Mailing-List: contact dev-help@apr.apache.org; run by ezmlm
Precedence: bulk
Received-SPF: pass (athena.apache.org: domain of ivan@visualsvn.com designates
 74.125.82.41 as permitted sender)
MIME-Version: 1.0
In-Reply-To: <20130823184823.059ff9c2@hub>
References: <002c01cae068$d7e5e970$87b1bc50$@qqmail.nl>
 <s2tcc67648e1004261219qdb99e608p4d9a408e3b4c545@mail.gmail.com>
 <4BD5F9C8.707@rowe-clan.net>
 <m2icc67648e1004261413u1c0d605em6996bdf3c0ef805b@mail.gmail.com>
 <036901cae5ee$9e68b5c0$db3a2140$@qqmail.nl>
 <x2rcc67648e1004270403r3adcf083p55c76ee3d9aacf76@mail.gmail.com>
 <CABw-3YdwzzRLD8xpMU2KJsGB49Yq3Adh9VyTndT-=gA0JAb-8A@mail.gmail.com>
 <20130823183935.0e956cd3@hub> <20130823184823.059ff9c2@hub>
From: Ivan Zhakov <ivan@visualsvn.com>
Date: Sat, 24 Aug 2013 08:29:59 +0400
Message-ID: 
 <CABw-3Yf0EMTzCrpPhnweoJjwYQmaYjwp54P8=3ezDUNE23aomg@mail.gmail.com>
Subject: Re: Why does apr_file_read() with !APR_XTHREAD use mutexes on Windows
To: "William A. Rowe Jr." <wrowe@rowe-clan.net>
Cc: Jeff Trawick <trawick@gmail.com>, APR Developer List <dev@apr.apache.org>,
 Bert Huijben <bert@qqmail.nl>
Content-Type: text/plain; charset=UTF-8

On Sat, Aug 24, 2013 at 3:48 AM, William A. Rowe Jr.
<wrowe@rowe-clan.net> wrote:
> On Fri, 23 Aug 2013 18:39:35 -0500
> "William A. Rowe Jr." <wrowe@rowe-clan.net> wrote:
>
>> On Sat, 24 Aug 2013 00:33:38 +0400
>> Ivan Zhakov <ivan@visualsvn.com> wrote:
>>
>> > Actually Windows supports atomic seek-to-end+write: file should be
>> > opened with FILE_APPEND_DATA access right only [1] or Offset and
>> > OffsetHigh should be 0xFFFFFFFF if overlapped I/O is used [2].
>> >
>> > I'm reopening this thread because in Subversion we found case where
>> > we need true atomic append across processes/threads. So I'm willing
>> > to create a patch implementing atomic append on Windows. Is right
>> > idea for APR or not? Any concerns will be very helpful.
>> >
>> > [1]
>> > http://msdn.microsoft.com/en-us/library/windows/desktop/aa363778%28v=vs.85%29.aspx
>> > [2]
>> > http://msdn.microsoft.com/en-us/library/windows/desktop/aa365747%28v=vs.85%29.aspx
>>
>> IIRC the difference is that you have writev on unix to atomically
>> write multiple buffers.  On Windows we fake writev, so your proposed
>> atomic writes are no longer atomic.
>
Subversion doesn't use writev for file I/O, so implementing atomic
writes is enough for our case.

> Now we might use WriteFileGather at this point (remember, this stuff
> was all written when Server 2000 was still around :)... but note that
> the compatibility is listed as "Windows Server 2003 [Desktop apps only].
> WTF would that mean?  Also not supported for 32 bit Itanium apps running
> under WOW64.  How bizarre.  Then read the description of the
> aSegmentArray arg to really break your brain.
>
WriteFileGather does not help anyway: provided buffers must be at
least the size of a system memory page and must be aligned on a system
memory page size boundary.


-- 
Ivan Zhakov