BLAHAJ
/
libsigrok
10
0
Fork 0
Code Issues Pull Requests Projects Releases Wiki Activity
libsigrok/Makefile.am

870 lines
25 KiB
Makefile
Raw Normal View History

libsigrok: More coding style fixes.
2010-04-15 18:16:53 +00:00
##
GPL headers: Use correct project name.
2013-04-23 20:24:30 +00:00
## This file is part of the libsigrok project.
libsigrok: More coding style fixes.
2010-04-15 18:16:53 +00:00
##
sr: adjust copyright year
2012-02-13 13:31:51 +00:00
## Copyright (C) 2010-2012 Bert Vermeulen <bert@biot.com>
radioshack-dmm: Add support for Radioshack 22-812 DMM Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
2012-10-15 06:17:32 +00:00
## Copyright (C) 2012 Alexandru Gagniuc <mr.nuke.me@gmail.com>
libsigrok: More coding style fixes.
2010-04-15 18:16:53 +00:00
##
Start of code base layout restructuring.
2010-04-02 18:18:27 +00:00
## This program is free software: you can redistribute it and/or modify
## it under the terms of the GNU General Public License as published by
## the Free Software Foundation, either version 3 of the License, or
## (at your option) any later version.
libsigrok: More coding style fixes.
2010-04-15 18:16:53 +00:00
##
Start of code base layout restructuring.
2010-04-02 18:18:27 +00:00
## This program is distributed in the hope that it will be useful,
## but WITHOUT ANY WARRANTY; without even the implied warranty of
## MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
## GNU General Public License for more details.
libsigrok: More coding style fixes.
2010-04-15 18:16:53 +00:00
##
Start of code base layout restructuring.
2010-04-02 18:18:27 +00:00
## You should have received a copy of the GNU General Public License
## along with this program. If not, see <http://www.gnu.org/licenses/>.
libsigrok: More coding style fixes.
2010-04-15 18:16:53 +00:00
##
Start of code base layout restructuring.
2010-04-02 18:18:27 +00:00
Build: Place Autoconf macros into m4/ directory This separates the M4 macros from the other cruft in autostuff/ to prepare for the introduction of custom macro files.
2015-08-22 18:58:11 +00:00
ACLOCAL_AMFLAGS = -I m4
Build: Reduce unnecessary config substitutions Variable substitutions with a constant value may as well be set directly in Makefile.am.
2015-08-24 20:20:30 +00:00
AM_LIBTOOLFLAGS = --silent
Build: Use GNUMAKEFLAGS for --no-print-directory
2015-09-10 17:35:17 +00:00
GNUMAKEFLAGS = --no-print-directory
Build: Reduce unnecessary config substitutions Variable substitutions with a constant value may as well be set directly in Makefile.am.
2015-08-24 20:20:30 +00:00
build: Skip Python bindings during distcheck setup.py attempts to relink the bindings at installation time, which breaks distcheck. Thus, disable the Python bindings during distcheck for now.
2015-11-07 21:08:45 +00:00
# distutils/setuptools cause trouble on distcheck. Disable for now.
DISTCHECK_CONFIGURE_FLAGS = --disable-python
Build: Reduce unnecessary config substitutions Variable substitutions with a constant value may as well be set directly in Makefile.am.
2015-08-24 20:20:30 +00:00
FIRMWARE_DIR = $(datadir)/sigrok-firmware
sr/srd: Move some files to their resp. dirs.
2011-12-30 09:42:39 +00:00
Build: Pass nostdinc option to automake It seems automake automatically adds the directory containing the generated version.h to the include path. Use nostdinc to disable default includes altogether.
2015-08-15 16:10:50 +00:00
local_includes = -Iinclude -I$(srcdir)/include -I$(srcdir)/src -I.
Build: Cleanup up and modernize autotools setup Note that some --enable options of configure have become --with or --without options.
2015-08-15 13:59:52 +00:00
if BINDINGS_CXX
Build: Make compiler warnings configurable Use the SR_ARG_ENABLE_WARNINGS macro to configure and check for the availability of compiler warning flags. Maintain separate sets of warning flags for C and C++. The configure option --enable-warnings=[min|max|fatal|no] can be used to set the compiler warning level for all languages. The default level is "max".
2015-08-23 13:02:30 +00:00
local_includes += -Ibindings/cxx/include -I$(srcdir)/bindings/cxx/include -Ibindings/cxx
Build: Cleanup up and modernize autotools setup Note that some --enable options of configure have become --with or --without options.
2015-08-15 13:59:52 +00:00
endif
build: Do not define FIRMWARE_DIR on Windows The hard-coded location is bound to be wrong anyway. Instead, rely on the new resource lookup code to find the firmware files in a location relative to the library or executable.
2015-10-03 12:08:32 +00:00
# Do not hard-code the firmware location on Windows.
if WIN32
global_defs =
else
global_defs = -DFIRMWARE_DIR='"$(FIRMWARE_DIR)"'
endif
Build: Cleanup up and modernize autotools setup Note that some --enable options of configure have become --with or --without options.
2015-08-15 13:59:52 +00:00
# Ensure that local include directories are always searched first.
build: Do not define FIRMWARE_DIR on Windows The hard-coded location is bound to be wrong anyway. Instead, rely on the new resource lookup code to find the firmware files in a location relative to the library or executable.
2015-10-03 12:08:32 +00:00
AM_CPPFLAGS = $(local_includes) $(global_defs)
Fix out-of-tree build. Use libtool "noinst" local helper libs and use one Makefile.am per subdir, which is the usual/preferred method. These helper libraries are purely local and will not be installed. This also fixes out-of-tree builds of sigrok, i.e. building in a directory other than the sigrok source directory, e.g. $ cd /home/user $ git clone ...sigrok $ cd sigrok $ ./autogen.sh $ mkdir /tmp/foo $ cd /tmp/foo $ /home/user/sigrok/configure $ make $ make install This will place all build results (.o files, .la files, etc) in the local build directory (/tmp/foo) instead of the source directory (/home/user/sigrok in this example). The installation directory is selected via the --prefix configure option (/usr/local per default).
2011-01-08 02:32:25 +00:00
Build: Use TESTS prefix instead of CHECK for flag variables In order to avoid confusion of the flags-gathering pkg-config result with the actual test for the availability of "check", change the pkg-config output variable prefix from CHECK to TESTS.
2015-08-26 11:13:39 +00:00
# The tests CFLAGS are a superset of the libsigrok CFLAGS, and the
Build: Cleanup up and modernize autotools setup Note that some --enable options of configure have become --with or --without options.
2015-08-15 13:59:52 +00:00
# python bindings CFLAGS are a superset of the C++ bindings CFLAGS.
Build: Use TESTS prefix instead of CHECK for flag variables In order to avoid confusion of the flags-gathering pkg-config result with the actual test for the availability of "check", change the pkg-config output variable prefix from CHECK to TESTS.
2015-08-26 11:13:39 +00:00
AM_CFLAGS = $(SR_EXTRA_CFLAGS) $(SR_WFLAGS) $(TESTS_CFLAGS)
Makefile: use PYSIGROK_FLAGS only for building python bindings
2016-01-11 13:32:22 +00:00
AM_CXXFLAGS = $(SR_WXXFLAGS) $(LIBSIGROKCXX_CFLAGS)
Build: Cleanup up and modernize autotools setup Note that some --enable options of configure have become --with or --without options.
2015-08-15 13:59:52 +00:00
Revert temporary changes for 0.2 release.
2011-04-03 21:18:46 +00:00
lib_LTLIBRARIES = libsigrok.la
Start of code base layout restructuring.
2010-04-02 18:18:27 +00:00
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
# Backend files
Rename libbackend to libsigrok.
2010-04-02 18:26:48 +00:00
libsigrok_la_SOURCES = \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/backend.c \
src/device.c \
src/session.c \
src/session_file.c \
src/session_driver.c \
src/hwdriver.c \
src/trigger.c \
src/soft-trigger.c \
Add sr_analog_to_float().
2014-11-12 16:05:13 +00:00
src/analog.c \
Add a public API to list available serial ports.
2014-12-06 23:56:22 +00:00
src/fallback.c \
resource: New internal API for accessing resource files The resource API provides a generic means for accessing resources that are bundled with sigrok, such as device firmware files. Since the manner of resource bundling is platform-dependent, users of libsigrok may override the functions used to open, close and read a resource. The default implementation accesses resources as files located in one of the XDG data directories or a directory defined at compile time.
2015-09-26 11:38:30 +00:00
src/resource.c \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/strutil.c \
src/log.c \
src/version.c \
src/error.c \
Add helper functions for software limits Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2016-05-01 11:50:08 +00:00
src/std.c \
src/sw_limits.c
Only build hardware plugins if requested by user. Per default all plugins will be built, though. The user can override this via the --disable-la-xxxx options.
2010-06-24 23:11:11 +00:00
Reorganize project tree.
2014-07-21 19:27:33 +00:00
# Input modules
libsigrok_la_SOURCES += \
src/input/input.c \
input: Convert binary module.
2014-08-19 23:29:37 +00:00
src/input/binary.c \
input: Convert chronovu-la8 module.
2014-08-19 22:43:13 +00:00
src/input/chronovu_la8.c \
input: Convert CSV module.
2014-08-18 22:14:47 +00:00
src/input/csv.c \
input/raw_analog: Add input module for raw analog signals
2015-11-28 18:55:12 +00:00
src/input/raw_analog.c \
Add the Lauterbach Trace32 logic analyzer data import module
2015-12-20 19:20:56 +00:00
src/input/trace32_ad.c \
input: Fixes and VCD.
2014-08-13 12:18:12 +00:00
src/input/vcd.c \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/input/wav.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
Reorganize project tree.
2014-07-21 19:27:33 +00:00
# Output modules
libsigrok_la_SOURCES += \
src/output/output.c \
src/output/analog.c \
src/output/ascii.c \
src/output/bits.c \
src/output/binary.c \
src/output/csv.c \
src/output/chronovu_la8.c \
output/wav: Initial module skeleton.
2014-07-26 09:00:51 +00:00
src/output/wav.c \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/output/hex.c \
src/output/ols.c \
output: Add srzip, the session file format. The 'filename' option is required: this module creates the zip file itself, and never actually outputs anything back to the calling frontend.
2014-11-11 22:23:09 +00:00
src/output/srzip.c \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/output/vcd.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
transform: Add a basic set of API calls.
2015-02-10 19:49:46 +00:00
# Transform modules
libsigrok_la_SOURCES += \
transform: Add a "nop" transform module. This doesn't do anything, just passes input packets on unmodified.
2015-02-10 20:38:21 +00:00
src/transform/transform.c \
transform: Add a "scale" transform module.
2015-02-10 21:25:52 +00:00
src/transform/nop.c \
transform: Add an "invert" transform module. This inverts the data values: - An analog value of x becomes 1/x. - A digital value of 0 becomes 1 (and vice versa).
2015-02-11 08:34:38 +00:00
src/transform/scale.c \
src/transform/invert.c
transform: Add a basic set of API calls.
2015-02-10 19:49:46 +00:00
Reorganize project tree.
2014-07-21 19:27:33 +00:00
# SCPI support
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
libsigrok_la_SOURCES += \
build: Distribute missing files
2015-11-07 20:57:20 +00:00
src/scpi.h \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/scpi/scpi.c \
scpi: Make helper functions from scpi-pps available library-wide.
2015-08-16 13:12:34 +00:00
src/scpi/helpers.c \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/scpi/scpi_tcp.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if NEED_RPC
libsigrok_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/scpi/scpi_vxi.c \
src/scpi/vxi_clnt.c \
src/scpi/vxi_xdr.c \
src/scpi/vxi.h
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if NEED_SERIAL
libsigrok_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/serial.c \
src/scpi/scpi_serial.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if NEED_USB
libsigrok_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/ezusb.c \
src/usb.c \
src/scpi/scpi_usbtmc_libusb.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if NEED_VISA
libsigrok_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/scpi/scpi_visa.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
Add GPIB SCPI backend using linux-gpib and libgpib.
2014-09-05 09:34:04 +00:00
if NEED_GPIB
libsigrok_la_SOURCES += \
scpi/gpib: Rename backend to scpi_libgpib. libgpib is the userspace component to linux-gpib's kernel modules that implement low-level interface drivers. When libsigrok gets userspace GPIB interface drivers, that backend will be the "official" scpi_gpib.
2014-09-05 16:33:48 +00:00
src/scpi/scpi_libgpib.c
Add GPIB SCPI backend using linux-gpib and libgpib.
2014-09-05 09:34:04 +00:00
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
modbus: Minor typos, cosmetics and consistency fixes.
2015-06-07 18:40:32 +00:00
# Modbus support
Add a modbus communication helper module.
2015-02-22 23:13:31 +00:00
libsigrok_la_SOURCES += \
src/modbus/modbus.c
modbus: add a serial RTU backend.
2015-02-22 23:20:26 +00:00
if NEED_SERIAL
libsigrok_la_SOURCES += \
src/modbus/modbus_serial_rtu.c
endif
Add a modbus communication helper module.
2015-02-22 23:13:31 +00:00
Reorganize project tree.
2014-07-21 19:27:33 +00:00
# Hardware (DMM chip parsers)
libsigrok_la_SOURCES += \
src/dmm/es519xx.c \
src/dmm/fs9721.c \
src/dmm/fs9922.c \
src/dmm/m2110.c \
src/dmm/metex14.c \
dmm: introduce support for ASYC-II 16-byte protocol in PRINT mode Introduce an asycii.c source file (modelled after metex14.c) which implements support for the 16-byte protocol of the ASYC-II multimeter chipset (RX only, when the PRINT button was pressed).
2016-11-28 00:01:15 +00:00
src/dmm/asycii.c \
Implement Brymen BM25x series as a serial DMM. The Brymen BM25x series supports the BC-20X that is an opto-isolated serial cable. The link seems to be unidirectional i.e. when activated the DMM periodically sends updates to the host while the host cannot control the DMM in any way. The protocol is documented in "6000-count-digital-multimeters-r1.pdf" that is available from the manufacturer. Every 15 byte packet consists of a bitmap where the bits correspond to segments or symbols on the LCD display i.e. the DMM essentially sends the contents of its screen to the host in every update. This driver then decodes the measured quantity, unit and its value from the bitmap.
2014-07-30 07:28:58 +00:00
src/dmm/rs9lcd.c \
Add UNI-T UT71x DMM parser.
2014-10-25 09:36:58 +00:00
src/dmm/bm25x.c \
ut372: Initial sub-driver skeleton.
2015-03-01 21:13:30 +00:00
src/dmm/ut71x.c \
Add initial Voltcraft VC-870 support. There are a few details that have yet to be implemented or reverse engineered and tested.
2014-12-29 22:30:41 +00:00
src/dmm/ut372.c \
Add dtm0660 15-byte DMM protocol parser.
2015-11-04 21:58:15 +00:00
src/dmm/vc870.c \
src/dmm/dtm0660.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
Add protocol decoder for Cyrustek ES51919 LCR meter chip. Add a protocol decoder for the Cyrustek ES51919 LCR meter chip. This chipset (together with ES51920 front-end) is supposedly used by multiple different portable LCR meters including at least DER EE DE-5000, Yihua V&A VA520, Mastech MS5308, Uni-T UT612, CEM DT-9935 and various OEM rebadges of them. The communication protocol seems to be implemented on the Cyrustek chip itself so all the different models are expected to use the same protocol if they implement a host connection. Unfortunately the protocol is not available in the public documentation of the chipset, so this implementation is based on reverse engineering it from traffic captures. The actual connection between the meter and the host computer may be different from meter to meter even when based on the same chip. This module implements a decoder for the protocol and some common helper functions for interfacing with the meter via an RS-232 serial port.
2014-10-11 12:47:34 +00:00
# Hardware (LCR chip parsers)
Makefile.am: Add NEED_SERIAL condition.
2015-01-31 21:41:04 +00:00
if NEED_SERIAL
Add protocol decoder for Cyrustek ES51919 LCR meter chip. Add a protocol decoder for the Cyrustek ES51919 LCR meter chip. This chipset (together with ES51920 front-end) is supposedly used by multiple different portable LCR meters including at least DER EE DE-5000, Yihua V&A VA520, Mastech MS5308, Uni-T UT612, CEM DT-9935 and various OEM rebadges of them. The communication protocol seems to be implemented on the Cyrustek chip itself so all the different models are expected to use the same protocol if they implement a host connection. Unfortunately the protocol is not available in the public documentation of the chipset, so this implementation is based on reverse engineering it from traffic captures. The actual connection between the meter and the host computer may be different from meter to meter even when based on the same chip. This module implements a decoder for the protocol and some common helper functions for interfacing with the meter via an RS-232 serial port.
2014-10-11 12:47:34 +00:00
libsigrok_la_SOURCES += \
src/lcr/es51919.c
Makefile.am: Add NEED_SERIAL condition.
2015-01-31 21:41:04 +00:00
endif
Add protocol decoder for Cyrustek ES51919 LCR meter chip. Add a protocol decoder for the Cyrustek ES51919 LCR meter chip. This chipset (together with ES51920 front-end) is supposedly used by multiple different portable LCR meters including at least DER EE DE-5000, Yihua V&A VA520, Mastech MS5308, Uni-T UT612, CEM DT-9935 and various OEM rebadges of them. The communication protocol seems to be implemented on the Cyrustek chip itself so all the different models are expected to use the same protocol if they implement a host connection. Unfortunately the protocol is not available in the public documentation of the chipset, so this implementation is based on reverse engineering it from traffic captures. The actual connection between the meter and the host computer may be different from meter to meter even when based on the same chip. This module implements a decoder for the protocol and some common helper functions for interfacing with the meter via an RS-232 serial port.
2014-10-11 12:47:34 +00:00
Add a protocol parser for KERN scales. (KERN & SOHN GmbH, http://www.kern-sohn.com/)
2015-08-27 19:21:03 +00:00
# Hardware (Scale protocol parsers)
libsigrok_la_SOURCES += \
src/scale/kern.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
# Hardware drivers
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
noinst_LTLIBRARIES = src/libdrivers.la
Put driver pointers into special section The sigrok core needs a list of all available drivers. Currently this list is manually maintained by updating a global list whenever a driver is added or removed. Introduce a new special section that contains the list of all drivers. The SR_REGISTER_DEV_DRIVER() and SR_REGISTER_DEV_DRIVER_LIST() macro is used to add drivers to this new list. This is done by placing the pointers to the driver into a special section. Since nothing else is in this section it is known that it is simply a list of driver pointers and the core can iterate over it as if it was an array. The advantage of this approach is that the code necessary to add a driver to the list is completely contained to the driver source and it is no longer necessary to maintain a global list. If a driver is built it will automatically appear in the list, if it is not built in won't. This means that the list is always correct, whereas the previous approach used ifdefs in the global driver list file which could get out-of-sync with the actual condition when the driver was built. Any sr_dev_driver structs that are no longer used outside the driver module are marked as static. Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2016-05-12 13:13:47 +00:00
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src/libdrivers.o: src/libdrivers.la
$(AM_V_CCLD)$(LINK) src/libdrivers.la
src/libdrivers.lo: src/libdrivers.o
$(AM_V_GEN)echo "# Generated by libtool" > $@
$(AM_V_at)echo "pic_object='libdrivers.o'" >> $@
$(AM_V_at)echo "non_pic_object='libdrivers.o'" >> $@
src_libdrivers_la_SOURCES = src/drivers.c
Put driver pointers into special section The sigrok core needs a list of all available drivers. Currently this list is manually maintained by updating a global list whenever a driver is added or removed. Introduce a new special section that contains the list of all drivers. The SR_REGISTER_DEV_DRIVER() and SR_REGISTER_DEV_DRIVER_LIST() macro is used to add drivers to this new list. This is done by placing the pointers to the driver into a special section. Since nothing else is in this section it is known that it is simply a list of driver pointers and the core can iterate over it as if it was an array. The advantage of this approach is that the code necessary to add a driver to the list is completely contained to the driver source and it is no longer necessary to maintain a global list. If a driver is built it will automatically appear in the list, if it is not built in won't. This means that the list is always correct, whereas the previous approach used ifdefs in the global driver list file which could get out-of-sync with the actual condition when the driver was built. Any sr_dev_driver structs that are no longer used outside the driver module are marked as static. Signed-off-by: Lars-Peter Clausen <lars@metafoo.de>
2016-05-12 13:13:47 +00:00
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_AGILENT_DMM
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
agilent-dmm: File naming consistency changes.
2016-09-18 16:50:11 +00:00
src/hardware/agilent-dmm/protocol.h \
src/hardware/agilent-dmm/protocol.c \
src/hardware/agilent-dmm/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_APPA_55II
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/appa-55ii/protocol.h \
src/hardware/appa-55ii/protocol.c \
src/hardware/appa-55ii/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
arachnid-labs-re-load-pro: Initial driver skeleton.
2015-12-29 00:58:05 +00:00
if HW_ARACHNID_LABS_RE_LOAD_PRO
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
arachnid-labs-re-load-pro: Initial driver skeleton.
2015-12-29 00:58:05 +00:00
src/hardware/arachnid-labs-re-load-pro/protocol.h \
src/hardware/arachnid-labs-re-load-pro/protocol.c \
src/hardware/arachnid-labs-re-load-pro/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_ASIX_SIGMA
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
asix-sigma: Split into api.c and protocol.[ch] modules.
2015-11-05 10:41:27 +00:00
src/hardware/asix-sigma/protocol.h \
src/hardware/asix-sigma/protocol.c \
src/hardware/asix-sigma/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_ATTEN_PPS3XXX
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/atten-pps3xxx/protocol.h \
src/hardware/atten-pps3xxx/protocol.c \
src/hardware/atten-pps3xxx/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
baylibre-acme: Initial driver skeleton. Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
2015-02-12 13:53:49 +00:00
if HW_BAYLIBRE_ACME
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
baylibre-acme: Initial driver skeleton. Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
2015-02-12 13:53:49 +00:00
src/hardware/baylibre-acme/protocol.h \
src/hardware/baylibre-acme/protocol.c \
baylibre-acme: Add Linux-specific GPIO helpers. These functions allow to export, read and set GPIOs using Linux sysfs interface. Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
2015-02-12 13:53:54 +00:00
src/hardware/baylibre-acme/api.c \
src/hardware/baylibre-acme/gpio.h \
src/hardware/baylibre-acme/gpio.c
baylibre-acme: Initial driver skeleton. Signed-off-by: Bartosz Golaszewski <bgolaszewski@baylibre.com>
2015-02-12 13:53:49 +00:00
endif
beaglelogic: Initial driver skeleton.
2014-07-21 20:11:52 +00:00
if HW_BEAGLELOGIC
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Makefile.am: Add beaglelogic.h so it ends up in the tarball.
2014-07-22 20:50:06 +00:00
src/hardware/beaglelogic/beaglelogic.h \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/beaglelogic/protocol.h \
src/hardware/beaglelogic/protocol.c \
src/hardware/beaglelogic/api.c
beaglelogic: Initial driver skeleton.
2014-07-21 20:11:52 +00:00
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_BRYMEN_BM86X
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/brymen-bm86x/protocol.h \
src/hardware/brymen-bm86x/protocol.c \
src/hardware/brymen-bm86x/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_BRYMEN_DMM
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/brymen-dmm/parser.c \
src/hardware/brymen-dmm/protocol.h \
src/hardware/brymen-dmm/protocol.c \
src/hardware/brymen-dmm/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_CEM_DT_885X
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/cem-dt-885x/protocol.h \
src/hardware/cem-dt-885x/protocol.c \
src/hardware/cem-dt-885x/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_CENTER_3XX
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/center-3xx/protocol.h \
src/hardware/center-3xx/protocol.c \
src/hardware/center-3xx/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
Rename 'chronovu-la8' driver to 'chronovu-la'. The driver now also supports the LA16 device (and possibly others in the future).
2014-04-01 19:47:11 +00:00
if HW_CHRONOVU_LA
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/chronovu-la/protocol.h \
src/hardware/chronovu-la/protocol.c \
src/hardware/chronovu-la/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_COLEAD_SLM
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/colead-slm/protocol.h \
src/hardware/colead-slm/protocol.c \
src/hardware/colead-slm/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_CONRAD_DIGI_35_CPU
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/conrad-digi-35-cpu/protocol.h \
src/hardware/conrad-digi-35-cpu/protocol.c \
src/hardware/conrad-digi-35-cpu/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_DEMO
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
demo: File naming consistency changes.
2016-09-18 23:48:56 +00:00
src/hardware/demo/protocol.h \
src/hardware/demo/protocol.c \
src/hardware/demo/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
dreamsourcelab-dslogic: Naming and other consistency fixes.
2017-06-20 21:07:27 +00:00
if HW_DREAMSOURCELAB_DSLOGIC
fx2lafw/dslogic: Split DSLogic into a separate driver
2017-06-10 00:02:29 +00:00
src_libdrivers_la_SOURCES += \
dreamsourcelab-dslogic: Naming and other consistency fixes.
2017-06-20 21:07:27 +00:00
src/hardware/dreamsourcelab-dslogic/protocol.h \
src/hardware/dreamsourcelab-dslogic/protocol.c \
src/hardware/dreamsourcelab-dslogic/api.c
fx2lafw/dslogic: Split DSLogic into a separate driver
2017-06-10 00:02:29 +00:00
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_FLUKE_DMM
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
fluke-dmm: File naming consistency changes.
2016-09-18 16:52:53 +00:00
src/hardware/fluke-dmm/protocol.h \
src/hardware/fluke-dmm/protocol.c \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/fluke-dmm/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
Rename 'ft2232h' driver to 'ftdi-la'. The driver already supports more than just the FT2232H chip.
2016-04-02 16:36:52 +00:00
if HW_FTDI_LA
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Rename 'ft2232h' driver to 'ftdi-la'. The driver already supports more than just the FT2232H chip.
2016-04-02 16:36:52 +00:00
src/hardware/ftdi-la/protocol.h \
src/hardware/ftdi-la/protocol.c \
src/hardware/ftdi-la/api.c
ft2232h: Add support for FTDI FT2232H/FT232R chip as LA. This closes bug #780.
2015-07-30 04:27:43 +00:00
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_FX2LAFW
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/fx2lafw/protocol.h \
src/hardware/fx2lafw/protocol.c \
fx2lafw/dslogic: Split DSLogic into a separate driver
2017-06-10 00:02:29 +00:00
src/hardware/fx2lafw/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_GMC_MH_1X_2X
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/gmc-mh-1x-2x/protocol.h \
src/hardware/gmc-mh-1x-2x/protocol.c \
src/hardware/gmc-mh-1x-2x/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
gwinstek-gds-800: Initial driver skeleton.
2015-08-16 12:52:49 +00:00
if HW_GWINSTEK_GDS_800
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
gwinstek-gds-800: Initial driver skeleton.
2015-08-16 12:52:49 +00:00
src/hardware/gwinstek-gds-800/protocol.h \
src/hardware/gwinstek-gds-800/protocol.c \
src/hardware/gwinstek-gds-800/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_HAMEG_HMO
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/hameg-hmo/protocol.h \
src/hardware/hameg-hmo/protocol.c \
src/hardware/hameg-hmo/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
hantek-6xxx: Initial driver skeleton.
2015-09-16 17:41:22 +00:00
if HW_HANTEK_6XXX
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
hantek-6xxx: Initial driver skeleton.
2015-09-16 17:41:22 +00:00
src/hardware/hantek-6xxx/protocol.h \
src/hardware/hantek-6xxx/protocol.c \
src/hardware/hantek-6xxx/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_HANTEK_DSO
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
hantek-dso: File naming consistency changes.
2016-09-18 16:54:56 +00:00
src/hardware/hantek-dso/protocol.h \
src/hardware/hantek-dso/protocol.c \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/hantek-dso/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
hp-3457a: Initial driver skeleton.
2016-02-28 06:05:27 +00:00
if HW_HP_3457A
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
hp-3457a: Initial driver skeleton.
2016-02-28 06:05:27 +00:00
src/hardware/hp-3457a/protocol.h \
src/hardware/hp-3457a/protocol.c \
src/hardware/hp-3457a/api.c
endif
hung-chang-dso-2100: Initial driver skeleton.
2015-09-07 00:18:13 +00:00
if HW_HUNG_CHANG_DSO_2100
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
hung-chang-dso-2100: Initial driver skeleton.
2015-09-07 00:18:13 +00:00
src/hardware/hung-chang-dso-2100/protocol.h \
src/hardware/hung-chang-dso-2100/protocol.c \
src/hardware/hung-chang-dso-2100/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_IKALOGIC_SCANALOGIC2
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/ikalogic-scanalogic2/protocol.h \
src/hardware/ikalogic-scanalogic2/protocol.c \
src/hardware/ikalogic-scanalogic2/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_IKALOGIC_SCANAPLUS
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/ikalogic-scanaplus/protocol.h \
src/hardware/ikalogic-scanaplus/protocol.c \
src/hardware/ikalogic-scanaplus/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_KECHENG_KC_330B
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/kecheng-kc-330b/protocol.h \
src/hardware/kecheng-kc-330b/protocol.c \
src/hardware/kecheng-kc-330b/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
kern-scale: Initial driver skeleton.
2015-08-27 19:42:46 +00:00
if HW_KERN_SCALE
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
kern-scale: Initial driver skeleton.
2015-08-27 19:42:46 +00:00
src/hardware/kern-scale/protocol.h \
src/hardware/kern-scale/protocol.c \
src/hardware/kern-scale/api.c
endif
korad-kdxxxxp: Rename driver to korad-kaxxxxp. This matches the supported / supportable devices better.
2015-11-26 23:16:05 +00:00
if HW_KORAD_KAXXXXP
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
korad-kdxxxxp: Rename driver to korad-kaxxxxp. This matches the supported / supportable devices better.
2015-11-26 23:16:05 +00:00
src/hardware/korad-kaxxxxp/protocol.h \
src/hardware/korad-kaxxxxp/protocol.c \
src/hardware/korad-kaxxxxp/api.c
korad-kdxxxxp: Initial driver skeleton.
2015-09-24 19:58:38 +00:00
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_LASCAR_EL_USB
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/lascar-el-usb/protocol.h \
src/hardware/lascar-el-usb/protocol.c \
src/hardware/lascar-el-usb/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
lecroy-logicstudio: Initial driver implementation. This supports both 8 and 16 channel modes with samplerates up to 500 MHz. Voltage thresholds are currently fixed at 1.58V.
2015-10-08 18:13:53 +00:00
if HW_LECROY_LOGICSTUDIO
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
lecroy-logicstudio: Initial driver implementation. This supports both 8 and 16 channel modes with samplerates up to 500 MHz. Voltage thresholds are currently fixed at 1.58V.
2015-10-08 18:13:53 +00:00
src/hardware/lecroy-logicstudio/protocol.h \
src/hardware/lecroy-logicstudio/protocol.c \
src/hardware/lecroy-logicstudio/api.c
endif
lecroy-xstream: Initial driver skeleton. Signed-off-by: Sven Schnelle <svens@stackframe.org>
2017-02-12 19:18:15 +00:00
if HW_LECROY_XSTREAM
src_libdrivers_la_SOURCES += \
src/hardware/lecroy-xstream/protocol.h \
src/hardware/lecroy-xstream/protocol.c \
src/hardware/lecroy-xstream/api.c
endif
manson-hcs-3xxx: Initial driver skeleton.
2014-06-22 22:31:22 +00:00
if HW_MANSON_HCS_3XXX
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/manson-hcs-3xxx/protocol.h \
src/hardware/manson-hcs-3xxx/protocol.c \
src/hardware/manson-hcs-3xxx/api.c
manson-hcs-3xxx: Initial driver skeleton.
2014-06-22 22:31:22 +00:00
endif
maynuo-m97: Initial driver skeleton.
2015-02-01 22:49:03 +00:00
if HW_MAYNUO_M97
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
maynuo-m97: Initial driver skeleton.
2015-02-01 22:49:03 +00:00
src/hardware/maynuo-m97/protocol.h \
src/hardware/maynuo-m97/protocol.c \
src/hardware/maynuo-m97/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_MIC_985XX
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/mic-985xx/protocol.h \
src/hardware/mic-985xx/protocol.c \
src/hardware/mic-985xx/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
motech-lps-30x: Initial driver framework.
2014-05-19 17:23:30 +00:00
if HW_MOTECH_LPS_30X
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/motech-lps-30x/protocol.h \
src/hardware/motech-lps-30x/protocol.c \
src/hardware/motech-lps-30x/api.c
motech-lps-30x: Initial driver framework.
2014-05-19 17:23:30 +00:00
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_NORMA_DMM
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/norma-dmm/protocol.h \
src/hardware/norma-dmm/protocol.c \
src/hardware/norma-dmm/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
autoconf: Use DRIVER macro to declare drivers.
2014-07-10 15:26:22 +00:00
if HW_OPENBENCH_LOGIC_SNIFFER
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/openbench-logic-sniffer/protocol.h \
src/hardware/openbench-logic-sniffer/protocol.c \
src/hardware/openbench-logic-sniffer/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
pce-322a: Initial driver implementation.
2016-04-30 20:34:15 +00:00
if HW_PCE_322A
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
pce-322a: Initial driver implementation.
2016-04-30 20:34:15 +00:00
src/hardware/pce-322a/protocol.h \
src/hardware/pce-322a/protocol.c \
src/hardware/pce-322a/api.c
endif
added pipistrello-ols Conflicts: configure.ac src/hwdriver.c
2014-05-10 22:44:13 +00:00
if HW_PIPISTRELLO_OLS
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
added pipistrello-ols Conflicts: configure.ac src/hwdriver.c
2014-05-10 22:44:13 +00:00
src/hardware/pipistrello-ols/protocol.h \
src/hardware/pipistrello-ols/protocol.c \
src/hardware/pipistrello-ols/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_RIGOL_DS
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/rigol-ds/protocol.h \
src/hardware/rigol-ds/protocol.c \
src/hardware/rigol-ds/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
rohde-schwarz-sme-0x: Initial device support Signed-off-by: Vlad Ivanov <vlad.ivanov@lab-systems.ru>
2016-10-27 13:22:55 +00:00
if HW_ROHDE_SCHWARZ_SME_0X
src_libdrivers_la_SOURCES += \
src/hardware/rohde-schwarz-sme-0x/protocol.h \
src/hardware/rohde-schwarz-sme-0x/protocol.c \
src/hardware/rohde-schwarz-sme-0x/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_SALEAE_LOGIC16
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/saleae-logic16/protocol.h \
src/hardware/saleae-logic16/protocol.c \
src/hardware/saleae-logic16/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
scpi-pps: Initial driver skeleton.
2014-08-31 15:20:39 +00:00
if HW_SCPI_PPS
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
scpi-pps: Initial driver skeleton.
2014-08-31 15:20:39 +00:00
src/hardware/scpi-pps/protocol.h \
src/hardware/scpi-pps/protocol.c \
scpi-pps: Add support for Rigol DP832.
2014-09-05 01:23:32 +00:00
src/hardware/scpi-pps/profiles.c \
scpi-pps: Initial driver skeleton.
2014-08-31 15:20:39 +00:00
src/hardware/scpi-pps/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_SERIAL_DMM
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/serial-dmm/protocol.h \
src/hardware/serial-dmm/protocol.c \
src/hardware/serial-dmm/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
deree-de5000: rename to serial-lcr, update configure logic The former DER EE DE-5000 driver was a very thin wrapper around the ES51919 LCR meter chipset. None of its source was specific to the deree-de5000 device. In fact it contained code for all currently supported LCR meters, and it's expected that all LCR meters which will get added in the future will fit in as well. Follow the serial-dmm model. Rename the src/hardware/deree-de5000/ directory to serial-lcr/. Update the configure logic. Although the source directory and the configure option are named serial-lcr, the LCR meter still is used by specifying the "deree-de5000" device driver (which just happens to reside in the serial-lcr driver sources, among others).
2016-12-26 12:12:35 +00:00
if HW_SERIAL_LCR
src_libdrivers_la_SOURCES += \
src/hardware/serial-lcr/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_SYSCLK_LWLA
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/sysclk-lwla/lwla.h \
src/hardware/sysclk-lwla/lwla.c \
sysclk-lwla: Implement support for LWLA1016 Refactor the sysclk-lwla driver to separate the generic logic from the model-specific implementation. Based on this, implement support for the SysClk LWLA1016 device.
2015-11-22 16:45:54 +00:00
src/hardware/sysclk-lwla/lwla1016.c \
src/hardware/sysclk-lwla/lwla1034.c \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/sysclk-lwla/protocol.h \
src/hardware/sysclk-lwla/protocol.c \
src/hardware/sysclk-lwla/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_TELEINFO
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/teleinfo/protocol.h \
src/hardware/teleinfo/protocol.c \
src/hardware/teleinfo/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
testo: Initial driver skeleton.
2014-06-30 23:34:29 +00:00
if HW_TESTO
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/testo/protocol.h \
src/hardware/testo/protocol.c \
src/hardware/testo/api.c
testo: Initial driver skeleton.
2014-06-30 23:34:29 +00:00
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_TONDAJ_SL_814
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/tondaj-sl-814/protocol.h \
src/hardware/tondaj-sl-814/protocol.c \
src/hardware/tondaj-sl-814/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_UNI_T_DMM
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/uni-t-dmm/protocol.h \
src/hardware/uni-t-dmm/protocol.c \
src/hardware/uni-t-dmm/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_UNI_T_UT32X
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/uni-t-ut32x/protocol.h \
src/hardware/uni-t-ut32x/protocol.c \
src/hardware/uni-t-ut32x/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
if HW_VICTOR_DMM
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/victor-dmm/protocol.h \
src/hardware/victor-dmm/protocol.c \
src/hardware/victor-dmm/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
yokogawa-dlm: Integrate driver skeleton
2014-08-25 11:09:42 +00:00
if HW_YOKOGAWA_DLM
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
yokogawa-dlm: Integrate driver skeleton
2014-08-25 11:09:42 +00:00
src/hardware/yokogawa-dlm/protocol.h \
src/hardware/yokogawa-dlm/protocol.c \
src/hardware/yokogawa-dlm/protocol_wrappers.h \
src/hardware/yokogawa-dlm/protocol_wrappers.c \
src/hardware/yokogawa-dlm/api.c
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
if HW_ZEROPLUS_LOGIC_CUBE
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
src_libdrivers_la_SOURCES += \
Reorganize project tree.
2014-07-21 19:27:33 +00:00
src/hardware/zeroplus-logic-cube/analyzer.c \
src/hardware/zeroplus-logic-cube/analyzer.h \
src/hardware/zeroplus-logic-cube/gl_usb.h \
src/hardware/zeroplus-logic-cube/gl_usb.c \
src/hardware/zeroplus-logic-cube/protocol.h \
src/hardware/zeroplus-logic-cube/protocol.c \
src/hardware/zeroplus-logic-cube/api.c
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
endif
group all drivers into a single object This single object also contains the sr_drivers_init function, that will always be referenced. That ensures that the drivers object files won't be optimized out during static linking due to the fact that they are not referenced directly. This addresses (parts of) bug #802.
2016-06-05 21:10:44 +00:00
libsigrok_la_LIBADD = src/libdrivers.lo $(SR_EXTRA_LIBS) $(LIBSIGROK_LIBS)
Build: Cleanup up and modernize autotools setup Note that some --enable options of configure have become --with or --without options.
2015-08-15 13:59:52 +00:00
libsigrok_la_LDFLAGS = -version-info $(SR_LIB_VERSION) -no-undefined
windows: Fix building shared library on MinGW.
2015-03-18 12:30:26 +00:00
sr: moved sigrok.h so libsigrok/libsigrok.h All frontends will have to include <libsigrok/libsigrok.h> from now on. This header includes proto.h and version.h, both installed from the distribution into $INCLUDE/libsigrok/ as well. The only dynamically changed header is now version.h, which has both libsigrok and libtool compile-time versions in it.
2012-07-04 22:55:07 +00:00
library_includedir = $(includedir)/libsigrok
Move public headers into include/libsigrok. This is needed so that the C++ bindings, the header for which references "libsigrok/libsigrok.h", can have a valid include directory passed to build them before the headers are installed.
2014-04-26 02:03:11 +00:00
library_include_HEADERS = \
include/libsigrok/libsigrok.h \
Build: Do not distribute generated headers version.h and enums.hpp are generated at build time and should not be included in the distribution tarball. This seems to have triggered a weird error when doing a VPATH build of the dist tarball.
2015-08-15 16:33:05 +00:00
include/libsigrok/proto.h
nodist_library_include_HEADERS = \
Move public headers into include/libsigrok. This is needed so that the C++ bindings, the header for which references "libsigrok/libsigrok.h", can have a valid include directory passed to build them before the headers are installed.
2014-04-26 02:03:11 +00:00
include/libsigrok/version.h
Reorganize project tree.
2014-07-21 19:27:33 +00:00
noinst_HEADERS = src/libsigrok-internal.h
Start of code base layout restructuring.
2010-04-02 18:18:27 +00:00
Revert temporary changes for 0.2 release.
2011-04-03 21:18:46 +00:00
pkgconfigdir = $(libdir)/pkgconfig
pkgconfig_DATA = libsigrok.pc
pkg-config: Add (semi-generated) libsigrok.pc.
2010-05-11 16:07:54 +00:00
Makefile.am: Install MIME info file in $(datadir)/mime/packages. This fixes bug #983.
2017-06-15 16:46:14 +00:00
mimeappdir = $(datadir)/mime/packages
Add a MIME info file (and icons) for sigrok session files. File template by Stefan Brüns, thanks! This fixes bug #857. (the XML file is moved from PulseView to libsigrok since this is not PulseView-specific) Add a 48x48 PNG and a scalable SVG for the MIME type as well. Install the XML file and in the icons in the respective standard paths.
2017-06-11 21:19:27 +00:00
mimeapp_DATA = contrib/vnd.sigrok.session.xml
mimeicondir = $(datadir)/icons/hicolor/48x48/mimetypes
mimeicon_DATA = contrib/libsigrok.png
mimeiconscalabledir = $(datadir)/icons/hicolor/scalable/mimetypes
mimeiconscalable_DATA = contrib/libsigrok.svg
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
EXTRA_DIST = \
Doxyfile \
HACKING \
README.devices \
Build: Distribute bindings sources A couple of files referenced during the build were missing from the distribution, causing a tarball build failure.
2015-08-15 14:59:47 +00:00
bindings/cxx/ConfigKey_methods.cpp \
bindings/cxx/ConfigKey_methods.hpp \
bindings/cxx/ConfigKey_methods.i \
bindings/cxx/Doxyfile \
bindings/cxx/QuantityFlag_methods.cpp \
bindings/cxx/QuantityFlag_methods.hpp \
bindings/cxx/enums.py \
bindings/python/Doxyfile \
bindings/python/setup.py \
bindings/python/sigrok/__init__.py \
bindings/python/sigrok/core/__init__.py \
bindings/python/sigrok/core/classes.i \
build: Fix distribution of Ruby bindings The bindings file was not listed in EXTRA_DIST and therefore not distributed. We also need to provide an target to uninstall the Ruby bindings and add it to UNINSTALL_EXTRA in order to make `make distcheck` happy. This fixes bug #741
2016-05-13 15:44:12 +00:00
bindings/ruby/classes.i \
Build: Distribute bindings sources A couple of files referenced during the build were missing from the distribution, causing a tarball build failure.
2015-08-15 14:59:47 +00:00
bindings/java/Doxyfile \
bindings/java/org/sigrok/core/classes/classes.i \
bindings/java/org/sigrok/core/interfaces/DatafeedCallback.java \
bindings/java/org/sigrok/core/interfaces/LogCallback.java \
bindings/swig/classes.i \
bindings/swig/doc.py \
build: Distribute missing files
2015-11-07 20:57:20 +00:00
bindings/swig/templates.i \
Rename sigrok-logo-notext.png to libsigrok_112x112.png. This is more specific and prevents any potential issues e.g. when multiple distro packages might ship with a generic file like sigrok-logo-notext.png that's supposed to be installed in the same place.
2017-06-11 21:03:09 +00:00
contrib/libsigrok_112x112.png \
Add a MIME info file (and icons) for sigrok session files. File template by Stefan Brüns, thanks! This fixes bug #857. (the XML file is moved from PulseView to libsigrok since this is not PulseView-specific) Add a 48x48 PNG and a scalable SVG for the MIME type as well. Install the XML file and in the icons in the respective standard paths.
2017-06-11 21:19:27 +00:00
contrib/libsigrok.png \
contrib/libsigrok.svg \
contrib/vnd.sigrok.session.xml \
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
contrib/z60_libsigrok.rules
if HAVE_CHECK
unit tests: Drop unneeded check_ filename prefix.
2015-02-16 21:39:19 +00:00
TESTS = tests/main
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
check_PROGRAMS = ${TESTS}
Build: Use TESTS prefix instead of CHECK for flag variables In order to avoid confusion of the flags-gathering pkg-config result with the actual test for the availability of "check", change the pkg-config output variable prefix from CHECK to TESTS.
2015-08-26 11:13:39 +00:00
endif
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
unit tests: Drop unneeded check_ filename prefix.
2015-02-16 21:39:19 +00:00
tests_main_SOURCES = \
Fix 'make check' invocation wrt new libsigrok.h location.
2014-07-20 10:03:29 +00:00
include/libsigrok/libsigrok.h \
Switch to a non-recursive automake setup. Instead of >= 44 Makefile.am's we now only have one top-level Makefile.am, and use the 'subdir-objects' automake option to handle the build via non-recursive (auto)make. This has the advantage of fewer (boilerplate or other) files and less clutter in general, as well as performance advantages since the new setup can build many files in parallel (with 'make -j'), not only 2 or 3 files within the same (e.g. hardware/xxxx/* subdirectory) and also since we no longer need to build intermediate libtool helper libs per subdirectory. A quick, non-scientific test build on a quad-core laptop with 'make -j 4' yields a build time reduction from 35s to 19s. All autotools features that worked before are still intact without any regressions, including the Make targets 'install', 'uninstall', 'check', 'dist', 'clean', 'distclean' and so on, as well as all the usual portability handling (build works on any OS, with any Make implementation such as GNU Make or BSD Make, with any shell such as sh/ksh/zsh/bash/dash, etc. etc.) and features such as out-of-tree build support, cross-compile support, testsuite support (also with colored output), "silent make rules", etc. etc.
2014-03-19 22:04:55 +00:00
tests/lib.c \
tests/lib.h \
unit tests: Drop unneeded check_ filename prefix.
2015-02-16 21:39:19 +00:00
tests/main.c \
tests/core.c \
tests/input_all.c \
tests/input_binary.c \
tests/output_all.c \
tests/transform_all.c \
tests/session.c \
tests/strutil.c \
tests/version.c \
tests/driver_all.c \
tests/device.c \
Add a few unit tests for the analog conversion functions.
2015-10-15 10:13:43 +00:00
tests/trigger.c \
tests/analog.c
unit tests: Drop unneeded check_ filename prefix.
2015-02-16 21:39:19 +00:00
Build: Use TESTS prefix instead of CHECK for flag variables In order to avoid confusion of the flags-gathering pkg-config result with the actual test for the availability of "check", change the pkg-config output variable prefix from CHECK to TESTS.
2015-08-26 11:13:39 +00:00
tests_main_LDADD = libsigrok.la $(SR_EXTRA_LIBS) $(TESTS_LIBS)
sr/srd: Add Doxyfile to the tarball.
2012-01-04 01:50:26 +00:00
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
BUILD_EXTRA =
INSTALL_EXTRA =
build: Enable uninstall of Java bindings This is needed for make distcheck to work.
2015-11-07 21:20:19 +00:00
UNINSTALL_EXTRA =
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
CLEAN_EXTRA =
uninstall: Remove empty libsigrok/libsigrokcxx include directories. Change uninstall-local to uninstall-hook, since the latter is guaranteed to run last (order is apparently not guaranteed for uninstall-local). This fixes bug #861.
2017-06-03 16:25:54 +00:00
libsigrok-uninstall:
-rmdir $(DESTDIR)$(includedir)/libsigrok
UNINSTALL_EXTRA += libsigrok-uninstall
Add C++ bindings.
2014-07-17 17:39:49 +00:00
if BINDINGS_CXX
Rename libsigrokxx to the more common libsigrokcxx.
2015-01-17 17:49:04 +00:00
lib_LTLIBRARIES += bindings/cxx/libsigrokcxx.la
Add C++ bindings.
2014-07-17 17:39:49 +00:00
Rename libsigrokxx to the more common libsigrokcxx.
2015-01-17 17:49:04 +00:00
bindings_cxx_libsigrokcxx_la_SOURCES = bindings/cxx/classes.cpp
Add C++ bindings.
2014-07-17 17:39:49 +00:00
Bindings: Link C++ code with gnustl_shared if it exists (Android)
2016-12-06 20:01:23 +00:00
bindings_cxx_libsigrokcxx_la_LIBADD = libsigrok.la $(SR_EXTRA_LIBS) $(LIBSIGROKCXX_LIBS) $(SR_EXTRA_CXX_LIBS)
Build: Cleanup up and modernize autotools setup Note that some --enable options of configure have become --with or --without options.
2015-08-15 13:59:52 +00:00
bindings_cxx_libsigrokcxx_la_LDFLAGS = -version-info $(SR_LIB_VERSION) -no-undefined
cxx: Makefile.am: Use libtool's -no-undefined option.
2015-04-07 00:29:09 +00:00
Install libsigrokcxx headers in separate directory. Don't mix libsigrokcxx header files into the $prefix/include/libsigrok directory. Use a separate $prefix/include/libsigrokcxx instead.
2015-01-17 19:29:05 +00:00
bindings_cxx_libsigrokcxx_la_includedir = $(includedir)/libsigrokcxx
bindings_cxx_libsigrokcxx_la_include_HEADERS = \
Build: Do not distribute generated headers version.h and enums.hpp are generated at build time and should not be included in the distribution tarball. This seems to have triggered a weird error when doing a VPATH build of the dist tarball.
2015-08-15 16:33:05 +00:00
bindings/cxx/include/libsigrokcxx/libsigrokcxx.hpp
nodist_bindings_cxx_libsigrokcxx_la_include_HEADERS = \
cxx: Rename include/libsigrok/ to include/libsigrokcxx/.
2015-01-17 19:11:53 +00:00
bindings/cxx/include/libsigrokcxx/enums.hpp
Add C++ bindings.
2014-07-17 17:39:49 +00:00
Rename libsigrokxx to the more common libsigrokcxx.
2015-01-17 17:49:04 +00:00
pkgconfig_DATA += bindings/cxx/libsigrokcxx.pc
Add C++ bindings.
2014-07-17 17:39:49 +00:00
C++ bindings: Reimplement enums.py using doxygen XML output instead of gccxml.
2014-08-01 16:18:09 +00:00
doxy/xml/index.xml: include/libsigrok/libsigrok.h
Makefile.am: Fix doxygen invocation in out-of-tree build
2014-08-15 12:42:18 +00:00
$(AM_V_GEN)cd $(srcdir) && BUILDDIR=$(abs_builddir)/ doxygen Doxyfile 2>/dev/null
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
Build: Depend on swig/enums.i where appropriate
2015-08-15 17:18:36 +00:00
bindings/swig/enums.i: bindings/cxx/enums.timestamp
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
bindings/cxx/enums.cpp: bindings/cxx/enums.timestamp
cxx: Rename include/libsigrok/ to include/libsigrokcxx/.
2015-01-17 19:11:53 +00:00
bindings/cxx/include/libsigrokcxx/enums.hpp: bindings/cxx/enums.timestamp
Add C++ bindings.
2014-07-17 17:39:49 +00:00
Fix problem building with BSD make This fixes bug #556: Bug 556 - libsigrok fails to build with BSD Make
2015-02-10 13:09:28 +00:00
bindings/cxx/enums.timestamp: $(srcdir)/bindings/cxx/enums.py doxy/xml/index.xml \
Add C++ bindings.
2014-07-17 17:39:49 +00:00
bindings/cxx/ConfigKey_methods.cpp bindings/cxx/QuantityFlag_methods.cpp
Fix problem building with BSD make This fixes bug #556: Bug 556 - libsigrok fails to build with BSD Make
2015-02-10 13:09:28 +00:00
$(AM_V_GEN)$(PYTHON) $(srcdir)/bindings/cxx/enums.py doxy/xml/index.xml
build: Silence language bindings building.
2014-07-19 18:21:19 +00:00
$(AM_V_at)touch $@
Add C++ bindings.
2014-07-17 17:39:49 +00:00
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
bindings/cxx/classes.lo: bindings/cxx/classes.cpp bindings/cxx/enums.cpp \
Build: Depend on swig/enums.i where appropriate
2015-08-15 17:18:36 +00:00
$(library_include_HEADERS) $(nodist_library_include_HEADERS)
Add C++ bindings.
2014-07-17 17:39:49 +00:00
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
cxx-clean:
C++ bindings: Reimplement enums.py using doxygen XML output instead of gccxml.
2014-08-01 16:18:09 +00:00
rm -rf doxy/
bindings: Transfer C++ documentation strings to Python and Java wrappers.
2014-08-01 17:17:33 +00:00
rm -rf bindings/cxx/doxy/
build: Delete generated files on make clean
2015-11-07 21:29:12 +00:00
rm -f bindings/swig/enums.i
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
rm -f bindings/cxx/enums.cpp
cxx: Rename include/libsigrok/ to include/libsigrokcxx/.
2015-01-17 19:11:53 +00:00
rm -f bindings/cxx/include/libsigrokcxx/enums.hpp
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
rm -f bindings/cxx/enums.timestamp
Add C++ bindings.
2014-07-17 17:39:49 +00:00
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
CLEAN_EXTRA += cxx-clean
uninstall: Remove empty libsigrok/libsigrokcxx include directories. Change uninstall-local to uninstall-hook, since the latter is guaranteed to run last (order is apparently not guaranteed for uninstall-local). This fixes bug #861.
2017-06-03 16:25:54 +00:00
cxx-uninstall:
-rmdir $(DESTDIR)$(includedir)/libsigrokcxx
UNINSTALL_EXTRA += cxx-uninstall
Additional build system fixes for C++/Python/Java bindings.
2014-07-19 16:05:29 +00:00
endif
Reimplement high-level Python bindings on top of SWIG/C++ bindings.
2014-07-17 17:45:29 +00:00
bindings: Transfer C++ documentation strings to Python and Java wrappers.
2014-08-01 17:17:33 +00:00
CPPXMLDOC = bindings/cxx/doxy/xml/index.xml
cxx: Rename include/libsigrok/ to include/libsigrokcxx/.
2015-01-17 19:11:53 +00:00
$(CPPXMLDOC): bindings/cxx/include/libsigrokcxx/libsigrokcxx.hpp \
bindings: Transfer C++ documentation strings to Python and Java wrappers.
2014-08-01 17:17:33 +00:00
bindings/cxx/enums.timestamp
Makefile.am: Fix out-of-tree build for C++ and Java bindings
2014-08-14 21:07:45 +00:00
$(AM_V_GEN)cd $(srcdir)/bindings/cxx && BUILDDIR=$(abs_builddir)/bindings/cxx/ doxygen Doxyfile 2>/dev/null
bindings: Transfer C++ documentation strings to Python and Java wrappers.
2014-08-01 17:17:33 +00:00
build: Use common SWIG defines for Python and Java Make the Python and Java bindings use the same set of preprocessor macros for the SWIG parsing stage, taken from a make variable. Add G_GNUC_{BEGIN,END}_IGNORE_DEPRECATIONS to that set.
2015-10-29 21:39:44 +00:00
# Macro definitions to be used by the SWIG parser.
swig_defs = -Dnoexcept= -Dprivate=protected -DG_GNUC_BEGIN_IGNORE_DEPRECATIONS= -DG_GNUC_END_IGNORE_DEPRECATIONS=
Reimplement high-level Python bindings on top of SWIG/C++ bindings.
2014-07-17 17:45:29 +00:00
if BINDINGS_PYTHON
Assorted build system fixes for C++/Python/Java bindings.
2014-07-19 15:23:45 +00:00
PDIR = bindings/python
python: Give all enum values __doc__ strings.
2015-10-25 14:04:49 +00:00
PDOC_START = bindings/python/sigrok/core/doc_start.i
PDOC_END = bindings/python/sigrok/core/doc_end.i
bindings: Transfer C++ documentation strings to Python and Java wrappers.
2014-08-01 17:17:33 +00:00
Build: fix #865 yet again It appears that the symmetry changes of setting CC and CFLAGS correctly for C code compilation in commit 104f02f broke things for people using some other version of setuptools which uses those vars instead of CXX and CXXFLAGS when compiling C++ code. In order to make this work everywhere, set _both_ sets of variables as required for C++ compilation. No C code is compiled by the python binding module anyway.
2017-02-01 17:50:26 +00:00
setup_vars = VERSION='$(PACKAGE_VERSION)' CC='$(CXX)' CXX='$(CXX)' CFLAGS='$(CXXFLAGS) $(SR_WXXFLAGS) $(PYSIGROK_CFLAGS)' CXXFLAGS='$(CXXFLAGS) $(SR_WXXFLAGS) $(PYSIGROK_CFLAGS)' LDADD='$(PYSIGROK_LIBS)'
Build: Pass compiler flags from make to setup.py Extend setup.py to allow environment variables to be set on the command line. Use that functionality to replace the pkg-config invocations with flags passed on from make. Suppress the annoying -Wstrict-prototypes warning by overriding the OPT variable. Also move the "cd bindings/python" from Makefile.am to setup.py to side-step problems with "cd" in make rules. This also fixes bug #628.
2015-08-28 23:54:17 +00:00
setup_quiet = --quiet
setup_py = $(PYTHON) $(srcdir)/$(PDIR)/setup.py $(setup_vars) $(setup_quiet)
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
python: Give all enum values __doc__ strings.
2015-10-25 14:04:49 +00:00
$(PDOC_START): $(srcdir)/bindings/swig/doc.py $(CPPXMLDOC)
Makefile.am: Use $(MKDIR_P) everywhere.
2015-01-12 00:19:12 +00:00
$(AM_V_at)test -d $(PDIR)/sigrok/core || $(MKDIR_P) $(PDIR)/sigrok/core
python: Give all enum values __doc__ strings.
2015-10-25 14:04:49 +00:00
$(AM_V_GEN)$(PYTHON) $(srcdir)/bindings/swig/doc.py python $(CPPXMLDOC) start > $@
$(PDOC_END): $(srcdir)/bindings/swig/doc.py $(CPPXMLDOC)
$(AM_V_at)test -d $(PDIR)/sigrok/core || $(MKDIR_P) $(PDIR)/sigrok/core
$(AM_V_GEN)$(PYTHON) $(srcdir)/bindings/swig/doc.py python $(CPPXMLDOC) end > $@
Assorted build system fixes for C++/Python/Java bindings.
2014-07-19 15:23:45 +00:00
python-build: $(PDIR)/timestamp
Don't rebuild Python or Java bindings unless C++ interface changes.
2015-09-07 11:32:05 +00:00
$(PDIR)/timestamp: $(PDIR)/sigrok/core/classes.i \
Makefile.am: Make SWIG wrappers also depend on new templates.i file.
2015-10-25 14:43:15 +00:00
bindings/swig/classes.i bindings/swig/templates.i \
bindings/swig/enums.i $(PDOC_START) $(PDOC_END) \
$(library_include_HEADERS) \
python: Give all enum values __doc__ strings.
2015-10-25 14:04:49 +00:00
$(nodist_library_include_HEADERS) \
Don't rebuild Python or Java bindings unless C++ interface changes.
2015-09-07 11:32:05 +00:00
$(bindings_cxx_libsigrokcxx_la_include_HEADERS) \
$(nodist_bindings_cxx_libsigrokcxx_la_include_HEADERS) \
Makefile.am: Use @ORDER@ instead of |. This is set to | (or left empty) by SR_PROG_MAKE_ORDER_ONLY for portability reasons, since not all Make implementations support order-only prerequisites.
2015-09-09 06:21:01 +00:00
@ORDER@ bindings/cxx/libsigrokcxx.la
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
$(AM_V_at)$(setup_py) clean --all 2>/dev/null
build: Use common SWIG defines for Python and Java Make the Python and Java bindings use the same set of preprocessor macros for the SWIG parsing stage, taken from a make variable. Add G_GNUC_{BEGIN,END}_IGNORE_DEPRECATIONS to that set.
2015-10-29 21:39:44 +00:00
$(AM_V_GEN)$(setup_py) build_ext --swig "$(SWIG)" --swig-opts '$(swig_defs)' build_py
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
$(AM_V_at): >$@
Reimplement high-level Python bindings on top of SWIG/C++ bindings.
2014-07-17 17:45:29 +00:00
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
python-install:
$(AM_V_at)$(MKDIR_P) "$(DESTDIR)$(prefix)" "$(DESTDIR)$(exec_prefix)"
Build: Fix installation of Python bindings Apparently setup.py install behaves differently when given the --root option, so omit that if DESTDIR is empty. Fixes #644.
2015-09-02 12:07:06 +00:00
destdir='$(DESTDIR)'; $(setup_py) install $${destdir:+"--root=$$destdir"} \
--prefix "$(prefix)" --exec-prefix "$(exec_prefix)"
Reimplement high-level Python bindings on top of SWIG/C++ bindings.
2014-07-17 17:45:29 +00:00
python: Fix build dependency. This also makes it more portable, notably to non-bash shells.
2014-07-22 02:31:54 +00:00
python-clean:
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
-$(AM_V_at)rm -f $(PDIR)/timestamp
-$(AM_V_at)rm -fr $(PDIR)/doxy
Build: Pass compiler flags from make to setup.py Extend setup.py to allow environment variables to be set on the command line. Use that functionality to replace the pkg-config invocations with flags passed on from make. Suppress the annoying -Wstrict-prototypes warning by overriding the OPT variable. Also move the "cd bindings/python" from Makefile.am to setup.py to side-step problems with "cd" in make rules. This also fixes bug #628.
2015-08-28 23:54:17 +00:00
-$(AM_V_at)$(setup_py) clean --all 2>/dev/null
python: Add documentation generation.
2014-08-01 18:22:24 +00:00
python-doc:
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
$(AM_V_at)cd $(srcdir)/$(PDIR) && BUILDDIR="$(abs_builddir)/$(PDIR)/" doxygen Doxyfile 2>/dev/null
Reimplement high-level Python bindings on top of SWIG/C++ bindings.
2014-07-17 17:45:29 +00:00
BUILD_EXTRA += python-build
INSTALL_EXTRA += python-install
CLEAN_EXTRA += python-clean
endif
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
if BINDINGS_RUBY
RDIR = bindings/ruby
RDOC = $(RDIR)/doc.i
RWRAP = $(RDIR)/classes_wrap.cpp
ROBJ = $(RWRAP:.cpp=.o)
configure: replace buggy AX_RUBY_EXT by a proper implementation This actually allows to get the proper DLEXT expected by Ruby on the target system.
2016-01-28 21:48:44 +00:00
REXT = $(RDIR)/sigrok.$(RUBY_DLEXT)
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
$(RDOC): $(srcdir)/bindings/swig/doc.py $(CPPXMLDOC)
ruby: Fix out-of-tree build of the bindings. This closes bug #797.
2016-05-12 21:08:21 +00:00
$(AM_V_at)test -d $(RDIR) || $(MKDIR_P) $(RDIR)
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
$(AM_V_GEN)$(PYTHON) $(srcdir)/bindings/swig/doc.py ruby $(CPPXMLDOC) > $@
Build: avoid $< in ruby bindings rules This fixes building with BSD make. This closes bug #801.
2016-05-25 20:38:41 +00:00
$(RWRAP): $(srcdir)/$(RDIR)/classes.i $(RDOC) \
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
bindings/swig/classes.i bindings/swig/templates.i \
bindings/swig/enums.i
Build: avoid $< in ruby bindings rules This fixes building with BSD make. This closes bug #801.
2016-05-25 20:38:41 +00:00
$(AM_V_GEN)$(SWIG) -ruby -c++ -Ibindings -Ibindings/cxx/include -I$(RDIR) $(local_includes) $(swig_defs) -o $@ $(srcdir)/$(RDIR)/classes.i
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
$(ROBJ): $(RWRAP) \
$(library_include_HEADERS) \
$(nodist_library_include_HEADERS) \
$(bindings_cxx_libsigrokcxx_la_include_HEADERS) \
$(nodist_bindings_cxx_libsigrokcxx_la_include_HEADERS)
Build: avoid $< in ruby bindings rules This fixes building with BSD make. This closes bug #801.
2016-05-25 20:38:41 +00:00
$(AM_V_CXX)$(CXX) $(RBSIGROK_CFLAGS) -I. -Iinclude -Ibindings/cxx/include $(local_includes) -fPIC -o $@ -c $(RWRAP)
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
$(REXT): $(ROBJ) @ORDER@ bindings/cxx/libsigrokcxx.la
Build: avoid $< in ruby bindings rules This fixes building with BSD make. This closes bug #801.
2016-05-25 20:38:41 +00:00
$(AM_V_CXXLD)$(CXX) -shared -o $@ $(ROBJ) -lsigrokcxx -Lbindings/cxx/.libs $(RBSIGROK_LIBS)
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
ruby-build: $(REXT)
ruby-install: $(REXT)
configure: replace buggy AX_RUBY_EXT by a proper implementation This actually allows to get the proper DLEXT expected by Ruby on the target system.
2016-01-28 21:48:44 +00:00
$(INSTALL) -d $(DESTDIR)$(prefix)/$(RBSIGROK_EXTDIR)
Build: avoid $< in ruby bindings rules This fixes building with BSD make. This closes bug #801.
2016-05-25 20:38:41 +00:00
$(INSTALL) $(REXT) $(DESTDIR)$(prefix)/$(RBSIGROK_EXTDIR)
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
build: Fix distribution of Ruby bindings The bindings file was not listed in EXTRA_DIST and therefore not distributed. We also need to provide an target to uninstall the Ruby bindings and add it to UNINSTALL_EXTRA in order to make `make distcheck` happy. This fixes bug #741
2016-05-13 15:44:12 +00:00
ruby-uninstall:
rm -f $(DESTDIR)$(prefix)/$(RBSIGROK_EXTDIR)/sigrok.$(RUBY_DLEXT)
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
ruby-clean:
-$(AM_V_at)rm -fr $(RDIR)/doc
ruby: Fix out-of-tree build of the bindings. This closes bug #797.
2016-05-12 21:08:21 +00:00
-$(AM_V_at)rm -f $(REXT) $(ROBJ) $(RWRAP) $(RDOC)
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
ruby-doc: $(RWRAP)
Build: avoid $< in ruby bindings rules This fixes building with BSD make. This closes bug #801.
2016-05-25 20:38:41 +00:00
$(AM_V_at)yard doc -o $(RDIR)/doc $(RWRAP)
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
BUILD_EXTRA += ruby-build
INSTALL_EXTRA += ruby-install
build: Fix distribution of Ruby bindings The bindings file was not listed in EXTRA_DIST and therefore not distributed. We also need to provide an target to uninstall the Ruby bindings and add it to UNINSTALL_EXTRA in order to make `make distcheck` happy. This fixes bug #741
2016-05-13 15:44:12 +00:00
UNINSTALL_EXTRA += ruby-uninstall
Implement Ruby bindings on top of SWIG/C++ bindings.
2016-01-11 13:31:47 +00:00
CLEAN_EXTRA += ruby-clean
endif
Add Java bindings.
2014-04-27 08:48:18 +00:00
if BINDINGS_JAVA
JDIR = bindings/java
Java: Move interfaces into org.sigrok.core.interfaces, tidy build system.
2014-07-26 18:11:11 +00:00
JPKG = org/sigrok/core
JCLS = $(JDIR)/$(JPKG)/classes
JINT = $(JDIR)/$(JPKG)/interfaces
Makefile.am: Fix out-of-tree build for C++ and Java bindings
2014-08-14 21:07:45 +00:00
JSRC = $(JCLS)/*.java $(srcdir)/$(JINT)/*.java
Java: Move interfaces into org.sigrok.core.interfaces, tidy build system.
2014-07-26 18:11:11 +00:00
JSWG = $(JCLS)/classes.i
bindings: Transfer C++ documentation strings to Python and Java wrappers.
2014-08-01 17:17:33 +00:00
JDOC = $(JCLS)/doc.i
Java: Move interfaces into org.sigrok.core.interfaces, tidy build system.
2014-07-26 18:11:11 +00:00
JCXX = $(JCLS)/classes_wrap.cxx
Add Java bindings.
2014-04-27 08:48:18 +00:00
JLIB = $(JDIR)/libsigrok_java_core_classes.so
JJAR = $(JDIR)/sigrok-core.jar
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
java_cleanfiles = $(JCXX) $(JCLS)/*.java $(JCLS)/*.class $(JINT)/*.class $(JJAR) $(JLIB)
Add Java bindings.
2014-04-27 08:48:18 +00:00
java-build: $(JJAR) $(JLIB)
Build: Avoid $< in explicit rules Although useful, makes other than GNU make do not like this. Name the prerequisite explicitly instead, and circumvent any VPATH substitution other makes may do.
2015-08-18 16:06:08 +00:00
$(JDOC): $(srcdir)/bindings/swig/doc.py $(CPPXMLDOC)
Makefile.am: Use $(MKDIR_P) everywhere.
2015-01-12 00:19:12 +00:00
$(AM_V_at)test -d $(JCLS) || $(MKDIR_P) $(JCLS)
Build: Avoid $< in explicit rules Although useful, makes other than GNU make do not like this. Name the prerequisite explicitly instead, and circumvent any VPATH substitution other makes may do.
2015-08-18 16:06:08 +00:00
$(AM_V_GEN)$(PYTHON) $(srcdir)/bindings/swig/doc.py java $(CPPXMLDOC) > $@
bindings: Transfer C++ documentation strings to Python and Java wrappers.
2014-08-01 17:17:33 +00:00
Makefile.am: Make SWIG wrappers also depend on new templates.i file.
2015-10-25 14:43:15 +00:00
$(JCXX): $(srcdir)/$(JSWG) $(JDOC) bindings/swig/classes.i \
bindings/swig/templates.i bindings/swig/enums.i \
Don't rebuild Python or Java bindings unless C++ interface changes.
2015-09-07 11:32:05 +00:00
$(bindings_cxx_libsigrokcxx_la_include_HEADERS) \
$(nodist_bindings_cxx_libsigrokcxx_la_include_HEADERS)
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
-$(AM_V_at)rm -f $(java_cleanfiles)
build: Use common SWIG defines for Python and Java Make the Python and Java bindings use the same set of preprocessor macros for the SWIG parsing stage, taken from a make variable. Add G_GNUC_{BEGIN,END}_IGNORE_DEPRECATIONS to that set.
2015-10-29 21:39:44 +00:00
$(AM_V_GEN)$(SWIG) -c++ $(swig_defs) \
SWIG: Hack around SWIG segfault on private destructors Apparently this problem has been fixed in SWIG 3.0.6. However, until we can require that version, define "private" as "protected" when running the SWIG parser.
2015-10-11 12:31:18 +00:00
-java -package org.sigrok.core.classes \
Build: Avoid $< in explicit rules Although useful, makes other than GNU make do not like this. Name the prerequisite explicitly instead, and circumvent any VPATH substitution other makes may do.
2015-08-18 16:06:08 +00:00
-Ibindings -I$(JCLS) $(local_includes) -I$(srcdir) $(JNI_CPPFLAGS) \
-outdir $(JCLS) -o $@ $(srcdir)/$(JSWG)
Add Java bindings.
2014-04-27 08:48:18 +00:00
Java: Move interfaces into org.sigrok.core.interfaces, tidy build system.
2014-07-26 18:11:11 +00:00
$(JJAR): $(JCXX)
Build: Cleanup up and modernize autotools setup Note that some --enable options of configure have become --with or --without options.
2015-08-15 13:59:52 +00:00
$(AM_V_GEN)$(JAVAC) -d $(JDIR) $(JSRC)
build: Silence language bindings building.
2014-07-19 18:21:19 +00:00
$(AM_V_at)jar cf $(JJAR) -C $(JDIR) $(JPKG)
Add Java bindings.
2014-04-27 08:48:18 +00:00
Don't rebuild Python or Java bindings unless C++ interface changes.
2015-09-07 11:32:05 +00:00
$(JLIB): $(JCXX) \
$(library_include_HEADERS) $(nodist_library_include_HEADERS) \
$(bindings_cxx_libsigrokcxx_la_include_HEADERS) \
$(nodist_bindings_cxx_libsigrokcxx_la_include_HEADERS) \
Makefile.am: Use @ORDER@ instead of |. This is set to | (or left empty) by SR_PROG_MAKE_ORDER_ONLY for portability reasons, since not all Make implementations support order-only prerequisites.
2015-09-09 06:21:01 +00:00
@ORDER@ bindings/cxx/libsigrokcxx.la
Bindings: Link C++ code with gnustl_shared if it exists (Android)
2016-12-06 20:01:23 +00:00
$(AM_V_GEN)$(CXXCOMPILE) $(JNI_CPPFLAGS) $(LDFLAGS) -L.libs -Lbindings/cxx/.libs \
Java: Move interfaces into org.sigrok.core.interfaces, tidy build system.
2014-07-26 18:11:11 +00:00
-fno-strict-aliasing -fPIC -shared $(JCLS)/classes_wrap.cxx \
Bindings: Link C++ code with gnustl_shared if it exists (Android)
2016-12-06 20:01:23 +00:00
-lsigrokcxx $(LIBSIGROKCXX_LIBS) $(SR_EXTRA_CXX_LIBS) -o $(JLIB)
Add Java bindings.
2014-04-27 08:48:18 +00:00
java-install:
Java: install files into DESTDIR This fixes bug #537.
2015-05-29 22:00:19 +00:00
$(INSTALL) -d $(DESTDIR)$(libdir)/jni
Makefile.am: Avoid non-portable -t option for $(INSTALL). The 'install' tool doesn't have a -t option on all platforms (e.g. Mac OS X or FreeBSD).
2015-10-29 22:59:12 +00:00
$(INSTALL) $(JLIB) $(DESTDIR)$(libdir)/jni
Java: install files into DESTDIR This fixes bug #537.
2015-05-29 22:00:19 +00:00
$(INSTALL) -d $(DESTDIR)$(datadir)/java
Makefile.am: Avoid non-portable -t option for $(INSTALL). The 'install' tool doesn't have a -t option on all platforms (e.g. Mac OS X or FreeBSD).
2015-10-29 22:59:12 +00:00
$(INSTALL) $(JJAR) $(DESTDIR)$(datadir)/java
Add Java bindings.
2014-04-27 08:48:18 +00:00
build: Enable uninstall of Java bindings This is needed for make distcheck to work.
2015-11-07 21:20:19 +00:00
java-uninstall:
-rm -f $(DESTDIR)$(datadir)/java/sigrok-core.jar
-rm -f $(DESTDIR)$(libdir)/jni/libsigrok_java_core_classes.so
Add Java bindings.
2014-04-27 08:48:18 +00:00
java-clean:
build: Delete generated files on make clean
2015-11-07 21:29:12 +00:00
-$(AM_V_at)rm -f $(java_cleanfiles) $(JDOC)
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
-$(AM_V_at)rm -fr $(JDIR)/doxy
java: Add documentation generation.
2014-08-01 18:29:48 +00:00
java-doc:
Build: Get rid of recursive make invocations Do not invoke $(MAKE) just to execute some clean rule. This gets rid of the "jobserver unavailable" warnings.
2015-08-18 19:24:39 +00:00
$(AM_V_at)cd $(srcdir)/$(JDIR) && BUILDDIR="$(abs_builddir)/$(JDIR)/" doxygen Doxyfile
Add Java bindings.
2014-04-27 08:48:18 +00:00
BUILD_EXTRA += java-build
INSTALL_EXTRA += java-install
build: Enable uninstall of Java bindings This is needed for make distcheck to work.
2015-11-07 21:20:19 +00:00
UNINSTALL_EXTRA += java-uninstall
Add Java bindings.
2014-04-27 08:48:18 +00:00
CLEAN_EXTRA += java-clean
endif
Reimplement high-level Python bindings on top of SWIG/C++ bindings.
2014-07-17 17:45:29 +00:00
all-local: $(BUILD_EXTRA)
install-exec-local: $(INSTALL_EXTRA)
uninstall: Remove empty libsigrok/libsigrokcxx include directories. Change uninstall-local to uninstall-hook, since the latter is guaranteed to run last (order is apparently not guaranteed for uninstall-local). This fixes bug #861.
2017-06-03 16:25:54 +00:00
uninstall-hook: $(UNINSTALL_EXTRA)
Assorted build system fixes for C++/Python/Java bindings.
2014-07-19 15:23:45 +00:00
clean-local: $(CLEAN_EXTRA)
Reimplement high-level Python bindings on top of SWIG/C++ bindings.
2014-07-17 17:45:29 +00:00
build: Make ChangeLog generation distcheck-safe
2015-11-07 20:51:40 +00:00
.PHONY: dist-changelog
dist-hook: dist-changelog
dist-changelog:
$(AM_V_at)if test ! -d '$(top_srcdir)/.git'; then \
cp -f '$(top_srcdir)/ChangeLog' "$(top_distdir)/ChangeLog"; \
elif git -C '$(top_srcdir)' log >.ChangeLog.tmp; then \
mv -f .ChangeLog.tmp "$(top_distdir)/ChangeLog"; \
else \
rm -f .ChangeLog.tmp; exit 1; \
fi
sr: Generate ChangeLog from git upon 'make dist'.
2012-03-28 00:32:52 +00:00