Rainer Toebbicke wrote:
> No, I'm afraid that's not sufficient: after dropping and re-acquiring
> the locks it may well be that the rxcon_client has changed again.
> 
> Therefore you need the
> 
> while (... && ... && ...) / continue instead of the if (...) construct.
> 
> I came up with the patch from the 1.4.1 base and wasn't aware
> of the rxr_CidOf() changes.

I don't believe this is an issue.  I do not see a scenario in which
rx_SetSpecific(tcon, rxcon_client_key, client) can be called with
any value other than 'oldClient' if the 'oldClient' exists and is
not deleted.

Jeffrey Altman


Attachment: smime.p7s
Description: S/MIME Cryptographic Signature

Reply via email to