<div dir="ltr">The overhead of doing separate XShmPutImages should be minimal. If you're seeing performance issues from that, you might be using the XCB Image library: you probably should look there for optimizations: I know that little care went into trying to make it fast back when it was written.<div><br></div><div><div class="gmail_quote"><div dir="ltr">On Tue, Dec 26, 2017 at 12:11 PM Clemens Eisserer <<a href="mailto:linuxhippy@gmail.com" target="_blank">linuxhippy@gmail.com</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Hi there,<br>
<br>
I am currently trying to solve a performance problem in Java2D's<br>
xrender backend.<br>
What I actually would like to do is to defer the actual command<br>
submission to the x-server in order to batch many small XPutImages to<br>
one large XShmPutImage.<br>
I could do this in my own code (log API calls to an application<br>
managed buffer and call the coresponding Xlib/xcb functions later from<br>
within a large protocol-decoding switch statement). However I wonder<br>
if there isn't a more efficient way, given the fact that X's network<br>
transparency introduces some kind of defering anyway.<br>
<br>
Is there any way to intercept the protocol before it is sent to the X-Server?<br>
Or in case interception is not supported, is there a way to let xcb<br>
generate all protocol to a user-accessible / user-controlled buffer<br>
and submit it manually to the x-server?<br>
<br>
Thank you in advance and best regards, Clemens<br>
_______________________________________________<br>
Xcb mailing list<br>
<a href="mailto:Xcb@lists.freedesktop.org" target="_blank">Xcb@lists.freedesktop.org</a><br>
<a href="https://lists.freedesktop.org/mailman/listinfo/xcb" rel="noreferrer" target="_blank">https://lists.freedesktop.org/mailman/listinfo/xcb</a></blockquote></div></div></div>