Quickly jump to event?

Robert O'Callahan robert at ocallahan.org
Wed May 24 12:34:59 UTC 2017

On Wed, May 24, 2017 at 6:48 PM, Benjamin King <benjaminking at web.de> wrote:

> Ok. I was hoping that I could achieve something like that without creating
> the checkpoint in the first place. Maybe this is a weird request. My mental
> model of how rr works is still mostly "hardcore engineering + lots and lots
> of pixie dust" so I have no intuition on what is easy or impossible to do.
> And I have already watched most of your talks about it :-)

restart is fast because restoring a checkpoint is fast. Going to an event
where you don't have a checkpoint necessarily takes time proportional to
the distance to the previous checkpoint (or start of trace).

I'll add just a few checkpoints then and navigate between them with restart.
> Your suggestion to keep track of them with eyeballing the event number from
> 'info checkpoints' will be helpful, thanks!
> One feature that really should be added to rr itself is the ability to run
>> forward or backward to a specific event without restarting.
> Yes, or to set breakpoints at events. Maybe a hamfisted approach by
> setting a
> conditional breakpoints at the event location + checking the event number
> would already work?

The problem there is that you can't easily use the current event number in
an expression. Maybe we should expose the current event number as an extra
pseudo-register to enable that.

Supporting actual breakpoints on events is a bit tricky because we'd have
to provide a stop reason; whatever we do there might confuse gdb. Also we'd
need to find a way to easily let you continue past the breakpoint.

lbir ye,ea yer.tnietoehr  rdn rdsme,anea lurpr  edna e hnysnenh hhe uresyf
selthor  stor  edna  siewaoeodm  or v sstvr  esBa  kbvted,t
o l euetiuruewFa  kbn e hnystoivateweh uresyf tulsa rehr  rdm  or rnea
.a war hsrer holsa rodvted,t  nenh hneireseoouot.tniesiewaoeivatewt sstvr
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.mozilla.org/pipermail/rr-dev/attachments/20170525/6a299651/attachment.html>

More information about the rr-dev mailing list