haproxy/Makefile

1011 lines
38 KiB
Makefile
Raw Normal View History

# This GNU Makefile supports different OS and CPU combinations.
#
# You should use it this way :
# [g]make TARGET=os ARCH=arch CPU=cpu USE_xxx=1 ...
#
# By default the detailed commands are hidden for a cleaner output, but you may
# see them by appending "V=1" to the make command.
#
# Valid USE_* options are enumerated in the "use_opts" variable and are listed
# below. Most of them are automatically set by the TARGET, others have to be
# explicitly specified :
# USE_EPOLL : enable epoll() on Linux 2.6. Automatic.
# USE_KQUEUE : enable kqueue() on BSD. Automatic.
# USE_MY_EPOLL : redefine epoll_* syscalls. Automatic.
# USE_MY_SPLICE : redefine the splice syscall if build fails without.
# USE_NETFILTER : enable netfilter on Linux. Automatic.
# USE_PCRE : enable use of libpcre for regex. Recommended.
# USE_PCRE_JIT : enable JIT for faster regex on libpcre >= 8.32
# USE_PCRE2 : enable use of libpcre2 for regex.
# USE_PCRE2_JIT : enable JIT for faster regex on libpcre2
# USE_POLL : enable poll(). Automatic.
# USE_PRIVATE_CACHE : disable shared memory cache of ssl sessions.
# USE_THREAD : enable threads support.
# USE_PTHREAD_PSHARED : enable pthread process shared mutex on sslcache.
# USE_REGPARM : enable regparm optimization. Recommended on x86.
# USE_STATIC_PCRE : enable static libpcre. Recommended.
# USE_STATIC_PCRE2 : enable static libpcre2.
# USE_TPROXY : enable transparent proxy. Automatic.
# USE_LINUX_TPROXY : enable full transparent proxy. Automatic.
# USE_LINUX_SPLICE : enable kernel 2.6 splicing. Automatic.
# USE_LIBCRYPT : enable crypted passwords using -lcrypt
# USE_CRYPT_H : set it if your system requires including crypt.h
# USE_VSYSCALL : enable vsyscall on Linux x86, bypassing libc
# USE_GETADDRINFO : use getaddrinfo() to resolve IPv6 host names.
# USE_OPENSSL : enable use of OpenSSL. Recommended, but see below.
# USE_LUA : enable Lua support.
# USE_FUTEX : enable use of futex on kernel 2.6. Automatic.
# USE_ACCEPT4 : enable use of accept4() on linux. Automatic.
# USE_MY_ACCEPT4 : use own implemention of accept4() if glibc < 2.10.
# USE_PRCTL : enable use of prctl(). Automatic.
MEDIUM: HTTP compression (zlib library support) This commit introduces HTTP compression using the zlib library. http_response_forward_body has been modified to call the compression functions. This feature includes 3 algorithms: identity, gzip and deflate: * identity: this is mostly for debugging, and it was useful for developping the compression feature. With Content-Length in input, it is making each chunk with the data available in the current buffer. With chunks in input, it is rechunking, the output chunks will be bigger or smaller depending of the size of the input chunk and the size of the buffer. Identity does not apply any change on data. * gzip: same as identity, but applying a gzip compression. The data are deflated using the Z_NO_FLUSH flag in zlib. When there is no more data in the input buffer, it flushes the data in the output buffer (Z_SYNC_FLUSH). At the end of data, when it receives the last chunk in input, or when there is no more data to read, it writes the end of data with Z_FINISH and the ending chunk. * deflate: same as gzip, but with deflate algorithm and zlib format. Note that this algorithm has ambiguous support on many browsers and no support at all from recent ones. It is strongly recommended not to use it for anything else than experimentation. You can't choose the compression ratio at the moment, it will be set to Z_BEST_SPEED (1), as tests have shown very little benefit in terms of compression ration when going above for HTML contents, at the cost of a massive CPU impact. Compression will be activated depending of the Accept-Encoding request header. With identity, it does not take care of that header. To build HAProxy with zlib support, use USE_ZLIB=1 in the make parameters. This work was initially started by David Du Colombier at Exceliance.
2012-10-23 08:25:10 +00:00
# USE_ZLIB : enable zlib library support.
MAJOR: compression: integrate support for libslz This library is designed to emit a zlib-compatible stream with no memory usage and to favor resource savings over compression ratio. While zlib requires 256 kB of RAM per compression context (and can only support 4000 connections per GB of RAM), the stateless compression offered by libslz does not need to retain buffers between subsequent calls. In theory this slightly reduces the compression ratio but in practice it does not have that much of an effect since the zlib window is limited to 32kB. Libslz is available at : http://git.1wt.eu/web?p=libslz.git It was designed for web compression and provides a lot of savings over zlib in haproxy. Here are the preliminary results on a single core of a core2-quad 3.0 GHz in 32-bit for only 300 concurrent sessions visiting the home page of www.haproxy.org (76 kB) with the default 16kB buffers : BW In BW Out BW Saved Ratio memory VSZ/RSS zlib 237 Mbps 92 Mbps 145 Mbps 2.58 84M / 69M slz 733 Mbps 380 Mbps 353 Mbps 1.93 5.9M / 4.2M So while the compression ratio is lower, the bandwidth savings are much more important due to the significantly lower compression cost which allows to consume even more data from the servers. In the example above, zlib became the bottleneck at 24% of the output bandwidth. Also the difference in memory usage is obvious. More tests run on a single core of a core i5-3320M, with 500 concurrent users and the default 16kB buffers : At 100% CPU (no limit) : BW In BW Out BW Saved Ratio memory VSZ/RSS hits/s zlib 480 Mbps 188 Mbps 292 Mbps 2.55 130M / 101M 744 slz 1700 Mbps 810 Mbps 890 Mbps 2.10 23.7M / 9.7M 2382 At 85% CPU (limited) : BW In BW Out BW Saved Ratio memory VSZ/RSS hits/s zlib 1240 Mbps 976 Mbps 264 Mbps 1.27 130M / 100M 1738 slz 1600 Mbps 976 Mbps 624 Mbps 1.64 23.7M / 9.7M 2210 The most important benefit really happens when the CPU usage is limited by "maxcompcpuusage" or the BW limited by "maxcomprate" : in order to preserve resources, haproxy throttles the compression ratio until usage is within limits. Since slz is much cheaper, the average compression ratio is much higher and the input bandwidth is quite higher for one Gbps output. Other tests made with some reference files : BW In BW Out BW Saved Ratio hits/s daniels.html zlib 1320 Mbps 163 Mbps 1157 Mbps 8.10 1925 slz 3600 Mbps 580 Mbps 3020 Mbps 6.20 5300 tv.com/listing zlib 980 Mbps 124 Mbps 856 Mbps 7.90 310 slz 3300 Mbps 553 Mbps 2747 Mbps 5.97 1100 jquery.min.js zlib 430 Mbps 180 Mbps 250 Mbps 2.39 547 slz 1470 Mbps 764 Mbps 706 Mbps 1.92 1815 bootstrap.min.css zlib 790 Mbps 165 Mbps 625 Mbps 4.79 777 slz 2450 Mbps 650 Mbps 1800 Mbps 3.77 2400 So on top of saving a lot of memory, slz is constantly 2.5-3.5 times faster than zlib and results in providing more savings for a fixed CPU usage. For links smaller than 100 Mbps, zlib still provides a better compression ratio, at the expense of a much higher CPU usage. Larger input files provide slightly higher bandwidth for both libs, at the expense of a bit more memory usage for zlib (it converges to 256kB per connection).
2015-03-29 01:32:06 +00:00
# USE_SLZ : enable slz library instead of zlib (pick at most one).
# USE_CPU_AFFINITY : enable pinning processes to CPU on Linux. Automatic.
# USE_TFO : enable TCP fast open. Supported on Linux >= 3.7.
MAJOR: namespace: add Linux network namespace support This patch makes it possible to create binds and servers in separate namespaces. This can be used to proxy between multiple completely independent virtual networks (with possibly overlapping IP addresses) and a non-namespace-aware proxy implementation that supports the proxy protocol (v2). The setup is something like this: net1 on VLAN 1 (namespace 1) -\ net2 on VLAN 2 (namespace 2) -- haproxy ==== proxy (namespace 0) net3 on VLAN 3 (namespace 3) -/ The proxy is configured to make server connections through haproxy and sending the expected source/target addresses to haproxy using the proxy protocol. The network namespace setup on the haproxy node is something like this: = 8< = $ cat setup.sh ip netns add 1 ip link add link eth1 type vlan id 1 ip link set eth1.1 netns 1 ip netns exec 1 ip addr add 192.168.91.2/24 dev eth1.1 ip netns exec 1 ip link set eth1.$id up ... = 8< = = 8< = $ cat haproxy.cfg frontend clients bind 127.0.0.1:50022 namespace 1 transparent default_backend scb backend server mode tcp server server1 192.168.122.4:2222 namespace 2 send-proxy-v2 = 8< = A bind line creates the listener in the specified namespace, and connections originating from that listener also have their network namespace set to that of the listener. A server line either forces the connection to be made in a specified namespace or may use the namespace from the client-side connection if that was set. For more documentation please read the documentation included in the patch itself. Signed-off-by: KOVACS Tamas <ktamas@balabit.com> Signed-off-by: Sarkozi Laszlo <laszlo.sarkozi@balabit.com> Signed-off-by: KOVACS Krisztian <hidden@balabit.com>
2014-11-17 14:11:45 +00:00
# USE_NS : enable network namespace support. Supported on Linux >= 2.6.24.
# USE_DL : enable it if your system requires -ldl. Automatic on Linux.
# USE_RT : enable it if your system requires -lrt. Automatic on Linux.
# USE_DEVICEATLAS : enable DeviceAtlas api.
# USE_51DEGREES : enable third party device detection library from 51Degrees
# USE_WURFL : enable WURFL detection library from Scientiamobile
MEDIUM: mworker: Add systemd `Type=notify` support This patch adds support for `Type=notify` to the systemd unit. Supporting `Type=notify` improves both starting as well as reloading of the unit, because systemd will be let known when the action completed. See this quote from `systemd.service(5)`: > Note however that reloading a daemon by sending a signal (as with the > example line above) is usually not a good choice, because this is an > asynchronous operation and hence not suitable to order reloads of > multiple services against each other. It is strongly recommended to > set ExecReload= to a command that not only triggers a configuration > reload of the daemon, but also synchronously waits for it to complete. By making systemd aware of a reload in progress it is able to wait until the reload actually succeeded. This patch introduces both a new `USE_SYSTEMD` build option which controls including the sd-daemon library as well as a `-Ws` runtime option which runs haproxy in master-worker mode with systemd support. When haproxy is running in master-worker mode with systemd support it will send status messages to systemd using `sd_notify(3)` in the following cases: - The master process forked off the worker processes (READY=1) - The master process entered the `mworker_reload()` function (RELOADING=1) - The master process received the SIGUSR1 or SIGTERM signal (STOPPING=1) Change the unit file to specify `Type=notify` and replace master-worker mode (`-W`) with master-worker mode with systemd support (`-Ws`). Future evolutions of this feature could include making use of the `STATUS` feature of `sd_notify()` to send information about the number of active connections to systemd. This would require bidirectional communication between the master and the workers and thus is left for future work.
2017-11-20 14:58:35 +00:00
# USE_SYSTEMD : enable sd_notify() support.
BUILD: re-implement an initcall variant without using executable sections The current initcall implementation relies on dedicated sections (one section per init stage) to store the initcall descriptors. Then upon startup, these sections are scanned from beginning to end and all items found there are called in sequence. On platforms like AIX or Cygwin it seems difficult to figure the beginning and end of sections as the linker doesn't seem to provide the corresponding symbols. In order to replace this, this patch simply implements an array of single linked (one per init stage) which are fed using constructors for each register call. These constructors are declared static, with a name depending on their line number in the file, in order to avoid name clashes. The final effect is the same, except that the method is slightly more expensive in that it explicitly produces code to register these initcalls : $ size haproxy.sections haproxy.constructor text data bss dec hex filename 4060312 249176 1457652 5767140 57ffe4 haproxy.sections 4062862 260408 1457652 5780922 5835ba haproxy.constructor This mechanism is enabled as an alternative to the default one when build option USE_OBSOLETE_LINKER is set. This option is currently enabled by default only on AIX and Cygwin, and may be attempted for any target which fails to build complaining about missing symbols __start_init_* and/or __stop_init_*. Once confirmed as a reliable fix, this will likely have to be backported to 1.9 where AIX and Cygwin do not build anymore.
2019-03-29 20:30:17 +00:00
# USE_OBSOLETE_LINKER : use when the linker fails to emit __start_init/__stop_init
#
# Options can be forced by specifying "USE_xxx=1" or can be disabled by using
# "USE_xxx=" (empty string).
#
# Variables useful for packagers :
# CC is set to "gcc" by default and is used for compilation only.
# LD is set to "gcc" by default and is used for linking only.
# ARCH may be useful to force build of 32-bit binary on 64-bit systems
# CFLAGS is automatically set for the specified CPU and may be overridden.
# LDFLAGS is automatically set to -g and may be overridden.
# DEP may be cleared to ignore changes to include files during development
# SMALL_OPTS may be used to specify some options to shrink memory usage.
# DEBUG may be used to set some internal debugging options.
# ERR may be set to non-empty to pass -Werror to the compiler
# ADDINC may be used to complete the include path in the form -Ipath.
# ADDLIB may be used to complete the library list in the form -Lpath -llib.
# DEFINE may be used to specify any additional define, which will be reported
# by "haproxy -vv" in CFLAGS.
# SILENT_DEFINE may be used to specify other defines which will not be
# reported by "haproxy -vv".
# EXTRA is used to force building or not building some extra tools.
# DESTDIR is not set by default and is used for installation only.
# It might be useful to set DESTDIR if you want to install haproxy
# in a sandbox.
# PREFIX is set to "/usr/local" by default and is used for installation only.
# SBINDIR is set to "$(PREFIX)/sbin" by default and is used for installation
# only.
# MANDIR is set to "$(PREFIX)/share/man" by default and is used for
# installation only.
# DOCDIR is set to "$(PREFIX)/doc/haproxy" by default and is used for
# installation only.
#
# Other variables :
# PCRE_CONFIG : force the binary path to get pcre config (by default
# pcre-config)
# PCREDIR : force the path to libpcre.
# PCRE_LIB : force the lib path to libpcre (defaults to $PCREDIR/lib).
# PCRE_INC : force the include path to libpcre ($PCREDIR/inc)
# PCRE2_CONFIG : force the binary path to get pcre2 config (by default
# pcre2-config)
# SSL_LIB : force the lib path to libssl/libcrypto
# SSL_INC : force the include path to libssl/libcrypto
# LUA_LIB : force the lib path to lua
# LUA_INC : force the include path to lua
# LUA_LIB_NAME : force the lib name (or automatically evaluated, by order of
# priority : lua5.3, lua53, lua).
# IGNOREGIT : ignore GIT commit versions if set.
# VERSION : force haproxy version reporting.
# SUBVERS : add a sub-version (eg: platform, model, ...).
# VERDATE : force haproxy's release date.
#
# VTEST_PROGRAM : location of the vtest program to run reg-tests.
# verbosity: pass V=1 for verbose shell invocation
V = 0
Q = @
ifeq ($V,1)
Q=
endif
# Function used to detect support of a given option by the compiler.
# Usage: CFLAGS += $(call cc-opt,option). Eg: $(call cc-opt,-fwrapv)
# Note: ensure the referencing variable is assigned using ":=" and not "=" to
# call it only once.
cc-opt = $(shell set -e; if $(CC) $(1) -E -xc - -o /dev/null </dev/null >&0 2>&0; then echo "$(1)"; fi;)
# same but emits $2 if $1 is not supported
cc-opt-alt = $(shell set -e; if $(CC) $(1) -E -xc - -o /dev/null </dev/null >&0 2>&0; then echo "$(1)"; else echo "$(2)"; fi;)
# Disable a warning when supported by the compiler. Don't put spaces around the
# warning! And don't use cc-opt which doesn't always report an error until
# another one is also returned.
# Usage: CFLAGS += $(call cc-nowarn,warning). Eg: $(call cc-opt,format-truncation)
cc-nowarn = $(shell set -e; if $(CC) -W$(1) -E -xc - -o /dev/null </dev/null >&0 2>&0; then echo "-Wno-$(1)"; fi;)
#### Installation options.
DESTDIR =
PREFIX = /usr/local
SBINDIR = $(PREFIX)/sbin
MANDIR = $(PREFIX)/share/man
DOCDIR = $(PREFIX)/doc/haproxy
#### TARGET system
# Use TARGET=<target_name> to optimize for a specifc target OS among the
# following list (use the default "generic" if uncertain) :
# generic, linux22, linux24, linux24e, linux26, solaris,
2015-10-19 23:01:16 +00:00
# freebsd, openbsd, netbsd, cygwin, haiku, custom, aix51, aix52
TARGET =
#### TARGET CPU
# Use CPU=<cpu_name> to optimize for a particular CPU, among the following
# list :
# generic, native, i586, i686, ultrasparc, custom
CPU = generic
2005-12-17 11:21:26 +00:00
#### Architecture, used when not building for native architecture
# Use ARCH=<arch_name> to force build for a specific architecture. Known
# architectures will lead to "-m32" or "-m64" being added to CFLAGS and
# LDFLAGS. This can be required to build 32-bit binaries on 64-bit targets.
# Currently, only 32, 64, x86_64, i386, i486, i586 and i686 are understood.
ARCH =
#### Toolchain options.
# GCC is normally used both for compiling and linking.
CC = gcc
LD = $(CC)
#### Debug flags (typically "-g").
# Those flags only feed CFLAGS so it is not mandatory to use this form.
DEBUG_CFLAGS = -g
#### Add -Werror when set to non-empty
ERR =
#### May be used to force running a specific set of reg-tests
REG_TEST_FILES =
REG_TEST_SCRIPT=./scripts/run-regtests.sh
#### Compiler-specific flags that may be used to disable some negative over-
# optimization or to silence some warnings. -fno-strict-aliasing is needed with
# gcc >= 4.4.
# We rely on signed integer wraparound on overflow, however clang think it
# can do whatever it wants since it's an undefined behavior, so use -fwrapv
# to be sure we get the intended behavior.
SPEC_CFLAGS := -fno-strict-aliasing -Wdeclaration-after-statement
SPEC_CFLAGS += $(call cc-opt-alt,-fwrapv,$(call cc-opt,-fno-strict-overflow))
SPEC_CFLAGS += $(call cc-nowarn,format-truncation)
SPEC_CFLAGS += $(call cc-nowarn,address-of-packed-member)
SPEC_CFLAGS += $(call cc-nowarn,unused-label)
SPEC_CFLAGS += $(call cc-nowarn,sign-compare)
SPEC_CFLAGS += $(call cc-nowarn,unused-parameter)
SPEC_CFLAGS += $(call cc-nowarn,old-style-declaration)
SPEC_CFLAGS += $(call cc-nowarn,ignored-qualifiers)
SPEC_CFLAGS += $(call cc-nowarn,clobbered)
SPEC_CFLAGS += $(call cc-nowarn,missing-field-initializers)
SPEC_CFLAGS += $(call cc-nowarn,implicit-fallthrough)
SPEC_CFLAGS += $(call cc-nowarn,stringop-overflow)
SPEC_CFLAGS += $(call cc-nowarn,cast-function-type)
SPEC_CFLAGS += $(call cc-opt,-Wtype-limits)
SPEC_CFLAGS += $(call cc-opt,-Wshift-negative-value)
SPEC_CFLAGS += $(call cc-opt,-Wshift-overflow=2)
SPEC_CFLAGS += $(call cc-opt,-Wduplicated-cond)
SPEC_CFLAGS += $(call cc-opt,-Wnull-dereference)
#### Memory usage tuning
# If small memory footprint is required, you can reduce the buffer size. There
# are 2 buffers per concurrent session, so 16 kB buffers will eat 32 MB memory
# with 1000 concurrent sessions. Putting it slightly lower than a page size
# will prevent the additional parameters to go beyond a page. 8030 bytes is
# exactly 5.5 TCP segments of 1460 bytes and is generally good. Useful tuning
# macros include :
# SYSTEM_MAXCONN, BUFSIZE, MAXREWRITE, REQURI_LEN, CAPTURE_LEN.
# Example: SMALL_OPTS = -DBUFSIZE=8030 -DMAXREWRITE=1030 -DSYSTEM_MAXCONN=1024
SMALL_OPTS =
2005-12-17 11:21:26 +00:00
#### Debug settings
# You can enable debugging on specific code parts by setting DEBUG=-DDEBUG_xxx.
# Currently defined DEBUG macros include DEBUG_FULL, DEBUG_MEMORY, DEBUG_FSM,
# DEBUG_HASH, DEBUG_AUTH, DEBUG_SPOE, DEBUG_UAF and DEBUG_THREAD. Please check
# sources for exact meaning or do not use at all.
DEBUG =
MINOR: add a new function call tracer for debugging purposes This feature relies on GCC's ability to call helpers at function entry/exit points. We define these helpers to quickly dump the minimum info into a trace file that can be converted to a human readable format using a script in the contrib/trace directory. This has only been implemented in the GNU makefile for now on as it is unsure whether it's supported on all OSes. The feature is enabled by building with "TRACE=1". The performance impact is huge, so this feature should only be used when debugging. To limit the loss of performance, fprintf() has been disabled and the output is hand-crafted and emitted using fwrite(), resulting in doubling the performance. Using the TSC instead of gettimeofday() also doubles the performance. Around 1200 conns/s may be achieved on a Pentium-M 1.7 GHz which leads to around 50 MB/s of traces. The entry and exits of all functions will be dumped into a file designated by the HAPROXY_TRACE environment variable, or by default "trace.out". If the trace file name is empty or "/dev/null", then traces are disabled. If opening the trace file fails, then stderr is used. If HAPROXY_TRACE_FAST is used, then the time is taken from the global <now> variable. Last, if HAPROXY_TRACE_TSC is used, then the machine's TSC is used instead of the real time (almost twice as fast). The output format is : <sec.usec> <level> <caller_ptr> <dir> <callee_ptr> or : <tsc> <level> <caller_ptr> <dir> <callee_ptr> where <dir> is '>' when entering a function and '<' when leaving. The awk script in contrib/trace provides a nicer indented output : 6f74989e6f8 ->->-> run_poll_loop > signal_process_queue [src/haproxy.c:1097:0x804bd69] > [include/proto/signal.h:32:0x8049cd0] 6f74989eb00 run_poll_loop < signal_process_queue [src/haproxy.c:1097:0x804bd69] < [include/proto/signal.h:32:0x8049cd0] 6f74989ef44 ->->-> run_poll_loop > wake_expired_tasks [src/haproxy.c:1100:0x804bd72] > [src/task.c:123:0x8055060] 6f74989f3a6 ->->->-> wake_expired_tasks > eb32_lookup_ge [src/task.c:128:0x8055091] > [ebtree/eb32tree.c:138:0x80a8c70] 6f74989f7e9 wake_expired_tasks < eb32_lookup_ge [src/task.c:128:0x8055091] < [ebtree/eb32tree.c:138:0x80a8c70] 6f74989fc0d ->->->-> wake_expired_tasks > eb32_first [src/task.c:134:0x80550d5] > [ebtree/eb32tree.h:55:0x8054ad0] 6f7498a003d ->->->->-> eb32_first > eb_first [ebtree/eb32tree.h:56:0x8054af1] > [ebtree/ebtree.h:520:0x8054a10] 6f7498a0436 ->->->->->-> eb_first > eb_walk_down [ebtree/ebtree.h:521:0x8054a33] > [ebtree/ebtree.h:442:0x80549a0] 6f7498a0843 ->->->->->->-> eb_walk_down > eb_gettag [ebtree/ebtree.h:445:0x80549d6] > [ebtree/ebtree.h:418:0x80548e0] 6f7498a0c2b eb_walk_down < eb_gettag [ebtree/ebtree.h:445:0x80549d6] < [ebtree/ebtree.h:418:0x80548e0] 6f7498a1042 ->->->->->->-> eb_walk_down > eb_untag [ebtree/ebtree.h:447:0x80549e2] > [ebtree/ebtree.h:412:0x80548a0] 6f7498a1498 eb_walk_down < eb_untag [ebtree/ebtree.h:447:0x80549e2] < [ebtree/ebtree.h:412:0x80548a0] 6f7498a18c6 ->->->->->->-> eb_walk_down > eb_root_to_node [ebtree/ebtree.h:448:0x80549e7] > [ebtree/ebtree.h:432:0x8054960] 6f7498a1cd4 eb_walk_down < eb_root_to_node [ebtree/ebtree.h:448:0x80549e7] < [ebtree/ebtree.h:432:0x8054960] 6f7498a20c4 eb_first < eb_walk_down [ebtree/ebtree.h:521:0x8054a33] < [ebtree/ebtree.h:442:0x80549a0] 6f7498a24b4 eb32_first < eb_first [ebtree/eb32tree.h:56:0x8054af1] < [ebtree/ebtree.h:520:0x8054a10] 6f7498a289c wake_expired_tasks < eb32_first [src/task.c:134:0x80550d5] < [ebtree/eb32tree.h:55:0x8054ad0] 6f7498a2c8c run_poll_loop < wake_expired_tasks [src/haproxy.c:1100:0x804bd72] < [src/task.c:123:0x8055060] 6f7498a3095 ->->-> run_poll_loop > process_runnable_tasks [src/haproxy.c:1103:0x804bd7a] > [src/task.c:190:0x8055150] A nice improvement would possibly consist in trying to get the function's arguments in the stack and to dump a few more infor for some well-known functions (eg: the session's status for process_session).
2012-05-25 21:53:16 +00:00
#### Trace options
# Use TRACE=1 to trace function calls to file "trace.out" or to stderr if not
# possible.
TRACE =
#### Additional include and library dirs
# Redefine this if you want to add some special PATH to include/libs
ADDINC =
ADDLIB =
#### Specific macro definitions
# Use DEFINE=-Dxxx to set any tunable macro. Anything declared here will appear
# in the build options reported by "haproxy -vv". Use SILENT_DEFINE if you do
# not want to pollute the report with complex defines.
# The following settings might be of interest when SSL is enabled :
# LISTEN_DEFAULT_CIPHERS is a cipher suite string used to set the default SSL
# ciphers on "bind" lines instead of using OpenSSL's defaults.
# CONNECT_DEFAULT_CIPHERS is a cipher suite string used to set the default
# SSL ciphers on "server" lines instead of using OpenSSL's defaults.
DEFINE =
SILENT_DEFINE =
#### extra programs to build
# Force this to enable building extra programs or to disable them.
# It's automatically appended depending on the targets.
EXTRA =
#### CPU dependant optimizations
# Some CFLAGS are set by default depending on the target CPU. Those flags only
# feed CPU_CFLAGS, which in turn feed CFLAGS, so it is not mandatory to use
# them. You should not have to change these options. Better use CPU_CFLAGS or
# even CFLAGS instead.
CPU_CFLAGS.generic = -O2
CPU_CFLAGS.native = -O2 -march=native
CPU_CFLAGS.i586 = -O2 -march=i586
CPU_CFLAGS.i686 = -O2 -march=i686
CPU_CFLAGS.ultrasparc = -O6 -mcpu=v9 -mtune=ultrasparc
CPU_CFLAGS = $(CPU_CFLAGS.$(CPU))
#### ARCH dependant flags, may be overridden by CPU flags
ARCH_FLAGS.32 = -m32
ARCH_FLAGS.64 = -m64
ARCH_FLAGS.i386 = -m32 -march=i386
ARCH_FLAGS.i486 = -m32 -march=i486
ARCH_FLAGS.i586 = -m32 -march=i586
ARCH_FLAGS.i686 = -m32 -march=i686
ARCH_FLAGS.x86_64 = -m64 -march=x86-64
ARCH_FLAGS = $(ARCH_FLAGS.$(ARCH))
#### Common CFLAGS
# These CFLAGS contain general optimization options, CPU-specific optimizations
# and debug flags. They may be overridden by some distributions which prefer to
# set all of them at once instead of playing with the CPU and DEBUG variables.
CFLAGS = $(ARCH_FLAGS) $(CPU_CFLAGS) $(DEBUG_CFLAGS) $(SPEC_CFLAGS)
#### Common LDFLAGS
# These LDFLAGS are used as the first "ld" options, regardless of any library
# path or any other option. They may be changed to add any linker-specific
# option at the beginning of the ld command line.
LDFLAGS = $(ARCH_FLAGS) -g
#### list of all "USE_*" options. These ones must be updated if new options are
# added, so that the relevant options are properly added to the CFLAGS and to
# the reported build options.
use_opts = USE_EPOLL USE_KQUEUE USE_MY_EPOLL USE_MY_SPLICE USE_NETFILTER \
USE_PCRE USE_PCRE_JIT USE_PCRE2 USE_PCRE2_JIT USE_POLL \
USE_PRIVATE_CACHE USE_THREAD USE_PTHREAD_PSHARED USE_REGPARM \
USE_STATIC_PCRE USE_STATIC_PCRE2 USE_TPROXY USE_LINUX_TPROXY \
USE_LINUX_SPLICE USE_LIBCRYPT USE_CRYPT_H USE_VSYSCALL \
USE_GETADDRINFO USE_OPENSSL USE_LUA USE_FUTEX USE_ACCEPT4 \
USE_MY_ACCEPT4 USE_ZLIB USE_SLZ USE_CPU_AFFINITY USE_TFO USE_NS \
USE_DL USE_RT USE_DEVICEATLAS USE_51DEGREES USE_WURFL USE_SYSTEMD \
USE_OBSOLETE_LINKER USE_PRCTL
#### Target system options
# Depending on the target platform, some options are set, as well as some
# CFLAGS and LDFLAGS. All variables pre-set here will not appear in the build
# options string. They may be set to any value, but are historically set to
# "implicit" which eases debugging. You should not have to change anything
# there unless you're adding support for a new platform.
default_opts = $(foreach name,$(1),$(eval $(name)=implicit))
# poll() is always supported, unless explicitly disabled by passing USE_POLL=""
# on the make command line.
USE_POLL = default
# Always enable threads support by default and let the Makefile detect if
# HAProxy can be compiled with threads or not.
# generic system target has nothing specific
ifeq ($(TARGET),generic)
set_target_defaults = $(call default_opts,USE_POLL USE_TPROXY)
endif
# Haiku
2015-10-19 23:01:16 +00:00
ifeq ($(TARGET),haiku)
TARGET_LDFLAGS = -lnetwork
set_target_defaults = $(call default_opts,USE_POLL USE_TPROXY)
endif
# Linux 2.2
ifeq ($(TARGET),linux22)
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_LIBCRYPT USE_DL USE_RT)
endif
# Standard Linux 2.4 with netfilter but without epoll()
ifeq ($(TARGET),linux24)
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_LIBCRYPT USE_DL USE_RT USE_CRYPT_H USE_NETFILTER)
endif
# Enhanced Linux 2.4 with netfilter and epoll() patch > 0.21
ifeq ($(TARGET),linux24e)
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_LIBCRYPT USE_DL USE_RT USE_CRYPT_H USE_NETFILTER \
USE_EPOLL USE_MY_EPOLL)
endif
# Standard Linux 2.6 with netfilter and standard epoll()
ifeq ($(TARGET),linux26)
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_LIBCRYPT USE_DL USE_RT USE_CRYPT_H USE_NETFILTER \
USE_EPOLL USE_FUTEX USE_PRCTL)
endif
# Standard Linux >= 2.6.28 with netfilter, epoll, tproxy and splice
ifeq ($(TARGET),linux2628)
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_LIBCRYPT USE_DL USE_RT USE_CRYPT_H USE_NETFILTER \
USE_CPU_AFFINITY USE_THREAD USE_EPOLL USE_FUTEX USE_LINUX_TPROXY \
USE_ACCEPT4 USE_LINUX_SPLICE USE_PRCTL ASSUME_SPLICE_WORKS)
endif
# Solaris 8 and above
ifeq ($(TARGET),solaris)
# We also enable getaddrinfo() which works since solaris 8.
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_LIBCRYPT USE_CRYPT_H USE_GETADDRINFO USE_THREAD)
TARGET_CFLAGS = -fomit-frame-pointer -DFD_SETSIZE=65536 -D_REENTRANT -D_XOPEN_SOURCE=500 -D__EXTENSIONS__
TARGET_LDFLAGS = -lnsl -lsocket
endif
# FreeBSD 5 and above
ifeq ($(TARGET),freebsd)
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_LIBCRYPT USE_THREAD USE_CPU_AFFINITY USE_KQUEUE \
USE_CLOSEFROM)
endif
# Mac OS/X
ifeq ($(TARGET),osx)
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_KQUEUE)
EXPORT_SYMBOL = -export_dynamic
endif
# OpenBSD 5.7 and above
ifeq ($(TARGET),openbsd)
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_THREAD USE_KQUEUE USE_ACCEPT4)
endif
# NetBSD
ifeq ($(TARGET),netbsd)
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_KQUEUE)
endif
# AIX 5.1 only
ifeq ($(TARGET),aix51)
set_target_defaults = $(call default_opts, \
USE_POLL USE_LIBCRYPT USE_OBSOLETE_LINKER USE_PRIVATE_CACHE)
TARGET_CFLAGS = -Dss_family=__ss_family -Dip6_hdr=ip6hdr -DSTEVENS_API -D_LINUX_SOURCE_COMPAT -Dunsetenv=my_unsetenv
DEBUG_CFLAGS =
endif
# AIX 5.2 and above
ifeq ($(TARGET),aix52)
set_target_defaults = $(call default_opts, \
BUILD: re-implement an initcall variant without using executable sections The current initcall implementation relies on dedicated sections (one section per init stage) to store the initcall descriptors. Then upon startup, these sections are scanned from beginning to end and all items found there are called in sequence. On platforms like AIX or Cygwin it seems difficult to figure the beginning and end of sections as the linker doesn't seem to provide the corresponding symbols. In order to replace this, this patch simply implements an array of single linked (one per init stage) which are fed using constructors for each register call. These constructors are declared static, with a name depending on their line number in the file, in order to avoid name clashes. The final effect is the same, except that the method is slightly more expensive in that it explicitly produces code to register these initcalls : $ size haproxy.sections haproxy.constructor text data bss dec hex filename 4060312 249176 1457652 5767140 57ffe4 haproxy.sections 4062862 260408 1457652 5780922 5835ba haproxy.constructor This mechanism is enabled as an alternative to the default one when build option USE_OBSOLETE_LINKER is set. This option is currently enabled by default only on AIX and Cygwin, and may be attempted for any target which fails to build complaining about missing symbols __start_init_* and/or __stop_init_*. Once confirmed as a reliable fix, this will likely have to be backported to 1.9 where AIX and Cygwin do not build anymore.
2019-03-29 20:30:17 +00:00
USE_POLL USE_LIBCRYPT USE_OBSOLETE_LINKER)
TARGET_CFLAGS = -D_MSGQSUPPORT
DEBUG_CFLAGS =
endif
# Cygwin
ifeq ($(TARGET),cygwin)
BUILD: re-implement an initcall variant without using executable sections The current initcall implementation relies on dedicated sections (one section per init stage) to store the initcall descriptors. Then upon startup, these sections are scanned from beginning to end and all items found there are called in sequence. On platforms like AIX or Cygwin it seems difficult to figure the beginning and end of sections as the linker doesn't seem to provide the corresponding symbols. In order to replace this, this patch simply implements an array of single linked (one per init stage) which are fed using constructors for each register call. These constructors are declared static, with a name depending on their line number in the file, in order to avoid name clashes. The final effect is the same, except that the method is slightly more expensive in that it explicitly produces code to register these initcalls : $ size haproxy.sections haproxy.constructor text data bss dec hex filename 4060312 249176 1457652 5767140 57ffe4 haproxy.sections 4062862 260408 1457652 5780922 5835ba haproxy.constructor This mechanism is enabled as an alternative to the default one when build option USE_OBSOLETE_LINKER is set. This option is currently enabled by default only on AIX and Cygwin, and may be attempted for any target which fails to build complaining about missing symbols __start_init_* and/or __stop_init_*. Once confirmed as a reliable fix, this will likely have to be backported to 1.9 where AIX and Cygwin do not build anymore.
2019-03-29 20:30:17 +00:00
set_target_defaults = $(call default_opts, \
USE_POLL USE_TPROXY USE_OBSOLETE_LINKER)
# Cygwin adds IPv6 support only in version 1.7 (in beta right now).
TARGET_CFLAGS = $(if $(filter 1.5.%, $(shell uname -r)), -DUSE_IPV6 -DAF_INET6=23 -DINET6_ADDRSTRLEN=46, )
endif
# set the default settings according to the target above
$(set_target_defaults)
#### Determine version, sub-version and release date.
# If GIT is found, and IGNOREGIT is not set, VERSION, SUBVERS and VERDATE are
# extracted from the last commit. Otherwise, use the contents of the files
# holding the same names in the current directory.
ifeq ($(IGNOREGIT),)
VERSION := $(shell [ -d .git/. ] && ref=`(git describe --tags --match 'v*' --abbrev=0) 2>/dev/null` && ref=$${ref%-g*} && echo "$${ref\#v}")
ifneq ($(VERSION),)
# OK git is there and works.
SUBVERS := $(shell comms=`git log --format=oneline --no-merges v$(VERSION).. 2>/dev/null | wc -l | tr -dc '0-9'`; commit=`(git log -1 --pretty=%h --abbrev=6) 2>/dev/null`; [ $$comms -gt 0 ] && echo "-$$commit-$$comms")
VERDATE := $(shell git log -1 --pretty=format:%ci | cut -f1 -d' ' | tr '-' '/')
endif
endif
# Last commit version not found, take it from the files.
ifeq ($(VERSION),)
VERSION := $(shell cat VERSION 2>/dev/null || touch VERSION)
endif
ifeq ($(SUBVERS),)
SUBVERS := $(shell (grep -v '\$$Format' SUBVERS 2>/dev/null || touch SUBVERS) | head -n 1)
endif
ifeq ($(VERDATE),)
VERDATE := $(shell (grep -v '^\$$Format' VERDATE 2>/dev/null || touch VERDATE) | head -n 1 | cut -f1 -d' ' | tr '-' '/')
endif
#### Build options
# Do not change these ones, enable USE_* variables instead.
OPTIONS_CFLAGS =
OPTIONS_LDFLAGS =
OPTIONS_OBJS =
#### Extra objects to be built and integrated (used only for development)
EXTRA_OBJS =
# Return USE_xxx=$(USE_xxx) if the variable was set from the environment or the
# command line.
# Usage:
# BUILD_OPTIONS += $(call ignore_implicit,USE_xxx)
ignore_implicit = $(if $(subst environment,,$(origin $(1))), \
$(if $(subst command line,,$(origin $(1))),, \
$(1)=$($(1))), \
$(1)=$($(1))) \
# This variable collects all USE_* values except those set to "implicit". This
# is used to report a list of all flags which were used to build this version.
# Do not assign anything to it.
BUILD_OPTIONS := $(foreach opt,$(use_opts),$(call ignore_implicit,$(opt)))
BUILD_FEATURES := $(foreach opt,$(patsubst USE_%,%,$(use_opts)),$(if $(USE_$(opt)),+$(opt),-$(opt)))
# All USE_* options have their equivalent macro defined in the code (some might
# possibly be unused though)
OPTIONS_CFLAGS += $(foreach opt,$(use_opts),$(if $($(opt)),-D$(opt),))
ifneq ($(USE_LINUX_SPLICE),)
OPTIONS_CFLAGS += -DCONFIG_HAP_LINUX_SPLICE
endif
ifneq ($(USE_TPROXY),)
OPTIONS_CFLAGS += -DTPROXY
endif
ifneq ($(USE_LINUX_TPROXY),)
OPTIONS_CFLAGS += -DCONFIG_HAP_LINUX_TPROXY
endif
ifneq ($(USE_LIBCRYPT),)
OPTIONS_CFLAGS += -DCONFIG_HAP_CRYPT
OPTIONS_LDFLAGS += -lcrypt
endif
ifneq ($(USE_CRYPT_H),)
OPTIONS_CFLAGS += -DNEED_CRYPT_H
endif
MAJOR: compression: integrate support for libslz This library is designed to emit a zlib-compatible stream with no memory usage and to favor resource savings over compression ratio. While zlib requires 256 kB of RAM per compression context (and can only support 4000 connections per GB of RAM), the stateless compression offered by libslz does not need to retain buffers between subsequent calls. In theory this slightly reduces the compression ratio but in practice it does not have that much of an effect since the zlib window is limited to 32kB. Libslz is available at : http://git.1wt.eu/web?p=libslz.git It was designed for web compression and provides a lot of savings over zlib in haproxy. Here are the preliminary results on a single core of a core2-quad 3.0 GHz in 32-bit for only 300 concurrent sessions visiting the home page of www.haproxy.org (76 kB) with the default 16kB buffers : BW In BW Out BW Saved Ratio memory VSZ/RSS zlib 237 Mbps 92 Mbps 145 Mbps 2.58 84M / 69M slz 733 Mbps 380 Mbps 353 Mbps 1.93 5.9M / 4.2M So while the compression ratio is lower, the bandwidth savings are much more important due to the significantly lower compression cost which allows to consume even more data from the servers. In the example above, zlib became the bottleneck at 24% of the output bandwidth. Also the difference in memory usage is obvious. More tests run on a single core of a core i5-3320M, with 500 concurrent users and the default 16kB buffers : At 100% CPU (no limit) : BW In BW Out BW Saved Ratio memory VSZ/RSS hits/s zlib 480 Mbps 188 Mbps 292 Mbps 2.55 130M / 101M 744 slz 1700 Mbps 810 Mbps 890 Mbps 2.10 23.7M / 9.7M 2382 At 85% CPU (limited) : BW In BW Out BW Saved Ratio memory VSZ/RSS hits/s zlib 1240 Mbps 976 Mbps 264 Mbps 1.27 130M / 100M 1738 slz 1600 Mbps 976 Mbps 624 Mbps 1.64 23.7M / 9.7M 2210 The most important benefit really happens when the CPU usage is limited by "maxcompcpuusage" or the BW limited by "maxcomprate" : in order to preserve resources, haproxy throttles the compression ratio until usage is within limits. Since slz is much cheaper, the average compression ratio is much higher and the input bandwidth is quite higher for one Gbps output. Other tests made with some reference files : BW In BW Out BW Saved Ratio hits/s daniels.html zlib 1320 Mbps 163 Mbps 1157 Mbps 8.10 1925 slz 3600 Mbps 580 Mbps 3020 Mbps 6.20 5300 tv.com/listing zlib 980 Mbps 124 Mbps 856 Mbps 7.90 310 slz 3300 Mbps 553 Mbps 2747 Mbps 5.97 1100 jquery.min.js zlib 430 Mbps 180 Mbps 250 Mbps 2.39 547 slz 1470 Mbps 764 Mbps 706 Mbps 1.92 1815 bootstrap.min.css zlib 790 Mbps 165 Mbps 625 Mbps 4.79 777 slz 2450 Mbps 650 Mbps 1800 Mbps 3.77 2400 So on top of saving a lot of memory, slz is constantly 2.5-3.5 times faster than zlib and results in providing more savings for a fixed CPU usage. For links smaller than 100 Mbps, zlib still provides a better compression ratio, at the expense of a much higher CPU usage. Larger input files provide slightly higher bandwidth for both libs, at the expense of a bit more memory usage for zlib (it converges to 256kB per connection).
2015-03-29 01:32:06 +00:00
ifneq ($(USE_SLZ),)
# Use SLZ_INC and SLZ_LIB to force path to zlib.h and libz.{a,so} if needed.
SLZ_INC =
SLZ_LIB =
OPTIONS_CFLAGS += $(if $(SLZ_INC),-I$(SLZ_INC))
MAJOR: compression: integrate support for libslz This library is designed to emit a zlib-compatible stream with no memory usage and to favor resource savings over compression ratio. While zlib requires 256 kB of RAM per compression context (and can only support 4000 connections per GB of RAM), the stateless compression offered by libslz does not need to retain buffers between subsequent calls. In theory this slightly reduces the compression ratio but in practice it does not have that much of an effect since the zlib window is limited to 32kB. Libslz is available at : http://git.1wt.eu/web?p=libslz.git It was designed for web compression and provides a lot of savings over zlib in haproxy. Here are the preliminary results on a single core of a core2-quad 3.0 GHz in 32-bit for only 300 concurrent sessions visiting the home page of www.haproxy.org (76 kB) with the default 16kB buffers : BW In BW Out BW Saved Ratio memory VSZ/RSS zlib 237 Mbps 92 Mbps 145 Mbps 2.58 84M / 69M slz 733 Mbps 380 Mbps 353 Mbps 1.93 5.9M / 4.2M So while the compression ratio is lower, the bandwidth savings are much more important due to the significantly lower compression cost which allows to consume even more data from the servers. In the example above, zlib became the bottleneck at 24% of the output bandwidth. Also the difference in memory usage is obvious. More tests run on a single core of a core i5-3320M, with 500 concurrent users and the default 16kB buffers : At 100% CPU (no limit) : BW In BW Out BW Saved Ratio memory VSZ/RSS hits/s zlib 480 Mbps 188 Mbps 292 Mbps 2.55 130M / 101M 744 slz 1700 Mbps 810 Mbps 890 Mbps 2.10 23.7M / 9.7M 2382 At 85% CPU (limited) : BW In BW Out BW Saved Ratio memory VSZ/RSS hits/s zlib 1240 Mbps 976 Mbps 264 Mbps 1.27 130M / 100M 1738 slz 1600 Mbps 976 Mbps 624 Mbps 1.64 23.7M / 9.7M 2210 The most important benefit really happens when the CPU usage is limited by "maxcompcpuusage" or the BW limited by "maxcomprate" : in order to preserve resources, haproxy throttles the compression ratio until usage is within limits. Since slz is much cheaper, the average compression ratio is much higher and the input bandwidth is quite higher for one Gbps output. Other tests made with some reference files : BW In BW Out BW Saved Ratio hits/s daniels.html zlib 1320 Mbps 163 Mbps 1157 Mbps 8.10 1925 slz 3600 Mbps 580 Mbps 3020 Mbps 6.20 5300 tv.com/listing zlib 980 Mbps 124 Mbps 856 Mbps 7.90 310 slz 3300 Mbps 553 Mbps 2747 Mbps 5.97 1100 jquery.min.js zlib 430 Mbps 180 Mbps 250 Mbps 2.39 547 slz 1470 Mbps 764 Mbps 706 Mbps 1.92 1815 bootstrap.min.css zlib 790 Mbps 165 Mbps 625 Mbps 4.79 777 slz 2450 Mbps 650 Mbps 1800 Mbps 3.77 2400 So on top of saving a lot of memory, slz is constantly 2.5-3.5 times faster than zlib and results in providing more savings for a fixed CPU usage. For links smaller than 100 Mbps, zlib still provides a better compression ratio, at the expense of a much higher CPU usage. Larger input files provide slightly higher bandwidth for both libs, at the expense of a bit more memory usage for zlib (it converges to 256kB per connection).
2015-03-29 01:32:06 +00:00
OPTIONS_LDFLAGS += $(if $(SLZ_LIB),-L$(SLZ_LIB)) -lslz
endif
MEDIUM: HTTP compression (zlib library support) This commit introduces HTTP compression using the zlib library. http_response_forward_body has been modified to call the compression functions. This feature includes 3 algorithms: identity, gzip and deflate: * identity: this is mostly for debugging, and it was useful for developping the compression feature. With Content-Length in input, it is making each chunk with the data available in the current buffer. With chunks in input, it is rechunking, the output chunks will be bigger or smaller depending of the size of the input chunk and the size of the buffer. Identity does not apply any change on data. * gzip: same as identity, but applying a gzip compression. The data are deflated using the Z_NO_FLUSH flag in zlib. When there is no more data in the input buffer, it flushes the data in the output buffer (Z_SYNC_FLUSH). At the end of data, when it receives the last chunk in input, or when there is no more data to read, it writes the end of data with Z_FINISH and the ending chunk. * deflate: same as gzip, but with deflate algorithm and zlib format. Note that this algorithm has ambiguous support on many browsers and no support at all from recent ones. It is strongly recommended not to use it for anything else than experimentation. You can't choose the compression ratio at the moment, it will be set to Z_BEST_SPEED (1), as tests have shown very little benefit in terms of compression ration when going above for HTML contents, at the cost of a massive CPU impact. Compression will be activated depending of the Accept-Encoding request header. With identity, it does not take care of that header. To build HAProxy with zlib support, use USE_ZLIB=1 in the make parameters. This work was initially started by David Du Colombier at Exceliance.
2012-10-23 08:25:10 +00:00
ifneq ($(USE_ZLIB),)
# Use ZLIB_INC and ZLIB_LIB to force path to zlib.h and libz.{a,so} if needed.
ZLIB_INC =
ZLIB_LIB =
OPTIONS_CFLAGS += $(if $(ZLIB_INC),-I$(ZLIB_INC))
OPTIONS_LDFLAGS += $(if $(ZLIB_LIB),-L$(ZLIB_LIB)) -lz
MEDIUM: HTTP compression (zlib library support) This commit introduces HTTP compression using the zlib library. http_response_forward_body has been modified to call the compression functions. This feature includes 3 algorithms: identity, gzip and deflate: * identity: this is mostly for debugging, and it was useful for developping the compression feature. With Content-Length in input, it is making each chunk with the data available in the current buffer. With chunks in input, it is rechunking, the output chunks will be bigger or smaller depending of the size of the input chunk and the size of the buffer. Identity does not apply any change on data. * gzip: same as identity, but applying a gzip compression. The data are deflated using the Z_NO_FLUSH flag in zlib. When there is no more data in the input buffer, it flushes the data in the output buffer (Z_SYNC_FLUSH). At the end of data, when it receives the last chunk in input, or when there is no more data to read, it writes the end of data with Z_FINISH and the ending chunk. * deflate: same as gzip, but with deflate algorithm and zlib format. Note that this algorithm has ambiguous support on many browsers and no support at all from recent ones. It is strongly recommended not to use it for anything else than experimentation. You can't choose the compression ratio at the moment, it will be set to Z_BEST_SPEED (1), as tests have shown very little benefit in terms of compression ration when going above for HTML contents, at the cost of a massive CPU impact. Compression will be activated depending of the Accept-Encoding request header. With identity, it does not take care of that header. To build HAProxy with zlib support, use USE_ZLIB=1 in the make parameters. This work was initially started by David Du Colombier at Exceliance.
2012-10-23 08:25:10 +00:00
endif
ifneq ($(USE_POLL),)
OPTIONS_CFLAGS += -DENABLE_POLL
OPTIONS_OBJS += src/ev_poll.o
endif
ifneq ($(USE_EPOLL),)
OPTIONS_CFLAGS += -DENABLE_EPOLL
OPTIONS_OBJS += src/ev_epoll.o
endif
ifneq ($(USE_KQUEUE),)
OPTIONS_CFLAGS += -DENABLE_KQUEUE
OPTIONS_OBJS += src/ev_kqueue.o
endif
ifneq ($(USE_VSYSCALL),)
OPTIONS_OBJS += src/i386-linux-vsys.o
OPTIONS_CFLAGS += -DCONFIG_HAP_LINUX_VSYSCALL
endif
ifneq ($(ASSUME_SPLICE_WORKS),)
OPTIONS_CFLAGS += -DASSUME_SPLICE_WORKS
endif
ifneq ($(USE_NETFILTER),)
OPTIONS_CFLAGS += -DNETFILTER
endif
ifneq ($(USE_REGPARM),)
OPTIONS_CFLAGS += -DCONFIG_REGPARM=3
endif
ifneq ($(USE_DL),)
OPTIONS_LDFLAGS += -ldl
endif
ifneq ($(USE_THREAD),)
OPTIONS_LDFLAGS += -lpthread
endif
ifneq ($(USE_RT),)
OPTIONS_LDFLAGS += -lrt
endif
ifneq ($(USE_OPENSSL),)
# OpenSSL is packaged in various forms and with various dependencies.
# In general -lssl is enough, but on some platforms, -lcrypto may be needed,
# reason why it's added by default. Some even need -lz, then you'll need to
# pass it in the "ADDLIB" variable if needed. If your SSL libraries are not
# in the usual path, use SSL_INC=/path/to/inc and SSL_LIB=/path/to/lib.
OPTIONS_CFLAGS += $(if $(SSL_INC),-I$(SSL_INC))
OPTIONS_LDFLAGS += $(if $(SSL_LIB),-L$(SSL_LIB)) -lssl -lcrypto
ifneq ($(USE_DL),)
OPTIONS_LDFLAGS += -ldl
endif
OPTIONS_OBJS += src/ssl_sock.o
endif
# The private cache option affect the way the shctx is built
ifeq ($(USE_PRIVATE_CACHE),)
ifneq ($(USE_PTHREAD_PSHARED),)
OPTIONS_LDFLAGS += -lpthread
endif
endif
ifneq ($(USE_LUA),)
check_lua_lib = $(shell echo "int main(){}" | $(CC) -o /dev/null -x c - $(2) -l$(1) 2>/dev/null && echo $(1))
check_lua_inc = $(shell if [ -d $(2)$(1) ]; then echo $(2)$(1); fi;)
OPTIONS_CFLAGS += $(if $(LUA_INC),-I$(LUA_INC))
LUA_LD_FLAGS := -Wl,$(if $(EXPORT_SYMBOL),$(EXPORT_SYMBOL),--export-dynamic) $(if $(LUA_LIB),-L$(LUA_LIB))
ifeq ($(LUA_LIB_NAME),)
# Try to automatically detect the Lua library
LUA_LIB_NAME := $(firstword $(foreach lib,lua5.3 lua53 lua,$(call check_lua_lib,$(lib),$(LUA_LD_FLAGS))))
ifeq ($(LUA_LIB_NAME),)
$(error unable to automatically detect the Lua library name, you can enforce its name with LUA_LIB_NAME=<name> (where <name> can be lua5.3, lua53, lua, ...))
endif
LUA_INC := $(firstword $(foreach lib,lua5.3 lua53 lua,$(call check_lua_inc,$(lib),"/usr/include/")))
ifneq ($(LUA_INC),)
OPTIONS_CFLAGS += -I$(LUA_INC)
endif
endif
OPTIONS_LDFLAGS += $(LUA_LD_FLAGS) -l$(LUA_LIB_NAME) -lm
ifneq ($(USE_DL),)
OPTIONS_LDFLAGS += -ldl
endif
OPTIONS_OBJS += src/hlua.o src/hlua_fcn.o
endif
ifneq ($(USE_DEVICEATLAS),)
ifeq ($(USE_PCRE),)
$(error the DeviceAtlas module needs the PCRE library in order to compile)
endif
# Use DEVICEATLAS_SRC and possibly DEVICEATLAS_INC and DEVICEATLAS_LIB to force path
# to DeviceAtlas headers and libraries if needed.
DEVICEATLAS_SRC =
DEVICEATLAS_INC = $(DEVICEATLAS_SRC)
DEVICEATLAS_LIB = $(DEVICEATLAS_SRC)
ifeq ($(DEVICEATLAS_SRC),)
OPTIONS_LDFLAGS += -lda
else
OPTIONS_OBJS += $(DEVICEATLAS_LIB)/json.o
OPTIONS_OBJS += $(DEVICEATLAS_LIB)/dac.o
endif
OPTIONS_OBJS += src/da.o
OPTIONS_CFLAGS += $(if $(DEVICEATLAS_INC),-I$(DEVICEATLAS_INC))
endif
ifneq ($(USE_51DEGREES),)
# Use 51DEGREES_SRC and possibly 51DEGREES_INC and 51DEGREES_LIB to force path
# to 51degrees headers and libraries if needed.
51DEGREES_SRC =
51DEGREES_INC = $(51DEGREES_SRC)
51DEGREES_LIB = $(51DEGREES_SRC)
OPTIONS_OBJS += $(51DEGREES_LIB)/../cityhash/city.o
OPTIONS_OBJS += $(51DEGREES_LIB)/51Degrees.o
OPTIONS_OBJS += src/51d.o
OPTIONS_CFLAGS += $(if $(51DEGREES_INC),-I$(51DEGREES_INC))
ifeq ($(USE_THREAD),)
OPTIONS_CFLAGS += -DFIFTYONEDEGREES_NO_THREADING
else
OPTIONS_OBJS += $(51DEGREES_LIB)/../threading.o
endif
OPTIONS_LDFLAGS += $(if $(51DEGREES_LIB),-L$(51DEGREES_LIB)) -lm
endif
ifneq ($(USE_WURFL),)
# Use WURFL_SRC and possibly WURFL_INC and WURFL_LIB to force path
# to WURFL headers and libraries if needed.
WURFL_SRC =
WURFL_INC = $(WURFL_SRC)
WURFL_LIB = $(WURFL_SRC)
OPTIONS_OBJS += src/wurfl.o
OPTIONS_CFLAGS += $(if $(WURFL_INC),-I$(WURFL_INC))
ifneq ($(WURFL_DEBUG),)
OPTIONS_CFLAGS += -DWURFL_DEBUG
endif
ifneq ($(WURFL_HEADER_WITH_DETAILS),)
OPTIONS_CFLAGS += -DWURFL_HEADER_WITH_DETAILS
endif
OPTIONS_LDFLAGS += $(if $(WURFL_LIB),-L$(WURFL_LIB)) -lwurfl
endif
MEDIUM: mworker: Add systemd `Type=notify` support This patch adds support for `Type=notify` to the systemd unit. Supporting `Type=notify` improves both starting as well as reloading of the unit, because systemd will be let known when the action completed. See this quote from `systemd.service(5)`: > Note however that reloading a daemon by sending a signal (as with the > example line above) is usually not a good choice, because this is an > asynchronous operation and hence not suitable to order reloads of > multiple services against each other. It is strongly recommended to > set ExecReload= to a command that not only triggers a configuration > reload of the daemon, but also synchronously waits for it to complete. By making systemd aware of a reload in progress it is able to wait until the reload actually succeeded. This patch introduces both a new `USE_SYSTEMD` build option which controls including the sd-daemon library as well as a `-Ws` runtime option which runs haproxy in master-worker mode with systemd support. When haproxy is running in master-worker mode with systemd support it will send status messages to systemd using `sd_notify(3)` in the following cases: - The master process forked off the worker processes (READY=1) - The master process entered the `mworker_reload()` function (RELOADING=1) - The master process received the SIGUSR1 or SIGTERM signal (STOPPING=1) Change the unit file to specify `Type=notify` and replace master-worker mode (`-W`) with master-worker mode with systemd support (`-Ws`). Future evolutions of this feature could include making use of the `STATUS` feature of `sd_notify()` to send information about the number of active connections to systemd. This would require bidirectional communication between the master and the workers and thus is left for future work.
2017-11-20 14:58:35 +00:00
ifneq ($(USE_SYSTEMD),)
OPTIONS_LDFLAGS += -lsystemd
endif
ifneq ($(USE_PCRE)$(USE_STATIC_PCRE)$(USE_PCRE_JIT),)
ifneq ($(USE_PCRE2)$(USE_STATIC_PCRE2)$(USE_PCRE2_JIT),)
$(error cannot compile both PCRE and PCRE2 support)
endif
# PCREDIR is used to automatically construct the PCRE_INC and PCRE_LIB paths,
# by appending /include and /lib respectively. If your system does not use the
# same sub-directories, simply force these variables instead of PCREDIR. It is
# automatically detected but can be forced if required (for cross-compiling).
# Forcing PCREDIR to an empty string will let the compiler use the default
# locations.
PCRE_CONFIG := pcre-config
PCREDIR := $(shell $(PCRE_CONFIG) --prefix 2>/dev/null || echo /usr/local)
ifneq ($(PCREDIR),)
PCRE_INC := $(PCREDIR)/include
PCRE_LIB := $(PCREDIR)/lib
endif
ifeq ($(USE_STATIC_PCRE),)
# dynamic PCRE
OPTIONS_CFLAGS += -DUSE_PCRE $(if $(PCRE_INC),-I$(PCRE_INC))
OPTIONS_LDFLAGS += $(if $(PCRE_LIB),-L$(PCRE_LIB)) -lpcreposix -lpcre
else
# static PCRE
OPTIONS_CFLAGS += -DUSE_PCRE $(if $(PCRE_INC),-I$(PCRE_INC))
OPTIONS_LDFLAGS += $(if $(PCRE_LIB),-L$(PCRE_LIB)) -Wl,-Bstatic -lpcreposix -lpcre -Wl,-Bdynamic
endif
endif
ifneq ($(USE_PCRE2)$(USE_STATIC_PCRE2)$(USE_PCRE2_JIT),)
PCRE2_CONFIG := pcre2-config
PCRE2DIR := $(shell $(PCRE2_CONFIG) --prefix 2>/dev/null || echo /usr/local)
ifneq ($(PCRE2DIR),)
PCRE2_INC := $(PCRE2DIR)/include
PCRE2_LIB := $(PCRE2DIR)/lib
ifeq ($(PCRE2_WIDTH),)
PCRE2_WIDTH = 8
endif
ifneq ($(PCRE2_WIDTH),8)
ifneq ($(PCRE2_WIDTH),16)
ifneq ($(PCRE2_WIDTH),32)
$(error PCRE2_WIDTH needs to be set to either 8,16 or 32)
endif
endif
endif
PCRE2_LDFLAGS := $(shell $(PCRE2_CONFIG) --libs$(PCRE2_WIDTH) 2>/dev/null || echo -L/usr/local/lib -lpcre2-$(PCRE2_WIDTH))
ifeq ($(PCRE2_LDFLAGS),)
$(error libpcre2-$(PCRE2_WIDTH) not found)
else
ifeq ($(PCRE2_WIDTH),8)
PCRE2_LDFLAGS += -lpcre2-posix
endif
endif
OPTIONS_CFLAGS += -DUSE_PCRE2 -DPCRE2_CODE_UNIT_WIDTH=$(PCRE2_WIDTH)
OPTIONS_CFLAGS += $(if $(PCRE2_INC), -I$(PCRE2_INC))
ifneq ($(USE_STATIC_PCRE2),)
OPTIONS_LDFLAGS += $(if $(PCRE2_LIB),-L$(PCRE2_LIB)) -Wl,-Bstatic -L$(PCRE2_LIB) $(PCRE2_LDFLAGS) -Wl,-Bdynamic
else
OPTIONS_LDFLAGS += $(if $(PCRE2_LIB),-L$(PCRE2_LIB)) -L$(PCRE2_LIB) $(PCRE2_LDFLAGS)
endif
endif
endif
# This one can be changed to look for ebtree files in an external directory
EBTREE_DIR := ebtree
#### Global compile options
VERBOSE_CFLAGS = $(CFLAGS) $(TARGET_CFLAGS) $(SMALL_OPTS) $(DEFINE)
COPTS = -Iinclude -I$(EBTREE_DIR) -Wall -Wextra
ifneq ($(ERR),)
COPTS += -Werror
endif
COPTS += $(CFLAGS) $(TARGET_CFLAGS) $(SMALL_OPTS) $(DEFINE) $(SILENT_DEFINE)
COPTS += $(DEBUG) $(OPTIONS_CFLAGS) $(ADDINC)
ifneq ($(VERSION)$(SUBVERS),)
COPTS += -DCONFIG_HAPROXY_VERSION=\"$(VERSION)$(SUBVERS)\"
endif
ifneq ($(VERDATE),)
COPTS += -DCONFIG_HAPROXY_DATE=\"$(VERDATE)\"
endif
MINOR: add a new function call tracer for debugging purposes This feature relies on GCC's ability to call helpers at function entry/exit points. We define these helpers to quickly dump the minimum info into a trace file that can be converted to a human readable format using a script in the contrib/trace directory. This has only been implemented in the GNU makefile for now on as it is unsure whether it's supported on all OSes. The feature is enabled by building with "TRACE=1". The performance impact is huge, so this feature should only be used when debugging. To limit the loss of performance, fprintf() has been disabled and the output is hand-crafted and emitted using fwrite(), resulting in doubling the performance. Using the TSC instead of gettimeofday() also doubles the performance. Around 1200 conns/s may be achieved on a Pentium-M 1.7 GHz which leads to around 50 MB/s of traces. The entry and exits of all functions will be dumped into a file designated by the HAPROXY_TRACE environment variable, or by default "trace.out". If the trace file name is empty or "/dev/null", then traces are disabled. If opening the trace file fails, then stderr is used. If HAPROXY_TRACE_FAST is used, then the time is taken from the global <now> variable. Last, if HAPROXY_TRACE_TSC is used, then the machine's TSC is used instead of the real time (almost twice as fast). The output format is : <sec.usec> <level> <caller_ptr> <dir> <callee_ptr> or : <tsc> <level> <caller_ptr> <dir> <callee_ptr> where <dir> is '>' when entering a function and '<' when leaving. The awk script in contrib/trace provides a nicer indented output : 6f74989e6f8 ->->-> run_poll_loop > signal_process_queue [src/haproxy.c:1097:0x804bd69] > [include/proto/signal.h:32:0x8049cd0] 6f74989eb00 run_poll_loop < signal_process_queue [src/haproxy.c:1097:0x804bd69] < [include/proto/signal.h:32:0x8049cd0] 6f74989ef44 ->->-> run_poll_loop > wake_expired_tasks [src/haproxy.c:1100:0x804bd72] > [src/task.c:123:0x8055060] 6f74989f3a6 ->->->-> wake_expired_tasks > eb32_lookup_ge [src/task.c:128:0x8055091] > [ebtree/eb32tree.c:138:0x80a8c70] 6f74989f7e9 wake_expired_tasks < eb32_lookup_ge [src/task.c:128:0x8055091] < [ebtree/eb32tree.c:138:0x80a8c70] 6f74989fc0d ->->->-> wake_expired_tasks > eb32_first [src/task.c:134:0x80550d5] > [ebtree/eb32tree.h:55:0x8054ad0] 6f7498a003d ->->->->-> eb32_first > eb_first [ebtree/eb32tree.h:56:0x8054af1] > [ebtree/ebtree.h:520:0x8054a10] 6f7498a0436 ->->->->->-> eb_first > eb_walk_down [ebtree/ebtree.h:521:0x8054a33] > [ebtree/ebtree.h:442:0x80549a0] 6f7498a0843 ->->->->->->-> eb_walk_down > eb_gettag [ebtree/ebtree.h:445:0x80549d6] > [ebtree/ebtree.h:418:0x80548e0] 6f7498a0c2b eb_walk_down < eb_gettag [ebtree/ebtree.h:445:0x80549d6] < [ebtree/ebtree.h:418:0x80548e0] 6f7498a1042 ->->->->->->-> eb_walk_down > eb_untag [ebtree/ebtree.h:447:0x80549e2] > [ebtree/ebtree.h:412:0x80548a0] 6f7498a1498 eb_walk_down < eb_untag [ebtree/ebtree.h:447:0x80549e2] < [ebtree/ebtree.h:412:0x80548a0] 6f7498a18c6 ->->->->->->-> eb_walk_down > eb_root_to_node [ebtree/ebtree.h:448:0x80549e7] > [ebtree/ebtree.h:432:0x8054960] 6f7498a1cd4 eb_walk_down < eb_root_to_node [ebtree/ebtree.h:448:0x80549e7] < [ebtree/ebtree.h:432:0x8054960] 6f7498a20c4 eb_first < eb_walk_down [ebtree/ebtree.h:521:0x8054a33] < [ebtree/ebtree.h:442:0x80549a0] 6f7498a24b4 eb32_first < eb_first [ebtree/eb32tree.h:56:0x8054af1] < [ebtree/ebtree.h:520:0x8054a10] 6f7498a289c wake_expired_tasks < eb32_first [src/task.c:134:0x80550d5] < [ebtree/eb32tree.h:55:0x8054ad0] 6f7498a2c8c run_poll_loop < wake_expired_tasks [src/haproxy.c:1100:0x804bd72] < [src/task.c:123:0x8055060] 6f7498a3095 ->->-> run_poll_loop > process_runnable_tasks [src/haproxy.c:1103:0x804bd7a] > [src/task.c:190:0x8055150] A nice improvement would possibly consist in trying to get the function's arguments in the stack and to dump a few more infor for some well-known functions (eg: the session's status for process_session).
2012-05-25 21:53:16 +00:00
ifneq ($(TRACE),)
# if tracing is enabled, we want it to be as fast as possible
TRACE_COPTS := $(filter-out -O0 -O1 -O2 -pg -finstrument-functions,$(COPTS)) -O3 -fomit-frame-pointer
COPTS += -finstrument-functions
endif
MAJOR: namespace: add Linux network namespace support This patch makes it possible to create binds and servers in separate namespaces. This can be used to proxy between multiple completely independent virtual networks (with possibly overlapping IP addresses) and a non-namespace-aware proxy implementation that supports the proxy protocol (v2). The setup is something like this: net1 on VLAN 1 (namespace 1) -\ net2 on VLAN 2 (namespace 2) -- haproxy ==== proxy (namespace 0) net3 on VLAN 3 (namespace 3) -/ The proxy is configured to make server connections through haproxy and sending the expected source/target addresses to haproxy using the proxy protocol. The network namespace setup on the haproxy node is something like this: = 8< = $ cat setup.sh ip netns add 1 ip link add link eth1 type vlan id 1 ip link set eth1.1 netns 1 ip netns exec 1 ip addr add 192.168.91.2/24 dev eth1.1 ip netns exec 1 ip link set eth1.$id up ... = 8< = = 8< = $ cat haproxy.cfg frontend clients bind 127.0.0.1:50022 namespace 1 transparent default_backend scb backend server mode tcp server server1 192.168.122.4:2222 namespace 2 send-proxy-v2 = 8< = A bind line creates the listener in the specified namespace, and connections originating from that listener also have their network namespace set to that of the listener. A server line either forces the connection to be made in a specified namespace or may use the namespace from the client-side connection if that was set. For more documentation please read the documentation included in the patch itself. Signed-off-by: KOVACS Tamas <ktamas@balabit.com> Signed-off-by: Sarkozi Laszlo <laszlo.sarkozi@balabit.com> Signed-off-by: KOVACS Krisztian <hidden@balabit.com>
2014-11-17 14:11:45 +00:00
ifneq ($(USE_NS),)
OPTIONS_CFLAGS += -DCONFIG_HAP_NS
OPTIONS_OBJS += src/namespace.o
MAJOR: namespace: add Linux network namespace support This patch makes it possible to create binds and servers in separate namespaces. This can be used to proxy between multiple completely independent virtual networks (with possibly overlapping IP addresses) and a non-namespace-aware proxy implementation that supports the proxy protocol (v2). The setup is something like this: net1 on VLAN 1 (namespace 1) -\ net2 on VLAN 2 (namespace 2) -- haproxy ==== proxy (namespace 0) net3 on VLAN 3 (namespace 3) -/ The proxy is configured to make server connections through haproxy and sending the expected source/target addresses to haproxy using the proxy protocol. The network namespace setup on the haproxy node is something like this: = 8< = $ cat setup.sh ip netns add 1 ip link add link eth1 type vlan id 1 ip link set eth1.1 netns 1 ip netns exec 1 ip addr add 192.168.91.2/24 dev eth1.1 ip netns exec 1 ip link set eth1.$id up ... = 8< = = 8< = $ cat haproxy.cfg frontend clients bind 127.0.0.1:50022 namespace 1 transparent default_backend scb backend server mode tcp server server1 192.168.122.4:2222 namespace 2 send-proxy-v2 = 8< = A bind line creates the listener in the specified namespace, and connections originating from that listener also have their network namespace set to that of the listener. A server line either forces the connection to be made in a specified namespace or may use the namespace from the client-side connection if that was set. For more documentation please read the documentation included in the patch itself. Signed-off-by: KOVACS Tamas <ktamas@balabit.com> Signed-off-by: Sarkozi Laszlo <laszlo.sarkozi@balabit.com> Signed-off-by: KOVACS Krisztian <hidden@balabit.com>
2014-11-17 14:11:45 +00:00
endif
#### Global link options
# These options are added at the end of the "ld" command line. Use LDFLAGS to
# add options at the beginning of the "ld" command line if needed.
LDOPTS = $(TARGET_LDFLAGS) $(OPTIONS_LDFLAGS) $(ADDLIB)
ifeq ($V,1)
cmd_CC = $(CC)
cmd_LD = $(LD)
cmd_AR = $(AR)
else
cmd_CC = $(Q)echo " CC $@";$(CC)
cmd_LD = $(Q)echo " LD $@";$(LD)
cmd_AR = $(Q)echo " AR $@";$(AR)
endif
ifeq ($(TARGET),)
all:
@echo
@echo "Due to too many reports of suboptimized setups, building without"
@echo "specifying the target is no longer supported. Please specify the"
@echo "target OS in the TARGET variable, in the following form:"
@echo
@echo " $ make TARGET=xxx"
@echo
@echo "Please choose the target among the following supported list :"
@echo
@echo " linux2628, linux26, linux24, linux24e, linux22, solaris,"
@echo " freebsd, netbsd, osx, openbsd, aix51, aix52, cygwin, haiku,"
@echo " generic, custom"
@echo
@echo "Use \"generic\" if you don't want any optimization, \"custom\" if you"
@echo "want to precisely tweak every option, or choose the target which"
@echo "matches your OS the most in order to gain the maximum performance"
@echo "out of it. Please check the Makefile in case of doubts."
@echo
@exit 1
else
all: haproxy $(EXTRA)
endif
2005-12-17 11:21:26 +00:00
OBJS = src/proto_http.o src/cfgparse-listen.o src/proto_htx.o src/stream.o \
src/mux_h2.o src/stats.o src/flt_spoe.o src/server.o src/checks.o \
src/haproxy.o src/cfgparse.o src/flt_http_comp.o src/http_fetch.o \
src/dns.o src/stick_table.o src/mux_h1.o src/peers.o src/standard.o \
src/proxy.o src/cli.o src/log.o src/backend.o src/pattern.o \
src/sample.o src/stream_interface.o src/proto_tcp.o src/listener.o \
src/h1.o src/cfgparse-global.o src/cache.o src/http_rules.o \
src/http_act.o src/tcp_rules.o src/filters.o src/connection.o \
src/session.o src/acl.o src/vars.o src/raw_sock.o src/map.o \
src/proto_uxst.o src/payload.o src/fd.o src/queue.o src/flt_trace.o \
src/task.o src/lb_chash.o src/frontend.o src/applet.o src/mux_pt.o \
src/signal.o src/ev_select.o src/proto_sockpair.o src/compression.o \
src/http_conv.o src/memory.o src/lb_fwrr.o src/channel.o src/htx.o \
src/uri_auth.o src/regex.o src/chunk.o src/pipe.o src/lb_fas.o \
src/lb_map.o src/lb_fwlc.o src/auth.o src/time.o src/hathreads.o \
src/http_htx.o src/buffer.o src/hpack-tbl.o src/shctx.o src/sha1.o \
src/http.o src/hpack-dec.o src/action.o src/proto_udp.o src/http_acl.o \
src/xxhash.o src/hpack-enc.o src/h2.o src/freq_ctr.o src/lru.o \
src/protocol.o src/arg.o src/hpack-huff.o src/hdr_idx.o src/base64.o \
src/hash.o src/mailers.o src/activity.o src/http_msg.o src/version.o \
src/mworker.o src/mworker-prog.o
EBTREE_OBJS = $(EBTREE_DIR)/ebtree.o $(EBTREE_DIR)/eb32sctree.o \
$(EBTREE_DIR)/eb32tree.o $(EBTREE_DIR)/eb64tree.o \
$(EBTREE_DIR)/ebmbtree.o $(EBTREE_DIR)/ebsttree.o \
$(EBTREE_DIR)/ebimtree.o $(EBTREE_DIR)/ebistree.o
MINOR: add a new function call tracer for debugging purposes This feature relies on GCC's ability to call helpers at function entry/exit points. We define these helpers to quickly dump the minimum info into a trace file that can be converted to a human readable format using a script in the contrib/trace directory. This has only been implemented in the GNU makefile for now on as it is unsure whether it's supported on all OSes. The feature is enabled by building with "TRACE=1". The performance impact is huge, so this feature should only be used when debugging. To limit the loss of performance, fprintf() has been disabled and the output is hand-crafted and emitted using fwrite(), resulting in doubling the performance. Using the TSC instead of gettimeofday() also doubles the performance. Around 1200 conns/s may be achieved on a Pentium-M 1.7 GHz which leads to around 50 MB/s of traces. The entry and exits of all functions will be dumped into a file designated by the HAPROXY_TRACE environment variable, or by default "trace.out". If the trace file name is empty or "/dev/null", then traces are disabled. If opening the trace file fails, then stderr is used. If HAPROXY_TRACE_FAST is used, then the time is taken from the global <now> variable. Last, if HAPROXY_TRACE_TSC is used, then the machine's TSC is used instead of the real time (almost twice as fast). The output format is : <sec.usec> <level> <caller_ptr> <dir> <callee_ptr> or : <tsc> <level> <caller_ptr> <dir> <callee_ptr> where <dir> is '>' when entering a function and '<' when leaving. The awk script in contrib/trace provides a nicer indented output : 6f74989e6f8 ->->-> run_poll_loop > signal_process_queue [src/haproxy.c:1097:0x804bd69] > [include/proto/signal.h:32:0x8049cd0] 6f74989eb00 run_poll_loop < signal_process_queue [src/haproxy.c:1097:0x804bd69] < [include/proto/signal.h:32:0x8049cd0] 6f74989ef44 ->->-> run_poll_loop > wake_expired_tasks [src/haproxy.c:1100:0x804bd72] > [src/task.c:123:0x8055060] 6f74989f3a6 ->->->-> wake_expired_tasks > eb32_lookup_ge [src/task.c:128:0x8055091] > [ebtree/eb32tree.c:138:0x80a8c70] 6f74989f7e9 wake_expired_tasks < eb32_lookup_ge [src/task.c:128:0x8055091] < [ebtree/eb32tree.c:138:0x80a8c70] 6f74989fc0d ->->->-> wake_expired_tasks > eb32_first [src/task.c:134:0x80550d5] > [ebtree/eb32tree.h:55:0x8054ad0] 6f7498a003d ->->->->-> eb32_first > eb_first [ebtree/eb32tree.h:56:0x8054af1] > [ebtree/ebtree.h:520:0x8054a10] 6f7498a0436 ->->->->->-> eb_first > eb_walk_down [ebtree/ebtree.h:521:0x8054a33] > [ebtree/ebtree.h:442:0x80549a0] 6f7498a0843 ->->->->->->-> eb_walk_down > eb_gettag [ebtree/ebtree.h:445:0x80549d6] > [ebtree/ebtree.h:418:0x80548e0] 6f7498a0c2b eb_walk_down < eb_gettag [ebtree/ebtree.h:445:0x80549d6] < [ebtree/ebtree.h:418:0x80548e0] 6f7498a1042 ->->->->->->-> eb_walk_down > eb_untag [ebtree/ebtree.h:447:0x80549e2] > [ebtree/ebtree.h:412:0x80548a0] 6f7498a1498 eb_walk_down < eb_untag [ebtree/ebtree.h:447:0x80549e2] < [ebtree/ebtree.h:412:0x80548a0] 6f7498a18c6 ->->->->->->-> eb_walk_down > eb_root_to_node [ebtree/ebtree.h:448:0x80549e7] > [ebtree/ebtree.h:432:0x8054960] 6f7498a1cd4 eb_walk_down < eb_root_to_node [ebtree/ebtree.h:448:0x80549e7] < [ebtree/ebtree.h:432:0x8054960] 6f7498a20c4 eb_first < eb_walk_down [ebtree/ebtree.h:521:0x8054a33] < [ebtree/ebtree.h:442:0x80549a0] 6f7498a24b4 eb32_first < eb_first [ebtree/eb32tree.h:56:0x8054af1] < [ebtree/ebtree.h:520:0x8054a10] 6f7498a289c wake_expired_tasks < eb32_first [src/task.c:134:0x80550d5] < [ebtree/eb32tree.h:55:0x8054ad0] 6f7498a2c8c run_poll_loop < wake_expired_tasks [src/haproxy.c:1100:0x804bd72] < [src/task.c:123:0x8055060] 6f7498a3095 ->->-> run_poll_loop > process_runnable_tasks [src/haproxy.c:1103:0x804bd7a] > [src/task.c:190:0x8055150] A nice improvement would possibly consist in trying to get the function's arguments in the stack and to dump a few more infor for some well-known functions (eg: the session's status for process_session).
2012-05-25 21:53:16 +00:00
ifneq ($(TRACE),)
OBJS += src/trace.o
endif
ifneq ($(EXTRA_OBJS),)
OBJS += $(EXTRA_OBJS)
endif
# Not used right now
LIB_EBTREE = $(EBTREE_DIR)/libebtree.a
# Used only for forced dependency checking. May be cleared during development.
INCLUDES = $(wildcard include/*/*.h ebtree/*.h)
DEP = $(INCLUDES) .build_opts
help:
$(Q)sed -ne "/^[^#]*$$/q;s/^#\(.*\)/\1/p" Makefile
# Used only to force a rebuild if some build options change
build_opts = $(shell rm -f .build_opts.new; echo \'$(TARGET) $(BUILD_OPTIONS) $(VERBOSE_CFLAGS)\' > .build_opts.new; if cmp -s .build_opts .build_opts.new; then rm -f .build_opts.new; else mv -f .build_opts.new .build_opts; fi)
.build_opts: $(build_opts)
haproxy: $(OPTIONS_OBJS) $(OBJS) $(EBTREE_OBJS)
$(cmd_LD) $(LDFLAGS) -o $@ $^ $(LDOPTS)
2005-12-17 11:21:26 +00:00
$(LIB_EBTREE): $(EBTREE_OBJS)
$(cmd_AR) rv $@ $^
objsize: haproxy
$(Q)objdump -t $^|grep ' g '|grep -F '.text'|awk '{print $$5 FS $$6}'|sort
%.o: %.c $(DEP)
$(cmd_CC) $(COPTS) -c -o $@ $<
2005-12-17 11:21:26 +00:00
# rebuild it every time
.PHONY: src/version.c
src/trace.o: src/trace.c $(DEP)
$(cmd_CC) $(TRACE_COPTS) -c -o $@ $<
MINOR: add a new function call tracer for debugging purposes This feature relies on GCC's ability to call helpers at function entry/exit points. We define these helpers to quickly dump the minimum info into a trace file that can be converted to a human readable format using a script in the contrib/trace directory. This has only been implemented in the GNU makefile for now on as it is unsure whether it's supported on all OSes. The feature is enabled by building with "TRACE=1". The performance impact is huge, so this feature should only be used when debugging. To limit the loss of performance, fprintf() has been disabled and the output is hand-crafted and emitted using fwrite(), resulting in doubling the performance. Using the TSC instead of gettimeofday() also doubles the performance. Around 1200 conns/s may be achieved on a Pentium-M 1.7 GHz which leads to around 50 MB/s of traces. The entry and exits of all functions will be dumped into a file designated by the HAPROXY_TRACE environment variable, or by default "trace.out". If the trace file name is empty or "/dev/null", then traces are disabled. If opening the trace file fails, then stderr is used. If HAPROXY_TRACE_FAST is used, then the time is taken from the global <now> variable. Last, if HAPROXY_TRACE_TSC is used, then the machine's TSC is used instead of the real time (almost twice as fast). The output format is : <sec.usec> <level> <caller_ptr> <dir> <callee_ptr> or : <tsc> <level> <caller_ptr> <dir> <callee_ptr> where <dir> is '>' when entering a function and '<' when leaving. The awk script in contrib/trace provides a nicer indented output : 6f74989e6f8 ->->-> run_poll_loop > signal_process_queue [src/haproxy.c:1097:0x804bd69] > [include/proto/signal.h:32:0x8049cd0] 6f74989eb00 run_poll_loop < signal_process_queue [src/haproxy.c:1097:0x804bd69] < [include/proto/signal.h:32:0x8049cd0] 6f74989ef44 ->->-> run_poll_loop > wake_expired_tasks [src/haproxy.c:1100:0x804bd72] > [src/task.c:123:0x8055060] 6f74989f3a6 ->->->-> wake_expired_tasks > eb32_lookup_ge [src/task.c:128:0x8055091] > [ebtree/eb32tree.c:138:0x80a8c70] 6f74989f7e9 wake_expired_tasks < eb32_lookup_ge [src/task.c:128:0x8055091] < [ebtree/eb32tree.c:138:0x80a8c70] 6f74989fc0d ->->->-> wake_expired_tasks > eb32_first [src/task.c:134:0x80550d5] > [ebtree/eb32tree.h:55:0x8054ad0] 6f7498a003d ->->->->-> eb32_first > eb_first [ebtree/eb32tree.h:56:0x8054af1] > [ebtree/ebtree.h:520:0x8054a10] 6f7498a0436 ->->->->->-> eb_first > eb_walk_down [ebtree/ebtree.h:521:0x8054a33] > [ebtree/ebtree.h:442:0x80549a0] 6f7498a0843 ->->->->->->-> eb_walk_down > eb_gettag [ebtree/ebtree.h:445:0x80549d6] > [ebtree/ebtree.h:418:0x80548e0] 6f7498a0c2b eb_walk_down < eb_gettag [ebtree/ebtree.h:445:0x80549d6] < [ebtree/ebtree.h:418:0x80548e0] 6f7498a1042 ->->->->->->-> eb_walk_down > eb_untag [ebtree/ebtree.h:447:0x80549e2] > [ebtree/ebtree.h:412:0x80548a0] 6f7498a1498 eb_walk_down < eb_untag [ebtree/ebtree.h:447:0x80549e2] < [ebtree/ebtree.h:412:0x80548a0] 6f7498a18c6 ->->->->->->-> eb_walk_down > eb_root_to_node [ebtree/ebtree.h:448:0x80549e7] > [ebtree/ebtree.h:432:0x8054960] 6f7498a1cd4 eb_walk_down < eb_root_to_node [ebtree/ebtree.h:448:0x80549e7] < [ebtree/ebtree.h:432:0x8054960] 6f7498a20c4 eb_first < eb_walk_down [ebtree/ebtree.h:521:0x8054a33] < [ebtree/ebtree.h:442:0x80549a0] 6f7498a24b4 eb32_first < eb_first [ebtree/eb32tree.h:56:0x8054af1] < [ebtree/ebtree.h:520:0x8054a10] 6f7498a289c wake_expired_tasks < eb32_first [src/task.c:134:0x80550d5] < [ebtree/eb32tree.h:55:0x8054ad0] 6f7498a2c8c run_poll_loop < wake_expired_tasks [src/haproxy.c:1100:0x804bd72] < [src/task.c:123:0x8055060] 6f7498a3095 ->->-> run_poll_loop > process_runnable_tasks [src/haproxy.c:1103:0x804bd7a] > [src/task.c:190:0x8055150] A nice improvement would possibly consist in trying to get the function's arguments in the stack and to dump a few more infor for some well-known functions (eg: the session's status for process_session).
2012-05-25 21:53:16 +00:00
src/haproxy.o: src/haproxy.c $(DEP)
$(cmd_CC) $(COPTS) \
-DBUILD_TARGET='"$(strip $(TARGET))"' \
-DBUILD_ARCH='"$(strip $(ARCH))"' \
-DBUILD_CPU='"$(strip $(CPU))"' \
-DBUILD_CC='"$(strip $(CC))"' \
-DBUILD_CFLAGS='"$(strip $(VERBOSE_CFLAGS))"' \
-DBUILD_OPTIONS='"$(strip $(BUILD_OPTIONS))"' \
-DBUILD_FEATURES='"$(strip $(BUILD_FEATURES))"' \
-c -o $@ $<
install-man:
$(Q)install -v -d "$(DESTDIR)$(MANDIR)"/man1
$(Q)install -v -m 644 doc/haproxy.1 "$(DESTDIR)$(MANDIR)"/man1
EXCLUDE_DOCUMENTATION = lgpl gpl coding-style
DOCUMENTATION = $(filter-out $(EXCLUDE_DOCUMENTATION),$(patsubst doc/%.txt,%,$(wildcard doc/*.txt)))
install-doc:
$(Q)install -v -d "$(DESTDIR)$(DOCDIR)"
$(Q)for x in $(DOCUMENTATION); do \
install -v -m 644 doc/$$x.txt "$(DESTDIR)$(DOCDIR)" ; \
done
install-bin:
$(Q)for i in haproxy $(EXTRA); do \
if ! [ -e "$$i" ]; then \
echo "Please run 'make' before 'make install'."; \
exit 1; \
fi; \
done
$(Q)install -v -d "$(DESTDIR)$(SBINDIR)"
$(Q)install -v haproxy $(EXTRA) "$(DESTDIR)$(SBINDIR)"
install: install-bin install-man install-doc
uninstall:
$(Q)rm -f "$(DESTDIR)$(MANDIR)"/man1/haproxy.1
$(Q)for x in $(DOCUMENTATION); do \
rm -f "$(DESTDIR)$(DOCDIR)"/$$x.txt ; \
done
$(Q)-rmdir "$(DESTDIR)$(DOCDIR)"
$(Q)rm -f "$(DESTDIR)$(SBINDIR)"/haproxy
2005-12-17 11:21:26 +00:00
clean:
$(Q)rm -f *.[oas] src/*.[oas] ebtree/*.[oas] haproxy test .build_opts .build_opts.new
$(Q)for dir in . src include/* doc ebtree; do rm -f $$dir/*~ $$dir/*.rej $$dir/core; done
$(Q)rm -f haproxy-$(VERSION).tar.gz haproxy-$(VERSION)$(SUBVERS).tar.gz
$(Q)rm -f haproxy-$(VERSION) haproxy-$(VERSION)$(SUBVERS) nohup.out gmon.out
2009-10-10 20:20:44 +00:00
tags:
$(Q)find src include \( -name '*.c' -o -name '*.h' \) -print0 | \
2009-10-10 20:20:44 +00:00
xargs -0 etags --declarations --members
cscope:
$(Q)find src include -name "*.[ch]" -print | cscope -q -b -i -
tar: clean
$(Q)ln -s . haproxy-$(VERSION)$(SUBVERS)
$(Q)tar --exclude=haproxy-$(VERSION)$(SUBVERS)/.git \
--exclude=haproxy-$(VERSION)$(SUBVERS)/haproxy-$(VERSION)$(SUBVERS) \
--exclude=haproxy-$(VERSION)$(SUBVERS)/haproxy-$(VERSION)$(SUBVERS).tar.gz \
-cf - haproxy-$(VERSION)$(SUBVERS)/* | gzip -c9 >haproxy-$(VERSION)$(SUBVERS).tar.gz
$(Q)echo haproxy-$(VERSION)$(SUBVERS).tar.gz
$(Q)rm -f haproxy-$(VERSION)$(SUBVERS)
git-tar:
$(Q)git archive --format=tar --prefix="haproxy-$(VERSION)$(SUBVERS)/" HEAD | gzip -9 > haproxy-$(VERSION)$(SUBVERS).tar.gz
$(Q)echo haproxy-$(VERSION)$(SUBVERS).tar.gz
version:
@echo "VERSION: $(VERSION)"
@echo "SUBVERS: $(SUBVERS)"
@echo "VERDATE: $(VERDATE)"
# never use this one if you don't know what it is used for.
update-version:
@echo "Ready to update the following versions :"
@echo "VERSION: $(VERSION)"
@echo "SUBVERS: $(SUBVERS)"
@echo "VERDATE: $(VERDATE)"
@echo "Press [ENTER] to continue or Ctrl-C to abort now.";read
echo "$(VERSION)" > VERSION
echo "$(SUBVERS)" > SUBVERS
echo "$(VERDATE)" > VERDATE
# just display the build options
opts:
@echo -n 'Using: '
@echo -n 'TARGET="$(strip $(TARGET))" '
@echo -n 'ARCH="$(strip $(ARCH))" '
@echo -n 'CPU="$(strip $(CPU))" '
@echo -n 'CC="$(strip $(CC))" '
@echo -n 'ARCH_FLAGS="$(strip $(ARCH_FLAGS))" '
@echo -n 'CPU_CFLAGS="$(strip $(CPU_CFLAGS))" '
@echo -n 'DEBUG_CFLAGS="$(strip $(DEBUG_CFLAGS))" '
@echo "$(strip $(BUILD_OPTIONS))"
@echo 'COPTS="$(strip $(COPTS))"'
@echo 'LDFLAGS="$(strip $(LDFLAGS))"'
@echo 'LDOPTS="$(strip $(LDOPTS))"'
@echo 'OPTIONS_OBJS="$(strip $(OPTIONS_OBJS))"'
@echo 'OBJS="$(strip $(OBJS))"'
ifeq (reg-tests, $(firstword $(MAKECMDGOALS)))
REGTEST_ARGS := $(wordlist 2, $(words $(MAKECMDGOALS)), $(MAKECMDGOALS))
$(eval $(REGTEST_ARGS):;@true)
endif
# Target to run the regression testing script files.
reg-tests:
$(Q)$(REG_TEST_SCRIPT) --type "$(REGTESTS_TYPES)" $(REGTEST_ARGS) $(REG_TEST_FILES)
.PHONY: $(REGTEST_ARGS)
reg-tests-help:
@echo
@echo "To launch the reg tests for haproxy, first export to your environment "
@echo "VTEST_PROGRAM variable to point to your vtest program:"
@echo " $$ export VTEST_PROGRAM=/opt/local/bin/vtest"
@echo "or"
@echo " $$ setenv VTEST_PROGRAM /opt/local/bin/vtest"
@echo
@echo "The same thing may be done to set your haproxy program with HAPROXY_PROGRAM "
@echo "but with ./haproxy as default value."
@echo
@echo "To run all the tests:"
@echo " $$ make reg-tests"
@echo
@echo "You can also set the programs to be used on the command line:"
@echo " $$ VTEST_PROGRAM=<...> HAPROXY_PROGRAM=<...> make reg-tests"
@echo
@echo "To run tests with specific types:"
@echo " $$ REGTESTS_TYPES=slow,default make reg-tests"
@echo
@echo "with 'any' as default value for REGTESTS_TYPES variable."
@echo
@echo "About the reg test types:"
@echo " any : all the tests without distinction (this is the default"
@echo " value of REGTESTS_TYPES."
@echo " default : dedicated to pure haproxy compliance tests."
@echo " slow : scripts which take non negligible time to run."
@echo " bug : scripts in relation with bugs they help to reproduce."
@echo " broken : scripts triggering known broken behaviors for which"
@echo " there is still no fix."
@echo " experimental: for scripts which are experimental, typically used to"
@echo " develop new scripts."
@echo
@echo "Note that 'reg-tests' target run '"$(REG_TEST_SCRIPT)"' script"
@echo "(see --help option of this script for more information)."
.PHONY: reg-tests reg-tests-help