At 08:47 AM 4/22/2001 -0700, Eric D. Friedman wrote:
>
>Using FSF emacs (will try xemacs shortly) and 2.2.7b10, I was able to
>debug the test code I wote about earlier.
>
><theory>
>I was interested to learn that previous releases of JDEbug had been
>setting a debug thread to max priority; I'm testing on linux and on that
>platform native threads are implemented as processes. I did notice
>(but did not report) that emacs became quite sluggish when I started
>JDEbug, which might meant that it was having trouble getting CPU cycles.
>You may recall that earlier yesterday, with the same release, I was
>able to debug my application. At that time I had many more processes
>running concurrently on my system than I did when I could not debug later
>that same day. It worked, in other words, when the scheduler wouldn't
>let the CPU be monopolized by the MAX_PRIORITY thread, and hung up when
>there weren't any non-emacs/jde processes around to challenge "mad max."
></theory>
>
Eric, I am a little confused. Are you saying that my thread priority "fix"
in beta10 is an improvement? If so, that is very interesting. I believer
there other debugger threads that JDEbug sets to max priority (actually
MAX_PRIORITY-1). I've never understood why the code, which was written by a
Sun summer intern, was setting the priority of these threads. However,
since the intern was working as a member of the JPDA team, I assumed he had
access to inside information so I was reluctant to touch his code in this
area. Recently, however, I have been reading Holub's "Taming Java Threads"
and found his discussion of how thread scheduling differs drastically from
platform to platform enlightening and horrifying. Holub's advice is
basically just not to monkey with thread prioirty since you can't set
priorities in a platform-independent way. So I'm inclined to comment out
all the set-priority calls in the JDEbug code.
>So, now that I can do my `basic' debug test, I'm trying some of
>the debugging features that I haven't used before and am having new
>problems. :-)
>
>I attempted to `watch' the fields in my test class and entered the fully
>qualified classname `test.Main' into the appropriate text entry field.
>I got the following error:
>
>Error: unable to enable watch request.
> Reason: '-sp' not understood: use either 'for_access' or 'for_modification'.
>
>in the *JDEbug* buffer, this corresponds to the following failed instruction:
>
>JDE> 1 10 watch test.Main for_modification -sp none
>
>(jde-dbo-command-error
>10 "'-sp' not understood: use either 'for_access' or 'for_modification'")
>
Thanks. I'll look into this right away. I really appreciate your detailed
and highly informed feedback, Eric.
- Paul
