Factor/Bug reports

If you find a bug in Factor, please report it with as much detail as possible by editing one of the pages above. We will work with you to ensure that the bug is resolved.

Known problems

Sometimes, bugs in compilers and operating systems cause problems for Factor. Generally, there is nothing we can do about such problems. Before reporting a bug in Factor, consult the below list.

Widcomm Bluetooth drivers

Several driver vendors inject SEH into other processes causing a problem with the way the current VM handles exceptions. Widcomm has been identified as being one of these vendors across multiple laptop brands.

windows seh issues -- < mnestic> erg: it's definitely BTTray.exe from the widcomm driver set


  • Kill BTTray.exe in Task Manager
  • Once BT device is setup, msconfig can be used to disable BTTray.exe startup (Start | Run | msconfig -> Startup Tab)
12:42:25 <doublec> sqiurid: edit windows.factor
12:42:30 <doublec> change the peek-message? definition to
12:42:31 --- join: thecardinal (n=peter@chello212017088160.1.15.vie.surfer.at) joined #concatenative
12:42:34 <doublec> : peek-message? ( msg -- ? ) f 0 0 PM_REMOVE [ PeekMessage ] [ 4drop 0 ] recover ;
12:42:41 <doublec> bootstrap again
12:42:44 <doublec> and things should work
12:43:10 <doublec> by bootstrap I mean, run from a cmd prompt
12:43:17 <doublec> factor.exe -i=boot.x86.32.image
12:43:21 <mnestic1> sqiurid: F2 wouldn't work?
12:43:34 <mnestic1> doublec: i mean
12:43:46 <doublec> mnestic1: any keypress caused a failure
12:43:56 <British0zzy> is there anyway to view the second stack in the interpreter
12:43:59 <doublec> putting it in .factor-rc might work
13:16:57 <sqiurid> doublec: there is a space between 4 and drop in "[ PeekMessage ] [ 4drop 0 ] recover", yeah?
13:17:18 <doublec> sqiurid: no
13:17:22 <mnestic1> sqiurid: it used to have an interpreter: http://factor-language.blogspot.com/2007/09/two-tier-compilation-comes-to-factor.html
13:17:28 <doublec> if 4drop doesn't exist use drop drop drop drop
13:17:32 --- join: jewel (n=jewel@dsl-242-172-01.telkomadsl.co.za) joined #concatenative
13:17:40 <sqiurid> doublec: ok, thanx

GCC bugs

  • Some builds of GCC 4.3 have problems compiling Factor on x86 and x86.64. Change the CFLAGS in the Makefile from -O3 to -O1 to work around the issue.
  • GCC 3.3 and earlier cannot be used to compile Factor on x86 due to bugs in the register allocator.

Video driver bugs

  • If you are running Windows with an Intel 9xx graphics adapter, you may notice rendering artifacts in the Factor UI. This is due to bugs in Intel's OpenGL implementation. There is no workaround.
  • You might get this error when running the Factor UI on Ubuntu Linux 8.04, after which the UI will crash:
File r300_mem.c function r300_mem_alloc line 225
Ran out of GART memory (for 1048576)!
Please consider adjusting GARTSize option.
Error: Could not get dma buffer... exiting

The problem is caused by a bug in the Ubuntu X11 configuration tool; any OpenGL program will be affected and there is nothing Factor can do about it. A fix is discussed at https://bugs.launchpad.net/ubuntu/+source/stellarium/+bug/200173.

Kernel bugs

  • User Mode Linux has a kernel bug which affects Factor -- Factor may hang instead of reporting certain errors. More details
  • Older glibc releases exported outdated struct layouts on Linux/ARM. If you get a compile error in vm/os-linux-arm.h, then you need to either upgrade your glibc or patch a header file, as outlined in this weblog entry

This revision created on Tue, 4 Nov 2008 21:18:15 by mnestic