89cbb95c2c
Merge appveyor changes from 3.14
2017-12-19 09:52:15 +01:00
006ce1a240
appveyor-ci: remove VS 2008, add VS 2017
2017-12-19 09:39:57 +01:00
276dee2c3e
appveyor-ci: use curl for download of make.zip
2017-12-19 09:39:10 +01:00
98a2871727
Reset SNAPSHOT to -DEV after tagging 3.14.12.7
2017-12-15 16:09:21 -06:00
5ca1bb3bd5
Set SNAPSHOT for 3.14.12.7 final release
2017-12-15 16:06:08 -06:00
f6be3c7f70
Set snapshot to -rc1-DEV
2017-12-08 16:40:08 -06:00
00924dcba0
About to tag 3.14.12.7-rc1
2017-12-08 16:37:03 -06:00
3b0f34e0be
Fix recursive check in Release.pm
2017-12-06 16:29:31 -06:00
db005ac3fe
RULES_TARGET: Define $(lib)_DIR before it is needed
...
TESTLIBRARY products were being installed, because an executable
that was linked to one added a dependency assuming it would be in
the INSTALL_LIB directory. The $(lib)_DIR was being set too late
for test libraries, so the fallback location was used instead.
This fix sets $(lib)_DIR for test libraries before any rules use it.
2017-11-17 14:55:50 -06:00
0f0deb8903
Merge fixups/db_close_events into 3.15
2017-11-15 21:23:28 -06:00
baf997193e
Merge aSubRecord fixes from 3.14 into 3.15
2017-11-15 21:15:51 -06:00
5278799575
aSubRecord: Copy the right amount of data
...
Initialize OVLx from VALx with NEVx elements, size from FTVx
2017-11-15 20:59:06 -06:00
031e755303
installEpics.pl: Make temporary install filename more different
...
A parallel build was seen matching the temporary name in a wildcard.
Now the temporary name both starts and ends differently.
2017-11-11 18:00:57 -06:00
e794639e31
ioc/db: avoid possible race in db_close_events()
...
lp:1730982
2017-11-08 14:42:08 -06:00
2af98c33c9
std/rec: aSub handle INAM changing outputs
...
Allow EFLG="On Change" to work with INAM.
2017-11-08 14:29:30 -06:00
e38252eecc
Fix osiSockOptMcastLoop_t for solaris
2017-10-26 16:29:03 -05:00
31c7d91884
libCom/test/Makefile: Clean up SYS_LIBS var's
...
Add PROD_SYS_LIBS_solaris global setting
Remove settings for individual tests
2017-10-26 16:28:24 -05:00
42d37eb19a
Merge branch 'osiSockTest' into 3.15
2017-10-26 14:30:38 -05:00
cec4f4e24b
Unify rsrv socket errors, add missing error string
2017-10-20 14:12:49 -05:00
58460d25b9
Release Notes
2017-10-13 10:57:21 +02:00
9e75b8b32d
libcom/test: osiSockTest fix for RTEMS
...
Instead of 0/1 RTEMS gives 0/32 (32==SO_BROADCAST).
So switch the test to zero/non-zero.
2017-10-13 10:42:19 +02:00
32a6f6c4f1
Fix DBD parsing slow-up with Perl 5.20 and later
2017-10-09 14:45:01 +02:00
169e9f2699
Merged 3.14 changes into 3.15
2017-09-25 15:45:53 -05:00
22debb3532
Fix for LP: #1702298
2017-09-25 15:41:31 -05:00
c441cdd5a4
Adjust comments about EPICS_TIMEZONE
2017-09-25 12:14:48 -05:00
6ee2cc6039
Merged 3.14 branch into 3.15
2017-09-07 17:48:05 -05:00
13fa1e2722
Travis: Use 'make test-results' for more compact output
2017-09-07 17:39:42 -05:00
fad89189da
Fix typo in CAref.html
2017-09-07 17:38:58 -05:00
2e4634853c
Don't inherit alarm severity from self-referential links
...
This behavior was present in 3.14 and got lost in my link changes.
Includes a new regression test to detect the broken behavior.
2017-09-02 23:35:07 -05:00
e0757204cc
More tests, check both true and false values for options.
2017-08-30 13:58:17 -05:00
78abb25884
Add osiSockOptMcastLoop_t and osiSockTest
2017-08-28 22:36:25 -05:00
73f3e76150
Merged 3.14 branch config changes into 3.15
2017-08-18 17:19:05 -05:00
4ab56518a0
Add readline support to linux-xscale_be rules
...
Also moves cross-build config to the right host-arch's
2017-08-18 15:14:42 -05:00
80dbc7aeef
Clean up the cross-build rules slightly
2017-08-18 15:10:13 -05:00
238f6772bd
Dirk's simpler VxWorks version rules
2017-08-18 15:06:38 -05:00
dc9859cee9
Fix RPATH and readline for linux-arm cross-builds
2017-08-18 11:59:17 -05:00
5d5f27a486
Don't set RPATH if STATIC_BUILD=YES
2017-08-18 11:57:50 -05:00
ad6a16d7c4
Update EPICS_TIMEZONE settings, extend to 2022
2017-08-18 11:33:17 -05:00
111cac8e47
merge 3.14
2017-08-01 17:14:34 +02:00
5c8e5c52ef
rsrv: fix recv() error handling on WIN32
...
For WIN32 osiSockIoctl_t is unsigned, so
> osiSockIoctl_t nchars = recv(...
is casting signed -> unsigned which treats
errors as success.
2017-08-01 17:02:41 +02:00
546df1c1f0
rsrv: export CASDEBUG to iocsh
2017-08-01 17:02:41 +02:00
603331e7a5
rsrv: flush any queued messages before forced disconnect
...
Avoid loss of various ERROR messages which camessage()
has queued.
2017-08-01 16:44:48 +02:00
4b272cc0cf
rsrv: locking in cas_send_bs_msg()
...
Must lock around "pclient->send.stk = 0u;"
2017-08-01 16:44:48 +02:00
619a99bf99
rsrv: missing send lock around send_err()
2017-08-01 16:44:48 +02:00
1f8cb740f1
rsrv: drop un-commited VERSION message
...
This is a no-op as cas_commit_msg() isn't called.
A VERSION message is already queued during create_tcp_client().
2017-08-01 16:44:29 +02:00
322f7a97de
rsrv: add some comments
2017-08-01 16:44:29 +02:00
0fc770166c
rsrv: avoid possible overflow in vsend_err()
...
Accounting of message size doesn't take into account
space used by header of failed message (16 or 24 bytes).
This would allow a theoretical really long error message
to overflow the send buffer by 16 or 24 bytes.
2017-08-01 16:44:08 +02:00
f04f17f9bf
Remove vpath for %.pl files
...
It isn't needed, and in some configurations it causes build failures.
Similar problems could still occur for other file types with a similar
vpath, but it's less likely that other file types would be installed in
an equivalent $(TOP)/../<dir> path.
The original issue with .pl files was caused by an installation of Base
where the parent directory had a bin/<host> soft-link pointing to the
same bin/<host> of a newer installation of the same version of Base, so
it had all the necessary .pl files in it. The vpath allowed the build in
src/tools to search $(TOP)/../bin/<host> when checking the buildInstall
dependency, and since it found those .pl files through the vpath it
didn't bother to install them. When they were run later on in the build
from the $(TOP)/bin/<host>/ path they weren't present so it failed.
2017-07-26 12:02:24 -05:00
1a70855e25
Use static strings for epicsInterruptContextMessage()
...
The callbackRequest() routine was passing a stack-allocated
string to epicsInterruptContextMessage() but on RTEMS the
pointer is queued without copying the string. This fix uses
static strings for the 3 messages instead.
Fixes LP: #1705219
2017-07-24 11:19:20 -05:00
745109e423
Add loop detection to Release.pm
...
GNUmake will *usually* spot recursive "include" loops first, but
not if the loop is in another module that we're just checking.
In that case Release.pm eats file descriptors until they run out
without this fix.
2017-06-22 18:38:07 -05:00
Ralph Lange