KST cpu hog (100%) on Gnome natty/lucid

[sdaau]

As I don't have the possibility right now to register an account at KDE Bug Tracking System, I'll try posting here.

I tried using Kst plot on Natty; I cannot remember if I could install the vanilla Lucid version on Natty - but eventually, I tried the Lucid version of the latest 2.0.7 ppa on both Natty and Lucid; and then I also built tag 2.0.7 from git source on Natty. And in all these cases, `kst` starts up fine; e.g. `top` reports:

  PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
 1101 root 20 0 54272 22m 10m S 4 1.1 128:41.95 Xorg
 4475 user 20 0 79428 26m 20m S 1 1.4 0:01.11 kst2

... (if kst2 even makes the `top` list) - but as soon as you put an empty plot in the window, CPU usage goes to nearly 100%; e.g. `top` reports:

  PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
 4475 user 20 0 79428 27m 21m R 97 1.4 0:24.87 kst2
 1101 root 20 0 56756 22m 10m S 93 1.1 129:16.00 Xorg

... and this is regardless if Pause in interface is clicked (this is adding an empty plot after startup, after all - before any data is loaded); I had OpenGL activated - and so the CPU is not exactly 99% but a little less, as above - but still, it's impossible to do anything with that after all.

Since I tried both ppa and source-built versions, I guess the fault is not with `kst` itself, but with whatever Gnome-Qt bridge my systems are using; but I thought it'd be best reported here first. Not sure if it helps, but on natty I have:

$ qmake --version
QMake version 2.01a
Using Qt version 4.7.2 in /usr/lib

I tried to run `kst2` a bit under `ltrace`, and can see massive output as soon as the CPU hog starts; messages like:

....
_ZNK11QMetaObject4castEP7QObject(0x831fbac, 0x96c8958, 402, 270, 215) = 0x96c8958
_ZNK11QMetaObject4castEP7QObject(0x831fbac, 0x96c8958, 402, 270, 215) = 0x96c8958
_ZNK13QGraphicsItem10isSelectedEv(0x94e96c4, 0x96c8958, 402, 270, 215) = 0
_ZNK17QGraphicsRectItem12boundingRectEv(0xbfc5a1f0, 0x94e96c4, 402, 270, 215) = 0xbfc5a1f0
_ZNK11QMetaObject4castEP7QObject(0x831fbac, 0x96c8958, 0xbfc59bf8, 0x182a841, 0x975eda0) = 0x96c8958
_ZNK11QMetaObject4castEP7QObject(0x831fbac, 0x96c8958, 0xbfc59bf8, 0x182a841, 0x975eda0) = 0x96c8958
_ZNK13QGraphicsItem10isSelectedEv(0x974ad0c, 0x96c8958, 0xbfc59bf8, 0x182a841, 0x975eda0) = 0
_ZNK17QGraphicsRectItem12boundingRectEv(0xbfc59cb0, 0x974ad0c, 0xbfc59bf8, 0x182a841, 0x975eda0) = 0xbfc59cb0
_ZN12QApplication6notifyEP7QObjectP6QEvent(0xbfc5b2a4, 0x9568de0, 0x9464ef8, 0x6958f51, 0 <unfinished ...>
_ZN11QMainWindow5eventEP6QEvent(0x9568de0, 0x9464ef8, 0x9464ef8, 0x1f5a844, 77 <unfinished ...>
_ZThn8_NK7QWidget7devTypeEv(0x96da9c0, 0xbfc59cb0, 32, 0xb5c6aad5, 16) = 1
_ZThn8_NK7QWidget7devTypeEv(0x96da9c0, 0x945d140, 0xbfc59aa8, 0x1739b9f, 0xbfc59a94) = 1
_ZThn8_NK7QWidget7devTypeEv(0x96da9c0, 0x96da9c0, 0xbfc59b98, 0x69c38b8, 0) = 1
_ZThn8_NK7QWidget7devTypeEv(0x96da9c0, 0x96ee8a4, 0x40000000, 0, 0x404f8000) = 1
_ZThn8_NK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE(0x96da9c0, 8, 0, 0x1610641, 0x96da9b8 <unfinished ...>
_ZNK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE(0x96c8958, 8, 0xbfc59ae8, 0x932b4d1, 0x974a690 <unfinished ...>
_ZNK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE(0x9577c78, 8, 0x9a61898, 264, 0x974a700 <unfinished ...>
_ZNK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE(0x9582180, 8, 0xbfc59ac8, 0x19b85c5, 0xb5d5f3c0 <unfinished ...>
_ZNK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE(0x9568de0, 8, 0xbfc59a58, 0x974c4f8, 0x946aca8) = 96
<... _ZNK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE resumed> ) = 96
<... _ZNK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE resumed> ) = 96
<... _ZNK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE resumed> ) = 96
<... _ZThn8_NK7QWidget6metricEN12QPaintDevice17PaintDeviceMetricE resumed> ) = 96
_ZThn8_NK7QWidget7devTypeEv(0x96da9c0, 0x9431fb8, 0x96da830, 0x1f5a844, 0x1f5a844) = 1
_ZN12QApplication6notifyEP7QObjectP6QEvent(0xbfc5b2a4, 0x96da9b8, ^C0xbfc59ccc, 0x15f2a7c, 0xbfc59ce8 <unfinished ...>
....

... looping all the time - so I guess this supports some problem with Qt libraries.

I'm aware this is a report for now obsolete versions of Ubuntu, but I thought it'd be best if it is logged; hope that's ok,
Cheers!

[sdaau (sd-imi)]

Forgot to say - what _does_ work for me, on natty at least, is - somewhat ironically - installing the Windows version of `kst` under `wine` (!) No CPU hog, and pretty decent responsivity - so that's what I'm using... Though - of course - given that the source seems to build nicely and all, of course I'd prefer to use a native build instead...

[dino99 (9d9)]

Natty has expired