Mailing-List: contact dev-help@harmony.apache.org; run by ezmlm
Precedence: bulk
Reply-To: dev@harmony.apache.org
Received-SPF: pass (nike.apache.org: domain of clraychen@gmail.com designates
 209.85.210.188 as permitted sender)
DomainKey-Signature: a=rsa-sha1; c=nofws;
        d=gmail.com; s=gamma;
        h=mime-version:in-reply-to:references:date:message-id:subject:from:to
         :content-type;
        b=omT5aQ3jC5V1P4su058fMvZB3oAAvA/NjWiynA5+yQanB/vunsdVU5YLfEGjNfYqa1
         FvaqvrJeoJ5ra69+aAQkKBj0spSHSeYNN90qa6Co6jh59c2Ny/H0jJSoETDwOIObXX0B
         R3BxclHBT3D4WZPqdhUaC9InlaAChPYx3lYD0=
MIME-Version: 1.0
In-Reply-To: <4A9FA881.5020402@gmail.com>
References: <290644005.1251857072710.JavaMail.jira@brutus>
	 <4A9F8871.4020605@gmail.com> <4A9F9976.7010003@gmail.com>
	 <4A9FA881.5020402@gmail.com>
Date: Sun, 6 Sep 2009 21:19:24 +0800
Message-ID: <4d9fb1a00909060619m41bf348du2130841d19656075@mail.gmail.com>
Subject: Re: [jira] Created: (HARMONY-6328) [classlib][nio] optimize
	SocketChannel.write(ByteBuffer[], int, int) by writev
From: Ray Chen <clraychen@gmail.com>
To: dev@harmony.apache.org
Content-Type: multipart/alternative; boundary=001636b2b0c60159480472e895f4

--001636b2b0c60159480472e895f4
Content-Type: text/plain; charset=ISO-8859-1

Hi Reigs,
I have tested your patch on z/OS,  it works OK.

Greate job.

On Thu, Sep 3, 2009 at 7:29 PM, Regis <xu.regis@gmail.com> wrote:

> Tim Ellison wrote:
>
>> A minor comment first :-) Please check return values for :
>>
>> + vect = (struct iovec*) hymem_allocate_memory(sizeof(struct iovec) *
>> length);
>> +
>> + message = (*env)->GetPrimitiveArrayCritical(env, addrs, &isCopyMessage);
>> + cts = (*env)->GetPrimitiveArrayCritical(env, counts, &isCopyCts);
>>
>
> Yes, will add it, thanks.
>
>
>>
>> A further enhancement is to have two versions of primitives, one that
>> deals with direct buffers and one that deals with java heap buffers, so
>> that there is (potentially) no data copying required for the java-heap
>> buffers version that you have got today:
>>
>
> One case stop me to do that is a buffer array contains direct and java heap
> buffers both. In this case, either copying java heap buffres to direct
> buffers or passing the whole array to native directly.
>
> If pass the array to native directly, it need to call isDirect and
> AddressUtil.getDirectBufferAddress in native code, as I understand, call
> java methods from native side is like reflection, even more expensive, so I
> chose copying all java heap buffers to direct buffers, at least it fast when
> most buffers are direct and size of java heap buffers are not too big.
>
> One possible solution I can think is adding a new native call
>
> writevDirect(JNIEnv *env, jobject thiz, jobject fd, jobject bufferArray,
> jobject addrs, jobject counts, jint length)
>
> passing buffer array and native addresses of direct buffers together, if
> element of addrs is 0, the corresponding buffer is not direct. Well, this
> method is a little bit hard to understand, it has 7 parameters....
>
>
>> + if (!buffer.isDirect()) {
>> + src[i] = ByteBuffer.allocateDirect(buffer.remaining());
>> + int oldPosition = buffer.position();
>> + src[i].put(buffer);
>>
>> some VMs will pin the java heap memory accessed in JNI
>> (GetByteArrayElements), so if you can get hold of the ByteBuffer#array()
>> you can pass a set of pointers to the backing arrays and maybe send the
>> data directly.
>>
>> Looking forward to seeing the readv impl too ;-)
>>
>
> Sure, it's on my TODO list :)
>
>
>> Regards,
>> Tim
>>
>>
>>
>
> --
> Best Regards,
> Regis.
>


-- 
Best Regards,
Ray Chen

--001636b2b0c60159480472e895f4--