From: Julien Grall <julien.grall@arm.com>
To: Benjamin Sanda <ben.sanda@dornerworks.com>,
xen-devel@lists.xenproject.org
Cc: sstabellini@kernel.org, Wei Liu <wei.liu2@citrix.com>,
George Dunlap <george.dunlap@eu.citrix.com>,
Ian Jackson <ian.jackson@eu.citrix.com>, Tim Deegan <tim@xen.org>,
Jan Beulich <jbeulich@suse.com>,
Andrew Cooper <andrew.cooper3@citrix.com>,
Keir Fraser <keir@xen.org>
Subject: Re: [PATCH v3 3/5] xentrace: Timestamp support for ARM platform
Date: Fri, 8 Apr 2016 11:50:56 +0100 [thread overview]
Message-ID: <57078D10.1000609@arm.com> (raw)
In-Reply-To: <1459795727-3116-4-git-send-email-ben.sanda@dornerworks.com>
Hello Benjamin,
On 04/04/16 19:48, Benjamin Sanda wrote:
> Moved get_cycles() to time.c and modified to return the core timestamp
> tick count for use by the trace buffer timestamping routines in
> xentrace. get_cycles() was moved to the C file to avoid including the
> register specific header file in time.h and to commonize it with the
> get_s_time() function. Also defined cycles_t as uint64_t to simplify
> casting.
I'm not sure what you mean by "simplify casting".
The type cycles_t is not correctly defined for ARM32 because "unsigned
long" is always 32-bits. However, the physical count register (CNTPCT)
is always 64-bits. So the number of cycles would have been truncated.
The rest of the patch looks good to me.
> get_s_time() was also modified to now use the updated get_cycles() to
> retrieve the tick count instead of directly reading it.
>
> Signed-off-by: Benjamin Sanda <ben.sanda@dornerworks.com>
>
> ---
> Changed since v2:
> * Combined v2 patches 7 and 6 into one patch in v3. No code change.
>
> ---
> Changed since v1:
> * Moved get_cycles() to time.c
> * Added function prototype for get_cycles()
> ---
> xen/arch/arm/time.c | 9 ++++++++-
> xen/include/asm-arm/time.h | 11 +++++------
> 2 files changed, 13 insertions(+), 7 deletions(-)
>
> diff --git a/xen/arch/arm/time.c b/xen/arch/arm/time.c
> index 7dae28b..9aface3 100644
> --- a/xen/arch/arm/time.c
> +++ b/xen/arch/arm/time.c
> @@ -192,10 +192,17 @@ int __init init_xen_time(void)
> /* Return number of nanoseconds since boot */
> s_time_t get_s_time(void)
> {
> - uint64_t ticks = READ_SYSREG64(CNTPCT_EL0) - boot_count;
> + cycles_t ticks = get_cycles();
> return ticks_to_ns(ticks);
> }
>
> +/* Return the number of ticks since boot */
> +cycles_t get_cycles(void)
> +{
> + /* return raw tick count of main timer */
> + return READ_SYSREG64(CNTPCT_EL0) - boot_count;
> +}
> +
> /* Set the timer to wake us up at a particular time.
> * Timeout is a Xen system time (nanoseconds since boot); 0 disables the timer.
> * Returns 1 on success; 0 if the timeout is too soon or is in the past. */
> diff --git a/xen/include/asm-arm/time.h b/xen/include/asm-arm/time.h
> index 5b9a31d..b57f4c1 100644
> --- a/xen/include/asm-arm/time.h
> +++ b/xen/include/asm-arm/time.h
> @@ -5,12 +5,8 @@
> DT_MATCH_COMPATIBLE("arm,armv7-timer"), \
> DT_MATCH_COMPATIBLE("arm,armv8-timer")
>
> -typedef unsigned long cycles_t;
> -
> -static inline cycles_t get_cycles (void)
> -{
> - return 0;
> -}
> +/* Tick count type */
> +typedef uint64_t cycles_t;
>
> /* List of timer's IRQ */
> enum timer_ppi
> @@ -37,6 +33,9 @@ extern void init_timer_interrupt(void);
> /* Counter value at boot time */
> extern uint64_t boot_count;
>
> +/* Get raw system tick count */
> +cycles_t get_cycles(void);
> +
> extern s_time_t ticks_to_ns(uint64_t ticks);
> extern uint64_t ns_to_ticks(s_time_t ns);
>
>
Regards,
--
Julien Grall
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
next prev parent reply other threads:[~2016-04-08 10:51 UTC|newest]
Thread overview: 27+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-04 18:48 [PATCH v3 0/5] xentrace/xenalyze Support on ARM Benjamin Sanda
2016-04-04 18:48 ` [PATCH v3 1/5] xentrace: Common Support for get_pg_owner/put_pg_owner on ARM and x86 Benjamin Sanda
2016-04-04 23:05 ` Andrew Cooper
2016-04-05 8:12 ` Jan Beulich
2016-04-14 19:59 ` Ben Sanda
2016-04-17 7:58 ` Jan Beulich
2016-04-04 18:48 ` [PATCH v3 2/5] xentrace: Memory/Page Mapping support for DOMID_XEN on ARM Benjamin Sanda
2016-04-08 10:42 ` Julien Grall
2016-04-08 15:49 ` Jan Beulich
2016-04-08 17:58 ` Andrew Cooper
2016-04-11 9:52 ` George Dunlap
2016-04-12 15:53 ` Julien Grall
2016-04-14 19:52 ` Ben Sanda
2016-04-20 12:48 ` Julien Grall
2016-04-22 9:42 ` Stefano Stabellini
2016-04-22 17:01 ` Julien Grall
2016-04-04 18:48 ` [PATCH v3 3/5] xentrace: Timestamp support for ARM platform Benjamin Sanda
2016-04-08 10:50 ` Julien Grall [this message]
2016-04-11 14:56 ` Konrad Rzeszutek Wilk
2016-04-04 18:48 ` [PATCH v3 4/5] xentrace: Trace Buffer Initialization on ARM Benjamin Sanda
2016-04-08 10:53 ` Julien Grall
2016-04-04 18:48 ` [PATCH v3 5/5] xenalyze: Build for Both ARM and x86 Platforms Benjamin Sanda
2016-04-05 8:09 ` [PATCH v3 0/5] xentrace/xenalyze Support on ARM Jan Beulich
2016-04-06 16:51 ` Ben Sanda
2016-04-06 16:59 ` Andrew Cooper
2016-04-06 17:03 ` Ben Sanda
2016-04-08 14:44 ` George Dunlap
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=57078D10.1000609@arm.com \
--to=julien.grall@arm.com \
--cc=andrew.cooper3@citrix.com \
--cc=ben.sanda@dornerworks.com \
--cc=george.dunlap@eu.citrix.com \
--cc=ian.jackson@eu.citrix.com \
--cc=jbeulich@suse.com \
--cc=keir@xen.org \
--cc=sstabellini@kernel.org \
--cc=tim@xen.org \
--cc=wei.liu2@citrix.com \
--cc=xen-devel@lists.xenproject.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).