Patch to fix glibc 2.1.2 so that new semaphore functions work in high signal env.

[Kevin Hendricks <khendricks@ivey.uwo.ca>]

[-- Attachment #1: Type: text/plain, Size: 572 bytes --]

Hi,

I had to make the following patch to get glibc 2.1.2 new semaphores to work in
high signal environments.  For some reason calls to sem_post made by the
pthread manager thread were *NOT* resolving to the __new_sem_post but instead
to somewhere else (__old_sem_post or to kernel sem_post ????).

I don't know why (the correct fix might be somewhere else) but this does the
trick.

The pthread manager only does the sem_post calls when the original thread was
in a signal handler.

I am not sure this will help others, but it does wonders for the JDK!!!

Thanks,

Kevin

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: patch to pthreads manager.c --]
[-- Type: text/x-c; name="glibc.patch", Size: 340 bytes --]

--- manager.c.prev	Thu Oct  7 23:40:04 1999
+++ manager.c	Thu Oct  7 23:40:47 1999
@@ -156,7 +156,7 @@
         }
         break;
       case REQ_POST:
-        sem_post(request.req_args.post);
+        __new_sem_post(request.req_args.post);
         break;
       case REQ_DEBUG:
 	/* Make gdb aware of new thread and gdb will restart the