|[4 earlier articles]|
|Debuggers firstname.lastname@example.org (1991-08-08)|
|How long does it take to build a compiler? email@example.com (Youngwhan Lee) (1993-10-27)|
|Re: debuggers firstname.lastname@example.org (1993-11-11)|
|Re: debuggers email@example.com (1993-11-11)|
|Re: debuggers firstname.lastname@example.org (1993-11-11)|
|debuggers email@example.com (1993-11-11)|
|Re: debuggers firstname.lastname@example.org (1993-11-11)|
|Re: debuggers sean@PICARD.TAMU.EDU (1993-11-12)|
|Re: debuggers email@example.com (1993-11-14)|
|Re: debuggers firstname.lastname@example.org (1993-11-15)|
|From:||email@example.com (Bill Sommerfeld)|
|Organization:||HP Apollo Systems Division, Chelmsford, MA.|
|Date:||Thu, 11 Nov 1993 18:33:41 GMT|
firstname.lastname@example.org (Shyamal Prasad) writes:
gdb on running on a DEC Station 5000 under ULTRIX allows watch points.
However, it does *not* use the hardware and runs about 10 to a 100
I have used them one time when I was really sunk...they *do* work.
... assuming you're lucky and the bug in question isn't timing
dependant (generally not an issue if you're debugging a compiler, but
often an issue for other things..).
BTW, if you have some idea of where the error is taking place, you can
often significantly increase the effectiveness of watchpoints/ breakpoints
by chaining breakpoints together using gdb's enable/disable commands
*from* commands executed at breakpoints or conditional breakpoints...
Return to the
Search the comp.compilers archives again.