Re: cvs, buffered_nextbyte

[Sam S. <sd...@gn...>]

> * Don Cohen <qba...@vf...3-vap.pbz> [2003-09-30 18:43:05 -0700]:
>
> Sam Steingold writes:
>  > > (How does errno work for multiple threads,
>  > #define errno *thread_specific_errno_address()
> 
> As you see from the previous message I have yet to actually test this
> stuff but my current implementation (hack?) is to just rely on errno.
> Is this code correct?  How would you rather see it done?
> Can I do
>  #define myerrno *thread_specific_errno_address()
> And if so, where should I put it?

no, the system header that defines errno (be it errno.h or stdlib.h)
will define it correctly for your platform.
e.g., modules/bindings/glibc/linux.lisp has a lot of problems making
errno available because of that.
the flip side is that in C you do not have to worry about errno.

> Of course, the really ugly part of this solution is that you have to
> worry about what other code you execute between this and the later
> check.  At the moment I think it's ok:
> 
>  buffered_nextbyte
> 
>       result = BufferedStreamLow_fill(stream)(stream,no_hang);
>      if (errno == EAGAIN){
>         printf("buffered_nextbyte returning -1\n"); //
>         return -1; /* hang case */
>         }

this looks kind of yuky.
maybe BufferedStreamLow_fill could return -1 instead of relying on
errno?

Bruno?

> local uintL low_fill_buffered_handle (object stream, bool no_hang) {
>   var sintL result = 0;
>   var Handle handle = TheHandle(BufferedStream_channel(stream));
>   if (!no_hang || ls_avail_p(listen_handle(handle,false,NULL))) {
>     begin_system_call();
>     result = read_helper(handle,BufferedStream_buffer_address(stream,0),
>                          strm_buffered_bufflen,no_hang);
>     end_system_call();
>     if (result<0) # error occurred?
>       OS_filestream_error(stream);
>   }
>   return result;
> }
> 
> local uintL low_fill_buffered_socket (object stream, bool no_hang) {
>   var sintL result;
>   /* *** doesn't use no_hang? */
>   SYSCALL(result,sock_read(TheSocket(BufferedStream_channel(stream)),
>                            BufferedStream_buffer_address(stream,0),
>                            strm_buffered_bufflen));
>   return result;
> }



-- 
Sam Steingold (http://www.podval.org/~sds) running w2k
<http://www.camera.org> <http://www.iris.org.il> <http://www.memri.org/>
<http://www.mideasttruth.com/> <http://www.honestreporting.com>
If I had known that it was harmless, I would have killed it myself.