qpid-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Kenneth Giusti" <kgiu...@apache.org>
Subject Re: Review Request 36098: Prevent pn_data_scan from stepping on default value.
Date Wed, 01 Jul 2015 18:56:50 GMT


> On July 1, 2015, 6:47 p.m., Ted Ross wrote:
> > I hate to say this, but shouldn't you just handle the defaults for max-frame-size
and idle-time-out while you're in there?  You're doing the timing belt, may as well change
out the water pump while you're in there.

that may not be necessary - remote_max_frame and remote_idle are initialized to zero locally,
and they are only considered provided by the remote if they are non-zero after the open arrives.


- Kenneth


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/36098/#review90100
-----------------------------------------------------------


On July 1, 2015, 6:02 p.m., michael goulish wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/36098/
> -----------------------------------------------------------
> 
> (Updated July 1, 2015, 6:02 p.m.)
> 
> 
> Review request for qpid, Gordon Sim and Kenneth Giusti.
> 
> 
> Repository: qpid-proton-git
> 
> 
> Description
> -------
> 
> PROTON-925
> Use a "this-variable-was-actually-scanned" flag to see if the value returned by pn_data_scan()
for remote_channel_max is real.  If it's not real don't use it.
> 
> 
> Diffs
> -----
> 
>   proton-c/src/transport/transport.c 2271f27 
> 
> Diff: https://reviews.apache.org/r/36098/diff/
> 
> 
> Testing
> -------
> 
> ctest -VV
> 
> and also -- the way gsim found this -- use a sender based on this new code against a
receiver based on 0.9 or older code.  WIthout this fix, the sender will hallucinate a "0"
from the receiver because ... pn_data_scan() overwrites addresses for which it received no
value.
> 
> 
> Thanks,
> 
> michael goulish
> 
>


Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message