On 2014-02-28 01:12, zhang json wrote:
2014-02-26 1:07 GMT+08:00 Gedare Bloom mailto:ged...@rtems.org>>:
You may like to read
http://wiki.rtems.org/wiki/index.php/SMP#Non-Preempt_Mode_for_Mutual_Exclusion
in addition to the Condition Variables project page. You can search
for "
On 2014-02-27 14:24, Jennifer Averett wrote:
/**
+ * @brief Copy POSIX Thread attribute structure
+ *
+ * This routine copies the attr2 thread attribute structure
+ * to the attr1 Thread Attribute structure.
+ *
+ * @param[out] attr1 is a pointer to the thread attribute
+ * structure to copy in
Hi Gedare,
2014-02-26 1:07 GMT+08:00 Gedare Bloom :
> You may like to read
>
> http://wiki.rtems.org/wiki/index.php/SMP#Non-Preempt_Mode_for_Mutual_Exclusion
> in addition to the Condition Variables project page. You can search
> for "condition variable" in a search engine to get some useful
> ba
Ahh, ok.
On Thu, Feb 27, 2014 at 10:53 AM, Jennifer Averett
wrote:
> I dug into this one a little more and on that path the object isn't
> found so there is no reason to call put.
>
>
>> -Original Message-
>> From: ged...@gwmail.gwu.edu [mailto:ged...@gwmail.gwu.edu] On
>> Behalf Of Gedar
I dug into this one a little more and on that path the object isn't
found so there is no reason to call put.
> -Original Message-
> From: ged...@gwmail.gwu.edu [mailto:ged...@gwmail.gwu.edu] On
> Behalf Of Gedare Bloom
> Sent: Thursday, February 27, 2014 9:18 AM
> To: Jennifer Averett
> C
On Thu, Feb 27, 2014 at 10:40 AM, Jennifer Averett
wrote:
> Ok... I see what you were saying now. Calling the
> pthread_attr_setaffinity_np from
> pthread_setaffinity_np seemed like overkill, when
> it was just one line added to the pthread_attr_setaffinity_np
> method.
>
I agree.
>> -Origi
Ok... I see what you were saying now. Calling the
pthread_attr_setaffinity_np from
pthread_setaffinity_np seemed like overkill, when
it was just one line added to the pthread_attr_setaffinity_np
method.
> -Original Message-
> From: rtems-devel-boun...@rtems.org [mailto:rtems-devel-
> b
Nevermind, you would not gain anything by what I was recommending.
What I was saying is that this patch implements
pthread_setaffinity_np(). You could have it call
pthread_attr_setaffinity_np() instead of directly writing to
api->Attributes, e.g. replace
CPU_COPY( api->Attributes.affinityset, cpuse
I'm sorry I don't follow this question. This patch is the
Implementation and test of pthread_attr_setaffinity_np.
Am I missing something? The method sets the api->Attributes affinity.
> -Original Message-
> From: ged...@gwmail.gwu.edu [mailto:ged...@gwmail.gwu.edu] On
> Behalf Of Gedare
The copy method in sys/cpuset assumes they are the same size,
if you have different sizes the non-used bits will not be set
to 0 unless the creator explicitly sets them.
> -Original Message-
> From: ged...@gwmail.gwu.edu [mailto:ged...@gwmail.gwu.edu] On
> Behalf Of Gedare Bloom
> Sent: Th
On Thu, Feb 27, 2014 at 10:09 AM, Jennifer Averett
wrote:
> I don't think you can make the default attributes static, as you
> can see in pthreadimpl.h it is used by several methods when
> to copy the default values into new attributes.
>
OK, it could be done by adding a "getter" function, but th
In case of an error ESRCH, shouldn't there be an _Objects_Put() call
before returning?
On Thu, Feb 27, 2014 at 8:30 AM, Jennifer Averett
wrote:
> Attached is the modified pthread get attribute
> patch and its corresponding test.
>
>
> Jennifer Averett
> On-Line Applications Research
> ___
This issue would apply in some of the other patches submitted that
switch on location.
On Thu, Feb 27, 2014 at 10:18 AM, Gedare Bloom wrote:
> In case of an error ESRCH, shouldn't there be an _Objects_Put() call
> before returning?
>
> On Thu, Feb 27, 2014 at 8:30 AM, Jennifer Averett
> wrote:
>
Same comments from my email on the other thread apply here. Why don't
you use the pthread_attr_setaffinity_np() to set the api->Attributes
affinity?
-Gedare
On Thu, Feb 27, 2014 at 8:28 AM, Jennifer Averett
wrote:
>
> Attached are patches for pthread get and set
> affinity, as well as a test for
I don't think you can make the default attributes static, as you
can see in pthreadimpl.h it is used by several methods when
to copy the default values into new attributes.
I'll add the copyright and doc :-)
> -Original Message-
> From: ged...@gwmail.gwu.edu [mailto:ged...@gwmail.gwu.
Do you need to check if the attr->setsize == cpusetsize in setaffinity?
Is it ever the case that the cpusetsize > attr->setsize in
getaffinity, but that is ok? similarly in setaffinity if attr->setsize
> cpusetsize? That is to say, is strict equality between the
cpusetsize and the destination of C
Since you drop const from _POSIX_Threads_Default_attributes it would
be good to change it to be statically scoped in pthread.c.
Please add documentation for the rtems_pthread_attribute_compare()
function, and fix the copyright year (and license url) in the new
pthreadattrcompare.c file.
-Gedare
Rename Cpuset_Control to CPU_set_Control. The rest looks good to me.
-Gedare
On Thu, Feb 27, 2014 at 8:22 AM, Jennifer Averett
wrote:
>
> Attached are the latest classic affinity patches.
> I believe all the requested modifications have
> been addressed.
>
> Jennifer Averett
> On-Line Application
I've seen the status page but I know Sebastian is spearheading the bulk of
the work and I'm not exactly sure what would be reasonable/useful as gsoc
project. What I'm specifically looking at:
Migration away from task variables.
Low level broadcasts.
Condition variables seems like a good topic bu
Attached is the modified pthread get attribute
patch and its corresponding test.
Jennifer Averett
On-Line Applications ResearchFrom 4f74ebdf993f09993355df27a2fab784997a8733 Mon Sep 17 00:00:00 2001
From: Jennifer Averett
Date: Tue, 14 Jan 2014 09:43:14 -0600
Subject: [PATCH 05/15] posix: Add pth
Attached are patches for pthread get and set
affinity, as well as a test for the new functions.
Jennifer Averett
On-Line Applications ResearchFrom c5d1bae56e5720232799844a76a444abad9008e3 Mon Sep 17 00:00:00 2001
From: Jennifer Averett
Date: Mon, 10 Feb 2014 10:01:31 -0600
Subject: [PATCH 10/15]
Attached patches are for pthread attribute get
and set affinity. As well as a test for the new
functions
Jennifer Averett
On-Line Applications ResearchFrom 16ea3cc81f9681af0c3102ef0d43720b4295cb6c Mon Sep 17 00:00:00 2001
From: Jennifer Averett
Date: Thu, 6 Feb 2014 12:59:08 -0600
Subject: [PAT
Attached are files for posix support of the affinity
attribute in pthread. The affinity methods and tests
will follow.
Jennifer Averett
On-Line Applications ResearchFrom d4214167cc08bfe6464a719468b0f7513a90ca42 Mon Sep 17 00:00:00 2001
From: Jennifer Averett
Date: Fri, 31 Jan 2014 08:54:45 -060
Attached are the latest classic affinity patches.
I believe all the requested modifications have
been addressed.
Jennifer Averett
On-Line Applications ResearchFrom 5b0ddb29ab7a677339ee7b2177941a06e31da423 Mon Sep 17 00:00:00 2001
From: Jennifer Averett
Date: Thu, 6 Feb 2014 12:42:24 -0600
Subjec
---
c/src/lib/libbsp/sparc/leon3/include/leon.h | 42 +++
c/src/lib/libbsp/sparc/leon3/smp/smp_leon3.c| 16 +---
c/src/lib/libbsp/sparc/leon3/startup/bspstart.c |9 +
3 files changed, 46 insertions(+), 21 deletions(-)
diff --git a/c/src/lib/libbsp/sparc
---
c/src/lib/libbsp/sparc/leon3/include/cache_.h | 88 +
1 files changed, 88 insertions(+), 0 deletions(-)
diff --git a/c/src/lib/libbsp/sparc/leon3/include/cache_.h
b/c/src/lib/libbsp/sparc/leon3/include/cache_.h
index eafbb48..a3f4af8 100644
--- a/c/src/lib/libbsp/sp
On 2014-02-27 09:16, Chris Johns wrote:
On 26/02/2014 7:43 pm, Sebastian Huber wrote:
First some words to this patch set. It is a gradual improvement of the
existing SMP low-level initialization and shutdown procedure.
The existing solution was able to start an SMP system. Only atomic
reads/w
On 26/02/2014 7:43 pm, Sebastian Huber wrote:
First some words to this patch set. It is a gradual improvement of the
existing SMP low-level initialization and shutdown procedure.
The existing solution was able to start an SMP system. Only atomic
reads/writes were used. It is impossible to imp
28 matches
Mail list logo
