otp.git, branch maint-r14b03

Prepare release

2011-07-11T13:40:44+00:00

Merge branch 'rickard/sbmbc/OTP-9339' into maint-r14

2011-07-11T13:40:42+00:00

* rickard/sbmbc/OTP-9339:
  Use separate memory carriers for small blocks

Merge branch 'sverker/ets_delete-deadlock-race/OTP-9423' into maint-r14

2011-07-11T13:40:41+00:00

* sverker/ets_delete-deadlock-race/OTP-9423:
  Fix bug in ets:delete for write_concurrency that could lead to deadlock

Merge branch 'sverker/trace-ms-invalid-op/OTP-9422' into maint-r14

2011-07-11T13:40:41+00:00

* sverker/trace-ms-invalid-op/OTP-9422:
  Fix a match-spec trace bug that could cause emulator crash

Use separate memory carriers for small blocks

2011-07-08T14:08:37+00:00

Fix a match-spec trace bug that could cause emulator crash

2011-07-07T15:38:08+00:00

A trace matchspec with 'enable_trace' or 'disable_trace' in body could
cause an emulator crash if a concurrent process altered the trace
setting of the traced function by calling erlang:trace_pattern.

The effect was a deallocation of the binary holding the matchspec
program while it was running. Fixed by increasing reference count of
ms-binary in the cases when 'enable_trace' or 'disable_trace' may
cause a system block that may alter the ongoing trace.

The paradox here is that db_prog_match() is using erts_smp_block_system()
to do 'enable_trace' and 'disable_trace' in a safe (atomic) way. But that
also have the (non-atomic) effect that racing thread might block the
system and change the trace settings with erlang:trace_pattern.

Fix bug in ets:delete for write_concurrency that could lead to deadlock

2011-07-07T15:33:05+00:00

Relocking in ets_delete_1() and remove_named_tab() was done by
unlocking the table without clearing the is_thread_safe flag. A racing
thread could then read-lock the table and then incorrectly
write-unlock the table as db_unlock() looked at is_thread_safe to
determine which kind of lock to unlock.

Several fixes:
1. Make db_unlock() use argument 'kind' instead of is_thread_safe to
   determine lock type.
2. Make relock logic use db_lock() and db_unlock() instead of directly
   accessing lock primitives.
3. Do ownership transfer earlier in ets_delete_1 to avoid racing owner
   process to also start deleting the same table.

Merge branch 'ia/ssl/session-table-clean-up/OTP-9346' into maint-r14

2011-07-06T10:17:35+00:00

* ia/ssl/session-table-clean-up/OTP-9346:
  The clean up of the session table now works as intended.

The clean up of the session table now works as intended.

2011-07-06T10:13:29+00:00

In ssl-4.1.5 temporary clean-up processes would crash resulting in
that the session table would not be cleaned up (e.i. using more and
more memory) and error reports would be printed, but connections would
not be affected.

Merge branch 'bmk/inets/inets57_integration2' into maint-r14

2011-06-22T12:16:22+00:00

* bmk/inets/inets57_integration2:
  Fixed non-related test case (ticket_6035).
  Corrected appup-file (missing ',').
  Fixed ipv6 support detection.
  (httpc) test case cleanups.
  Uppdated appup-file.
  Fixed IPv6 test case selection. That is if a IPv6 test case should be run or not.
  Stopping httpc client...
  Added test cases for httpd.
  Clients started stand-alone not properly handled.
  Clients started stand-alone not properly handled.
  Temporary solution for profile_name stuff. What about Pids???
  SSL with IPv6 now works "in principle".
  [httpc] Remove unnecessary usage of iolist_to_binary when processing body (for PUT and POST). Filipe David Manana OTP-9317
  Set proper version (5.7).
  Set proper version (5.7).
  Updated release notes.
  Peer/sockname resolv doesn't work with IPv6 addrs in HTTP. OTP-9343
  OTP-9342: FTP client doesn't work with IPv6 OTP-9342: IpFamily config option was not handled OTP-9342: Release notes remain... OTP-9342: attila rajmund nohl