Dominik Vogt <fvwm-workers@fvwm.org> writes: > On Thu, Dec 12, 2002 at 09:02:32AM -0500, Dan Espen wrote: > > Dominik Vogt <fvwm-workers@fvwm.org> writes: > > > On Wed, Dec 11, 2002 at 02:40:02PM -0500, Dan Espen wrote: > > > > Here are the UMRs: > > > > > > > > UMR: Uninitialized memory read: > > > > * This is occurring while in: > > > > __execute_function [functions.c:545] > > > > execute_function [functions.c:1176] > > > > StartupStuff [fvwm.c:1498] > > > > My_XNextEvent [events.c:3320] > > > > HandleEvents [events.c:3240] > > > > main [fvwm.c:2412] > > > > _start [crt1.o] > > > > * Reading 4 bytes from 0x2c7480 in the heap. > > > > * Address 0x2c7480 is 16 bytes into a malloc'd block at 0x2c7470 of 1 > 24 b > > > ytes. > > > > * This block was allocated from: > > > > malloc [rtlib.o] > > > > calloc [rtlib.o] > > > > safecalloc [safemalloc.c:64] > > > > exc_create_null_context [execcontext.c:96] > > > > exc_create_context [execcontext.c:117] > > > > StartupStuff [fvwm.c:1465] > > > > My_XNextEvent [events.c:3320] > > > > HandleEvents [events.c:3240] > > > > main [fvwm.c:2412] > > > > _start [crt1.o] > > > > > > I don't understand this one. What code do you have in > > > functions.c, line 545? > > > > if (expaction[0] == '*') > > Hm. Perhaps it's because the string is somthing between 13 and 15 > characters, but expaction[0] reads four bytes?
If the compiler caused code to be generated that accessed 4 bytes, I think it generated bad code. If you don't see anything wrong then I think you should forget it. Its not a perfect tool. > > > > UMR: Uninitialized memory read: > > > > * This is occurring while in: > > > > _SetFocusWindow [focus.c:922] > > > > > > > > UMR: Uninitialized memory read: > > > > * This is occurring while in: > > > > DrawIconPixmapWindow [icons.c:1012] > > > > > > Should both the "fixed". They are not really UMRs. Purify seems > > > to be unable to handle btructs with bit fields properly. > > > > Thats right, there are a few things Purify gets wrong. > > I'm not surprised that it doesn't track things at the bit level. > > Maybe using bit fileds isn't this clever at all. Sure it saves > several hundred bytes per window, but we pay this with additional > code and slower execution. If I remember correctly, we checked the code generated by the compilers and bit field access generated code comparable to byte access. -- Dan Espen E-mail: [EMAIL PROTECTED] -- Visit the official FVWM web page at <URL:http://www.fvwm.org/>. To unsubscribe from the list, send "unsubscribe fvwm-workers" in the body of a message to [EMAIL PROTECTED] To report problems, send mail to [EMAIL PROTECTED]
