1berry/wireshark
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
wireshark/ConfigureChecks.cmake

269 lines
8.3 KiB
CMake
Raw Permalink Normal View History

Set svn:keywords and svn:eol-style. Fix the EOL style. Add the copyright comment and give it an SVN ID. svn path=/trunk/; revision=38381
2011-08-06 19:10:25 +00:00
# ConfigureChecks.cmake
#
# Wireshark - Network traffic analyzer
# By Gerald Combs <gerald@wireshark.org>
# Copyright 1998 Gerald Combs
#
More licenses converted to SPDX. Change-Id: Id4f987dcdacf06622d70263f4659a4400e30dc39 Reviewed-on: https://code.wireshark.org/review/26332 Petri-Dish: Dario Lombardo <lomato@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Anders Broman <a.broman58@gmail.com>
2018-03-06 15:31:02 +01:00
# SPDX-License-Identifier: GPL-2.0-or-later
Set svn:keywords and svn:eol-style. Fix the EOL style. Add the copyright comment and give it an SVN ID. svn path=/trunk/; revision=38381
2011-08-06 19:10:25 +00:00
#
Set some values for Windows manually where autodetection does not yet work. svn path=/trunk/; revision=52406
2013-10-06 16:11:50 +00:00
include(CMakePushCheckState)
Set svn:keywords and svn:eol-style. Fix the EOL style. Add the copyright comment and give it an SVN ID. svn path=/trunk/; revision=38381
2011-08-06 19:10:25 +00:00
#check system for includes
include(CheckIncludeFile)
ifaddrs.h may require sys/types.h to be included first. That's the case on DragonFly BSD 5.2.1, at least. Change-Id: I8bbd51462d74380004c611183f4b9229f4d20ff6 Reviewed-on: https://code.wireshark.org/review/27913 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-05-30 13:35:04 -07:00
include(CheckIncludeFiles)
check_include_file("arpa/inet.h" HAVE_ARPA_INET_H)
check_include_file("grp.h" HAVE_GRP_H)
#
# This may require <sys/types.h> to be included
#
check_include_files("sys/types.h;ifaddrs.h" HAVE_IFADDRS_H)
check_include_file("netinet/in.h" HAVE_NETINET_IN_H)
check_include_file("netdb.h" HAVE_NETDB_H)
check_include_file("pwd.h" HAVE_PWD_H)
check_include_file("sys/select.h" HAVE_SYS_SELECT_H)
check_include_file("sys/socket.h" HAVE_SYS_SOCKET_H)
check_include_file("sys/time.h" HAVE_SYS_TIME_H)
check_include_file("sys/utsname.h" HAVE_SYS_UTSNAME_H)
check_include_file("sys/wait.h" HAVE_SYS_WAIT_H)
check_include_file("unistd.h" HAVE_UNISTD_H)
Set svn:keywords and svn:eol-style. Fix the EOL style. Add the copyright comment and give it an SVN ID. svn path=/trunk/; revision=38381
2011-08-06 19:10:25 +00:00
CMake: Test that a trivial program compiles and runs Test that the simplest possible C program runs. If this fails, quit and tell the user. There's probably a permissions error or antivirus or other security software not allowing unsigned code to run that will cause problems with the build. Fix #20170
2025-04-21 21:19:06 -04:00
# Check that the C compiler works on a trivial program.
# Do this early on before more specific tests so we can give
# an appropriate error message.
if(NOT CMAKE_CROSSCOMPILING)
check_c_source_runs("
int main(void)
{
return 0;
}
"
COMPILER_RUNS
)
if(NOT COMPILER_RUNS)
if(WIN32)
message(FATAL_ERROR
"${CMAKE_C_COMPILER} failed to compile and run a trivial test program. \
This is likely a permissions error or the result of security software like \
Windows Defender preventing untrusted code from running."
)
else()
message(FATAL_ERROR
"${CMAKE_C_COMPILER} failed to compile and run a trivial test program. \
This is likely a permissions error or the result of security software \
preventing untrusted code from running."
)
endif()
endif()
endif()
Work around a Linux bonding driver bug (and the lack of a libpcap workaround). The bonding driver does not properly handle unknown ioctls; it returns ENODEV rather than ENOTSUP, EOPNOTSUPP, ENOTTY, or a "not supported" error of that type. This causes problems detailed in bug 11058. On Linux, check for bonding devices before checking for monitor-mode support. While we're at it, get rid of a commented-out include of CheckCSourceCompiles (it's presumably already been implicitly included by other functions that use it). Bug: 11058 Change-Id: I13035de0650634c51a52f262829b2b6fb86b39e9 Reviewed-on: https://code.wireshark.org/review/7856 Petri-Dish: Guy Harris <guy@alum.mit.edu> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Guy Harris <guy@alum.mit.edu>
2015-03-30 11:21:17 -07:00
#
# On Linux, check for some additional headers, which we need as a
# workaround for a bonding driver bug and for libpcap's current lack
# of its own workaround for that bug.
#
if(CMAKE_SYSTEM_NAME STREQUAL "Linux")
#
# Those header files require <sys/socket.h>.
#
check_c_source_compiles(
"#include <sys/socket.h>
#include <linux/sockios.h>
int main(void)
{
return 0;
}"
HAVE_LINUX_SOCKIOS_H
)
check_c_source_compiles(
"#include <sys/socket.h>
#include <linux/if_bonding.h>
int main(void)
{
return 0;
}"
HAVE_LINUX_IF_BONDING_H
)
endif()
Set svn:keywords and svn:eol-style. Fix the EOL style. Add the copyright comment and give it an SVN ID. svn path=/trunk/; revision=38381
2011-08-06 19:10:25 +00:00
#Functions
include(CheckFunctionExists)
When checking for floorl(), include math.h. It's not being found with MSVC, but is apparently present; perhaps math.h does something magic, e.g. defining it as a macro, so try using check_symbol_exists with "math.h". Change-Id: Idaecac641fb0b87d399807d8bf1fcb74e7251a4e Reviewed-on: https://code.wireshark.org/review/5911 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2014-12-20 00:23:05 -08:00
include(CheckSymbolExists)
Implement in cmake: http://anonsvn.wireshark.org/viewvc/viewvc.cgi?view=rev&revision=51322 > User: guy > Date: 2013/08/12 01:01 PM > > Log: > Just check whether dladdr() is available; we check in the code whether > it succeeds, so there's no point in checking whether it succeeds in a > sample program. svn path=/trunk/; revision=51334
2013-08-13 09:58:09 +00:00
On Solaris, check for getexecname in CMake. We expect it to be checked for in wsutil/filesystem.c, so we should check for it. It's a Solarisism, so check for it only on Solaris. Change-Id: I09104c17d2ec91c74862b63e735c32a9d188f2a6 Reviewed-on: https://code.wireshark.org/review/27351 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-05-04 23:00:29 -07:00
#
# Platform-specific functions used in platform-specific code.
# We check for them only on the platform on which we use them.
#
Use dlget() and dlgetname() to get the executable path name on HP-UX. That leaves only AIX (and, if we're looking at dead UN*Xes, IRIX and Tru64 UNIX) as platforms on which we can't fetch that. Change-Id: If7a6a425aba30e1abf82ecc66f6c28dc532a227c Reviewed-on: https://code.wireshark.org/review/27358 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-05-04 23:39:46 -07:00
if(CMAKE_SYSTEM_NAME STREQUAL "HP-UX")
#
# HP-UX
#
cmake_push_check_state()
set(CMAKE_REQUIRED_LIBRARIES ${CMAKE_DL_LIBS})
check_function_exists("dlget" HAVE_DLGET)
cmake_pop_check_state()
elseif(CMAKE_SYSTEM_NAME STREQUAL "SunOS" AND CMAKE_SYSTEM_VERSION MATCHES "5[.][0-9.]*")
On Solaris, check for getexecname in CMake. We expect it to be checked for in wsutil/filesystem.c, so we should check for it. It's a Solarisism, so check for it only on Solaris. Change-Id: I09104c17d2ec91c74862b63e735c32a9d188f2a6 Reviewed-on: https://code.wireshark.org/review/27351 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-05-04 23:00:29 -07:00
#
# Solaris
#
check_function_exists("getexecname" HAVE_GETEXECNAME)
endif()
Implement in cmake: http://anonsvn.wireshark.org/viewvc/viewvc.cgi?view=rev&revision=51322 > User: guy > Date: 2013/08/12 01:01 PM > > Log: > Just check whether dladdr() is available; we check in the code whether > it succeeds, so there's no point in checking whether it succeeds in a > sample program. svn path=/trunk/; revision=51334
2013-08-13 09:58:09 +00:00
CMake: Fix timespec_get() detection on Windows Replace some instances of check_function_exists() with check_symbol_exists(). From the CMake documentation: - check_function_exists() can't detect functions that are inlined in headers or specified as a macro. - check_function_exists() can't detect anything in the 32-bit versions of the Win32 API, because of a mismatch in calling conventions. - check_function_exists() only verifies linking, it does not verify that the function is declared in system headers. This fixes timespec_get() detection on Windows.
2021-12-27 10:47:16 +00:00
check_symbol_exists("clock_gettime" "time.h" HAVE_CLOCK_GETTIME)
wsutil: Add portability wrapper for clock_gettime(CLOCK_REALTIME)
2021-12-15 13:06:57 +00:00
# Some platforms (macOS pre 10.15) are non-conformant with C11 and lack timespec_get()
CMake: Fix timespec_get() detection on Windows Replace some instances of check_function_exists() with check_symbol_exists(). From the CMake documentation: - check_function_exists() can't detect functions that are inlined in headers or specified as a macro. - check_function_exists() can't detect anything in the 32-bit versions of the Win32 API, because of a mismatch in calling conventions. - check_function_exists() only verifies linking, it does not verify that the function is declared in system headers. This fixes timespec_get() detection on Windows.
2021-12-27 10:47:16 +00:00
check_symbol_exists("timespec_get" "time.h" HAVE_TIMESPEC_GET)
Add portability wrapper for localtime_r()
2023-09-03 01:35:31 +01:00
if(NOT MSVC)
check_symbol_exists("localtime_r" "time.h" HAVE_LOCALTIME_R)
Add a routine to format an nstime as an ISO 8601-style date/time. This replaces code in both absolute_time_string(), in capinfos, and set_abs_ymd_time(), in libwireshark. Like absolute_time_string(), it reports, in the formatted string, an error if the seconds value in the nstime can't be converted to a struct tm (looking at *you*, Visual Studio C library), Have that routine handle all time precision values from 0 (seconds) to 9 (nanoseconds). Add a ws_gmtime_r(), matching ws_localtime_r(), for use by that routine.
2023-09-11 02:21:59 -07:00
check_symbol_exists("gmtime_r" "time.h" HAVE_GMTIME_R)
CMake: Add missing configure check for timegm() We have code that conditionally uses timegm() in time_util.c, but no configure check for it. Fix that.
2023-09-04 19:59:52 +01:00
check_symbol_exists("timegm" "time.h" HAVE_TIMEGM)
Add portability wrapper for localtime_r()
2023-09-03 01:35:31 +01:00
check_symbol_exists("tzset" "time.h" HAVE_TZSET)
Build with NetBSD's strptime() Replace our strptime code, which is from gnulib, with the simpler and better NetBSD implementation. This changes the ws_strptime() stub to unconditionally use the internal implementation. Previously it would use the system implementation of available. This is still possible but is opt-in, i.e., code should add the necessary #ifdefs and assume responsability for handling non-portable formats or providing limited functionality on some platforms. Text import allows the user to specify the strptime() format freely, so in that case it makes sense to use the system's implementation, and pass the responsability for understanding the implementation and the supported specifiers to the user. Only fall back to our implementation if the system libc lacks a strptime().
2023-09-02 22:57:53 +01:00
check_symbol_exists("tzname" "time.h" HAVE_TZNAME)
dftest: Read (list of) filters from a file (inc. stdin) dftest is very fast about compiling and testing filters; almost all of its runtime is the startup procedure. Sometimes you might have a large number of filters all of which need to be checked for validity. The existing dftest, like some of the other command line programs, after processing options concatenates all the positional arguments into one big string, with a single space between them. (Essentially undoing word splitting.) That's convenient for not having to quote filters, but makes it somewhat annoying to test multiple filters. (Of course, multiple filters can be connected with logical operators.) Add the ability to read a list of filters from a file line-by-line, similar to grep and other programs. Allow specifying stdin with '-'. (printf or other methods can be used to add newlines.) Perhaps an ability to have a different delimiter could also be implemented. For now, quit after the first failure. An option to keep going might also have use cases.
2024-10-16 20:29:04 -04:00
check_symbol_exists("getline" "stdio.h" HAVE_GETLINE)
Add portability wrapper for localtime_r()
2023-09-03 01:35:31 +01:00
endif()
extcap: add local_interfaces_to_list() to wsutil. This new function abstracts the creation of a list of the local interfaces that will be used by future extcaps to generate specific filters. Sshdump now uses it to create a pcap filter. Change-Id: I5b75a266f81104b3c9bcb3e51de246b7cc8785ce Reviewed-on: https://code.wireshark.org/review/14092 Reviewed-by: João Valverde <j@v6e.pt> Petri-Dish: João Valverde <j@v6e.pt> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Roland Knall <rknall@gmail.com>
2016-02-23 14:33:54 +01:00
check_function_exists("getifaddrs" HAVE_GETIFADDRS)
Set svn:keywords and svn:eol-style. Fix the EOL style. Add the copyright comment and give it an SVN ID. svn path=/trunk/; revision=38381
2011-08-06 19:10:25 +00:00
check_function_exists("issetugid" HAVE_ISSETUGID)
Move the QT_MACEXTRAS_LIB cmakedefine in cmakeconfig.h.in to the equivalent location in config.h.in, to simplify comparing the two files. Add some tests from the autoconf script to CMake: check for setresgid() and setresuid(); check for struct stat having an st_flags member. Get rid of INTTYPES_H_DEFINES_FORMATS; we don't appear to check for it, and we don't use it (we're using GLib's formatting functions, which have their own way of handling 64-bit integers). Get rid of STDC_HEADERS; it's in autoconf because autoconf was originally developed back when you couldn't rely in ANSI C and has never been removed, and something we're using for other purposes checks for it, not because we explicitly test for it, but we don't test it in the code, and CMake doesn't even check for it. svn path=/trunk/; revision=52522
2013-10-10 23:54:53 +00:00
check_function_exists("setresgid" HAVE_SETRESGID)
check_function_exists("setresuid" HAVE_SETRESUID)
Allow building under OSX Yosemite using cmake. CFPropertyListCreateFromStream() has been deprecated in Yosemite, in favor of CFPropertyListCreateWithStream(). The autotools build checks for the existence of the new function, but the cmake build does not. So, add a check for it to CMakeLists.txt, and update cmakeconfig.h.in to provide the "template" for HAVE_CFPROPERTYLISTCREATEWITHSTREAM. Change-Id: I1a4e3e1ce83092ecf7725527b921eeca80b18d4b Note: CFPropertyListCreateWithStream() works under Mavericks as well. Reviewed-on: https://code.wireshark.org/review/5226 Reviewed-by: Stig Bjørlykke <stig@bjorlykke.org> Tested-by: Stig Bjørlykke <stig@bjorlykke.org>
2014-11-10 14:24:18 -06:00
if (APPLE)
cmake_push_check_state()
set(CMAKE_REQUIRED_LIBRARIES ${APPLE_CORE_FOUNDATION_LIBRARY})
check_function_exists("CFPropertyListCreateWithStream" HAVE_CFPROPERTYLISTCREATEWITHSTREAM)
cmake_pop_check_state()
endif()
Move two functions from epan to wsutil/str_util Move epan_memmem() and epan_strcasestr() to wsutil/str_util. Rename to ws_memmem() and ws_strcasestr(). Add compile time check for a system implementation and use that if available. We invoke those functions using a wrapper to avoid exposing _GNU_SOURCE outside of the implementation.
2021-11-27 17:57:46 +00:00
if(UNIX)
cmake_push_check_state()
list(APPEND CMAKE_REQUIRED_DEFINITIONS -D_GNU_SOURCE)
CMake: Fix timespec_get() detection on Windows Replace some instances of check_function_exists() with check_symbol_exists(). From the CMake documentation: - check_function_exists() can't detect functions that are inlined in headers or specified as a macro. - check_function_exists() can't detect anything in the 32-bit versions of the Win32 API, because of a mismatch in calling conventions. - check_function_exists() only verifies linking, it does not verify that the function is declared in system headers. This fixes timespec_get() detection on Windows.
2021-12-27 10:47:16 +00:00
check_symbol_exists("memmem" "string.h" HAVE_MEMMEM)
wsutil: Add a memrchr implementation Use the system memrchr if found, otherwise use a very basic non optimized implementation
2023-10-11 08:43:35 -04:00
check_symbol_exists("memrchr" "string.h" HAVE_MEMRCHR)
wsutil/inet_addr: Refactor to use C99/POSIX types Rewrite ws_inet_pton{4,6} and ws_inet_ntop{4,6} without GLib types. Check for strerrorname_np() and use that is available, to simplify error handling. Add some minimal tests.
2021-12-27 13:28:29 +00:00
check_symbol_exists("strerrorname_np" "string.h" HAVE_STRERRORNAME_NP)
Use a wrapper function to call strptime() Encapsulate the feature requirements for strptime() in a portability wrapper. Use _GNU_SOURCE to expose strptime. It should be enough on glibc without the side-effect of selecting a particular SUS version, which we don't need and might hide other definitions.
2021-12-27 02:03:26 +00:00
check_symbol_exists("strptime" "time.h" HAVE_STRPTIME)
CMake: Fix timespec_get() detection on Windows Replace some instances of check_function_exists() with check_symbol_exists(). From the CMake documentation: - check_function_exists() can't detect functions that are inlined in headers or specified as a macro. - check_function_exists() can't detect anything in the 32-bit versions of the Win32 API, because of a mismatch in calling conventions. - check_function_exists() only verifies linking, it does not verify that the function is declared in system headers. This fixes timespec_get() detection on Windows.
2021-12-27 10:47:16 +00:00
check_symbol_exists("vasprintf" "stdio.h" HAVE_VASPRINTF)
Move two functions from epan to wsutil/str_util Move epan_memmem() and epan_strcasestr() to wsutil/str_util. Rename to ws_memmem() and ws_strcasestr(). Add compile time check for a system implementation and use that if available. We invoke those functions using a wrapper to avoid exposing _GNU_SOURCE outside of the implementation.
2021-11-27 17:57:46 +00:00
cmake_pop_check_state()
endif()
Get netlink detection working with cmake (not complete) svn path=/trunk/; revision=44791
2012-09-06 10:37:17 +00:00
Add HAVE_TM_ZONE and HAVE_TZNAME detection to fix timezone display in cmake builds. svn path=/trunk/; revision=48932
2013-04-19 16:38:28 +00:00
#Struct members
include(CheckStructHasMember)
Cmake: Make it possible to check on HAVE_STRUCT_STAT_ST_BLKSIZE. Change-Id: Id84adc85c1fbdef8e39240f55128cdec4ee9ca2f Reviewed-on: https://code.wireshark.org/review/31324 Petri-Dish: Anders Broman <a.broman58@gmail.com> Tested-by: Petri Dish Buildbot Reviewed-by: Martin Kaiser <wireshark@kaiser.cx> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2019-01-03 14:56:31 +01:00
check_struct_has_member("struct stat" st_blksize sys/stat.h HAVE_STRUCT_STAT_ST_BLKSIZE)
Fix the #defines for the presence of structure names. AC_CHECK_MEMBER() and AC_CHECK_MEMBERS() use a standard name for the {structurename} being the name of the structure type, complete with "struct" if a typedef wasn't used, and with all letters mapped to upper case, and with {membername} being the name of the structure member, with all letters mapped to upper case. check_struct_has_member() lets you choose the name; choose the same name that the autoconf macros use, and fix the code to check for them. Change-Id: Ifb3cf65e7e94907ad0a2f8aacca0c21a531f0c5b Reviewed-on: https://code.wireshark.org/review/18382 Petri-Dish: Guy Harris <guy@alum.mit.edu> Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-10-22 01:56:48 -07:00
check_struct_has_member("struct stat" st_birthtime sys/stat.h HAVE_STRUCT_STAT_ST_BIRTHTIME)
check_struct_has_member("struct stat" __st_birthtime sys/stat.h HAVE_STRUCT_STAT___ST_BIRTHTIME)
On UN*X, st_ctime is the last status change time, not the creation time. That's the time the file's inode last changed, so size changes, permission changes, etc. affect it. It's *not* the time the file was created; most UN*Xes don't provide that. Newer versions of FreeBSD, NetBSD, OpenBSD, and macOS do, but other UN*Xes don't appear to. On Windows, at least according to Microsoft's documentation, st_ctime *is* the creation time. Hopefully that's not the result of confusion on the part of somebody at Microsoft. Change-Id: I20743703f6ef66e40dff9004dc91bed46af6fad0 Reviewed-on: https://code.wireshark.org/review/18378 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2016-10-21 22:19:46 -07:00
check_struct_has_member("struct tm" tm_zone time.h HAVE_STRUCT_TM_TM_ZONE)
Build with NetBSD's strptime() Replace our strptime code, which is from gnulib, with the simpler and better NetBSD implementation. This changes the ws_strptime() stub to unconditionally use the internal implementation. Previously it would use the system implementation of available. This is still possible but is opt-in, i.e., code should add the necessary #ifdefs and assume responsability for handling non-portable formats or providing limited functionality on some platforms. Text import allows the user to specify the strptime() format freely, so in that case it makes sense to use the system's implementation, and pass the responsability for understanding the implementation and the supported specifiers to the user. Only fall back to our implementation if the system libc lacks a strptime().
2023-09-02 22:57:53 +01:00
check_struct_has_member("struct tm" tm_gmtoff time.h HAVE_STRUCT_TM_TM_GMTOFF)
Make the check for NL80211_CMD_SET_CHANNEL work by copying the logic of the autofoo test. Using Cmake's check_symbol_exists is not an option as it will not work for enums and types. As this is documented in the manpage this is to be considered a feature and not a bug. svn path=/trunk/; revision=44851
2012-09-10 18:35:56 +00:00
regex: Prefer C99/POSIX types Replace 'gssize' with 'ssize_t'. Add a CMake configure check for ssize_t. Fix missing "config.h" includes.
2021-11-12 17:23:04 +00:00
# Types
include(CheckTypeSize)
check_type_size("ssize_t" SSIZE_T)
CMake: Unconditionally use 64-bit time_t on 32-bit glibc platforms The GNU C library has support for 64-bit time_t on all architectures, but default to 32-bit time_t on some 32-bit architectures (x86 and ARM) unless the _TIME_BITS feature test macro is defined: https://www.gnu.org/software/libc/manual/html_node/Feature-Test-Macros.html Debian already does this when building packages. Most other platforms that Wireshark can build on have 64-bit time_t already on all architectures; one exception is FreeBSD on i386 (but not 32-bit ARM): https://man.freebsd.org/cgi/man.cgi?arch FreeBSD is dropping support for i386 for Release 15; Release 14 has an EOL of Nov 2028. So at some point we could consider requiring 64-bit time_t. Ping #11744
2025-05-13 10:59:13 -04:00
# Unconditionally force use of 64-bit time_t on 32-bit platforms with glibc.
check_symbol_exists(__GLIBC__ "time.h" HAVE___GLIBC__)
if(HAVE___GLIBC__ AND CMAKE_SIZEOF_VOID_P EQUAL 4)
# _TIME_BITS=64 requires _FILE_OFFSET_BITS=64, which we should have
# already ensured is set by FindLFS, but this doesn't hurt
check_symbol_exists(_FILE_OFFSET_BITS "time.h" HAVE__FILE_OFFSET_BITS)
if (NOT HAVE__FILE_OFFSET_BITS)
list(APPEND CMAKE_REQUIRED_DEFINITIONS -D_FILE_OFFSET_BITS=64)
endif()
check_symbol_exists(_TIME_BITS "time.h" HAVE__TIME_BITS)
if (NOT HAVE__TIME_BITS)
list(APPEND CMAKE_REQUIRED_DEFINITIONS -D_TIME_BITS=64)
endif()
endif()
# Do we want to eventually just fail for non Y2038 compliant hosts (or all
# less than 64-bit time_t hosts, if we don't want to consider theoretical
# cases like unsigned 32 bit time_t?)
# The only way to truly test this is to compile a program, especially
# due to multiple architectures.
#check_type_size("time_t" TIME_T_SIZE)
#if(TIME_T_SIZE_64 EQUAL 0)
# message("multiple architectures with possibly different time_t sizes.")
#elseif(TIME_T_SIZE_64 LESS 8)
# message(FATAL_ERROR "time_t is less than 64-bit")
#endif()
CMake: Check if vsnprinf() is C99-compliant
2021-09-21 18:54:55 +01:00
#
# Check if the libc vsnprintf() conforms to C99. If this fails we may
# need to fall-back on GLib I/O.
#
CMake: Fix a try_run() test when cross-compiling
2023-05-26 13:29:23 +01:00
# If cross-compiling we can't check so just assume this requirement is met.
#
if(NOT CMAKE_CROSSCOMPILING)
check_c_source_runs("
#include <stdio.h>
CMake: Hardening compiler options Hardened compiler option should give us a more secure application against potential attacks and/or misbehavior This is according to Best Practice from Open SSF https://best.openssf.org/Compiler-Hardening-Guides/Compiler-Options-Hardening-Guide-for-C-and-C++
2024-08-15 13:59:24 +02:00
#pragma GCC diagnostic push
#pragma GCC diagnostic ignored \"-Wall\"
CMake: Fix a try_run() test when cross-compiling
2023-05-26 13:29:23 +01:00
int main(void)
{
/* Check that snprintf() and vsnprintf() don't return
* -1 if the buffer is too small. C99 says this value
* is the length that would be written not including
* the nul byte. */
char buf[3];
return snprintf(buf, sizeof(buf), \"%s\", \"ABCDEF\") > 0 ? 0 : 1;
CMake: Hardening compiler options Hardened compiler option should give us a more secure application against potential attacks and/or misbehavior This is according to Best Practice from Open SSF https://best.openssf.org/Compiler-Hardening-Guides/Compiler-Options-Hardening-Guide-for-C-and-C++
2024-08-15 13:59:24 +02:00
}
#pragma GCC diagnostic pop"
CMake: Fix a try_run() test when cross-compiling
2023-05-26 13:29:23 +01:00
HAVE_C99_VSNPRINTF
)
if (NOT HAVE_C99_VSNPRINTF)
message(FATAL_ERROR
CMake: Check if vsnprinf() is C99-compliant
2021-09-21 18:54:55 +01:00
"Building Wireshark requires a C99 compliant vsnprintf() and this \
target does not meet that requirement. Compiling for ${CMAKE_SYSTEM} \
using ${CMAKE_C_COMPILER_ID}. Please report this issue to the Wireshark \
developers at wireshark-dev@wireshark.org."
CMake: Fix a try_run() test when cross-compiling
2023-05-26 13:29:23 +01:00
)
endif()
CMake: Check if vsnprinf() is C99-compliant
2021-09-21 18:54:55 +01:00
endif()
Don't check for nl80211 stuff if we don't have libnl. While we're at it, do the checks in the same order that we do in autotools. Change-Id: Ie253cdeb740a5be4a5817ce4f7ed683aeea868c8 Reviewed-on: https://code.wireshark.org/review/6316 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2015-01-04 16:31:49 -08:00
#
# *If* we found libnl, check if we can use nl80211 stuff with it.
#
if (NL_FOUND)
check_c_source_compiles(
"#include <linux/nl80211.h>
cmake: remove warnings -Werror=old-style-definition see https://cmake.org/Bug/bug_relationship_graph.php?bug_id=15058 Change-Id: I325f476b145a542e987a13bedd1f95a7d8faba94 Reviewed-on: https://code.wireshark.org/review/27121 Petri-Dish: Guy Harris <guy@alum.mit.edu> Tested-by: Petri Dish Buildbot Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-04-24 20:20:33 +02:00
int main(void) {
Don't check for nl80211 stuff if we don't have libnl. While we're at it, do the checks in the same order that we do in autotools. Change-Id: Ie253cdeb740a5be4a5817ce4f7ed683aeea868c8 Reviewed-on: https://code.wireshark.org/review/6316 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2015-01-04 16:31:49 -08:00
int x = NL80211_FREQUENCY_ATTR_MAX_TX_POWER;
Trivial Dead assignment fixes in configure Fixes a Dead assignment and unused variable warning. With this patch and http://www.cmake.org/Bug/view.php?id=15203, `scan-build cmake` will give zero warnings! Change-Id: Idc7a03fbf1e8196bc139a2c6663ec48b897f2897 Reviewed-on: https://code.wireshark.org/review/7144 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com>
2015-02-16 00:14:18 +01:00
x |= NL80211_ATTR_SUPPORTED_IFTYPES;
x |= NL80211_ATTR_SUPPORTED_COMMANDS;
x |= NL80211_ATTR_WIPHY_FREQ;
x |= NL80211_CHAN_NO_HT;
(void)x;
Don't check for nl80211 stuff if we don't have libnl. While we're at it, do the checks in the same order that we do in autotools. Change-Id: Ie253cdeb740a5be4a5817ce4f7ed683aeea868c8 Reviewed-on: https://code.wireshark.org/review/6316 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2015-01-04 16:31:49 -08:00
}"
HAVE_NL80211
)
check_c_source_compiles(
"#include <linux/nl80211.h>
cmake: remove warnings -Werror=old-style-definition see https://cmake.org/Bug/bug_relationship_graph.php?bug_id=15058 Change-Id: I325f476b145a542e987a13bedd1f95a7d8faba94 Reviewed-on: https://code.wireshark.org/review/27121 Petri-Dish: Guy Harris <guy@alum.mit.edu> Tested-by: Petri Dish Buildbot Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-04-24 20:20:33 +02:00
int main(void) {
Don't check for nl80211 stuff if we don't have libnl. While we're at it, do the checks in the same order that we do in autotools. Change-Id: Ie253cdeb740a5be4a5817ce4f7ed683aeea868c8 Reviewed-on: https://code.wireshark.org/review/6316 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2015-01-04 16:31:49 -08:00
enum nl80211_commands x = NL80211_CMD_SET_CHANNEL;
}"
HAVE_NL80211_CMD_SET_CHANNEL
)
check_c_source_compiles(
"#include <linux/nl80211.h>
cmake: remove warnings -Werror=old-style-definition see https://cmake.org/Bug/bug_relationship_graph.php?bug_id=15058 Change-Id: I325f476b145a542e987a13bedd1f95a7d8faba94 Reviewed-on: https://code.wireshark.org/review/27121 Petri-Dish: Guy Harris <guy@alum.mit.edu> Tested-by: Petri Dish Buildbot Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-04-24 20:20:33 +02:00
int main(void) {
Don't check for nl80211 stuff if we don't have libnl. While we're at it, do the checks in the same order that we do in autotools. Change-Id: Ie253cdeb740a5be4a5817ce4f7ed683aeea868c8 Reviewed-on: https://code.wireshark.org/review/6316 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2015-01-04 16:31:49 -08:00
enum nl80211_protocol_features x = NL80211_PROTOCOL_FEATURE_SPLIT_WIPHY_DUMP;
}"
HAVE_NL80211_SPLIT_WIPHY_DUMP
)
dumpcap: Add support for 802.11ac monitor modes Add dumpcap support for configuring 80MHz, 80+80MHz, 160MHz monitor modes via nl80211. Change-Id: I2ae8955670c2a9b5051e2223d45ce522459f2c5f Reviewed-on: https://code.wireshark.org/review/13964 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-02-07 00:40:51 +01:00
check_c_source_compiles(
"#include <linux/nl80211.h>
cmake: remove warnings -Werror=old-style-definition see https://cmake.org/Bug/bug_relationship_graph.php?bug_id=15058 Change-Id: I325f476b145a542e987a13bedd1f95a7d8faba94 Reviewed-on: https://code.wireshark.org/review/27121 Petri-Dish: Guy Harris <guy@alum.mit.edu> Tested-by: Petri Dish Buildbot Reviewed-by: Guy Harris <guy@alum.mit.edu>
2018-04-24 20:20:33 +02:00
int main(void) {
dumpcap: Add support for 802.11ac monitor modes Add dumpcap support for configuring 80MHz, 80+80MHz, 160MHz monitor modes via nl80211. Change-Id: I2ae8955670c2a9b5051e2223d45ce522459f2c5f Reviewed-on: https://code.wireshark.org/review/13964 Petri-Dish: Alexis La Goutte <alexis.lagoutte@gmail.com> Petri-Dish: Gerald Combs <gerald@wireshark.org> Tested-by: Petri Dish Buildbot <buildbot-no-reply@wireshark.org> Reviewed-by: Michael Mann <mmann78@netscape.net>
2016-02-07 00:40:51 +01:00
enum nl80211_attrs x = NL80211_ATTR_VHT_CAPABILITY;
}"
HAVE_NL80211_VHT_CAPABILITY
)
Don't check for nl80211 stuff if we don't have libnl. While we're at it, do the checks in the same order that we do in autotools. Change-Id: Ie253cdeb740a5be4a5817ce4f7ed683aeea868c8 Reviewed-on: https://code.wireshark.org/review/6316 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2015-01-04 16:31:49 -08:00
endif()
In CMake, check for thousands-grouping support in the GLib printf routines. We do that with the autotools; do it with CMake as well. We could, in theory, handle thosands-grouping ourselves, on all platforms supporting ANSI C (for which read "all platforms we care about") by using localeconv()'s thousands_sep and grouping items, but that's a bit more work. Fix autotools' comment for that item while we're at it (it checks the GLib printf routines, not the system printf routines). Change-Id: I000f0f3b955d9b192ade15e3fabc46d6b48a052e Reviewed-on: https://code.wireshark.org/review/6317 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2015-01-04 17:43:19 -08:00
ConfigureChecks (cmake): Fix indent (use tabs) and modelines Change-Id: I905c10583baf71b105c65c25f43fb09dd2279b8e Reviewed-on: https://code.wireshark.org/review/12609 Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-12-14 11:53:27 +01:00
#
HTTPS (almost) everywhere. Change all wireshark.org URLs to use https. Fix some broken links while we're at it. Change-Id: I161bf8eeca43b8027605acea666032da86f5ea1c Reviewed-on: https://code.wireshark.org/review/34089 Reviewed-by: Guy Harris <guy@alum.mit.edu>
2019-07-26 11:43:17 -07:00
# Editor modelines - https://www.wireshark.org/tools/modelines.html
ConfigureChecks (cmake): Fix indent (use tabs) and modelines Change-Id: I905c10583baf71b105c65c25f43fb09dd2279b8e Reviewed-on: https://code.wireshark.org/review/12609 Reviewed-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Tested-by: Alexis La Goutte <alexis.lagoutte@gmail.com> Reviewed-by: Anders Broman <a.broman58@gmail.com>
2015-12-14 11:53:27 +01:00
#
# Local variables:
# c-basic-offset: 8
# tab-width: 8
# indent-tabs-mode: t
# End:
#
# vi: set shiftwidth=8 tabstop=8 noexpandtab:
# :indentSize=8:tabSize=8:noTabs=false:
#
Reference in New Issue Copy Permalink