FYI
http://www.netbsd.org/~ozaki-r/pub/AsiaBSDCon2014-WIP-ozaki-r.pdf
This is my slides about this work, which was presented in WIP session
of AsiaBSDCon 2014 last week.
Regards,
ozaki-r
On Mon, Mar 17, 2014 at 8:02 PM, Ryota Ozaki ozak...@netbsd.org wrote:
I will commit the patch during
I will commit the patch during AsiaBSDCon 2014 if there is
no objection :)
I did it.
So you can try DTrace on ARM now! You can run some commands
and scripts described in
http://wiki.netbsd.org/tutorials/how_to_enable_and_run_dtrace/ .
Beaglebone (Black) is a recommended machine for it.
On Mon, Mar 10, 2014 at 5:04 PM, Ryota Ozaki ozak...@netbsd.org wrote:
Hi,
I've updated my patch.
https://gist.github.com/ozaki-r/8535846
http://www.netbsd.org/~ozaki-r/dtrace-arm.patch
The patch now consists of a few files:
$ diffstat dtrace-arm.patch
Hi,
I've updated my patch.
https://gist.github.com/ozaki-r/8535846
http://www.netbsd.org/~ozaki-r/dtrace-arm.patch
The patch now consists of a few files:
$ diffstat dtrace-arm.patch
external/cddl/osnet/dev/dtrace/arm/dtrace_asm.S | 13
external/cddl/osnet/dev/dtrace/arm/dtrace_isa.c |
Hi Matt,
So my patch for sys/arch/arm is now rather small than
so far (just two files); it's minimum code for the core part
to run DTrace on ARM.
Can you please check it?
The patch for sys/arch/arm is
http://www.netbsd.org/~ozaki-r/dtrace-arm-sys.patch
The full patch is available at the usual
On Mar 7, 2014, at 2:27 AM, Ryota Ozaki ozak...@netbsd.org wrote:
Hi Matt,
So my patch for sys/arch/arm is now rather small than
so far (just two files); it's minimum code for the core part
to run DTrace on ARM.
Can you please check it?
The patch for sys/arch/arm is
On Thu, Mar 06, 2014 at 03:34:18PM +0900, Ryota Ozaki wrote:
On Thu, Mar 6, 2014 at 2:21 PM, Masao Uebayashi uebay...@gmail.com wrote:
Ah. I misread that schedstate_percpu uses percpu(9)'s fast path,
which doesn't exist...
Anyway if it's assumed that cpu is not attached at run-time,
On Mar 6, 1:25pm, ozak...@netbsd.org (Ryota Ozaki) wrote:
-- Subject: Re: Porting DTrace to ARM
| The problem is that all the functions in cpufunc.h are cpu_xxx
| cpuid would be an outlier.
change it to static inline?
christos
On Mar 6, 1:55pm, ozak...@netbsd.org (Ryota Ozaki) wrote:
-- Subject: Re: Porting DTrace to ARM
| On Thu, Mar 6, 2014 at 1:26 PM, Matt Thomas m...@3am-software.com wrote:
|
| On Mar 5, 2014, at 8:25 PM, Ryota Ozaki ozak...@netbsd.org wrote:
|
| On Thu, Mar 6, 2014 at 12:48 PM, Matt Thomas m
On Thu, Mar 6, 2014 at 3:50 PM, Masao Uebayashi uebay...@gmail.com wrote:
Well, more type is better. :) See [1].
My question was more like, what happens if CPU can be attached at
run-time in the future? Per-CPU initialization operations have to be
remembered somewhere.
There is no concern
On Thu, Mar 6, 2014 at 5:32 PM, David Laight da...@l8s.co.uk wrote:
On Thu, Mar 06, 2014 at 03:34:18PM +0900, Ryota Ozaki wrote:
On Thu, Mar 6, 2014 at 2:21 PM, Masao Uebayashi uebay...@gmail.com wrote:
Ah. I misread that schedstate_percpu uses percpu(9)'s fast path,
which doesn't exist...
On Thu, Mar 6, 2014 at 10:46 PM, Christos Zoulas chris...@zoulas.com wrote:
On Mar 6, 1:55pm, ozak...@netbsd.org (Ryota Ozaki) wrote:
-- Subject: Re: Porting DTrace to ARM
| On Thu, Mar 6, 2014 at 1:26 PM, Matt Thomas m...@3am-software.com wrote:
|
| On Mar 5, 2014, at 8:25 PM, Ryota
Hi,
So I've updated my patch: https://gist.github.com/ozaki-r/8535846
The new patch dropped committed changes, followed KNF (hopefully)
and did some tweaks.
At this point, what I would like to get feedbacks are some changes
under sys/arch/arm:
- Replace cpu_id with cpuid in sys/arch/arm
-
On Mar 5, 2014, at 3:12 AM, Ryota Ozaki ozak...@netbsd.org wrote:
- Replace cpu_id with cpuid in sys/arch/arm
- Can I commit the change?
Why? It's just churn for no reason I can see.
- Move INKERNEL and FR_* macros from arm/arm/db_trace.c
to arm/include/db_machdep.h to refer them in
On Thu, Mar 6, 2014 at 12:48 PM, Matt Thomas m...@3am-software.com wrote:
On Mar 5, 2014, at 7:33 PM, Ryota Ozaki ozak...@netbsd.org wrote:
On Thu, Mar 6, 2014 at 1:39 AM, Matt Thomas m...@3am-software.com wrote:
On Mar 5, 2014, at 3:12 AM, Ryota Ozaki ozak...@netbsd.org wrote:
- Replace
Ah. I misread that schedstate_percpu uses percpu(9)'s fast path,
which doesn't exist...
Anyway if it's assumed that cpu is not attached at run-time, assigning
struct cpu_data::void *cpu_dtraceinfo at module attachment would be
just fine.
On Thu, Mar 6, 2014 at 2:08 PM, Matt Thomas
On Thu, Mar 6, 2014 at 2:21 PM, Masao Uebayashi uebay...@gmail.com wrote:
Ah. I misread that schedstate_percpu uses percpu(9)'s fast path,
which doesn't exist...
Anyway if it's assumed that cpu is not attached at run-time, assigning
struct cpu_data::void *cpu_dtraceinfo at module attachment
Well, more type is better. :) See [1].
My question was more like, what happens if CPU can be attached at
run-time in the future? Per-CPU initialization operations have to be
remembered somewhere.
[1] http://mail-index.netbsd.org/tech-kern/2014/03/03/msg016699.html
On Thu, Mar 6, 2014 at 3:34
(2014/01/27 19:57), Ryota Ozaki wrote:
(2014/01/27 14:43), Ryota Ozaki wrote:
Hi Matt,
Thank you very much for your kind reviewing.
(2014/01/24 23:03), Matt Thomas wrote:
On Jan 24, 2014, at 2:34 AM, Ryota Ozaki ozak...@iij.ad.jp wrote:
Updated as I planned, there is no curious update
(2014/01/27 14:43), Ryota Ozaki wrote:
Hi Matt,
Thank you very much for your kind reviewing.
(2014/01/24 23:03), Matt Thomas wrote:
On Jan 24, 2014, at 2:34 AM, Ryota Ozaki ozak...@iij.ad.jp wrote:
Updated as I planned, there is no curious update though.
Use of curcpu() or cpu_number()
Hi Matt,
Thank you very much for your kind reviewing.
(2014/01/24 23:03), Matt Thomas wrote:
On Jan 24, 2014, at 2:34 AM, Ryota Ozaki ozak...@iij.ad.jp wrote:
Updated as I planned, there is no curious update though.
Use of curcpu() or cpu_number() without having done kpreempt_disable() is
(2014/01/23 20:15), Ryota Ozaki wrote:
(2014/01/22 21:36), Christos Zoulas wrote:
On Jan 22, 11:00am, ozak...@iij.ad.jp (Ryota Ozaki) wrote:
-- Subject: Re: Porting DTrace to ARM
| 1. there seem to be some whitespace only changes
|
| Sorry for the messy code. Will fix.
Not a problem
Hi,
I'm working on porting DTrace to ARM.
The patch (*) is my WIP product on the work. It's not
still mature but it works anyway in some cases.
(*) https://gist.github.com/ozaki-r/8535846
I'm sending the patch because I would like to ask comments
and suggestions whether I'm going to the right
23 matches
Mail list logo
