From: Joel Fernandes <joelf@ti.com>
To: Joachim Eastwood <manabian@gmail.com>
Cc: "linux-omap@vger.kernel.org" <linux-omap@vger.kernel.org>
Subject: Re: L3 custom error from dmtimer.c
Date: Mon, 21 Apr 2014 11:44:45 -0500 [thread overview]
Message-ID: <53554AFD.5070504@ti.com> (raw)
In-Reply-To: <535548FE.3030808@ti.com>
On 04/21/2014 11:36 AM, Joel Fernandes wrote:
> On 04/21/2014 10:57 AM, Joachim Eastwood wrote:
>> On 21 April 2014 17:40, Joel Fernandes <joelf@ti.com> wrote:
>>> On 04/19/2014 05:25 PM, Joachim Eastwood wrote:
>>>> Hello,
>>>>
>>>> Playing around with an old OMAP pwm driver from NeilBrown. I get the
>>>> following warning:
>>>> [ 0.979522] omap-pwm omap-pwm.14: omap_dm_timer_set_load
>>>> [ 0.979553] ------------[ cut here ]------------
>>>> [ 0.979583] WARNING: CPU: 0 PID: 0 at drivers/bus/omap_l3_noc.c:113
>>>> l3_interrupt_handler+0xf4/0x154()
>>>> [ 0.979583] L3 custom error: MASTER:MPU TARGET:L4 PER2
>>>> [ 0.979614] Modules linked in:
>>>> [ 0.979614] CPU: 0 PID: 0 Comm: swapper/0 Not tainted
>>>> 3.15.0-rc1-00370-gd9d79f4b7b7d-dirty #65
>>>> [ 0.979644] [<c0014f48>] (unwind_backtrace) from [<c0011c2c>]
>>>> (show_stack+0x10/0x14)
>>>> [ 0.979675] [<c0011c2c>] (show_stack) from [<c05ad3bc>] (dump_stack+0x84/0x94)
>>>> [ 0.979705] [<c05ad3bc>] (dump_stack) from [<c0036ba8>]
>>>> (warn_slowpath_common+0x70/0x8c)
>>>> [ 0.979705] omap-pwm omap-pwm.14: omap_dm_timer_set_match
>>>> [ 0.979736] [<c0036ba8>] (warn_slowpath_common) from [<c0036bf4>]
>>>> (warn_slowpath_fmt+0x30/0x40)
>>>> [ 0.979736] [<c0036bf4>] (warn_slowpath_fmt) from [<c0286c90>]
>>>> (l3_interrupt_handler+0xf4/0x154)
>>>> [ 0.979766] [<c0286c90>] (l3_interrupt_handler) from [<c0085d1c>]
>>>> (handle_irq_event_percpu+0x54/0x1cc)
>>>> [ 0.979766] [<c0085d1c>] (handle_irq_event_percpu) from [<c0085ed0>]
>>>> (handle_irq_event+0x3c/0x5c)
>>>> [ 0.979797] [<c0085ed0>] (handle_irq_event) from [<c0088ed0>]
>>>> (handle_fasteoi_irq+0xac/0x1a0)
>>>> [ 0.979797] [<c0088ed0>] (handle_fasteoi_irq) from [<c00853fc>]
>>>> (generic_handle_irq+0x2c/0x3c)
>>>> [ 0.979827] [<c00853fc>] (generic_handle_irq) from [<c000eb20>]
>>>> (handle_IRQ+0x40/0x90)
>>>> [ 0.979827] [<c000eb20>] (handle_IRQ) from [<c0008594>]
>>>> (gic_handle_irq+0x2c/0x5c)
>>>> [ 0.979858] [<c0008594>] (gic_handle_irq) from [<c05b4a44>]
>>>> (__irq_svc+0x44/0x58)
>>>> [ 0.979858] Exception stack(0xc0889f58 to 0xc0889fa0)
>>>> [ 0.979858] 9f40: 00000001 00000001
>>>> [ 0.979888] 9f60: 00000000 c0893678 c0888000 c0888000 c08e75a4
>>>> c0890548 c0888000 ee7ffc00
>>>> [ 0.979888] omap-pwm omap-pwm.14: load value: 0xfffd11fe (-192002),
>>>> match value: 0xfffffffe (-2)
>>>> [ 0.979888] 9f80: c08904e0 c05bdcec 00000000 c0889fa0 c007a190
>>>> c000ee48 20000113 ffffffff
>>>> [ 0.979919] [<c05b4a44>] (__irq_svc) from [<c000ee48>] (arch_cpu_idle+0x24/0x30)
>>>> [ 0.979919] [<c000ee48>] (arch_cpu_idle) from [<c0071950>]
>>>> (cpu_startup_entry+0x138/0x204)
>>>> [ 0.979949] [<c0071950>] (cpu_startup_entry) from [<c0822b10>]
>>>> (start_kernel+0x370/0x37c)
>>>> [ 0.979980] [<c0822b10>] (start_kernel) from [<80008074>] (0x80008074)
>>>> [ 0.979980] omap-pwm omap-pwm.14: omap_dm_timer_set_pwm
>>>> [ 0.980010] omap-pwm omap-pwm.14: omap_pwm_enable
>>>> [ 0.980010] omap-pwm omap-pwm.14: omap_dm_timer_start
>>>> [ 0.980010] omap-pwm omap-pwm.14: omap_dm_timer_write_counter
>>>> [ 0.980041] ---[ end trace 5d002a14ec98c2ad ]---
>>>>
>>>> This seems to be caused by the call into omap_dm_timer_set_load.
>>>>
>>>
>>> Are you talking about this?
>>> https://lkml.org/lkml/2012/12/12/51
>>>
>>> It doesn't seem to be in mainline though.
>>
>> No, it isn't. I am trying to revive it and get it upstream.
>>
>>> I'll add a check for the enable/disable, thanks.
>>
>> Note that the warning I got might be cause by the pwm-omap and it's
>> use of dmtimer api. I will investigate further.
>>
>> But either way add a check to enable/disable is a good idea.
>>
>
> Sure, I added the check. The patch depends on a few other pending
> patches so I just zipped all together. It would be great if you could
> apply them and confirm it fixes the problem, then I can add your
> tested-by and include it in the next series.
Lost track that the mailing list was CC'd so included the latest patch
(check for pm_runtime_get_sync) inline below. Dependent patches are at [1].
Regards,
-Joel
[1] https://lkml.org/lkml/2014/4/16/737
---8<----------
>From b5884b869a21ffe5472edc59bd07d58b104284b9 Mon Sep 17 00:00:00 2001
From: Joel Fernandes <joelf@ti.com>
Date: Mon, 21 Apr 2014 11:05:10 -0500
Subject: [PATCH 10/10] ARM: OMAP: dmtimer: Check return of
pm_runtime_get_sync
omap_timer_enable doesn't check return value, this can lead to nasty bus
errors. Add a check and report issues.
Signed-off-by: Joel Fernandes <joelf@ti.com>
---
arch/arm/plat-omap/dmtimer.c | 61
+++++++++++++++++++++++------
arch/arm/plat-omap/include/plat/dmtimer.h | 2 +-
2 files changed, 50 insertions(+), 13 deletions(-)
diff --git a/arch/arm/plat-omap/dmtimer.c b/arch/arm/plat-omap/dmtimer.c
index 1fd30fa..9403d450 100644
--- a/arch/arm/plat-omap/dmtimer.c
+++ b/arch/arm/plat-omap/dmtimer.c
@@ -164,7 +164,9 @@ static int omap_dm_timer_prepare(struct
omap_dm_timer *timer)
}
}
- omap_dm_timer_enable(timer);
+ rc = omap_dm_timer_enable(timer);
+ if(rc)
+ return rc;
if (timer->capability & OMAP_TIMER_NEEDS_RESET) {
rc = omap_dm_timer_reset(timer);
@@ -375,11 +377,15 @@ int omap_dm_timer_free(struct omap_dm_timer *timer)
}
EXPORT_SYMBOL_GPL(omap_dm_timer_free);
-void omap_dm_timer_enable(struct omap_dm_timer *timer)
+int omap_dm_timer_enable(struct omap_dm_timer *timer)
{
- int c;
+ int c, err;
- pm_runtime_get_sync(&timer->pdev->dev);
+ err = pm_runtime_get_sync(&timer->pdev->dev);
+ if (err < 0) {
+ pr_err("%s: timer could not be get_synced.\n", __func__);
+ return err;
+ }
if (!(timer->capability & OMAP_TIMER_ALWON)) {
if (timer->get_context_loss_count) {
@@ -392,6 +398,7 @@ void omap_dm_timer_enable(struct omap_dm_timer *timer)
omap_timer_restore_context(timer);
}
}
+ return 0;
}
EXPORT_SYMBOL_GPL(omap_dm_timer_enable);
@@ -432,11 +439,14 @@ EXPORT_SYMBOL_GPL(omap_dm_timer_trigger);
int omap_dm_timer_start(struct omap_dm_timer *timer)
{
u32 l;
+ int rc;
if (unlikely(!timer))
return -EINVAL;
- omap_dm_timer_enable(timer);
+ rc = omap_dm_timer_enable(timer);
+ if (rc)
+ return rc;
l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
if (!(l & OMAP_TIMER_CTRL_ST)) {
@@ -534,11 +544,15 @@ int omap_dm_timer_set_load(struct omap_dm_timer
*timer, int autoreload,
unsigned int load)
{
u32 mask = ~0, val = 0;
+ int rc;
if (unlikely(!timer))
return -EINVAL;
- omap_dm_timer_enable(timer);
+ rc = omap_dm_timer_enable(timer);
+ if (rc)
+ return rc;
+
if (autoreload)
val |= OMAP_TIMER_CTRL_AR;
else
@@ -560,11 +574,14 @@ int omap_dm_timer_set_load_start(struct
omap_dm_timer *timer, int autoreload,
unsigned int load)
{
u32 l;
+ int rc;
if (unlikely(!timer))
return -EINVAL;
- omap_dm_timer_enable(timer);
+ rc = omap_dm_timer_enable(timer);
+ if (rc)
+ return rc;
l = omap_dm_timer_read_reg(timer, OMAP_TIMER_CTRL_REG);
if (autoreload) {
@@ -588,11 +605,15 @@ int omap_dm_timer_set_match(struct omap_dm_timer
*timer, int enable,
unsigned int match)
{
u32 mask = ~0, val = 0;
+ int rc;
if (unlikely(!timer))
return -EINVAL;
- omap_dm_timer_enable(timer);
+ rc = omap_dm_timer_enable(timer);
+ if (rc)
+ return rc;
+
if (enable)
val |= OMAP_TIMER_CTRL_CE;
else
@@ -613,11 +634,15 @@ int omap_dm_timer_set_pwm(struct omap_dm_timer
*timer, int def_on,
int toggle, int trigger)
{
u32 mask = ~0, val = 0;
+ int rc;
if (unlikely(!timer))
return -EINVAL;
- omap_dm_timer_enable(timer);
+ rc = omap_dm_timer_enable(timer);
+ if (rc)
+ return rc;
+
mask &= ~(OMAP_TIMER_CTRL_GPOCFG | OMAP_TIMER_CTRL_SCPWM |
OMAP_TIMER_CTRL_PT | (0x03 << 10));
if (def_on)
@@ -637,11 +662,15 @@ EXPORT_SYMBOL_GPL(omap_dm_timer_set_pwm);
int omap_dm_timer_set_prescaler(struct omap_dm_timer *timer, int prescaler)
{
u32 mask = ~0, val = 0;
+ int rc;
if (unlikely(!timer))
return -EINVAL;
- omap_dm_timer_enable(timer);
+ rc = omap_dm_timer_enable(timer);
+ if (rc)
+ return rc;
+
mask &= ~(OMAP_TIMER_CTRL_PRE | (0x07 << 2));
if (prescaler >= 0x00 && prescaler <= 0x07) {
val |= OMAP_TIMER_CTRL_PRE;
@@ -659,10 +688,15 @@ EXPORT_SYMBOL_GPL(omap_dm_timer_set_prescaler);
int omap_dm_timer_set_int_enable(struct omap_dm_timer *timer,
unsigned int value)
{
+ int rc;
+
if (unlikely(!timer))
return -EINVAL;
- omap_dm_timer_enable(timer);
+ rc = omap_dm_timer_enable(timer);
+ if (rc)
+ return rc;
+
__omap_dm_timer_int_enable(timer, value);
/* Save the context */
@@ -683,11 +717,14 @@ EXPORT_SYMBOL_GPL(omap_dm_timer_set_int_enable);
int omap_dm_timer_set_int_disable(struct omap_dm_timer *timer, u32 mask)
{
u32 l = mask;
+ int rc;
if (unlikely(!timer))
return -EINVAL;
- omap_dm_timer_enable(timer);
+ rc = omap_dm_timer_enable(timer);
+ if (rc)
+ return rc;
if (timer->revision == 1)
l = __raw_readl(timer->irq_ena) & ~mask;
diff --git a/arch/arm/plat-omap/include/plat/dmtimer.h
b/arch/arm/plat-omap/include/plat/dmtimer.h
index fe3780a..6b6fbd2 100644
--- a/arch/arm/plat-omap/include/plat/dmtimer.h
+++ b/arch/arm/plat-omap/include/plat/dmtimer.h
@@ -130,7 +130,7 @@ struct omap_dm_timer
*omap_dm_timer_request_specific(int timer_id);
struct omap_dm_timer *omap_dm_timer_request_by_cap(u32 cap);
struct omap_dm_timer *omap_dm_timer_request_by_node(struct device_node
*np);
int omap_dm_timer_free(struct omap_dm_timer *timer);
-void omap_dm_timer_enable(struct omap_dm_timer *timer);
+int omap_dm_timer_enable(struct omap_dm_timer *timer);
void omap_dm_timer_disable(struct omap_dm_timer *timer);
int omap_dm_timer_get_irq(struct omap_dm_timer *timer);
--
1.7.9.5
next prev parent reply other threads:[~2014-04-21 16:44 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-04-19 22:25 L3 custom error from dmtimer.c Joachim Eastwood
2014-04-21 15:40 ` Joel Fernandes
2014-04-21 15:57 ` Joachim Eastwood
2014-04-21 16:36 ` Joel Fernandes
2014-04-21 16:44 ` Joel Fernandes [this message]
2014-04-21 17:04 ` Joachim Eastwood
2014-04-21 19:33 ` Joel Fernandes
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=53554AFD.5070504@ti.com \
--to=joelf@ti.com \
--cc=linux-omap@vger.kernel.org \
--cc=manabian@gmail.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).