diff options
| author | Olli Savia <ops@iki.fi> | 2004-05-29 14:17:58 +0000 |
|---|---|---|
| committer | Olli Savia <ops@iki.fi> | 2004-05-29 14:17:58 +0000 |
| commit | 3c2d57b656f475f4c19ae1fe9234838493b2ef36 (patch) | |
| tree | 516581d17ee9dc746ae7e1398c92a724cb364761 | |
| parent | 0bb49b584e85047e60f3c08ae8f75e39ee193aa0 (diff) | |
| download | ATCD-3c2d57b656f475f4c19ae1fe9234838493b2ef36.tar.gz | |
ChangeLogTag: Sat May 29 16:12:28 2004 Olli Savia <ops@iki.fi>
| -rw-r--r-- | ChangeLog | 36 | ||||
| -rw-r--r-- | include/makeinclude/platform_lynxos.GNU | 8 |
2 files changed, 29 insertions, 15 deletions
diff --git a/ChangeLog b/ChangeLog index e13deac051a..76166bf9738 100644 --- a/ChangeLog +++ b/ChangeLog @@ -1,26 +1,32 @@ +Sat May 29 16:12:28 2004 Olli Savia <ops@iki.fi> + + * include/makeinclude/platform_lynxos.GNU: + + Added X11 related libs needed to build XtReactor. + Fri May 28 18:35:41 2004 Steve Huston <shuston@riverace.com> - * tests/FIFO_Test.cpp (server): The select() trick to wait for the - writer to open the FIFO doesn't seem to work on AIX. It seems to - select() the fifo handle as a normal file handle, always readable. - So on AIX, sleep 1 second before proceeding. + * tests/FIFO_Test.cpp (server): The select() trick to wait for the + writer to open the FIFO doesn't seem to work on AIX. It seems to + select() the fifo handle as a normal file handle, always readable. + So on AIX, sleep 1 second before proceeding. Fri May 28 15:54:26 2004 Steve Huston <shuston@riverace.com> - * tests/MT_Reference_Counted_Event_Handler_Test.cpp: In - Close_Socket_Thread::svc(), do a ACE_OS::shutdown() before closing - the socket. This is enough to pop another thread off a blocked - read/write on platforms that block the close() until an in-progress - operation completes, such as AIX and Linux. With this, remove the - #if linux checks that disabled some types of test. - It could be argued that the shutdown() should be done as part of - ACE_OS::closesocket() (it already does this for HP-UX), but I - hesitate to rock the boat here... + * tests/MT_Reference_Counted_Event_Handler_Test.cpp: In + Close_Socket_Thread::svc(), do a ACE_OS::shutdown() before closing + the socket. This is enough to pop another thread off a blocked + read/write on platforms that block the close() until an in-progress + operation completes, such as AIX and Linux. With this, remove the + #if linux checks that disabled some types of test. + It could be argued that the shutdown() should be done as part of + ACE_OS::closesocket() (it already does this for HP-UX), but I + hesitate to rock the boat here... Thu May 27 19:02:18 2004 Steve Huston <shuston@riverace.com> - * ace/Timer_Hash_T.cpp (schedule_i): For Win64, reinterpret_cast from - ptrdiff_t to long is incorrect; static_cast is more appropriate. + * ace/Timer_Hash_T.cpp (schedule_i): For Win64, reinterpret_cast from + ptrdiff_t to long is incorrect; static_cast is more appropriate. Thu May 27 20:00:29 2004 Olli Savia <ops@iki.fi> diff --git a/include/makeinclude/platform_lynxos.GNU b/include/makeinclude/platform_lynxos.GNU index e3309b575fc..717c4d25b04 100644 --- a/include/makeinclude/platform_lynxos.GNU +++ b/include/makeinclude/platform_lynxos.GNU @@ -97,6 +97,14 @@ else ACE_HAS_GNUG_PRE_2_8 = 0 endif # VERSION +PLATFORM_XT_CPPFLAGS= +PLATFORM_XT_LIBS=-lXm -lXt +PLATFORM_XT_LDFLAGS= + +PLATFORM_X11_CPPFLAGS= +PLATFORM_X11_LIBS=-lXpm -lXext -lX11 -lSM -lICE +PLATFORM_X11_LDFLAGS= + pipes ?= 1 CC = gcc |