<html>
<body>
<div style="font-family: Verdana, Arial, Helvetica, Sans-Serif;">
<table bgcolor="#f9f3c9" width="100%" cellpadding="12" style="border: 1px #c9c399 solid; border-radius: 6px; -moz-border-radius: 6px; -webkit-border-radius: 6px;">
<tr>
<td>
This is an automatically generated e-mail. To reply, visit:
<a href="https://git.reviewboard.kde.org/r/128480/">https://git.reviewboard.kde.org/r/128480/</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 20th, 2016, 10:17 p.m. UTC, <b>Milian Wolff</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;"><p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">I agree, let's go for it. And we all should report if we still reproduce the bug, or not (also, remember to reference it in your commit message, and close it for now).</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">Could you do me a favor and have a look at the maximum resident set sizes reported by</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;"><div class="codehilite" style="background: #f8f8f8"><pre style="line-height: 125%">CLEAR_DUCHAIN_DIR=1 \time -v kdevelop
</pre></div>
</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">before and after this patch, for a session that has a bunch of files opened? I simply wonder whether we now consume gigs of memory more, or just a couple of megs (I think the latter).</p></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;"><p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">Not much of a difference.</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">After:</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;"><div class="codehilite" style="background: #f8f8f8"><pre style="line-height: 125%"> User time (seconds): 731.34
System time (seconds): 15.96
Percent of CPU this job got: 536%
Elapsed (wall clock) time (h:mm:ss or m:ss): 2:19.30
Average shared text size (kbytes): 0
Average unshared data size (kbytes): 0
Average stack size (kbytes): 0
Average total size (kbytes): 0
Maximum resident set size (kbytes): 1872812
Average resident set size (kbytes): 0
Major (requiring I/O) page faults: 0
Minor (reclaiming a frame) page faults: 2952649
Voluntary context switches: 133801
Involuntary context switches: 60886
Swaps: 0
File system inputs: 88
File system outputs: 251952
Socket messages sent: 0
Socket messages received: 0
Signals delivered: 0
Page size (bytes): 4096
Exit status: 0
</pre></div>
</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">Before:</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;"><div class="codehilite" style="background: #f8f8f8"><pre style="line-height: 125%"> User time (seconds): 770.12
System time (seconds): 16.27
Percent of CPU this job got: 366%
Elapsed (wall clock) time (h:mm:ss or m:ss): 3:34.68
Average shared text size (kbytes): 0
Average unshared data size (kbytes): 0
Average stack size (kbytes): 0
Average total size (kbytes): 0
Maximum resident set size (kbytes): 1900812
Average resident set size (kbytes): 0
Major (requiring I/O) page faults: 1
Minor (reclaiming a frame) page faults: 2485091
Voluntary context switches: 228263
Involuntary context switches: 93074
Swaps: 0
File system inputs: 1760
File system outputs: 260736
Socket messages sent: 0
Socket messages received: 0
Signals delivered: 0
Page size (bytes): 4096
Exit status: 0
</pre></div>
</p></pre>
<br />
<p>- Sven</p>
<br />
<p>On July 20th, 2016, 9:49 p.m. UTC, Sven Brauch wrote:</p>
<table bgcolor="#fefadf" width="100%" cellspacing="0" cellpadding="12" style="border: 1px #888a85 solid; border-radius: 6px; -moz-border-radius: 6px; -webkit-border-radius: 6px;">
<tr>
<td>
<div>Review request for KDevelop, Kevin Funk and Milian Wolff.</div>
<div>By Sven Brauch.</div>
<p style="color: grey;"><i>Updated July 20, 2016, 9:49 p.m.</i></p>
<div style="margin-top: 1.5em;">
<b style="color: #575012; font-size: 10pt;">Repository: </b>
kdevelop
</div>
<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;"><p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">Terrible patch, I know. Sorry. It cost a lot of nerves and time though and I want somebody to look at / try out whether this is a working fix in principle.</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">In the end, the issue(s) are simple:</p>
<ol style="padding: 0;text-rendering: inherit;margin: 0 0 0 2em;line-height: inherit;white-space: normal;">
<li style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: normal;">
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">we need to hold the foreground lock between reading the modification revision and reading the contents. Otherwise they can mismatch.</p>
</li>
<li style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: normal;">
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">in buildDUChain(), there was</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">envFile->setModificationRevision(ModificationRevision::revisionForFile(context->url()));</p>
</li>
</ol>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">which is wrong. It sets the modification revision of the parsing environment file to the revision the file has <em style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: normal;">right now</em>. The contents, however, were read before in the constructor, and several milliseconds might have passed until this line is reached, with no locks held to prevent anything.
Instead, we have to set the revision to the one which was stored when the tracker was last reset (which happened right after reading the contents, coupled to the contents by the foreground lock).</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">Actually I'm not sure if the foreground lock is required in this case, because clang for some reason reads the contents in the constructor, which afaik runs in the main thread anyways.</p></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;"><p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">Very hard to reproduce. Apply this patch to kdevplatform:</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;"><div class="codehilite" style="background: #f8f8f8"><pre style="line-height: 125%"><span style="color: #000080; font-weight: bold">diff --git a/language/backgroundparser/documentchangetracker.cpp b/language/backgroundparser/documentchangetracker.cpp</span>
<span style="color: #000080; font-weight: bold">index 294bc14..6f56250 100644</span>
<span style="color: #A00000">--- a/language/backgroundparser/documentchangetracker.cpp</span>
<span style="color: #00A000">+++ b/language/backgroundparser/documentchangetracker.cpp</span>
<span style="color: #800080; font-weight: bold">@@ -237,6 +237,11 @@ KDevelop::RangeInRevision DocumentChangeTracker::transformBetweenRevisions(KDeve</span>
{
VERIFY_FOREGROUND_LOCKED
<span style="color: #00A000">+ if ( !((fromRevision == -1 || holdingRevision(fromRevision)) && (toRevision == -1 || holdingRevision(toRevision)) ) ) {</span>
<span style="color: #00A000">+ qWarning() << "invalid transform: from" << fromRevision << "to" << toRevision</span>
<span style="color: #00A000">+ << "but not both revisions held: [from/to]:" << holdingRevision(fromRevision) << holdingRevision(toRevision);</span>
<span style="color: #00A000">+ };</span>
<span style="color: #00A000">+</span>
if((fromRevision == -1 || holdingRevision(fromRevision)) && (toRevision == -1 || holdingRevision(toRevision)))
{
m_moving->transformCursor(range.start.line, range.start.column, KTextEditor::MovingCursor::MoveOnInsert, fromRevision, toRevision);
<span style="color: #800080; font-weight: bold">@@ -348,6 +353,8 @@ void DocumentChangeTracker::unlockRevision(qint64 revision)</span>
m_moving->unlockRevision(revision);
m_revisionLocks.erase(it);
}
<span style="color: #00A000">+</span>
<span style="color: #00A000">+ qDebug() << "** clearing revision" << revision;</span>
}
qint64 RevisionLockerAndClearer::revision() const
</pre></div>
</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">You get <em style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: normal;">lots</em> of those warnings iff you manage to trigger the issue. The best bet seems to be open a relatively large cpp file, and keep removing text and readding it with Ctrl+Z with varying wait times in between (500ms-parse-delay-ish). Do that for a minute or two and it will trigger eventually.</p>
<p style="padding: 0;text-rendering: inherit;margin: 0;line-height: inherit;white-space: inherit;">You still get a few of the warning messages sometimes even with the patch applied (but far more without). I think they are from the problem reporter plugin, and I'm not sure if the plugin does something wrong or the parse job.</p></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>languages/clang/clangparsejob.cpp <span style="color: grey">(8375eb5)</span></li>
</ul>
<p><a href="https://git.reviewboard.kde.org/r/128480/diff/" style="margin-left: 3em;">View Diff</a></p>
</td>
</tr>
</table>
</div>
</body>
</html>