Apache OpenOffice (AOO) Bugzilla – Issue 1298
Build 633: Installation on Linux 2.4.0 / XFree86 4.1.0 fails
Last modified: 2004-11-25 17:36:11 UTC
I tried to install Build 633 from the Linux binary package. But running from the commandline only shows the message: "glibc version: 2.2". After that the setup simply terminates. There is not even an error message. I have a glibc 2.2-7 installed on my machine. It's a SuSE 7.1 with Kernel 2.4.0 and XFree86 4.1.0, KDE 2.1.1 (both from SuSE rpm packages). The Kernel is patched to allow the use of DRI with AGP cards. I'm using a Voodoo 3 AGP card with the tdfx.o driver. DRI is enabled.
As far as I can see it seems to be a problem with glibc newer than 2.2.2. I use mandrake8, kernel 2.4.3-20, kde 2.1.1 and xfree 4.0.3 => no problem. *** This issue has been marked as a duplicate of 916 ***
I'm not using a glibc newer than 2.2.2. Actually it is an 2.2. But I have been successful when installing OO Build 633 on a SuSE Linux 7.1 with XFree86 4.0.3 (and running glibc 2.2). So I think it must be a problem with the new XFree86 4.1.0.
Hi Olli, seems to be a problem with xfree 4.1.0
Using build 633 Debian testing on XFree86 4.0.2 install will depending on how truetype is installed. Using the old XF86Config with FontPath it installs fine. If I install using the new XF96Config-4 method (http://www.paulandlesley.org/linux/xfree4_tt.html ) I get : failProgram received signal SIGSEGV, Segmentation fault. 0x412e420b in free () from /lib/libc.so.6 (gdb) where #0 0x412e420b in free () from /lib/libc.so.6 #1 0x415bd55c in CloseTTFont () from /tmp/sv001.tmp/./libpsp633li.so #2 0x41595a6a in psp::PrintFontManager::analyzeTrueTypeFile () from /tmp/sv001.tmp/./libpsp633li.so #3 0x415934c9 in psp::PrintFontManager::analyzeFontFile () from /tmp/sv001.tmp/./libpsp633li.so #4 0x415970d4 in psp::PrintFontManager::initialize () from /tmp/sv001.tmp/./libpsp633li.so #5 0x415910e4 in psp::PrintFontManager::get () from /tmp/sv001.tmp/./libpsp633li.so #6 0x4041fd7b in FontLookup::BuildSet () from /tmp/sv001.tmp/./libvcl633li.so #7 0x4042047f in SalDisplay::GetXlfdList () from /tmp/sv001.tmp/./libvcl633li.so #8 0x40423047 in SalGraphics::GetDevFontList () from /tmp/sv001.tmp/./libvcl633li.so #9 0x403ae913 in Window::ImplInit () from /tmp/sv001.tmp/./libvcl633li.so #10 0x40381fc6 in ImplBorderWindow::ImplInit () from /tmp/sv001.tmp/./libvcl633li.so #11 0x403820bc in ImplBorderWindow::ImplBorderWindow () from /tmp/sv001.tmp/./libvcl633li.so #12 0x403c04c4 in WorkWindow::ImplInit () from /tmp/sv001.tmp/./libvcl633li.so #13 0x403c0739 in WorkWindow::WorkWindow () from /tmp/sv001.tmp/./libvcl633li.so ---Type <return> to continue, or q <return> to quit--- #14 0x402f216f in ImplGetDefaultWindow () from /tmp/sv001.tmp/./libvcl633li.so #15 0x402f11e2 in Application::EnterMultiThread () from /tmp/sv001.tmp/./libvcl633li.so #16 0x805531e in SetupApp::Main () #17 0x402f2480 in SVMain () from /tmp/sv001.tmp/./libvcl633li.so #18 0x404289d5 in main () from /tmp/sv001.tmp/./libvcl633li.so #19 0x412902db in __libc_start_main () from /lib/libc.so.6
Now the installation works on SuSE 7.2 with Build 638. I'm using (a patched) Kernel 2.4.4 with XFree86 4.1.0 and KDE 2.2.
*** Issue 1807 has been marked as a duplicate of this issue. ***
Herbert you are working on this.
There is also a bug in libXrender of 4.1.0, which causes a similar crash. It seems to occur only for some Xfree 4.1.0 drivers for specific display settings (Rage128 under PPC with 24bit), so changing the settings e.g. to the fbdev driver helped for 4.1.0. Details of the libXrender bug are below: > Will you please look at the following diff from Xrender.c in XF 4.0.2 to > Xrender.c in XF410. It looks very strange. Starting in lie 190. > It seems they no longer Xmalloc a value for xFormat (they now use xData) > but just one line later they check if xFormat is zero. > The code before that in XRenderQueryFormats() doesNOT set a value for > xFormat at all after its is declared and before it is tested. > I think someone just missed changing and xFormat to an xData unless I am > missing something here. > Kevin > >@@ -190,19 +191,20 @@ > rep.numScreens * sizeof (xPictScreen) + > rep.numDepths * sizeof (xPictDepth) + > rep.numVisuals * sizeof (xPictVisual)); >- xFormat = (xPictFormInfo *) Xmalloc (rlength); >+ xData = (void *) Xmalloc (rlength); > > if (!xri || !xFormat) > { > if (xri) Xfree (xri); >- if (xFormat) Xfree (xFormat); >+ if (xData) Xfree (xData); > _XEatData (dpy, rlength); > UnlockDisplay (dpy); > SyncHandle (); > return 0;
The stacktrace also hints to an issue fixed by PL? Reassigning to Philipp.
This specific problem was already fixed on they way to SO6 beta. The problem occured with certain corrupt fonts which would lead to heap corruption. Fixed in 638.
closing