Michael Davidsaver
3d70e70640
Add dbPvt2Rec() cast
...
Reverse of dbRec2Pvt()
pacify -D_FORTIFY_SOURCE=3 and __builtin_object_size() as
"&precord->common" does not know than common as actually
the prefix of a variable sized struct.
2024-08-21 10:01:26 -05:00
Michael Davidsaver
f9e53dded6
show buffered line along with DB parse error
2024-08-21 09:53:38 -05:00
Hinko Kocevar
8e62ea4965
handle deletion of the non-existant record with field
2024-08-21 09:48:23 -05:00
Hinko Kocevar
f00de26be1
directory path name is free'd before it is used in testAbort()
2024-08-21 09:48:23 -05:00
Hinko Kocevar
50b6a3801a
check if precordType is null in dbGetRecordTypeName()
2024-08-21 09:48:23 -05:00
Michael Davidsaver
2a5def7fc6
test removal of non-existant record
2024-08-21 09:48:23 -05:00
Michael Davidsaver
fd072e4429
add more information to record(# warning
2024-08-21 09:48:23 -05:00
Hinko Kocevar
3a5c9ab0cd
add a warning if deleting a non-existent record
2024-08-21 09:48:23 -05:00
Hinko Kocevar
c735de4ff8
fix number of tests to execute
2024-08-21 09:48:23 -05:00
Hinko Kocevar
bb5423171d
remove warning and error messages, fix duplicate flag setting
2024-08-21 09:48:23 -05:00
Hinko Kocevar
4e4e55ca89
invoke dbFreeLinkContents() to clean up link related allocations
2024-08-21 09:48:23 -05:00
Hinko Kocevar
e9748881cd
make codacy happy
2024-08-21 09:48:23 -05:00
Hinko Kocevar
39b0301062
plug memory leak resulting from record deletion
2024-08-21 09:48:23 -05:00
Hinko Kocevar
2f98251c9f
add unit tests for record deletion
2024-08-21 09:48:23 -05:00
Hinko Kocevar
535c9c2a06
Allow deleting a record at database creation.
...
Using a magical record type "#" will allow the user to
delete previously created record from the database.
2024-08-21 09:48:23 -05:00
b6fffc2225
fix string to epicsUInt32 conversion via double
2024-08-21 09:28:45 -05:00
86cdfc596f
fix wrong (unsigned) comparison
2024-08-12 10:29:08 -04:00
Andrew Johnson
3dae29b7e8
Oops, needed stdint.h
2024-08-10 14:45:28 -05:00
Andrew Johnson
9d393c4437
Cast sizeOffset to uintptr_t for ordered comparison
...
Issue report & fix by Dirk Zimoch.
2024-08-10 13:21:30 -05:00
Andrew Johnson
856f345d2c
Fix Codacy complaint about C-style casts
2024-08-07 09:54:16 -05:00
Andrew Johnson
3f4d080260
Make dbFastGet/PutConvertRoutine arrays const
...
Also added Doxygen annotations for them.
2024-08-07 09:54:16 -05:00
Andrew Johnson
1e8d49f2a6
Drop debugging diagnostics
2024-08-07 09:54:16 -05:00
Andrew Johnson
31b22fd253
Add test of jlinkz::putValue() to dbPutLinkTest
...
Add OUTP link field to xRecord
xRecord::process() puts VAL to the OUTP link
jlinkz writes the output value to the record's own PHAS field
Correct the dbFastPutConvertRoutine lookup
Test sets OUTP link, processes record and confirms that PHAS was set
2024-08-07 09:54:16 -05:00
Andrew Johnson
916b17ef3f
Fix review comment
2024-08-07 09:54:16 -05:00
Andrew Johnson
437320926b
Move declaration back
2024-08-07 09:54:16 -05:00
Andrew Johnson
7890e67d37
Various misc warnings
2024-08-07 09:54:16 -05:00
Andrew Johnson
918a188285
Define USE_TYPED_DRVET, use drvet * instead of struct drvet *
2024-08-07 09:54:16 -05:00
Andrew Johnson
2f730b8e9f
Add arg's to function pointer typedefs and prototypes
2024-08-07 09:54:16 -05:00
Andrew Johnson
1835187a86
Change fast convert routines to match prototype
...
Most of these edits were make using regexes in find/replace.
2024-08-07 09:54:16 -05:00
Andrew Johnson
f287cfa2ac
Introduce & use FASTCONVERTFUNC with full prototype
2024-08-07 09:54:16 -05:00
Andrew Johnson
785237e41a
Remove duplicate link libraries
2024-08-07 09:54:16 -05:00
Andrew Johnson
172bfce1f3
Merge 'Release 7.0.8.1' branch into 7.0
2024-06-27 21:07:50 -05:00
Andrew Johnson
d0d15ee911
Replace UNRELEASED with 7.0.8.1
2024-06-26 14:59:18 -05:00
Tynan Ford
772c10d904
Fix issue with RSRV_SERVER_PORT above 9999
...
don't worry about null termination on epicsSnprintf call
2024-06-19 09:38:23 -07:00
Michael Davidsaver
72d50ce274
fix dblsr()
...
clearly doesn't get called very often...
2024-06-14 16:45:07 -07:00
Michael Davidsaver
0a6b9e4511
dbScan: handle scanStop() before start
2024-06-14 16:45:07 -07:00
Simon Rose
1b46077096
Fix off by one error in constant link fetch
...
For long string buffers, we currently write a null terminator one byte
past the end of the buffer. This can be seen with a record of the type
```
record(aai, foo) {
field(NELM, 1)
field(FTVL, CHAR)
field(INP, {const: "foo"})
}
```
where the buffer is only of size 1, but then we write at index 1 (aka
past the end of the buffer).
Co-authored-by: Lucas A. M. Magalhães <lucmaga@gmail.com >
2024-06-12 16:19:36 -07:00
Chris Guerrero
ede745cc34
Update aSubRecord.dbd.pod
2024-06-12 09:14:30 -07:00
Michael Davidsaver
a864f16318
dbCa test sync. improvements
2024-06-11 19:37:23 -07:00
seifalrahman
4c20518864
modifying the condition from (status!=0) to (status>0) to skip the block in case the status variable ==-1
2024-06-11 18:45:33 -07:00
Michael Davidsaver
aa77b1c04a
iocInit errors say ERROR
2024-06-11 11:14:07 -07:00
Michael Davidsaver
fe4247d57d
Send .db parser errors to stderr
2024-06-11 11:14:07 -07:00
Michael Davidsaver
0495ac3bc5
WARN for FLNK uses CA without PROC
2024-06-11 11:14:07 -07:00
Michael Davidsaver
4a305a42a7
softMain log iocInit() failure.
2024-06-11 11:14:07 -07:00
DW
11fba63d18
Fix histogram doc
2024-05-29 09:17:03 -05:00
Simon Rose
beec00b403
Fix issue with compress record
...
The handling of N-to-M array compression was broken with the addition
of the partial buffer option, which broke the bounds check that was
being used.
Note that this also makes the partial buffer option more consistent;
if, for example, you have
```
record(compress, foo) {
field(ALG, "N to 1 Average")
field(INP, "bar NPP")
field(NSAM, 2)
field(N, 2)
field(PBUF, YES)
}
```
(with `bar` having, e.g. length 3), then this will now behave as
expected on both of the samples.
2024-05-29 09:15:35 -05:00
DW
4966baf423
fix sizv for printf & fix doc
2024-05-20 09:18:26 -04:00
Michael Davidsaver
e5b4829074
bound lso/lsi to limit of dbAddr::field_size
2024-05-19 11:57:19 -04:00
Simon Rose
839f764bcb
Clean up some potential memory leaks
...
The watchdog tasks are allocated, but not consistently removed. In
general this doesn't matter: they run in threads that will only
end when the process actually quits. For consistency and for the
purpose of future-proofing, I think there is value in having the
cleanup added in each case.
2024-05-15 09:26:09 -05:00
Simon Rose
4bb50fe664
Memory leak in caservertask.c
2024-05-15 09:26:09 -05:00
