Herring <robh+dt@kernel.org>
CC: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Pankaj Dubey <pankaj.dubey@samsung.com>
---
drivers/soc/samsung/Kconfig | 5 ++
drivers/soc/samsung/Makefile | 1 +
drivers/soc/samsung/exynos-chipid.c | 148 ++++++++++++++++++++++++++++++++++++
3 files changed, 154 insertions(+)
create mode 100644 drivers/soc/samsung/exynos-chipid.c
diff --git a/drivers/soc/samsung/Kconfig b/drivers/soc/samsung/Kconfig
index 2455339..f9ab858 100644
--- a/drivers/soc/samsung/Kconfig
+++ b/drivers/soc/samsung/Kconfig
@@ -14,4 +14,9 @@ config EXYNOS_PM_DOMAINS
bool "Exynos PM domains" if COMPILE_TEST
depends on PM_GENERIC_DOMAINS || COMPILE_TEST
+config EXYNOS_CHIPID
+ bool "Exynos Chipid controller driver" if COMPILE_TEST
+ depends on (ARM && ARCH_EXYNOS) || ((ARM || ARM64) && COMPILE_TEST)
+ select SOC_BUS
+
endif
diff --git a/drivers/soc/samsung/Makefile b/drivers/soc/samsung/Makefile
index 3619f2e..2a8a85e 100644
--- a/drivers/soc/samsung/Makefile
+++ b/drivers/soc/samsung/Makefile
@@ -1,3 +1,4 @@
obj-$(CONFIG_EXYNOS_PMU) += exynos-pmu.o exynos3250-pmu.o exynos4-pmu.o \
exynos5250-pmu.o exynos5420-pmu.o
obj-$(CONFIG_EXYNOS_PM_DOMAINS) += pm_domains.o
+obj-$(CONFIG_EXYNOS_CHIPID) += exynos-chipid.o
diff --git a/drivers/soc/samsung/exynos-chipid.c b/drivers/soc/samsung/exynos-chipid.c
new file mode 100644
index 0000000..9761e54
--- /dev/null
+++ b/drivers/soc/samsung/exynos-chipid.c
@@ -0,0 +1,148 @@
+/*
+ * Copyright (c) 2016 Samsung Electronics Co., Ltd.
+ * http://www.samsung.com/
+ *
+ * EXYNOS - CHIP ID support
+ * Author: Pankaj Dubey <pankaj.dubey@samsung.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/io.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_platform.h>
+#include <linux/platform_device.h>
+#include <linux/slab.h>
+#include <linux/sys_soc.h>
+
+#define EXYNOS3250_SOC_ID 0xE3472000
+#define EXYNOS4210_SOC_ID 0x43210000
+#define EXYNOS4212_SOC_ID 0x43220000
+#define EXYNOS4412_SOC_ID 0xE4412000
+#define EXYNOS4415_SOC_ID 0xE4415000
+#define EXYNOS5250_SOC_ID 0x43520000
+#define EXYNOS5260_SOC_ID 0xE5260000
+#define EXYNOS5410_SOC_ID 0xE5410000
+#define EXYNOS5420_SOC_ID 0xE5420000
+#define EXYNOS5440_SOC_ID 0xE5440000
+#define EXYNOS5800_SOC_ID 0xE5422000
+
+#define EXYNOS_SOC_MASK 0xFFFFF000
+
+#define EXYNOS4210_REV_0 0x0
+#define EXYNOS4210_REV_1_0 0x10
+#define EXYNOS4210_REV_1_1 0x11
+
+#define EXYNOS_SUBREV_MASK (0xF << 4)
+#define EXYNOS_MAINREV_MASK (0xF << 0)
+#define EXYNOS_REV_MASK (EXYNOS_SUBREV_MASK | EXYNOS_MAINREV_MASK)
+
+
+static const char * __init product_id_to_soc_id(unsigned int product_id)
+{
+ const char *soc_name;
+ unsigned int soc_id = product_id & EXYNOS_SOC_MASK;
+
+ switch (soc_id) {
+ case EXYNOS3250_SOC_ID:
+ soc_name = "EXYNOS3250";
+ break;
+ case EXYNOS4210_SOC_ID:
+ soc_name = "EXYNOS4210";
+ break;
+ case EXYNOS4212_SOC_ID:
+ soc_name = "EXYNOS4212";
+ break;
+ case EXYNOS4412_SOC_ID:
+ soc_name = "EXYNOS4412";
+ break;
+ case EXYNOS4415_SOC_ID:
+ soc_name = "EXYNOS4415";
+ break;
+ case EXYNOS5250_SOC_ID:
+ soc_name = "EXYNOS5250";
+ break;
+ case EXYNOS5260_SOC_ID:
+ soc_name = "EXYNOS5260";
+ break;
+ case EXYNOS5420_SOC_ID:
+ soc_name = "EXYNOS5420";
+ break;
+ case EXYNOS5440_SOC_ID:
+ soc_name = "EXYNOS5440";
+ break;
+ case EXYNOS5800_SOC_ID:
+ soc_name = "EXYNOS5800";
+ break;
+ default:
+ soc_name = "UNKNOWN";
+ }
+ return soc_name;
+}
+
+static const struct of_device_id of_exynos_chipid_ids[] = {
+ {
+ .compatible = "samsung,exynos4210-chipid",
+ },
+ {},
+};
+
+/**
+ * exynos_chipid_early_init: Early chipid initialization
+ */
+int __init exynos_chipid_early_init(void)
+{
+ struct soc_device_attribute *soc_dev_attr;
+ struct soc_device *soc_dev;
+ struct device_node *root;
+ struct device_node *np;
+ void __iomem *exynos_chipid_base;
+ const struct of_device_id *match;
+ u32 product_id;
+ u32 revision;
+
+ np = of_find_matching_node_and_match(NULL,
+ of_exynos_chipid_ids, &match);
+ if (!np)
+ return -ENODEV;
+
+ exynos_chipid_base = of_iomap(np, 0);
+
+ if (!exynos_chipid_base)
+ return PTR_ERR(exynos_chipid_base);
+
+ product_id = __raw_readl(exynos_chipid_base);
+ revision = product_id & EXYNOS_REV_MASK;
+ iounmap(exynos_chipid_base);
+
+ soc_dev_attr = kzalloc(sizeof(*soc_dev_attr), GFP_KERNEL);
+ if (!soc_dev_attr)
+ return -ENODEV;
+
+ soc_dev_attr->family = "Samsung Exynos";
+
+ root = of_find_node_by_path("/");
+ of_property_read_string(root, "model", &soc_dev_attr->machine);
+ of_node_put(root);
+
+ soc_dev_attr->revision = kasprintf(GFP_KERNEL, "%x", revision);
+ soc_dev_attr->soc_id = product_id_to_soc_id(product_id);
+
+
+ pr_info("Exynos: CPU[%s] CPU_REV[0x%x] Detected\n",
+ product_id_to_soc_id(product_id), revision);
+
+ soc_dev = soc_device_register(soc_dev_attr);
+ if (IS_ERR(soc_dev)) {
+ kfree(soc_dev_attr->revision);
+ kfree_const(soc_dev_attr->soc_id);
+ kfree(soc_dev_attr);
+ return PTR_ERR(soc_dev);
+ }
+
+ return 0;
+}
+early_initcall(exynos_chipid_early_init);
--
2.7.4
^ permalink raw reply related
* [PATCH v7 2/2] ARM: EXYNOS: refactoring of mach-exynos to enable chipid driver
From: Pankaj Dubey @ 2016-11-05 12:03 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478347427-28409-1-git-send-email-pankaj.dubey@samsung.com>
This patch enables chipid driver for ARCH_EXYNOS and refactors
machine code for using chipid driver for identification of
SoC ID and SoC rev.
Signed-off-by: Pankaj Dubey <pankaj.dubey@samsung.com>
---
arch/arm/mach-exynos/Kconfig | 1 +
arch/arm/mach-exynos/common.h | 92 ----------------------------
arch/arm/mach-exynos/exynos.c | 31 ----------
arch/arm/mach-exynos/firmware.c | 10 +--
arch/arm/mach-exynos/include/mach/map.h | 21 -------
arch/arm/mach-exynos/platsmp.c | 22 ++++---
arch/arm/mach-exynos/pm.c | 41 ++++++++-----
arch/arm/plat-samsung/cpu.c | 14 -----
arch/arm/plat-samsung/include/plat/cpu.h | 2 -
arch/arm/plat-samsung/include/plat/map-s5p.h | 2 -
10 files changed, 47 insertions(+), 189 deletions(-)
delete mode 100644 arch/arm/mach-exynos/include/mach/map.h
diff --git a/arch/arm/mach-exynos/Kconfig b/arch/arm/mach-exynos/Kconfig
index b085855..a76c679 100644
--- a/arch/arm/mach-exynos/Kconfig
+++ b/arch/arm/mach-exynos/Kconfig
@@ -16,6 +16,7 @@ menuconfig ARCH_EXYNOS
select ARM_AMBA
select ARM_GIC
select COMMON_CLK_SAMSUNG
+ select EXYNOS_CHIPID
select EXYNOS_THERMAL
select EXYNOS_PMU
select EXYNOS_SROM
diff --git a/arch/arm/mach-exynos/common.h b/arch/arm/mach-exynos/common.h
index d19064b..9d76cf8 100644
--- a/arch/arm/mach-exynos/common.h
+++ b/arch/arm/mach-exynos/common.h
@@ -14,97 +14,6 @@
#include <linux/platform_data/cpuidle-exynos.h>
-#define EXYNOS3250_SOC_ID 0xE3472000
-#define EXYNOS3_SOC_MASK 0xFFFFF000
-
-#define EXYNOS4210_CPU_ID 0x43210000
-#define EXYNOS4212_CPU_ID 0x43220000
-#define EXYNOS4412_CPU_ID 0xE4412200
-#define EXYNOS4_CPU_MASK 0xFFFE0000
-
-#define EXYNOS5250_SOC_ID 0x43520000
-#define EXYNOS5410_SOC_ID 0xE5410000
-#define EXYNOS5420_SOC_ID 0xE5420000
-#define EXYNOS5440_SOC_ID 0xE5440000
-#define EXYNOS5800_SOC_ID 0xE5422000
-#define EXYNOS5_SOC_MASK 0xFFFFF000
-
-extern unsigned long samsung_cpu_id;
-
-#define IS_SAMSUNG_CPU(name, id, mask) \
-static inline int is_samsung_##name(void) \
-{ \
- return ((samsung_cpu_id & mask) == (id & mask)); \
-}
-
-IS_SAMSUNG_CPU(exynos3250, EXYNOS3250_SOC_ID, EXYNOS3_SOC_MASK)
-IS_SAMSUNG_CPU(exynos4210, EXYNOS4210_CPU_ID, EXYNOS4_CPU_MASK)
-IS_SAMSUNG_CPU(exynos4212, EXYNOS4212_CPU_ID, EXYNOS4_CPU_MASK)
-IS_SAMSUNG_CPU(exynos4412, EXYNOS4412_CPU_ID, EXYNOS4_CPU_MASK)
-IS_SAMSUNG_CPU(exynos5250, EXYNOS5250_SOC_ID, EXYNOS5_SOC_MASK)
-IS_SAMSUNG_CPU(exynos5410, EXYNOS5410_SOC_ID, EXYNOS5_SOC_MASK)
-IS_SAMSUNG_CPU(exynos5420, EXYNOS5420_SOC_ID, EXYNOS5_SOC_MASK)
-IS_SAMSUNG_CPU(exynos5440, EXYNOS5440_SOC_ID, EXYNOS5_SOC_MASK)
-IS_SAMSUNG_CPU(exynos5800, EXYNOS5800_SOC_ID, EXYNOS5_SOC_MASK)
-
-#if defined(CONFIG_SOC_EXYNOS3250)
-# define soc_is_exynos3250() is_samsung_exynos3250()
-#else
-# define soc_is_exynos3250() 0
-#endif
-
-#if defined(CONFIG_CPU_EXYNOS4210)
-# define soc_is_exynos4210() is_samsung_exynos4210()
-#else
-# define soc_is_exynos4210() 0
-#endif
-
-#if defined(CONFIG_SOC_EXYNOS4212)
-# define soc_is_exynos4212() is_samsung_exynos4212()
-#else
-# define soc_is_exynos4212() 0
-#endif
-
-#if defined(CONFIG_SOC_EXYNOS4412)
-# define soc_is_exynos4412() is_samsung_exynos4412()
-#else
-# define soc_is_exynos4412() 0
-#endif
-
-#define EXYNOS4210_REV_0 (0x0)
-#define EXYNOS4210_REV_1_0 (0x10)
-#define EXYNOS4210_REV_1_1 (0x11)
-
-#if defined(CONFIG_SOC_EXYNOS5250)
-# define soc_is_exynos5250() is_samsung_exynos5250()
-#else
-# define soc_is_exynos5250() 0
-#endif
-
-#if defined(CONFIG_SOC_EXYNOS5410)
-# define soc_is_exynos5410() is_samsung_exynos5410()
-#else
-# define soc_is_exynos5410() 0
-#endif
-
-#if defined(CONFIG_SOC_EXYNOS5420)
-# define soc_is_exynos5420() is_samsung_exynos5420()
-#else
-# define soc_is_exynos5420() 0
-#endif
-
-#if defined(CONFIG_SOC_EXYNOS5440)
-# define soc_is_exynos5440() is_samsung_exynos5440()
-#else
-# define soc_is_exynos5440() 0
-#endif
-
-#if defined(CONFIG_SOC_EXYNOS5800)
-# define soc_is_exynos5800() is_samsung_exynos5800()
-#else
-# define soc_is_exynos5800() 0
-#endif
-
extern u32 cp15_save_diag;
extern u32 cp15_save_power;
@@ -161,7 +70,6 @@ extern struct cpuidle_exynos_data cpuidle_coupled_exynos_data;
extern void exynos_set_delayed_reset_assertion(bool enable);
-extern unsigned int samsung_rev(void);
extern void exynos_core_restart(u32 core_id);
extern int exynos_set_boot_addr(u32 core_id, unsigned long boot_addr);
extern int exynos_get_boot_addr(u32 core_id, unsigned long *boot_addr);
diff --git a/arch/arm/mach-exynos/exynos.c b/arch/arm/mach-exynos/exynos.c
index fa08ef9..942131e 100644
--- a/arch/arm/mach-exynos/exynos.c
+++ b/arch/arm/mach-exynos/exynos.c
@@ -23,9 +23,6 @@
#include <asm/mach/arch.h>
#include <asm/mach/map.h>
-#include <mach/map.h>
-#include <plat/cpu.h>
-
#include "common.h"
static struct platform_device exynos_cpuidle = {
@@ -67,37 +64,9 @@ static void __init exynos_init_late(void)
exynos_pm_init();
}
-static int __init exynos_fdt_map_chipid(unsigned long node, const char *uname,
- int depth, void *data)
-{
- struct map_desc iodesc;
- const __be32 *reg;
- int len;
-
- if (!of_flat_dt_is_compatible(node, "samsung,exynos4210-chipid") &&
- !of_flat_dt_is_compatible(node, "samsung,exynos5440-clock"))
- return 0;
-
- reg = of_get_flat_dt_prop(node, "reg", &len);
- if (reg == NULL || len != (sizeof(unsigned long) * 2))
- return 0;
-
- iodesc.pfn = __phys_to_pfn(be32_to_cpu(reg[0]));
- iodesc.length = be32_to_cpu(reg[1]) - 1;
- iodesc.virtual = (unsigned long)S5P_VA_CHIPID;
- iodesc.type = MT_DEVICE;
- iotable_init(&iodesc, 1);
- return 1;
-}
-
static void __init exynos_init_io(void)
{
debug_ll_io_init();
-
- of_scan_flat_dt(exynos_fdt_map_chipid, NULL);
-
- /* detect cpu id and rev. */
- s5p_init_cpu(S5P_VA_CHIPID);
}
/*
diff --git a/arch/arm/mach-exynos/firmware.c b/arch/arm/mach-exynos/firmware.c
index fd6da54..a9f8504e 100644
--- a/arch/arm/mach-exynos/firmware.c
+++ b/arch/arm/mach-exynos/firmware.c
@@ -44,7 +44,7 @@ static int exynos_do_idle(unsigned long mode)
writel_relaxed(virt_to_phys(exynos_cpu_resume_ns),
sysram_ns_base_addr + 0x24);
writel_relaxed(EXYNOS_AFTR_MAGIC, sysram_ns_base_addr + 0x20);
- if (soc_is_exynos3250()) {
+ if (of_machine_is_compatible("samsung,exynos3250")) {
flush_cache_all();
exynos_smc(SMC_CMD_SAVE, OP_TYPE_CORE,
SMC_POWERSTATE_IDLE, 0);
@@ -65,7 +65,7 @@ static int exynos_cpu_boot(int cpu)
* Exynos3250 doesn't need to send smc command for secondary CPU boot
* because Exynos3250 removes WFE in secure mode.
*/
- if (soc_is_exynos3250())
+ if (of_machine_is_compatible("samsung,exynos3250"))
return 0;
/*
@@ -73,7 +73,7 @@ static int exynos_cpu_boot(int cpu)
* But, Exynos4212 has only one secondary CPU so second parameter
* isn't used for informing secure firmware about CPU id.
*/
- if (soc_is_exynos4212())
+ if (of_machine_is_compatible("samsung,exynos4212"))
cpu = 0;
exynos_smc(SMC_CMD_CPU1BOOT, cpu, 0, 0);
@@ -94,7 +94,7 @@ static int exynos_set_cpu_boot_addr(int cpu, unsigned long boot_addr)
* additional offset for every CPU, with Exynos4412 being the only
* exception.
*/
- if (soc_is_exynos4412())
+ if (of_machine_is_compatible("samsung,exynos4412"))
boot_reg += 4 * cpu;
writel_relaxed(boot_addr, boot_reg);
@@ -110,7 +110,7 @@ static int exynos_get_cpu_boot_addr(int cpu, unsigned long *boot_addr)
boot_reg = sysram_ns_base_addr + 0x1c;
- if (soc_is_exynos4412())
+ if (of_machine_is_compatible("samsung,exynos4412"))
boot_reg += 4 * cpu;
*boot_addr = readl_relaxed(boot_reg);
diff --git a/arch/arm/mach-exynos/include/mach/map.h b/arch/arm/mach-exynos/include/mach/map.h
deleted file mode 100644
index 0eef407..0000000
--- a/arch/arm/mach-exynos/include/mach/map.h
+++ /dev/null
@@ -1,21 +0,0 @@
-/*
- * Copyright (c) 2010-2011 Samsung Electronics Co., Ltd.
- * http://www.samsung.com/
- *
- * EXYNOS - Memory map definitions
- *
- * 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.
-*/
-
-#ifndef __ASM_ARCH_MAP_H
-#define __ASM_ARCH_MAP_H __FILE__
-
-#include <plat/map-base.h>
-
-#include <plat/map-s5p.h>
-
-#define EXYNOS_PA_CHIPID 0x10000000
-
-#endif /* __ASM_ARCH_MAP_H */
diff --git a/arch/arm/mach-exynos/platsmp.c b/arch/arm/mach-exynos/platsmp.c
index 553d0d9..884e885 100644
--- a/arch/arm/mach-exynos/platsmp.c
+++ b/arch/arm/mach-exynos/platsmp.c
@@ -19,6 +19,7 @@
#include <linux/smp.h>
#include <linux/io.h>
#include <linux/of_address.h>
+#include <linux/sys_soc.h>
#include <linux/soc/samsung/exynos-regs-pmu.h>
#include <asm/cacheflush.h>
@@ -27,8 +28,6 @@
#include <asm/smp_scu.h>
#include <asm/firmware.h>
-#include <mach/map.h>
-
#include "common.h"
extern void exynos4_secondary_startup(void);
@@ -93,7 +92,8 @@ void exynos_cpu_power_down(int cpu)
{
u32 core_conf;
- if (cpu == 0 && (soc_is_exynos5420() || soc_is_exynos5800())) {
+ if (cpu == 0 && (of_machine_is_compatible("samsung,exynos5420") ||
+ of_machine_is_compatible("samsung,exynos5800"))) {
/*
* Bypass power down for CPU0 during suspend. Check for
* the SYS_PWR_REG value to decide if we are suspending
@@ -120,7 +120,7 @@ void exynos_cpu_power_up(int cpu)
{
u32 core_conf = S5P_CORE_LOCAL_PWR_EN;
- if (soc_is_exynos3250())
+ if (of_machine_is_compatible("samsung,exynos3250"))
core_conf |= S5P_CORE_AUTOWAKEUP_EN;
pmu_raw_writel(core_conf,
@@ -168,9 +168,14 @@ int exynos_cluster_power_state(int cluster)
S5P_CORE_LOCAL_PWR_EN);
}
+static struct soc_device_attribute exynos4210_rev11[] = {
+ { .soc_id = "EXYNOS4210", .revision = "11", },
+ { },
+};
+
static void __iomem *cpu_boot_reg_base(void)
{
- if (soc_is_exynos4210() && samsung_rev() == EXYNOS4210_REV_1_1)
+ if (soc_device_match(exynos4210_rev11))
return pmu_base_addr + S5P_INFORM5;
return sysram_base_addr;
}
@@ -182,9 +187,10 @@ static inline void __iomem *cpu_boot_reg(int cpu)
boot_reg = cpu_boot_reg_base();
if (!boot_reg)
return IOMEM_ERR_PTR(-ENODEV);
- if (soc_is_exynos4412())
+ if (of_machine_is_compatible("samsung,exynos4412"))
boot_reg += 4*cpu;
- else if (soc_is_exynos5420() || soc_is_exynos5800())
+ else if (of_machine_is_compatible("samsung,exynos5420") ||
+ of_machine_is_compatible("samsung,exynos5800"))
boot_reg += 4;
return boot_reg;
}
@@ -356,7 +362,7 @@ static int exynos_boot_secondary(unsigned int cpu, struct task_struct *idle)
call_firmware_op(cpu_boot, core_id);
- if (soc_is_exynos3250())
+ if (of_machine_is_compatible("samsung,exynos3250"))
dsb_sev();
else
arch_send_wakeup_ipi_mask(cpumask_of(cpu));
diff --git a/arch/arm/mach-exynos/pm.c b/arch/arm/mach-exynos/pm.c
index 60e6827..430b3e2 100644
--- a/arch/arm/mach-exynos/pm.c
+++ b/arch/arm/mach-exynos/pm.c
@@ -19,6 +19,7 @@
#include <linux/io.h>
#include <linux/of.h>
#include <linux/of_address.h>
+#include <linux/sys_soc.h>
#include <linux/soc/samsung/exynos-regs-pmu.h>
#include <linux/soc/samsung/exynos-pmu.h>
@@ -29,20 +30,30 @@
#include "common.h"
+static struct soc_device_attribute exynos4210_rev11[] = {
+ { .soc_id = "EXYNOS4210", .revision = "11", },
+ { },
+};
+
+static struct soc_device_attribute exynos4210_rev10[] = {
+ { .soc_id = "EXYNOS4210", .revision = "10", },
+ { },
+};
+
static inline void __iomem *exynos_boot_vector_addr(void)
{
- if (samsung_rev() == EXYNOS4210_REV_1_1)
+ if (soc_device_match(exynos4210_rev11))
return pmu_base_addr + S5P_INFORM7;
- else if (samsung_rev() == EXYNOS4210_REV_1_0)
+ else if (soc_device_match(exynos4210_rev10))
return sysram_base_addr + 0x24;
return pmu_base_addr + S5P_INFORM0;
}
static inline void __iomem *exynos_boot_vector_flag(void)
{
- if (samsung_rev() == EXYNOS4210_REV_1_1)
+ if (soc_device_match(exynos4210_rev11))
return pmu_base_addr + S5P_INFORM6;
- else if (samsung_rev() == EXYNOS4210_REV_1_0)
+ else if (soc_device_match(exynos4210_rev10))
return sysram_base_addr + 0x20;
return pmu_base_addr + S5P_INFORM1;
}
@@ -122,11 +133,13 @@ int exynos_pm_central_resume(void)
}
/* Ext-GIC nIRQ/nFIQ is the only wakeup source in AFTR */
-static void exynos_set_wakeupmask(long mask)
+static void exynos_set_wakeupmask(void)
{
- pmu_raw_writel(mask, S5P_WAKEUP_MASK);
- if (soc_is_exynos3250())
+ if (of_machine_is_compatible("samsung,exynos3250")) {
+ pmu_raw_writel(0x40003ffe, S5P_WAKEUP_MASK);
pmu_raw_writel(0x0, S5P_WAKEUP_MASK2);
+ } else
+ pmu_raw_writel(0x0000ff3e, S5P_WAKEUP_MASK);
}
static void exynos_cpu_set_boot_vector(long flags)
@@ -140,7 +153,7 @@ static int exynos_aftr_finisher(unsigned long flags)
{
int ret;
- exynos_set_wakeupmask(soc_is_exynos3250() ? 0x40003ffe : 0x0000ff3e);
+ exynos_set_wakeupmask();
/* Set value of power down register for aftr mode */
exynos_sys_powerdown_conf(SYS_AFTR);
@@ -163,7 +176,7 @@ void exynos_enter_aftr(void)
cpu_pm_enter();
- if (soc_is_exynos3250())
+ if (of_machine_is_compatible("samsung,exynos3250"))
exynos_set_boot_flag(cpuid, C2_STATE);
exynos_pm_central_suspend();
@@ -192,7 +205,7 @@ void exynos_enter_aftr(void)
exynos_pm_central_resume();
- if (soc_is_exynos3250())
+ if (of_machine_is_compatible("samsung,exynos3250"))
exynos_clear_boot_flag(cpuid, C2_STATE);
cpu_pm_exit();
@@ -263,7 +276,7 @@ abort:
while (exynos_cpu_power_state(1) != S5P_CORE_LOCAL_PWR_EN)
cpu_relax();
- if (soc_is_exynos3250()) {
+ if (of_machine_is_compatible("samsung,exynos3250")) {
while (!pmu_raw_readl(S5P_PMU_SPARE2) &&
!atomic_read(&cpu1_wakeup))
cpu_relax();
@@ -285,7 +298,7 @@ abort:
call_firmware_op(cpu_boot, 1);
- if (soc_is_exynos3250())
+ if (of_machine_is_compatible("samsung,exynos3250"))
dsb_sev();
else
arch_send_wakeup_ipi_mask(cpumask_of(1));
@@ -297,7 +310,7 @@ fail:
static int exynos_wfi_finisher(unsigned long flags)
{
- if (soc_is_exynos3250())
+ if (of_machine_is_compatible("samsung,exynos3250"))
flush_cache_all();
cpu_do_idle();
@@ -319,7 +332,7 @@ static int exynos_cpu1_powerdown(void)
*/
exynos_cpu_power_down(1);
- if (soc_is_exynos3250())
+ if (of_machine_is_compatible("samsung,exynos3250"))
pmu_raw_writel(0, S5P_PMU_SPARE2);
ret = cpu_suspend(0, exynos_wfi_finisher);
diff --git a/arch/arm/plat-samsung/cpu.c b/arch/arm/plat-samsung/cpu.c
index a107b3a..e58f0f6 100644
--- a/arch/arm/plat-samsung/cpu.c
+++ b/arch/arm/plat-samsung/cpu.c
@@ -21,12 +21,6 @@
unsigned long samsung_cpu_id;
static unsigned int samsung_cpu_rev;
-unsigned int samsung_rev(void)
-{
- return samsung_cpu_rev;
-}
-EXPORT_SYMBOL(samsung_rev);
-
void __init s3c64xx_init_cpu(void)
{
samsung_cpu_id = readl_relaxed(S3C_VA_SYS + 0x118);
@@ -43,11 +37,3 @@ void __init s3c64xx_init_cpu(void)
pr_info("Samsung CPU ID: 0x%08lx\n", samsung_cpu_id);
}
-
-void __init s5p_init_cpu(const void __iomem *cpuid_addr)
-{
- samsung_cpu_id = readl_relaxed(cpuid_addr);
- samsung_cpu_rev = samsung_cpu_id & 0xFF;
-
- pr_info("Samsung CPU ID: 0x%08lx\n", samsung_cpu_id);
-}
diff --git a/arch/arm/plat-samsung/include/plat/cpu.h b/arch/arm/plat-samsung/include/plat/cpu.h
index b7b702a..913c176 100644
--- a/arch/arm/plat-samsung/include/plat/cpu.h
+++ b/arch/arm/plat-samsung/include/plat/cpu.h
@@ -115,8 +115,6 @@ extern void s3c24xx_init_io(struct map_desc *mach_desc, int size);
extern void s3c64xx_init_cpu(void);
extern void s5p_init_cpu(const void __iomem *cpuid_addr);
-extern unsigned int samsung_rev(void);
-
extern void s3c24xx_init_uarts(struct s3c2410_uartcfg *cfg, int no);
extern void s3c24xx_init_clocks(int xtal);
diff --git a/arch/arm/plat-samsung/include/plat/map-s5p.h b/arch/arm/plat-samsung/include/plat/map-s5p.h
index 512ed1f..d6853f1 100644
--- a/arch/arm/plat-samsung/include/plat/map-s5p.h
+++ b/arch/arm/plat-samsung/include/plat/map-s5p.h
@@ -13,8 +13,6 @@
#ifndef __ASM_PLAT_MAP_S5P_H
#define __ASM_PLAT_MAP_S5P_H __FILE__
-#define S5P_VA_CHIPID S3C_ADDR(0x02000000)
-
#define VA_VIC(x) (S3C_VA_IRQ + ((x) * 0x10000))
#define VA_VIC0 VA_VIC(0)
#define VA_VIC1 VA_VIC(1)
--
2.7.4
^ permalink raw reply related
* Low network throughput on i.MX28
From: Jörg Krause @ 2016-11-05 12:06 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <963717394.159124.9867e3e7-5710-4844-a098-6f44bd852a6d.open-xchange@email.1und1.de>
Hello Vinod,
as recommanded by Stefan Wahren I'm turning on you about this issue.
Please see below...
On Sat, 2016-11-05 at 12:33 +0100, Stefan Wahren wrote:
> Hi J?rg,
>
> > J?rg Krause <joerg.krause@embedded.rocks> hat am 4. November 2016
> > um 23:42
> > geschrieben:
> >
> >
> > Hi Stefan,
> >
> > sorry, I forget the link in the previous mail.
> >
> > On Fri, 2016-11-04 at 20:30 +0100, Stefan Wahren wrote:
> > > Hi J?rg,
> > >
> > > > J?rg Krause <joerg.krause@embedded.rocks> hat am 4. November
> > > > 2016
> > > > um 19:44
> > > > geschrieben:
> > > >
> > > >
> > > > Hi Shawn,
> > > >
> > > > On Wed, 2016-11-02 at 09:24 +0100, Stefan Wahren wrote:
> > > > > Am 02.11.2016 um 09:14 schrieb J?rg Krause:
> > > > > > On Sat, 2016-10-29 at 11:08 +0200, Stefan Wahren wrote:
> > > > > > > > J?rg Krause <joerg.krause@embedded.rocks> hat am 29.
> > > > > > > > Oktober
> > > > > > > > 2016
> > > > > > > > um 01:07
> > > > > > > > geschrieben:
> > > > > > > >
> > > > > > > >
> > > > > > > > You mentioned [1] an optimization in the Freescale
> > > > > > > > vendor
> > > > > > > > Linux
> > > > > > > > kernel
> > > > > > > > [2]. I would really like to see this optimization in
> > > > > > > > the
> > > > > > > > mainline
> > > > > > > > kernel.
> > > > > > > >
> > > > > > > > Did you ever tried to port this code from Freescale to
> > > > > > > > mainline?
> > > > > > >
> > > > > > > Yes, i tried once but i was frustrated soon because of
> > > > > > > the
> > > > > > > lot of
> > > > > > > required
> > > > > > > changes and resulting issues.
> > > > > >
> > > > > > I got the PIO mode working for the mxs-mmc driver. For this
> > > > > > I
> > > > > > ported
> > > > > > the PIO code from the vendor kernel and removed the usage
> > > > > > of
> > > > > > the
> > > > > > DMA
> > > > > > engine entirely.
> > > > >
> > > > > Good job
> > > > >
> > > > > >
> > > > > > Testing network bandwidth with iperf, I get about
> > > > > > ~10Mbit/sec
> > > > > > with
> > > > > > PIO
> > > > > > mode compared to ~6.5Mbit/sec with DMA mode for UDP and
> > > > > > about
> > > > > > ~6.5Mbit/sec compared to ~4.5Mbit/sec with DMA mode for
> > > > > > TCP.
> > > > >
> > > > > And how about MMC / sd card performance?
> > > >
> > > > I noticed poor performance with the i.MX28 MMC and/or DMA
> > > > driver
> > > > using
> > > > the mainline kernel compared to the vendor Freescale kernel
> > > > 2.6.35.
> > > > I've seen that hou have added the drivers to mainline some
> > > > years
> > > > ago.
> > > >
> > > > My custom i.MX28 board has a wifi chip attached to the SSP2
> > > > interface.
> > > > Comparing the bandwith with iperf I get >20Mbits/sec on the
> > > > vendor
> > > > kernel and <5Mbits/sec on the mainline kernel.
> > >
> > > there is one thing about the clock handling. I noticed that the
> > > Vendor Kernel
> > > round up the clock frequency and the Mainline Kernel round down
> > > the
> > > clock
> > > frequency [1]. So don't trust the clock ratings from DT / board
> > > code.
> > > Better
> > > verify the register settings or check it with an osci.
> > >
> > > [1] - http://www.spinics.net/lists/linux-mmc/msg09132.html
> >
> > I checked the clock rate setting by reading the register 0x80014070
> > (HW_SSP2_TIMING). CLOCK_DIVIDE is 0x2 and CLOCK_RATE is 0x0. As SSP
> > CLK
> > is 96MHz this makes a clock rate of 48MHz.
> >
> > There was a discussion on the mailing list [1] about that tasklets
> > might be slow.
> >
> > [1] http://lists.infradead.org/pipermail/linux-arm-kernel/2011-Febr
> > uary
> > /043395.html
>
> if i unterstand it right the tasklet is not the problem, but the
> design of the
> MXS DMA driver. Please refer to the chapter "General Design Notes" to
> the
> documentation of the DMA provider [2].
> I think the MXS DMA driver is affected. Maybe you should ask Vinod
> Koul about
> this.
>
> [2] - https://www.kernel.org/doc/Documentation/dmaengine/provider.txt
@ Vinod
In short, I noticed poor performance in the SSP2 (MMC/SD/SDIO)
interface on a custom i.MX28 board with a wifi chip attached. Comparing
the bandwith with iperf I get >20Mbits/sec on the vendor kernel and
<5Mbits/sec on the mainline kernel. I am trying to investigate what the
bottleneck is.
@ Stefan, all
My understanding is that the tasklet in this case is responsible for
reading the response registers of the DMA controller and return the
response to the MMC host driver.
The vendor kernel does this in the interrupt routine of mxs-mmc by
issueing a complete whereas the mainline kernel does this in the
interrupt routine in mxs-dma by scheduling the tasklet.
To check if this makes any difference I replaced the tasklet() usage
with using the complete() infrastructure. For this I hacked the DMA
engine and the MXS DMA driver. However, the performance stays the same.
So, if I understand correctly, this is not an issue here, right? So if
not the tasklet, what do you suspect?
J?rg
^ permalink raw reply
* Low network throughput on i.MX28
From: Koul, Vinod @ 2016-11-05 12:39 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478347610.353.2.camel@embedded.rocks>
On Sat, 2016-11-05 at 13:06 +0100, J?rg Krause wrote:
> @ Vinod
> In short, I noticed poor performance in the SSP2 (MMC/SD/SDIO)
> interface on a custom i.MX28 board with a wifi chip attached.
> Comparing
> the bandwith with iperf I get >20Mbits/sec on the vendor kernel and
> <5Mbits/sec on the mainline kernel. I am trying to investigate what
> the
> bottleneck is.
is this imx-dma or imx-sdma..
>
> @ Stefan, all
> My understanding is that the tasklet in this case is responsible for
> reading the response registers of the DMA controller and return the
> response to the MMC host driver.
>
> The vendor kernel does this in the interrupt routine of mxs-mmc by
> issueing a complete whereas the mainline kernel does this in the
> interrupt routine in mxs-dma by scheduling the tasklet.
Is vendor kernel using dmaengine APIs or not?
Okay, if we talk about getting best performance, I always advise folks
to issue next transaction in the interrupt routine and then do
descriptor management and callback in tasklet.
Some drivers do that correctly but some don't..
Tasklet can be an issue but only if there is a huge scheduling delay for
the tasklet. You can check using tracing tools and confirm.
>
> To check if this makes any difference I replaced the tasklet() usage
> with using the complete() infrastructure. For this I hacked the DMA
> engine and the MXS DMA driver. However, the performance stays the
> same.
>
> So, if I understand correctly, this is not an issue here, right? So if
> not the tasklet, what do you suspect?
--
~Vinod
^ permalink raw reply
* Low network throughput on i.MX28
From: Jörg Krause @ 2016-11-05 12:47 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478349578.3405.5.camel@intel.com>
On Sat, 2016-11-05 at 12:39 +0000, Koul, Vinod wrote:
> On Sat, 2016-11-05 at 13:06 +0100, J?rg Krause wrote:
> > @ Vinod
> > In short, I noticed poor performance in the SSP2 (MMC/SD/SDIO)
> > interface on a custom i.MX28 board with a wifi chip attached.
> > Comparing
> > the bandwith with iperf I get >20Mbits/sec on the vendor kernel and
> > <5Mbits/sec on the mainline kernel. I am trying to investigate what
> > the
> > bottleneck is.
>
> is this imx-dma or imx-sdma..
Its' mxs-dma.
>
> >
> > @ Stefan, all
> > My understanding is that the tasklet in this case is responsible
> > for
> > reading the response registers of the DMA controller and return the
> > response to the MMC host driver.
> >
> > The vendor kernel does this in the interrupt routine of mxs-mmc by
> > issueing a complete whereas the mainline kernel does this in the
> > interrupt routine in mxs-dma by scheduling the tasklet.
>
> Is vendor kernel using dmaengine APIs or not?
No. It's using a custom dmaengine.
>
> Okay, if we talk about getting best performance, I always advise
> folks
> to issue next transaction in the interrupt routine and then do
> descriptor management and callback in tasklet.
>
> Some drivers do that correctly but some don't..
Do you have an example for a driver doing it correctly?
> Tasklet can be an issue but only if there is a huge scheduling delay
> for
> the tasklet. You can check using tracing tools and confirm.
Don't think the tasklets is an issue here as I replaced the tasklets in
the dmaengine API by completion (which the vendor kernel uses) and
there are no performance benefits. However, I am not a Linux kernel
developer...
J?rg
^ permalink raw reply
* Low network throughput on i.MX28
From: Fabio Estevam @ 2016-11-05 12:48 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478349578.3405.5.camel@intel.com>
Hi Vinod,
On Sat, Nov 5, 2016 at 10:39 AM, Koul, Vinod <vinod.koul@intel.com> wrote:
> On Sat, 2016-11-05 at 13:06 +0100, J?rg Krause wrote:
>> @ Vinod
>> In short, I noticed poor performance in the SSP2 (MMC/SD/SDIO)
>> interface on a custom i.MX28 board with a wifi chip attached.
>> Comparing
>> the bandwith with iperf I get >20Mbits/sec on the vendor kernel and
>> <5Mbits/sec on the mainline kernel. I am trying to investigate what
>> the
>> bottleneck is.
>
> is this imx-dma or imx-sdma..
This is drivers/dma/mxs-dma.c, thanks.
^ permalink raw reply
* [Bug] ARM: mxs: STI: console can't wake up from freeze
From: Daniel Lezcano @ 2016-11-05 13:00 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <122295766.159553.9867e3e7-5710-4844-a098-6f44bd852a6d.open-xchange@email.1und1.de>
On Sat, Nov 05, 2016 at 01:01:26PM +0100, Stefan Wahren wrote:
[?... ]
> > > CONFIG_SUSPEND=y
> > > CONFIG_SUSPEND_FREEZER=y
> > > CONFIG_PM=y
> > > CONFIG_CPU_IDLE is not set
> >
> > hmmm, why not have CONFIG_CPU_IDLE set?
>
> I'm using mxs_defconfig which doesn't select the ARM CPU idle. Is this
> necessary?
Very likely :) suspend_freezer and cpuidle are tied together.
Moreover, while reading the code, it appears without CONFIG_CPU_IDLE the
function cpuidle_idle_call is constantly failing, ending up by having the
cpu looping again and again in the idle loop. The function stubs return
-ENODEV in the cpuidle's header when CONFIG_CPU_IDLE is not set.
If I'm not wrong the traces should show the cpu actually does never go
to suspend. As soon as it enters the state, it should exit immediately
without an interrupt event.
Probably there is an inconsistent configuration leaving the kernel with
a strange wakeup condition or a race with the short the suspend/resume
cycle delay.
First test would be to enable CONFIG_CPU_IDLE.
If it is confirmed, it would be nice to feed bugzilla.
^ permalink raw reply
* Low network throughput on i.MX28
From: Jörg Krause @ 2016-11-05 13:14 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478349578.3405.5.camel@intel.com>
On Sat, 2016-11-05 at 12:39 +0000, Koul, Vinod wrote:
> On Sat, 2016-11-05 at 13:06 +0100, J?rg Krause wrote:
> > @ Vinod
> > In short, I noticed poor performance in the SSP2 (MMC/SD/SDIO)
> > interface on a custom i.MX28 board with a wifi chip attached.
> > Comparing
> > the bandwith with iperf I get >20Mbits/sec on the vendor kernel and
> > <5Mbits/sec on the mainline kernel. I am trying to investigate what
> > the
> > bottleneck is.
>
> is this imx-dma or imx-sdma..
>
> >
> > @ Stefan, all
> > My understanding is that the tasklet in this case is responsible
> > for
> > reading the response registers of the DMA controller and return the
> > response to the MMC host driver.
> >
> > The vendor kernel does this in the interrupt routine of mxs-mmc by
> > issueing a complete whereas the mainline kernel does this in the
> > interrupt routine in mxs-dma by scheduling the tasklet.
>
> Is vendor kernel using dmaengine APIs or not?
It's this engine [1].
[1] http://git.freescale.com/git/cgit.cgi/imx/linux-2.6-imx.git/tree/ar
ch/arm/plat-mxs/dmaengine.c?h=imx_2.6.35_1.1.0
^ permalink raw reply
* [PATCH 1/1] pinctrl: st: st_pinconf_dbg_show wrong format string
From: Heinrich Schuchardt @ 2016-11-05 14:25 UTC (permalink / raw)
To: linux-arm-kernel
function is defined as unsigned int.
So we need %u to print it.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
---
drivers/pinctrl/pinctrl-st.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/pinctrl/pinctrl-st.c b/drivers/pinctrl/pinctrl-st.c
index b7bb371..8308f15 100644
--- a/drivers/pinctrl/pinctrl-st.c
+++ b/drivers/pinctrl/pinctrl-st.c
@@ -1006,7 +1006,7 @@ static void st_pinconf_dbg_show(struct pinctrl_dev *pctldev,
function = st_pctl_get_pin_function(pc, offset);
if (function)
- snprintf(f, 10, "Alt Fn %d", function);
+ snprintf(f, 10, "Alt Fn %u", function);
else
snprintf(f, 5, "GPIO");
--
2.10.1
^ permalink raw reply related
* [PATCH 1/1] pinctrl: st: st_pctl_dt_parse_groups simplify expression
From: Heinrich Schuchardt @ 2016-11-05 14:35 UTC (permalink / raw)
To: linux-arm-kernel
for_each_property_of_node(pins, pp) checks that pp is not NULL.
So there is no need to check it inside the loop.
Signed-off-by: Heinrich Schuchardt <xypron.glpk@gmx.de>
---
drivers/pinctrl/pinctrl-st.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/pinctrl/pinctrl-st.c b/drivers/pinctrl/pinctrl-st.c
index 8308f15..676efcc 100644
--- a/drivers/pinctrl/pinctrl-st.c
+++ b/drivers/pinctrl/pinctrl-st.c
@@ -1181,7 +1181,7 @@ static int st_pctl_dt_parse_groups(struct device_node *np,
if (!strcmp(pp->name, "name"))
continue;
- if (pp && (pp->length/sizeof(__be32)) >= OF_GPIO_ARGS_MIN) {
+ if (pp->length / sizeof(__be32) >= OF_GPIO_ARGS_MIN) {
npins++;
} else {
pr_warn("Invalid st,pins in %s node\n", np->name);
--
2.10.1
^ permalink raw reply related
* [PATCH 1/2] arm64: dts: add USB1-related nodes of Allwinner A64
From: Icenowy Zheng @ 2016-11-05 14:38 UTC (permalink / raw)
To: linux-arm-kernel
Allwinner A64 have two HCI USB controllers, a OTG controller and a USB
PHY device which have two ports. One of the port is wired to both a HCI
USB controller and the OTG controller, which is currently not supported.
The another one is only wired to a HCI controller, and the device node of
OHCI/EHCI controller of the port can be added now.
Also the A64 USB PHY device node is also added for the HCI controllers to
work.
Signed-off-by: Icenowy Zheng <icenowy@aosc.xyz>
---
arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi | 50 +++++++++++++++++++++++++++
1 file changed, 50 insertions(+)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi
index 3d70be3..c2b6dc8 100644
--- a/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi
+++ b/arch/arm64/boot/dts/allwinner/sun50i-a64.dtsi
@@ -259,5 +259,55 @@
interrupts = <GIC_SPI 40 IRQ_TYPE_LEVEL_HIGH>,
<GIC_SPI 41 IRQ_TYPE_LEVEL_HIGH>;
};
+
+ usbphy: phy at 01c19400 {
+ compatible = "allwinner,sun50i-a64-usb-phy";
+ reg = <0x01c19400 0x14>,
+ <0x01c1b800 0x4>;
+ reg-names = "phy_ctrl",
+ "pmu1";
+ clocks = <&ccu CLK_USB_PHY0>,
+ <&ccu CLK_USB_PHY1>;
+ clock-names = "usb0_phy",
+ "usb1_phy";
+ resets = <&ccu RST_USB_PHY0>,
+ <&ccu RST_USB_PHY1>;
+ reset-names = "usb0_reset",
+ "usb1_reset";
+ status = "disabled";
+ #phy-cells = <1>;
+ };
+
+ ohci1: usb at 01c1a400 {
+ compatible = "allwinner,sun50i-a64-ohci", "generic-ohci";
+ reg = <0x01c1b400 0x100>;
+ interrupts = <GIC_SPI 75 IRQ_TYPE_LEVEL_HIGH>;
+ /*
+ * According to the user manual, OHCI1 USB clock
+ * depends on OHCI0 clock.
+ */
+ clocks = <&ccu CLK_BUS_OHCI1>,
+ <&ccu CLK_USB_OHCI0>,
+ <&ccu CLK_USB_OHCI1>;
+ resets = <&ccu RST_BUS_OHCI1>;
+ phys = <&usbphy 1>;
+ phy-names = "usb";
+ status = "disabled";
+ };
+
+ ehci1: usb at 01c1a000 {
+ compatible = "allwinner,sun50i-a64-ehci", "generic-ehci";
+ reg = <0x01c1b000 0x100>;
+ interrupts = <GIC_SPI 74 IRQ_TYPE_LEVEL_HIGH>;
+ clocks = <&ccu CLK_BUS_OHCI1>,
+ <&ccu CLK_BUS_EHCI1>,
+ <&ccu CLK_USB_OHCI0>,
+ <&ccu CLK_USB_OHCI1>;
+ resets = <&ccu RST_BUS_OHCI1>,
+ <&ccu RST_BUS_EHCI1>;
+ phys = <&usbphy 1>;
+ phy-names = "usb";
+ status = "disabled";
+ };
};
};
--
2.10.1
^ permalink raw reply related
* [PATCH 2/2] arm64: dts: sunxi: enable EHCI1, OHCI1 and USB PHY nodes in Pine64
From: Icenowy Zheng @ 2016-11-05 14:38 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20161105143830.10099-1-icenowy@aosc.xyz>
Pine64 have two USB Type-A ports, which are wired to the two ports of
A64 USB PHY, and the lower port is the EHCI/OHCI1 port.
Enable the necessary nodes to enable the lower USB port to work.
Signed-off-by: Icenowy Zheng <icenowy@aosc.xyz>
---
arch/arm64/boot/dts/allwinner/sun50i-a64-pine64.dts | 12 ++++++++++++
1 file changed, 12 insertions(+)
diff --git a/arch/arm64/boot/dts/allwinner/sun50i-a64-pine64.dts b/arch/arm64/boot/dts/allwinner/sun50i-a64-pine64.dts
index 4709590..d836995 100644
--- a/arch/arm64/boot/dts/allwinner/sun50i-a64-pine64.dts
+++ b/arch/arm64/boot/dts/allwinner/sun50i-a64-pine64.dts
@@ -72,3 +72,15 @@
&i2c1_pins {
bias-pull-up;
};
+
+&usbphy {
+ status = "okay";
+};
+
+&ehci1 {
+ status = "okay";
+};
+
+&ohci1 {
+ status = "okay";
+};
--
2.10.1
^ permalink raw reply related
* [GIT PULL] ARM: dts: uniphier: UniPhier DT updates for v4.10
From: Masahiro Yamada @ 2016-11-05 15:18 UTC (permalink / raw)
To: linux-arm-kernel
Hi Arnd, Olof,
Here are UniPhier DT (32bit) updates for the v4.10 merge window.
Please pull!
(This time, I based this pull-request on -rc3 instead of -rc1
because I want to make sure to have no conflict
between fixes that had already been pulled-in and
new development commits in this series.)
The following changes since commit a909d3e636995ba7c349e2ca5dbb528154d4ac30:
Linux 4.9-rc3 (2016-10-29 13:52:02 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-uniphier.git
tags/uniphier-dt-v4.10
for you to fetch changes up to 13b4a6190b25c7bf69239bbc212d256656ada7b9:
ARM: dts: uniphier: make compatible of syscon nodes SoC-specific
(2016-11-05 23:33:36 +0900)
----------------------------------------------------------------
UniPhier ARM SoC DT updates for v4.10
- Add OPP tables to support generic cpufreq driver
- Use more clocks/resets properties
- Misc fixes and cleanups
----------------------------------------------------------------
Masahiro Yamada (7):
ARM: dts: uniphier: increase register region size of sysctrl node
ARM: dts: uniphier: add CPU clocks and OPP table for Pro5 SoC
ARM: dts: uniphier: add CPU clocks and OPP table for PXs2 SoC
ARM: dts: uniphier: make 32bit SoC DTSI linear
ARM: dts: uniphier: remove redundant serial fifo-size properties
ARM: dts: uniphier: add clocks/resets to EHCI nodes of sLD3 SoC
ARM: dts: uniphier: make compatible of syscon nodes SoC-specific
arch/arm/boot/dts/uniphier-common32.dtsi | 199 --------------
arch/arm/boot/dts/uniphier-ld4.dtsi | 358 +++++++++++++++++--------
arch/arm/boot/dts/uniphier-pro4.dtsi | 378 +++++++++++++++++---------
arch/arm/boot/dts/uniphier-pro5.dtsi | 432 ++++++++++++++++++++++--------
arch/arm/boot/dts/uniphier-pxs2.dtsi | 399 +++++++++++++++++++--------
arch/arm/boot/dts/uniphier-sld3.dtsi | 21 +-
arch/arm/boot/dts/uniphier-sld8.dtsi | 359 +++++++++++++++++--------
7 files changed, 1344 insertions(+), 802 deletions(-)
delete mode 100644 arch/arm/boot/dts/uniphier-common32.dtsi
--
Best Regards
Masahiro Yamada
^ permalink raw reply
* [GIT PULL] arm64: dts: uniphier: UniPhier DT updates (64bit) for v4.10
From: Masahiro Yamada @ 2016-11-05 15:25 UTC (permalink / raw)
To: linux-arm-kernel
Hi Arnd, Olof,
Here are UniPhier DT (64bit) updates for the v4.10 merge window.
Please pull!
The following changes since commit a909d3e636995ba7c349e2ca5dbb528154d4ac30:
Linux 4.9-rc3 (2016-10-29 13:52:02 -0700)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-uniphier.git
tags/uniphier-dt64-v4.10
for you to fetch changes up to fb28cef06a9065748fbc6949dd7fca3c320c8e0d:
arm64: dts: uniphier: make compatible of syscon nodes SoC-specific
(2016-11-05 23:30:11 +0900)
----------------------------------------------------------------
UniPhier ARM64 SoC DT updates for v4.10
- Switch CPU enable-method from spin-table to PSCI
- Add OPP tables to support generic cpufreq driver
- Misc fixes
----------------------------------------------------------------
Masahiro Yamada (5):
arm64: dts: uniphier: switch over to PSCI enable method
arm64: dts: uniphier: increase register region size of sysctrl node
arm64: dts: uniphier: add CPU clock and OPP table for LD11 SoC
arm64: dts: uniphier: add CPU clocks and OPP tables for LD20 SoC
arm64: dts: uniphier: make compatible of syscon nodes SoC-specific
arch/arm64/boot/dts/socionext/uniphier-ld11.dtsi | 59 ++++++++++--
arch/arm64/boot/dts/socionext/uniphier-ld20.dtsi | 111 +++++++++++++++++++---
2 files changed, 148 insertions(+), 22 deletions(-)
--
Best Regards
Masahiro Yamada
^ permalink raw reply
* [Bug] ARM: mxs: STI: console can't wake up from freeze
From: Stefan Wahren @ 2016-11-05 15:28 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <20161105130034.GG15759@mai>
Hi Daniel,
> Daniel Lezcano <daniel.lezcano@linaro.org> hat am 5. November 2016 um 14:00
> geschrieben:
>
>
> On Sat, Nov 05, 2016 at 01:01:26PM +0100, Stefan Wahren wrote:
>
> [?... ]
>
> > > > CONFIG_SUSPEND=y
> > > > CONFIG_SUSPEND_FREEZER=y
> > > > CONFIG_PM=y
> > > > CONFIG_CPU_IDLE is not set
> > >
> > > hmmm, why not have CONFIG_CPU_IDLE set?
> >
> > I'm using mxs_defconfig which doesn't select the ARM CPU idle. Is this
> > necessary?
>
> Very likely :) suspend_freezer and cpuidle are tied together.
>
> Moreover, while reading the code, it appears without CONFIG_CPU_IDLE the
> function cpuidle_idle_call is constantly failing, ending up by having the
> cpu looping again and again in the idle loop. The function stubs return
> -ENODEV in the cpuidle's header when CONFIG_CPU_IDLE is not set.
>
> If I'm not wrong the traces should show the cpu actually does never go
> to suspend. As soon as it enters the state, it should exit immediately
> without an interrupt event.
>
> Probably there is an inconsistent configuration leaving the kernel with
> a strange wakeup condition or a race with the short the suspend/resume
> cycle delay.
>
> First test would be to enable CONFIG_CPU_IDLE.
>
> If it is confirmed, it would be nice to feed bugzilla.
first i tried with CONFIG_CPU_IDLE=Y and then additionally with
CONFIG_ARM_CPUIDLE=y.
In both cases the issue is still reproducible.
As i wrote in my email before, i added a pr_info() into freeze_wake. But i never
see the output of this message. So i assume freeze_wake is never called. Again,
how could this happen?
^ permalink raw reply
* [PATCH 1/4] ARM: EXYNOS: Remove smp_init_cpus hook from platsmp.c
From: Krzysztof Kozlowski @ 2016-11-05 15:41 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478230764-13748-2-git-send-email-pankaj.dubey@samsung.com>
On Fri, Nov 04, 2016 at 09:09:21AM +0530, Pankaj Dubey wrote:
> We can safely remove exynos_smp_init_cpus() hook from mach-exynos/platsmp.c,
> as all SMP platforms in mach-exynos can rely on DT for CPU core description
> instead of determining number of cores from the SCU.
>
> Signed-off-by: Pankaj Dubey <pankaj.dubey@samsung.com>
> ---
> arch/arm/mach-exynos/platsmp.c | 31 -------------------------------
> 1 file changed, 31 deletions(-)
>
Thanks, applied. Somehow your patchset did not reach linux-samsung-soc's
Parchwork. I don't have a clue why... It is on linux-arm-kernel's
Patchwork.
Best regards,
Krzysztof
^ permalink raw reply
* [PATCH 2/4] ARM: dts: exynos: Add SCU device node to exynos4.dtsi
From: Krzysztof Kozlowski @ 2016-11-05 15:41 UTC (permalink / raw)
To: linux-arm-kernel
In-Reply-To: <1478230764-13748-3-git-send-email-pankaj.dubey@samsung.com>
On Fri, Nov 04, 2016 at 09:09:22AM +0530, Pankaj Dubey wrote:
> Exynos4 like other Cortex-A9 SoC's has a Snoop Control Unit(SCU)
> and its SFR are used during SMP boot and S2R. Add SCU node to the device tree.
>
> Signed-off-by: Pankaj Dubey <pankaj.dubey@samsung.com>
> ---
> arch/arm/boot/dts/exynos4.dtsi | 5 +++++
> 1 file changed, 5 insertions(+)
>
Thanks, applied.
Best regards,
Krzysztof
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox