<html>
<body>
<div style="font-family: Verdana, Arial, Helvetica, Sans-Serif;">
<table bgcolor="#f9f3c9" width="100%" cellpadding="8" style="border: 1px #c9c399 solid;">
<tr>
<td>
This is an automatically generated e-mail. To reply, visit:
<a href="http://git.reviewboard.kde.org/r/111462/">http://git.reviewboard.kde.org/r/111462/</a>
</td>
</tr>
</table>
<br />
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<p style="margin-top: 0;">On July 9th, 2013, 1:38 p.m. UTC, <b>David Faure</b> wrote:</p>
<blockquote style="margin-left: 1em; border-left: 2px solid #d0d0d0; padding-left: 10px;">
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">To save 2 stat() per second, you want the user to not be notified anymore that his download is going well and the local file is growing? I don't think this is a good compromise at all. How is 2 stat calls per second "a flood" ?
(I'm thinking of the use case of downloading one very big file; copying tons of small files is a different issue)</pre>
</blockquote>
</blockquote>
<pre style="white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">To me the number stats, which btw are 4 and not 2 because of the additional destination directory stat, simply seem to be an overkill. If you copy just 4 files that is 16 stats every second until these downloads complete. How is this any different in terms of impact to users with their home partition on nfs/smb mounts? In fact this code is exactly in the same path as the stat problem I fixed recently with https://git.reviewboard.kde.org/r/110834/. Granted this current problem only applies if the destination a local file.
Anyhow, I understand the implication of my patch and have said as much when I posted it. It would mean the user would not see the size increase by simply looking at the size column in Konqueror/Dolphin details view. Having said that, would it really matter to the user if the update happened every second instead of every half second? That by itself would cut down the # of stats by half. And if I could figure out the source of the continuous destination directory stat and stop that, we would have reduced the number of stats by 3/4 without really impacting the user's ability to check on the progress of file downloads.</pre>
<br />
<p>- Dawit</p>
<br />
<p>On July 9th, 2013, 1:05 p.m. UTC, Dawit Alemayehu wrote:</p>
<table bgcolor="#fefadf" width="100%" cellspacing="0" cellpadding="8" style="background-image: url('http://git.reviewboard.kde.org/static/rb/images/review_request_box_top_bg.ab6f3b1072c9.png'); background-position: left top; background-repeat: repeat-x; border: 1px black solid;">
<tr>
<td>
<div>Review request for kdelibs and David Faure.</div>
<div>By Dawit Alemayehu.</div>
<p style="color: grey;"><i>Updated July 9, 2013, 1:05 p.m.</i></p>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Description </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" style="border: 1px solid #b8b5a0">
<tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">The attached patch changes the logic that attempts to deal with a flood of the dirty signals received by KDirLister when existing files are change. The current code simply puts the file in an update pending list and starts a delayed timer (500 ms) to process the request. As a result, every half a second or so the pending update request is processed. This would have been fine were it not for the fact that the processing results in a call to KFileItem::refersh which does a stat the local file. Hence, a stat is done on the file being downloaded every 500 ms. Additionally, some other code is also doing a stat every half second on the
download destination directory as well.
This patch attempts to prevent the flood of stat calls by restarting the update timer if we get another dirty signal on a file that is already in the update pending list. IOW, we only do the last stat call. The downside of doing that is the information about the file being download will be stale until the download finishes. Unfortunately I still have not been able to figure out what is doing the stat on the destination directory itself and hence do not have a solution for that yet. Any ideas?</pre>
</td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Testing </h1>
<table width="100%" bgcolor="#ffffff" cellspacing="0" cellpadding="10" style="border: 1px solid #b8b5a0">
<tr>
<td>
<pre style="margin: 0; padding: 0; white-space: pre-wrap; white-space: -moz-pre-wrap; white-space: -pre-wrap; white-space: -o-pre-wrap; word-wrap: break-word;">- strace an instance of Konqueror or Dolphin.
- copy a file from a remote location (ftp,sftp, smb) to your local file system.
- check if there is a flood of stat calls on both the file being copied and its destination directory.</pre>
</td>
</tr>
</table>
<h1 style="color: #575012; font-size: 10pt; margin-top: 1.5em;">Diffs</b> </h1>
<ul style="margin-left: 3em; padding-left: 0;">
<li>kio/kio/kdirlister.cpp <span style="color: grey">(aad6893)</span></li>
</ul>
<p><a href="http://git.reviewboard.kde.org/r/111462/diff/" style="margin-left: 3em;">View Diff</a></p>
</td>
</tr>
</table>
</div>
</body>
</html>