Arup Mukherjee wrote:
> Hi,
>
> under the 2.1.x kernels, large write()'s to a socket never come
> back if they block, unless I/O occurs on another socket. I've seen
> this with 1.1.3 thru 1.1.6v2, with both the Steve Byrne and the Sergei
> Nikitin ports under at least 2.1.65 and 2.1.114. The problem is VERY
> reproducible and does not occur under the 2.0.x kernels.
>
This may not be related, but I am also experiencing extremely odd behavior with
sockets running Java v1.1.6v2 JRE, kernel 2.0.30, AMD 586 (? runs more like a
386 - stay far away from this chip), libc 5.4.44. I have built a client-server
system for multiplayer games and what happens is mainly that neither select(),
read() nor write() to a socket can reliably tell whether the other system is
even present or not. The Ethernet chip is a PCI version of NE2000. I can't
tell whether this is a driver problem, a Java problem (the infamous green
threads) or what. When I run client and server on the same machine, the sockets
work correctly; i.e.: when a client exits, the server immediately gets a broken
pipe on write() and eof on read; same for client if the server exits.
Thanks in advance for any help,
Peter Johnson
[EMAIL PROTECTED]
