From: Stefano Babic <sbabic@denx.de>
To: u-boot@lists.denx.de
Subject: [U-Boot] [PATCH] imx-common: timer: fix 32-bit overflow
Date: Wed, 20 Mar 2013 11:19:33 +0100 [thread overview]
Message-ID: <51498D35.2040608@denx.de> (raw)
In-Reply-To: <513D7D5B.3050602@de.bosch.com>
On 11/03/2013 07:44, Dirk Behme wrote:
> On 04.03.2013 15:16, Dirk Behme wrote:
>> From: Knut Wohlrab <knut.wohlrab@de.bosch.com>
>>
>> The i.MX6 common timer uses the 32-bit variable tbl (time base lower)
>> to record the overflow of the 32-bit counter. I.e. if the counter
>> overflows, the variable tbl does overflow, too.
>>
>> To capture this overflow, use the variable tbu (time base upper), too.
>> Return the combined value of tbl and tbu.
>>
>> lastinc is unused then, remove it.
>>
>> Signed-off-by: Knut Wohlrab <knut.wohlrab@de.bosch.com>
>> Signed-off-by: Dirk Behme <dirk.behme@de.bosch.com>
>> ---
>> Note: This replaces the patch
>> http://patchwork.ozlabs.org/patch/224646/
>>
>> arch/arm/imx-common/timer.c | 26 +++++++-------------------
>> 1 file changed, 7 insertions(+), 19 deletions(-)
>>
>> Index: freescale-u-boot-imx.git/arch/arm/imx-common/timer.c
>> ===================================================================
>> --- freescale-u-boot-imx.git.orig/arch/arm/imx-common/timer.c
>> +++ freescale-u-boot-imx.git/arch/arm/imx-common/timer.c
>> @@ -48,9 +48,6 @@ static struct mxc_gpt *cur_gpt = (struct
>>
>> DECLARE_GLOBAL_DATA_PTR;
>>
>> -#define timestamp (gd->arch.tbl)
>> -#define lastinc (gd->arch.lastinc)
>> -
>> static inline unsigned long long tick_to_time(unsigned long long tick)
>> {
>> tick *= CONFIG_SYS_HZ;
>> @@ -70,7 +67,6 @@ static inline unsigned long long us_to_t
>> int timer_init(void)
>> {
>> int i;
>> - ulong val;
>>
>> /* setup GP Timer 1 */
>> __raw_writel(GPTCR_SWR, &cur_gpt->control);
>> @@ -85,9 +81,8 @@ int timer_init(void)
>> i = __raw_readl(&cur_gpt->control);
>> __raw_writel(i | GPTCR_CLKSOURCE_32 | GPTCR_TEN, &cur_gpt->control);
>>
>> - val = __raw_readl(&cur_gpt->counter);
>> - lastinc = val / (MXC_CLK32 / CONFIG_SYS_HZ);
>> - timestamp = 0;
>> + gd->arch.tbl = __raw_readl(&cur_gpt->counter);
>> + gd->arch.tbu = 0;
>>
>> return 0;
>> }
>> @@ -96,18 +91,11 @@ unsigned long long get_ticks(void)
>> {
>> ulong now = __raw_readl(&cur_gpt->counter); /* current tick value */
>>
>> - if (now >= lastinc) {
>> - /*
>> - * normal mode (non roll)
>> - * move stamp forward with absolut diff ticks
>> - */
>> - timestamp += (now - lastinc);
>> - } else {
>> - /* we have rollover of incrementer */
>> - timestamp += (0xFFFFFFFF - lastinc) + now;
>> - }
>> - lastinc = now;
>> - return timestamp;
>> + /* increment tbu if tbl has rolled over */
>> + if (now < gd->arch.tbl)
>> + gd->arch.tbu++;
>> + gd->arch.tbl = now;
>> + return (((unsigned long long)gd->arch.tbu) << 32) | gd->arch.tbl;
>> }
>>
>> ulong get_timer_masked(void)
>>
>
> Any comments on this?
>
> Many thanks
Hi Dirk,
no thanks to fix this coherently with other SOCs. Checking your patch I
get convinced that we have a lot of duplicated code (get_ticks() for
example) in most SOCs, but let it for a further improvement.
Acked-by: Stefano Babic <sbabic@denx.de>
Best regards,
Stefano
--
=====================================================================
DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Phone: +49-8142-66989-53 Fax: +49-8142-66989-80 Email: sbabic at denx.de
=====================================================================
next prev parent reply other threads:[~2013-03-20 10:19 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-03-04 14:16 [U-Boot] [PATCH] imx-common: timer: fix 32-bit overflow Dirk Behme
2013-03-11 6:44 ` Dirk Behme
2013-03-20 10:19 ` Stefano Babic [this message]
2013-03-20 10:39 ` Stefano Babic
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=51498D35.2040608@denx.de \
--to=sbabic@denx.de \
--cc=u-boot@lists.denx.de \
/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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.