[Bug server/17148] New: problem in transition to noack mode

[Bug server/17148] New: problem in transition to noack mode

[xdje42 at gmail dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=17148

            Bug ID: 17148
           Summary: problem in transition to noack mode
           Product: gdb
           Version: HEAD
            Status: NEW
          Severity: normal
          Priority: P2
         Component: server
          Assignee: unassigned at sourceware dot org
          Reporter: xdje42 at gmail dot com

I thought I had filed a bug for this, but searching for "noack" returns zero
bugs.  I'm not sure this has been fixed yet, so filing this so it's not
dropped.

If debugging a remote target that is a fair distance from gdb, an appropriately
long delay during the transition to noack mode causes problems (connection
doesn't get established).

It's easy enough to see with a hacked gdb.
Need to submit an example.

ref:
https://www.sourceware.org/ml/gdb-patches/2010-04/msg00884.html

"btw, after fixing this I found another bug: timeout handling in noack mode.
What happens if gdb times out waiting for the ack after sending
QStartNoAckMode?  gdb and gdbserver will again get out of sync.
I think the solution there is to either have no timeout in noack mode
or at least have a *much* longer timeout than the default of 2 seconds."

-- 
You are receiving this mail because:
You are on the CC list for the bug.