- Fix documentation examples of
modbus_get_float_*functions.
- Fix FD_SET overflow when socket fd >= FD_SETSIZE.
- Check dest pointer not null and nb in read functions.
- NULL check for src and nb < 1 validation in write functions.
modbus_reply: don't compute address for FC 0x07/0x11.- Use O_NONBLOCK instead of deprecated O_NDELAY (closes #710).
- Explicit cast for Coverity CID 416366.
- Document required buffer size of
modbus_receive. - Document macros for error codes corresponding to Modbus exceptions (#758).
- Fix example of
modbus_rtu_set_serial_mode(closes #552). - Test filesystem provides symlink in autogen.sh (closes #414).
- Sync API signatures with the documentation.
- Many documentation fixes and typo corrections.
- Add coverage target and helper script.
- RTU - Check CRC before filtering on slave ID
- HAVE_NETINET_IN_H as guard around header (closes #765)
- Use default port 502 in documentation.
- Fix float endianness issue introduced in v3.1.8 (49af73d). Thank you @ghorwin for your excellent work on the subject. Closes #665, #694, #757, #770.
- Proper display of used CFLAGS.
- Don't build with debug flag by default anymore.
- Check request length in
modbus_replywhen used inmemcpy. Thank you Nozomi Networks Labs Advisory for the report. - Fix insecure data handling. CID 416366: INTEGER_OVERFLOW found with Coverity Scan.
- Remove useless cast in setsockopt call (closes #721)
- Link against socket and nsl libs when building on SunOS. Thank you @lanurmi.
- Update documentation about tcp-pi requiring 1 KiB of extra memory (closes #715). Thank you @psychon.
- Documentation improvements by @cedricboudinet and @mhei.
- Revert TCP checks for recovery (closes #711).
- Don't use loop initial declaration (closes #752).
- Fix wrong generated version for double number in win32 (closes #675).
- Don't use gai_strerror if not available.
- Add checks for netinet/ip.h and gai_strerror (#745)
- Log error in read input registers if debug (closes #755).
- Fix errno value on timeout with TCP connect (closes #736, #756). Thank you @kyllingstad for the issue and @psychon for the fix.
- Free addrinfo struct on getaddrinfo() gerrors.
- Fix doc of modbus_mapping_new_start_address (#615).
Urgent release to fix a regression.
- Fix baud rate handling with RTU under Linux
- Fix modbus_disable_quirks signature
- Remove Travis CI
- Move content about migration to libmodbus.org
- Avoid negative value in FD_SET call
- Test socket against positive value instead of -1
- Improvements to autotools by @ndim
- Fix doc links by @jordanjohnson56
- Test the protocol id for 0 by @metapsychologe
- Fix double negative in tests
- New quirks handler (closes #38 #533)
- Fix bitwise OR in documentation
- Improve doc about slave ID in RTU
- Add .clang-format and format code (closes #394)
- Remove constraints on baud rate values
- Accept IP or device in arg of unit test progs
- Avoid compilation issue with VS2022 with strdup
- Display created files in configure.js
- Use strcpy_s under Windows
- Replace inet_addr by inet_pton calls
- Replace inet_ntoa by inet_ptop
- Update configure.ac/config.h.win32 for new inet functions
- Instructions to build libmodbus in a VS project
- Fix connection check for Windows RTU (closes #660, #662)
- Add CIFuzz Github action
- Convert a few int to unsigned int (#402)
- Major rewrite of documentation with Material for mkdocs
- Reduce memory use of TCP PI backend (closes #621)
- Fixed MODBUS_ERROR_RECOVERY_LINK not working on Windows (@embeddedmz)
- Replace Travis CI by GitHub CI
- Fix linker error for Windows (VCLinkerTool)
- Address check in single register / coil responses added (#463)
- Swap CRC bytes in request data but not at CRC computing (#397)
- Fix float endianness issue on big endian architecture
- Fix comment about EMBUNKEXC (closes #566)
- Fix network library detection on Haiku
- Fix typos (closes #620)
- modbus_reply: fix copy & paste error in sanity check (fixes #614)
- Add SECURITY.md (closes #613)
- Fix typo in comment (closes #588)
- Replace obsolete AC_PROG_CC_STDC by AC_PROG_CC
- Fix position of CC flags in documentation (closes #599)
- Remove duplicate ';' (closes #602)
- Add the baud rate of 256k for Windows (closes #603)
- cosmetic changes in man page standardizing itemization
- Fix many typos
- Replace .dir-locals.el (Emacs) by .editorconfig
- Include the test LICENSE in tarball
- Install the NEWS and AUTHORS files
- Update README.md
- docs: fix simple typo, reponse -> response
- Add modbus_[get|set]_indication_timeout to doc build
- Fix warning issues
- Move malloc before starting unit tests
- Fixed MODBUSGET* macros in case of negative values
- SPDX: change LGPL-2.1+ to LGPL-2.1-or-later
Thank you to @yegorich, @i-ky, @jobol, @timgates42, @anton-bondarev, @richardash1981, @stefannilsson, @Zugschlus, @qqq89513 and @mhei.
- Fix awful typo in fix for VD-1301 vulnerability. Thank you @karlp.
- Fix URL
- Add appveyor.com config
- Update libmodbus.txt
- Change arg value from int to const uint16_t in modbus_write_register
- Mark raw_req argument as const in modbus_send_raw_request (closes #479)
- Add unit tests for VD-1301 and VD-1302 vulnerabilities
- Fix VD-1301 and VD-1302 vulnerabilities
- Create CODE_OF_CONDUCT.md
- modbus_mapping_t is now a named typedef (to allow forward declaration)
- Rename type as flags for consistency between functions
- Add missing SOCK_CLOEXEC flag on socket creation
- Remove wrong and harmful checks in configure.ac
- Oops fix OR on RS485 settings (1c5d969)
- Only set SER_RS485_ENABLED bit of existing RS485 settings
- Add .vscode/ to .gitignore
- Fix wrong function name
- Remove a duplicated semi-colon
- Fix a small typo in test message
- Fix modbus_reply for TCP when unit id == 0 (fixes #376)
- New functions to define the indication timeout (#95)
- Remove CYGWIN condition to provide bswap16 fallback (#383)
- typo fix
- Use builtin bswap16 of gcc 4.8 (#377)
- Handle out-of-memory conditions more gracefully
- [doc] Add entry points for accept/listen in libmodbus index
- Fix typo
- Refine issue template for Github
- Move WINVER definition before other includes (#350)
- Add new function modbus_get_slave()
- Fix LIBMODBUS_VERSION_HEX encoding (closes #345)
- tcp_modbus_accept mustn't close socket on error (closes #333)
- C_PROG_RANLIB is rendered obsolete by LT_INIT
- Improve ifdef around bswap_16 for CYGWIN
- Improve Cygwin support. Thanks to StalderT.
- Another round of DRY in modbus_reply()
- Rename raw_rep to raw_rsp in unit-test-client
- Fix wrong function name in debug message
- Fix handling of invalid function code (closes #315)
- Add debug message on unknown function and new unit test
- DRY in modbus_reply by improving response_exception()
- Fix typo in 3053bd0adb
- CID 69145 - Argument cannot be negative in unit-test-server
- CID 69142 - Unchecked return value in unit-test-server
- Fix CID 69140 - Bad bit shift operation (coverity) in tests
- Rewrite and rename README as README.md in tests/
- Github's contributing and issue template files
- Fix small leak (64 bytes in TCP) in unit-test-client
- Add unit-tests.sh to tarball
- Add ./configure.scan to .gitignore
- Move setting of option inside the relevant conditional group
- Add links to new modbus**_float** functions in index
- Slight change to modbus_report_slave_id doc.
- New bswap macros for Max OSX by Jakob Bysewski.
- Fix "wildcard address" in TCP IPv6 by Shoichi Sakane.
- Introduce offsets in modbus mappings with modbus_mapping_new_start_address. Thanks to Michael Heimpold and Stéphane Raimbault.
- Fix address range in random-test-client. Thanks to Martin Galvan.
- Add an option to disable tests compilation by Yegor Yefremov.
- Define MSG_DONTWAIT to MSG_NONBLOCK on AIX (#294). Thanks to Fabrice Cantos.
- Fix building when byteswap.h is not defined by Tomasz Mon.
- Add some more macros for data manipulation and documentation.
- Remove duplicate install of modbus.h (closes #290). Thanks to Daniel Sutcliffe.
- Move MIGRATION and README.md to dist_doc_DATA target.
- Change order of few functions in modbus RTU code.
- Add entries for modbusrtu[get|set]_delay in documentation index.
- Implemented runtime configurable RTS delay by Jimmy Bergström.
- Add an entry in libmodbus index page for modbus_rtu_set_custom_rts.
- Add support for user defined RTS toggle function by Jimmy Bergström.
- Added ILLEGAL_DATA_ADDRESS tests for modbus_write_register[|s]. Thanks to Andrey Skvortsov.
- Update documentation of modbus_rtu_set_rts
- Fix rts signal switch time by Hiromasa Ihara.
- Improve new_rtu and set_slave documentation (related to #276).
- Fix late check of ctx in modbus_reply[|_exception] (closes #269).
- Wait the server for 1 second before running tests (help Travis).
- A libmodbus context isn't thread safe and won't be (closes #246).
- Fix buffer overflow in modbus_mask_write_register (#265).
- Minor adjustments to README about AsciiDoc.
- Export MODBUS_MAX_ADU_LENGTH and documentation (ref #241).
- Explicit check against Modbus broadcast address.
- Do not reply on broadcast requests (fixes #153). Thanks to Michael.
- Add Travis support.
- Run unit tests with standard: make check (closes #205, closes #238). This patch has been developed by Andrey Skvortsov, Michael Heimpold and Stéphane Raimbault.
- modbus_send_raw_request: limit request length (fixes #207). Thanks to Hanno Neuer for spotting this security flaw.
- Add new contributors to AUTHORS
- Introduce SPDX license identifiers. Thanks to Michael Heimpold.
If you still want to listen any addresses in your TCP IPv4 server, you must now set the IP address to NULL in modbus_new_tcp before listening. Now, modbus_tcp_listen only listen the IP address set in the Modbus context (see documentation).
This release introduces API changes on modbus_get_byte_timeout, modbus_get_response_timeout, modbus_set_byte_timeout, modbus_set_response_timeout to ease writing of language bindings.
- Rewrite documentation building system
- Fix timeouts in unit tests
- Don't flush on illegal address errors in single write functions
- Fix compilation on compilers not supporting c99 mode. Thanks to Michael Heimpold.
- Update license for the tests in Debian packaging (#221)
- Move check of device earlier to avoid a free call
- Unit test for baud rate check and error message.
- Fix crash modbus_new_rtu when baud is 0. Thank you to Daniel Schürmann.
- Removed function prototype without implementation Thank you Andrej Skvortzov.
- Switch test programs to a BSD license
- Fix remote buffer overflow vulnerability on write requests
- Avoid twice connect() in source code (closes #194)
- Fix compilation with MinGW (GCC 4.8.1) under Win7 (closes #163) Thank you MarjanTomas and SwissKnife.
- Fix TCP IPv4 modbus_connect() on win32 (closes #100 and #165) Thank you Petr Gladkiy and Marjan Tomas.
- Fix 24a05ebd3c0 - win32: init of modbus_tcp_pi_listen (#187)
- INADDR_* macros are defined in host byte order
- Filter of IP addresses in IPv4 server (closes #190)
- Allow to listen any hosts in IPv6 (closes #32)
- Define and public export of MODBUS_MAX_PDU_LENGTH (closes #167)
- Truncate data from response in report_slave_id to new max arg (closes #167)
- Fix response timeout modification on connect (closes #80)
- New API to set/get response and byte timeouts. New unit tests and updated documentation.
- Export Modbus function codes supported by libmodbus
- Fix bandwidth-server-one (closes #152)
- Check debug flag in RTU code
- Remove warnings caused by shadowed 'index' variable. Thanks to Åke Forslund.
- Use accept4 in TCP PI if available
- Add documentation for tcp[_pi]_accept (closes #31)
- Fix mistake in modbus_tcp_listen documentation
- Add documentation for modbus_tcp_pi_listen
- Fix for MinGW and Windows (#144, #169, #175, #180, #181, #187) Thanks to Marjan Tomas.
- Many other fixes (#134, #157, #158, #183, #184) and improvements.
More robust to exploits, more compliant and better handling of connect/close sequences. Windows support still broken.
- New unit tests
- Fix remote buffer overflow vulnerability (closes #25, #105)
- Explain how to define response timeouts when many RTU slaves
- Fix receiving of incorrect queries in write_single and mask_write_register Thanks to James Nutaro.
- Check return value of autoreconf. Thanks to Lauri Nurmi
- Constant for broacast and test ordering
- Fix the fix of device string check
- Various changes to try to improve broken Windows support
- Try to fix MinGW compilation
- Portable use of bswap_32
- Improve support of MacOS X
- Fix socket value on init/close
- Returns -1 on invalid mode in modbus_rtu_set_rts
- Protect all public functions against invalid context
- Sleep for delay of response timeout before reconnect (closes #77). Thanks to Karl Palsson.
- Baud rate until 4,000,000 (POSIX), 1,000,000 (Windows) (closes #93)
- New modbus_get|set_float_dcba to get|set float in inversed byte order
- Remove unsupported -Wtype-limits for GCC < 4.3.5 (closes #109)
- Enable out-of-source build. Thanks to Yegor Yefremov.
- Fix alignment problem on ARMv5 platform
- Improvement to Debian package. Thanks to Alexander Klauer.
- Improve support of VS 2005. Thanks to Petr Gladkiy.
- Add documentation for modbus_mask_write_register (closes #91). Thanks to Martijn de Gouw.
- Avoid C99 declaration in win32 section code (closes #92). Thanks to oldfaber and endrelovas.
- Add a windows scripting host configure file. Thanks to oldfaber and Stéphane Raimbault.
- Fix typo in modbus_strerror documentation. Thanks to Mirko Rajkovaca.
- Rename reserved C++ keywords of modbus_mask_write_register. Thanks Tobias Doerffel.
- Another quick workaround for deficient OS (closes #63)
- Add support for Mask Write Register
- Fix missing close on socket in random-test-server. Thanks to Damian Zieliński.
- Use nonblocking sockets on Win32 and OS X/iOS too. Thanks to Julian Raschke.
- Fix all compilations warnings spotted by new compilation flags
- Major update of build system
- Calculate RTS activation time by send length
- Dynamic memory allocation of device name (closes #11)
- Add unit tests for modbus_mapping_new
- Add Visual Studio 2008 project files by oldfaber
- Fix missing argument in synopsis section of modbus_rtu_set_serial_mode
- Fix wrong constant names to create version number
- More compilation fixes for Windows by oldfaber.
- Fix wrong constant names to create version number. Thanks to Denis Davydov.
Major changes to handle many slaves in RTU mode, non blocking connections and RTS flow control.
- Fixes for Microsoft Visual C++ compiler
- Fixes for Windows. Thanks to oldfaber
- Return value of _modbus_tcp_pi_connect() on failure (closes #61)
- Avoid ioctl call on non-RS485 ports. Thanks to Michael Haberler
- Display node and service in PI and port in IPv4 at connection
- Return -1 on getaddrinfo error and print error in debug mode
- More robust way to establish the connection in non blocking mode
- TCP - Socket in non blocking mode by default. Thanks to Thomas Stalder
- Apply CLOEXEC flag for TCP protocol independent too (IPv6)
- New RTU receive() to ignore confirmation from other slaves (closes #18)
- Move RTU filtering in CRC check to avoid useless call to modbus_reply
- Unique transaction identifier by TCP connection
- Use accept4 with SOCK_CLOEXEC when available (Linux)
- Open fd and socket with the CLOEXEC flag when available
- Exception response on report slave ID wasn't detected (closes #27)
- Provides a way to disable the byte timeout (Alex Stapleton)
- Added slave ID check for response messages (Alex Stapleton)
- RTS flow control with modbus_rtu_set_rts and modbus_rtu_get_rts functions by Torello Querci tquerci@gmail.com and Stéphane Raimbault.
- Fix awful typo in fix for VD-1301 vulnerability. Thank you @karlp.
- Backport fixes for VD-1301 and VD-1302 vulnerabilities
- Move WINVER definition before other includes (closes #350)
- Replace signed int by unsigned
- Backport fix remote buffer overflow vulnerability on write requests
- Replace deprecated INCLUDES by AM_CPPFLAGS
- Fix remote buffer overflow vulnerability
- Fix receiving of incorrect queries in write_single
- autogen.sh creates symlinks instead of copies
- Add missing m4 directory (closes #103)
- Fix alignment problem on ARMv5 platform Thanks to Alexander Dahl
- Remove useless test on error_recovery argument Thanks to Carlos Tangerino
- Fix another Visual C++ 2008/2010 deficiency (closes #53)
- Add -lsocket to compile on QNX
- Fix TCP PI init under Windows. Thanks to oldfaber.
- Fix a missing free in random-test-client Thanks again to Stefan Finzel.
- Fix OMG bug in modbus_mapping_free not freeing memory. Thanks to Stefan Finzel for the bug report.
- Fix semicolon typo and unistd.h include under Windows. Thanks to Andrew Kravchuk.
- Update Debian package
- Documentation fixes and additions
- Add missing C++ macros in public headers. Thanks to Bernhard Agthe.
- Protects modbus_mapping_free against NULL argument. Thanks to Andrea Mattia
- Fix check on file doc/libmodbus.7 in acinclude.m4 (closes #28)
- Close file descriptor when the settings don't apply in RTU. Original patch provided by Thomas Stalder.
- unit-test.h is now generated to avoid config.h dependency.
- Request for Windows Sockets specification version 2.2 instead of 2.0 Thanks to Pavel Mazniker for the report.
- Avoid useless serial_mode integer when TIOCSRS485 isn't supported
- Fix compilation failure on Windows (RS485 support) by Tobias Doerffel tobias.doerffel@gmail.com
- Properly check TIOCSRS485 define by Matthijs Kool
- Rename package to libmodbus5 to fix lintian warning
- Revert libmodbus licence from LGPLv3 to LGPLv2.1 to avoid incompatibility with GPLv2 program. This change has been approved by Tobias Doerffel, Florian octo Forster and Hannu Vuolasaho.
- Enable RS485 support only when available
- Export modbus_set/get_serial_mode functions on all platforms
- API change for read/write multiple registers function:
- modbus_read_and_write_registers -> modbus_write_and_read_registers The function name was confusing because the write operation is performed before the read. Take care to swap the arguments in the migration process.
- Documentation of modbuswrite_and_read_registers, modbus_mapping_new/free, report_slave_id. modbus_get_byte_from_bits, modbus_set_bits_from_byte(s), modbus[gs]et_float, modbus_reply and modbus_reply_exception.
- Enhanced report slave ID
- New RPM spec file to package documentation and development files
- IPv6 support Make the TCP implementation "protocol independent" by Florian Forster and Stéphane Raimbault.
- Fix compilation on Windows 7 (x64) with MinGW/MSYS and GCC 4.5 Reported by Patsy Kaye.
- Documentation of libmodbus functions with AsciiDoc (man and HTML) by Stéphane Raimbault
- Avoid an iteration in flush function
- New functions to send and receive raw requests (modbus_send_raw_request, modbus_receive_confirmation)
- Fix flush function of TCP backend on Windows
- API changes for server/slave:
- modbus_receive doesn't take socket/fd argument anymore
- new function modbus_set_socket to set socket/fd
- API changes for timeout functions:
- modbus_get_timeout_begin -> modbus_get_response_timeout
- modbus_set_timeout_begin -> modbus_set_response_timeout
- modbus_get_timeout_end -> modbus_get_byte_timeout
- modbus_set_timeout_end -> modbus_set_byte_timeout
- Fix longstanding limitation of server to wait forever
- New functions modbus_set/get_serial_mode by Manfred Gruber and Stéphane Raimbault for RS485 communications
- Improved recovery mode (see modbus_set_error_recovery documentation) for data link and protocol errors.
- Fix compilation issue with Microsoft Visual Studio 2008. Reported by Allan Cornet.
- Major rewriting of the message reading (no more timeouts on exception) by Stéphane Raimbault
- New function to reply to an indication with an exception message modbus_reply_exception()
- New function modbus_get_header_length(modbus_t *ctx)
- New functions to manipulate data:
- MODBUS_GET_INT32_FROM_INT16
- MODBUS_GET_INT16_FROM_INT8
- MODBUS_SET_INT16_TO_INT8
- Fix GH-2. Read/write were swapped in _FC_READ_AND_WRITE_REGISTERS
- Install an ignore handler for SIGPIPE on *BSD Original patch by Jason Oster.
- Fix closing of Win32 socket. Reported by Petr Parýzek.
- Fix unit identifier not copied by the TCP server. Reported by Antti Manninen.
- Fix missing modbus_flush() in unit tests
- Fixes for OpenBSD by Barry Grumbine and Jason Oster
- Win32 support by Tobias Doerffel
- Split source code around RTU and TCP (backends)
- Rename modbus*[listen|accept] to modbus_tcp*[listen|accept]
- Remove slave argument from modbus_new_rtu()
- Check received function code
- Fix segfault in bandwidth-server-many-up on inet_ntoa() call
- Fix unit test of report slave ID in RTU
- Fix GH-3. Remove inclusion of config.h in modbus.h
- Correctly detect if we are cross-compiling for win32 by Kirill Smelkov.
- Fix setting of the broadcast address
- Brand new API (see MIGRATION notes)!
- Remove the internal function set_message_length_tcp
- Restore slave ID (server ID) argument in functions
- Error conventions of POSIX systems and error recover
- Parity setting is now a single char ('N', 'E' or 'O')
- Report slave ID server side
- OpenBSD support by Anibal Limón.
- New read and write registers function by Hannu Vuolasaho.
- Versioning infrastructure Inspired by the Clutter project and the work done by Florian Forster.
- Fix the broadcast constant (255 -> 0) Reported by David Olivari.
- Fix #463299 - New functions to define the timeouts of begin and end of trame Original patch by Sisyph (eric-paul).
- Fix #591142 - Slave id check should be disabled in TCP connection Reported by aladdinwu.
- New API to read and write float values by Stéphane Raimbault and Florian Forster.
- New API for slave server (see MIGRATION)
- New slave server able to handle multiple connections
- Slave only replies to broadcast queries or queries with its slave ID
- Improved Modbus protocol conformance
- modbus_param_t is smaller (2 int removed)
- Better error management and SIGPIPE signal is blocked
- Faster
- Fix #333455 - Cygwin IPTOS_LOWDELAY not supported on cygwin Reported by Jeff Laughlin and Yishin Li.
- Fix #375926 - modbus.c:164: error: `MSG_DONTWAIT' undeclared Reported and tested by Yishin Li.
- Fix #378981 - CRC error on RTU response doesn't return negative value Reported by Henrik Munktell.
- Fix report slave ID request Patch (bzr) provided by Paul Fertser.
- Fix #425604 - Conditional jump or move depends on uninitialised value(s) Occurs on first occurrence of slave timeout. Reported by Henrik Munktell.
- Fix #457200 - FreeBSD support Patch provided by Norbert Koch.
Other changes:
- The code is now published and developed on http://github.com/stephane/libmodbus
- Waf support has been removed
- Fix CRC error when a slave RTU send a response. Thanks to Justin Carroll to have reported and tested my patch.
- Remove an assignment in compute_response_length()
- Remove duplicate counter in read_io_status()
- Fix #274511 reported by 'Kylesch' Invalid error check in modbus_init_listen_tcp
- Fix a bug reported by email by Davide Pippa The function modbus_receive must check the number of values indicated in the response accordingly to the query.
- Fix #241006 reported by Jesus Hernandez Tapia modbus_check_response() crashes on an invalid exception code
- Reduce the number of function calls (read_reg_response and preset_response have been removed)
- Add a new unit test for bad response
- Catch the timeout even if the length is equal to a exception trame
- Test only msg_length_computed on change
- Many comments and cleanups
- Include libmodbus.spec in the tarball
- Fix #241006 reported by Jesus Hernandez Tapia modbus_check_response() crashes on an invalid exception code
- Slave API https://blueprints.launchpad.net/libmodbus/+spec/slave-api
- No more glib dependency https://blueprints.launchpad.net/libmodbus/+spec/glib-dependency
- Unit testing and many test programs
- Waf build support https://blueprints.launchpad.net/libmodbus/+spec/waf-support
- MacOS X support by Matthew Butch https://blueprints.launchpad.net/libmodbus/+spec/macosx-support
- Unit testing (unit-test-slave and unit-test-master)
- Port number is now defined at initialisation by Dirk Reusch
- Better memory management (uint8_t *data and packing of modbus_param_t)
- Better error management
- Declare many static functions and const arrays
- Enhance an integer division
- The GNU licences LGPL and GPL are in version 3
- Debian and RPM packages (#224496)
- Many cleanups
- Fix #159443 reported by Stefan Bisanz Index of incoming data in force multiple coils function
- Fix #161989 reported by Konstantinos Togias Serial device paths more than 10 chars long (eg. /dev/ttyUSB0) don't fit to modbus_param_t -> device char[11] var.
- Fix #188189 reported by Chris Hellyar Compute_response_size() no entry for read_input_status()
- Fix #191039 reported by Todd Denniston modbus.h is not installed at prefix.
- Fix #211460 reported by Todd Denniston With TCP, automatic reconnect on error may not be desired.
- Fix #224485 reported by Todd Denniston libmodbus does not link with c++ code.
- Fix #224496 reported by Todd Denniston It is easier to install on rpm based systems with a spec file.
- Fix #224485 reported by Todd Denniston libmodbus does not link with c++ code.
- Fix #191039 reported by Todd Denniston modbus.h is not installed at prefix.
- Fix #188189 reported by Chris Hellyar Compute_response_size() no entry for read_input_status()
- Fix #181887 reported by Jesus Hernandez Tapia. Slave address in build_request_packet_tcp() is hardcoded as 0xFF.
- Fix #161989 reported by Konstantinos Togias Serial device paths more than 10 chars long (eg. /dev/ttyUSB0) don't fit to modbus_param_t -> device char[11] var.
- Structure is also bit better 'packed' to conserve memory (see the trunk for a real enhancement).
- Fix #159443 reported by Stefan Bisanz Index of incoming data in force multiple coils function
- Deleted useless code in check_crc16()
- Untabify source code
- Changed author's email to Stéphane Raimbault
- FIX Compilation GCC-4.0
- Project name in autogen.sh