|
Re: Why try to reuse breakpoint structures?: msg#00076linux.kernel.debugging.kgdb.bugs
Amit Kale wrote: On Tuesday 10 Jan 2006 6:15 pm, Amit Kale wrote:I wonder if this is the best way to do this. On the x86, at least, the BP instruction causes a unique trap into KGDB. So, it seems to me a check, early on, if such a trap occured AND there is no BP instruction at the fault location should do the trick. The reason I bring this up is that it is not clear that we can EVER remove an old BP address from the list, i.e. we just don't know how long it will take a given cpu to finally get its BP serviced. I suppose we could keep track of cpu entries to KGDB since the BP was removed, but this seems rather complex. George Removed breakpoints can be placed again, in which case they can't be ignored on other processors. That's why we reuse those entries. -- George Anzinger george@xxxxxxxxxx HRT (High-res-timers): http://sourceforge.net/projects/high-res-timers/ ------------------------------------------------------- This SF.net email is sponsored by: Splunk Inc. Do you grep through log files for problems? Stop! Download the new AJAX search engine that makes searching your log files as easy as surfing the web. DOWNLOAD SPLUNK! http://ads.osdn.com/?ad_id=7637&alloc_id=16865&op=click |
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| Previous by Date: | Re: kgdb on arm for 2.6: 00076, George Anzinger |
|---|---|
| Next by Date: | Re: kgdb on arm for 2.6: 00076, Tom Rini |
| Previous by Thread: | Re: Why try to reuse breakpoint structures?i: 00076, Amit Kale |
| Next by Thread: | Re: Why try to reuse breakpoint structures?: 00076, Jim Blandy |
| Indexes: | [Date] [Thread] [Top] [All Lists] |
| News | FAQ | advertise |