Hi!
apr_thread_join for win32 is implemented incorrectly:
thread handle is destroyed too early (in apr_thread_exit).
If apr_thread_exit() is called before apr_thread_join() and
new object is created (thread handle is reused) before
calling apr_thread_join(), apr_thread_join() will possibly wait
on
--On Friday, July 30, 2004 8:33 AM -0400 Ryan Bloom <[EMAIL PROTECTED]> wrote:
However, we need to remove the APR_STATUS_IS_SUCCESS macro before 1.0 goes
out, because otherwise we are stuck with it for a very long time.
It's gone now. ;-)
I've got to fix up httpd 2.1 now and then I'll review Max'
this never showed up for some reason???
--- Begin Message ---
A couple of questions come up from an application perspective:
am I leaking memory? if so, on what operation?
how much memory does it take to perform a certain operation?
If the application can find out how much heap memory is presently
The attached replacement patch differs from the previous version only in
that a missing acceptable-majors parameter is now reported earlier
(autoconf-time, rather than configure-time).
IIRC, this, and APR_STATUS_IS_SUCCESS are the only remaining API issues
requiring resolution before 1.0.0 final.
On Jul 30, 2004, at 7:33 AM, Ryan Bloom wrote:
I don't understand why this is still being discussed. The patch makes
sense, it solves a real problem and just needs to be committed and
tested.
+1 on adding it to 1.0.0RC5 so that we can get the release out.
However, we need to remove the APR_STATU
While porting Subversion to OS/2 I came across a failure where it tries to
open a file using only APR_READ | APR_CREATE. This results in an error on
OS/2, it doesn't seem to like the idea of creating a new file with
read-only access.
Is this supposed to work or is it just lucky that happens to wo
