[PATCH] (Re: Query concerning use of flockfile/funlockfile in stdio FILE related functions)

[Jeff Johnston]

Patch ripped apart and checked in bit by bit.  In the future, please do not 
submit multiple patches in one.  How you broke them up in the ChangeLog was 
fine.  Also please do not include diffs to the ChangeLog;  just put the 
ChangeLog entry in your posting.  Thanks.

-- Jeff J.

Antony KING wrote:
> Please find attached a patch to newlib (CVS snapshot 10-Sep-2004) which 
> solves the problem I was having with being able to use sprintf/sscanf et 
> al from RTOS contexts where using locking was not permissible and where 
> no REENT structure exists.
> 
> Please note that I have re-enabled the definition of _GLOBAL_REENT to be 
> _global_impure_ptr when not CYGWIN; I hope this compromise is OK.
> 
> I have also made a number of other minor changes which I also hope are 
> OK (missing documentation, missing initialisations, default tz_lock 
> implementation a la env_lock, walking GLOBAL_REENT multiple times).
> 
> Cheers,
> 
> Antony.
>