On 10/10/05, Alan DuBoff <[EMAIL PROTECTED]> wrote: > How about an apples to apples comparision on same-like systems?
Being fortunate enough in having to deal with this comparison on a daily basis, i would like to share with this forum a few of my relevant experiences. i consider them relevant because they have changed me. i am a better person now, of stronger patience, will, and character. 1. (gdb) where ?? (gdb) Weeeeeeee!!! :-) No Stack! Cool! No Stack -- No Problem! 2. (gdb) break foo::bar::baz (gdb) run <flies past breakpoint like a bat out of hell> Hmmm. doesnt work. i guess you have to do the song and dance of setting a breakpoint in main() first. then you can set other breakpoints. maybe. 3. get a core dump from an app built on one kernel version, and try to load it in gdb on a *slightly* different kernel version (maybe because this other box has more RAM and it's not used by so many people). nono. does not work. My real life experience has also taught me a few simple facts: 1. noone cares about spec numbers, in production systems. the only thing "they" care about is for the software to run reliably, and to do what it is supposed to do, when it is supposed to do it. 2. if i ain't got a working debugger, i ain't got working production quality software. 3. an operating system should really not be an application of the Principle Of Maximum Astonishment. 4. disk I/O latency is really not an acceptable form of thread synchronization. Just my 0.02. --Stefan -- Stefan Teleman [EMAIL PROTECTED] _______________________________________________ opensolaris-discuss mailing list opensolaris-discuss@opensolaris.org
