<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN"> <HTML><HEAD> <META content="text/html; charset=utf-8" http-equiv=Content-Type> <STYLE> BLOCKQUOTE { MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px; MARGIN-LEFT: 2em } OL { MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px } UL { MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px } P { MARGIN-TOP: 0px; MARGIN-BOTTOM: 0px } BODY { LINE-HEIGHT: 1.5; FONT-FAMILY: 微软雅黑; COLOR: #000080; FONT-SIZE: 10.5pt } </STYLE> <META name=GENERATOR content="MSHTML 9.00.8112.16441"></HEAD> <BODY style="MARGIN: 10px"> <DIV>Thank you very much.</DIV> <DIV> </DIV> <DIV>1) I used Inspector XE 2011 on RHEL6.1 analysis the memory leak problem.</DIV> <DIV>    Inspector XE 2011 can give the leak location but no call stack.</DIV> <DIV> </DIV> <DIV>2) I used most of the spice-gtk code except the spicy.c, and modify no code below the spice-widget.c.</DIV> <DIV>    Successfully compiled the code using MingW gcc on Windows.</DIV> <DIV>    Then I use C++ memory leak detector(<A href="http://www.softwareverify.com/cpp-memory.php">http://www.softwareverify.com/cpp-memory.php</A>).</DIV> <DIV>    Finally I get the memory leak call stack. </DIV> <DIV>    I also use C++ memory leak detector analysis spicec, spices has less memory leak as you said.</DIV> <DIV> </DIV> <DIV>I will continue to spend time on tracing and analysis the image cache mechanism.</DIV> <DIV> </DIV> <DIV>But as far as I known, as one image has been created, there is on place to reference the allocated image.</DIV> <DIV> </DIV> <DIV>So :</DIV> <DIV> </DIV> <DIV>"""</DIV> <DIV>src_image = canvas_get_image(canvas, copy->src_bitmap, FALSE);<BR clear=all> <DIV> </DIV> <DIV>pixman_image_unref(src_image);</DIV> <DIV>"""<BR></DIV> <DIV>pixman_image_unref will free the image and associated image data.</DIV> <DIV> </DIV> <DIV>Am I right?</DIV></DIV> <DIV> </DIV> <HR style="WIDTH: 210px; HEIGHT: 1px" align=left color=#b5c4df SIZE=1> <DIV><SPAN>xufango</SPAN></DIV> <DIV> </DIV> <DIV style="BORDER-BOTTOM: medium none; BORDER-LEFT: medium none; PADDING-BOTTOM: 0cm; PADDING-LEFT: 0cm; PADDING-RIGHT: 0cm; BORDER-TOP: #b5c4df 1pt solid; BORDER-RIGHT: medium none; PADDING-TOP: 3pt"> <DIV style="PADDING-BOTTOM: 8px; PADDING-LEFT: 8px; PADDING-RIGHT: 8px; BACKGROUND: #efefef; COLOR: #000000; FONT-SIZE: 12px; PADDING-TOP: 8px"> <DIV><B>From:</B> <A href="mailto:marcandre.lureau@gmail.com">Marc-André Lureau</A></DIV> <DIV><B>Date:</B> 2012-03-16 23:24</DIV> <DIV><B>To:</B> <A href="mailto:xufango@gmail.com">徐舫</A></DIV> <DIV><B>CC:</B> <A href="mailto:spice-devel@lists.freedesktop.org">spice-devel</A></DIV> <DIV><B>Subject:</B> Re: [Spice-devel] [bug spice-gtk] spice-gtk memory leak</DIV></DIV></DIV> <DIV> <DIV>On Fri, Mar 16, 2012 at 10:27 AM, 徐舫 <xufango@gmail.com> wrote:</DIV> <DIV>> But the allocated memory does not free.</DIV> <DIV> </DIV> <DIV>According to massif, the memory usage becomes stable, but your</DIV> <DIV>backtrace is correct, the biggest offender is image cache, which can</DIV> <DIV>top 150mb. I am not very familiar with the caching logic, so I can't</DIV> <DIV>say whether this can be tuned to limit to a given amount. It could</DIV> <DIV>also be that spice-gtk cache is buggy, because spicec seems to use a</DIV> <DIV>little bit less (50mb). But I do tests manually, so my empirical</DIV> <DIV>results may vary.</DIV> <DIV> </DIV> <DIV>Also, running spicy under valgrind leak check tool shows quite clearly</DIV> <DIV>that there is nothing being leaked directly from spice-gtk, it seems.</DIV> <DIV> </DIV> <DIV>regards</DIV> <DIV> </DIV> <DIV>-- </DIV> <DIV>Marc-André Lureau</DIV></DIV></BODY></HTML>