Sat, 03 Jan 2015 20:18:00 +0100
Conditionally enable double key logic according to:
private browsing mode or privacy.thirdparty.isolate preference and
implement in GetCookieStringCommon and FindCookie where it counts...
With some reservations of how to convince FindCookie users to test
condition and pass a nullptr when disabling double key logic.
michael@0 | 1 | Installation Instructions |
michael@0 | 2 | ************************* |
michael@0 | 3 | |
michael@0 | 4 | Copyright (C) 1994, 1995, 1996, 1999, 2000, 2001, 2002, 2004, 2005, |
michael@0 | 5 | 2006, 2007 Free Software Foundation, Inc. |
michael@0 | 6 | |
michael@0 | 7 | This file is free documentation; the Free Software Foundation gives |
michael@0 | 8 | unlimited permission to copy, distribute and modify it. |
michael@0 | 9 | |
michael@0 | 10 | Glog-Specific Install Notes |
michael@0 | 11 | ================================ |
michael@0 | 12 | |
michael@0 | 13 | *** NOTE FOR 64-BIT LINUX SYSTEMS |
michael@0 | 14 | |
michael@0 | 15 | The glibc built-in stack-unwinder on 64-bit systems has some problems |
michael@0 | 16 | with the glog libraries. (In particular, if you are using |
michael@0 | 17 | InstallFailureSignalHandler(), the signal may be raised in the middle |
michael@0 | 18 | of malloc, holding some malloc-related locks when they invoke the |
michael@0 | 19 | stack unwinder. The built-in stack unwinder may call malloc |
michael@0 | 20 | recursively, which may require the thread to acquire a lock it already |
michael@0 | 21 | holds: deadlock.) |
michael@0 | 22 | |
michael@0 | 23 | For that reason, if you use a 64-bit system and you need |
michael@0 | 24 | InstallFailureSignalHandler(), we strongly recommend you install |
michael@0 | 25 | libunwind before trying to configure or install google glog. |
michael@0 | 26 | libunwind can be found at |
michael@0 | 27 | |
michael@0 | 28 | http://download.savannah.nongnu.org/releases/libunwind/libunwind-snap-070410.tar.gz |
michael@0 | 29 | |
michael@0 | 30 | Even if you already have libunwind installed, you will probably still |
michael@0 | 31 | need to install from the snapshot to get the latest version. |
michael@0 | 32 | |
michael@0 | 33 | CAUTION: if you install libunwind from the URL above, be aware that |
michael@0 | 34 | you may have trouble if you try to statically link your binary with |
michael@0 | 35 | glog: that is, if you link with 'gcc -static -lgcc_eh ...'. This |
michael@0 | 36 | is because both libunwind and libgcc implement the same C++ exception |
michael@0 | 37 | handling APIs, but they implement them differently on some platforms. |
michael@0 | 38 | This is not likely to be a problem on ia64, but may be on x86-64. |
michael@0 | 39 | |
michael@0 | 40 | Also, if you link binaries statically, make sure that you add |
michael@0 | 41 | -Wl,--eh-frame-hdr to your linker options. This is required so that |
michael@0 | 42 | libunwind can find the information generated by the compiler required |
michael@0 | 43 | for stack unwinding. |
michael@0 | 44 | |
michael@0 | 45 | Using -static is rare, though, so unless you know this will affect you |
michael@0 | 46 | it probably won't. |
michael@0 | 47 | |
michael@0 | 48 | If you cannot or do not wish to install libunwind, you can still try |
michael@0 | 49 | to use two kinds of stack-unwinder: 1. glibc built-in stack-unwinder |
michael@0 | 50 | and 2. frame pointer based stack-unwinder. |
michael@0 | 51 | |
michael@0 | 52 | 1. As we already mentioned, glibc's unwinder has a deadlock issue. |
michael@0 | 53 | However, if you don't use InstallFailureSignalHandler() or you don't |
michael@0 | 54 | worry about the rare possibilities of deadlocks, you can use this |
michael@0 | 55 | stack-unwinder. If you specify no options and libunwind isn't |
michael@0 | 56 | detected on your system, the configure script chooses this unwinder by |
michael@0 | 57 | default. |
michael@0 | 58 | |
michael@0 | 59 | 2. The frame pointer based stack unwinder requires that your |
michael@0 | 60 | application, the glog library, and system libraries like libc, all be |
michael@0 | 61 | compiled with a frame pointer. This is *not* the default for x86-64. |
michael@0 | 62 | |
michael@0 | 63 | If you are on x86-64 system, know that you have a set of system |
michael@0 | 64 | libraries with frame-pointers enabled, and compile all your |
michael@0 | 65 | applications with -fno-omit-frame-pointer, then you can enable the |
michael@0 | 66 | frame pointer based stack unwinder by passing the |
michael@0 | 67 | --enable-frame-pointers flag to configure. |
michael@0 | 68 | |
michael@0 | 69 | |
michael@0 | 70 | Basic Installation |
michael@0 | 71 | ================== |
michael@0 | 72 | |
michael@0 | 73 | Briefly, the shell commands `./configure; make; make install' should |
michael@0 | 74 | configure, build, and install this package. The following |
michael@0 | 75 | more-detailed instructions are generic; see the `README' file for |
michael@0 | 76 | instructions specific to this package. |
michael@0 | 77 | |
michael@0 | 78 | The `configure' shell script attempts to guess correct values for |
michael@0 | 79 | various system-dependent variables used during compilation. It uses |
michael@0 | 80 | those values to create a `Makefile' in each directory of the package. |
michael@0 | 81 | It may also create one or more `.h' files containing system-dependent |
michael@0 | 82 | definitions. Finally, it creates a shell script `config.status' that |
michael@0 | 83 | you can run in the future to recreate the current configuration, and a |
michael@0 | 84 | file `config.log' containing compiler output (useful mainly for |
michael@0 | 85 | debugging `configure'). |
michael@0 | 86 | |
michael@0 | 87 | It can also use an optional file (typically called `config.cache' |
michael@0 | 88 | and enabled with `--cache-file=config.cache' or simply `-C') that saves |
michael@0 | 89 | the results of its tests to speed up reconfiguring. Caching is |
michael@0 | 90 | disabled by default to prevent problems with accidental use of stale |
michael@0 | 91 | cache files. |
michael@0 | 92 | |
michael@0 | 93 | If you need to do unusual things to compile the package, please try |
michael@0 | 94 | to figure out how `configure' could check whether to do them, and mail |
michael@0 | 95 | diffs or instructions to the address given in the `README' so they can |
michael@0 | 96 | be considered for the next release. If you are using the cache, and at |
michael@0 | 97 | some point `config.cache' contains results you don't want to keep, you |
michael@0 | 98 | may remove or edit it. |
michael@0 | 99 | |
michael@0 | 100 | The file `configure.ac' (or `configure.in') is used to create |
michael@0 | 101 | `configure' by a program called `autoconf'. You need `configure.ac' if |
michael@0 | 102 | you want to change it or regenerate `configure' using a newer version |
michael@0 | 103 | of `autoconf'. |
michael@0 | 104 | |
michael@0 | 105 | The simplest way to compile this package is: |
michael@0 | 106 | |
michael@0 | 107 | 1. `cd' to the directory containing the package's source code and type |
michael@0 | 108 | `./configure' to configure the package for your system. |
michael@0 | 109 | |
michael@0 | 110 | Running `configure' might take a while. While running, it prints |
michael@0 | 111 | some messages telling which features it is checking for. |
michael@0 | 112 | |
michael@0 | 113 | 2. Type `make' to compile the package. |
michael@0 | 114 | |
michael@0 | 115 | 3. Optionally, type `make check' to run any self-tests that come with |
michael@0 | 116 | the package. |
michael@0 | 117 | |
michael@0 | 118 | 4. Type `make install' to install the programs and any data files and |
michael@0 | 119 | documentation. |
michael@0 | 120 | |
michael@0 | 121 | 5. You can remove the program binaries and object files from the |
michael@0 | 122 | source code directory by typing `make clean'. To also remove the |
michael@0 | 123 | files that `configure' created (so you can compile the package for |
michael@0 | 124 | a different kind of computer), type `make distclean'. There is |
michael@0 | 125 | also a `make maintainer-clean' target, but that is intended mainly |
michael@0 | 126 | for the package's developers. If you use it, you may have to get |
michael@0 | 127 | all sorts of other programs in order to regenerate files that came |
michael@0 | 128 | with the distribution. |
michael@0 | 129 | |
michael@0 | 130 | 6. Often, you can also type `make uninstall' to remove the installed |
michael@0 | 131 | files again. |
michael@0 | 132 | |
michael@0 | 133 | Compilers and Options |
michael@0 | 134 | ===================== |
michael@0 | 135 | |
michael@0 | 136 | Some systems require unusual options for compilation or linking that the |
michael@0 | 137 | `configure' script does not know about. Run `./configure --help' for |
michael@0 | 138 | details on some of the pertinent environment variables. |
michael@0 | 139 | |
michael@0 | 140 | You can give `configure' initial values for configuration parameters |
michael@0 | 141 | by setting variables in the command line or in the environment. Here |
michael@0 | 142 | is an example: |
michael@0 | 143 | |
michael@0 | 144 | ./configure CC=c99 CFLAGS=-g LIBS=-lposix |
michael@0 | 145 | |
michael@0 | 146 | *Note Defining Variables::, for more details. |
michael@0 | 147 | |
michael@0 | 148 | Compiling For Multiple Architectures |
michael@0 | 149 | ==================================== |
michael@0 | 150 | |
michael@0 | 151 | You can compile the package for more than one kind of computer at the |
michael@0 | 152 | same time, by placing the object files for each architecture in their |
michael@0 | 153 | own directory. To do this, you can use GNU `make'. `cd' to the |
michael@0 | 154 | directory where you want the object files and executables to go and run |
michael@0 | 155 | the `configure' script. `configure' automatically checks for the |
michael@0 | 156 | source code in the directory that `configure' is in and in `..'. |
michael@0 | 157 | |
michael@0 | 158 | With a non-GNU `make', it is safer to compile the package for one |
michael@0 | 159 | architecture at a time in the source code directory. After you have |
michael@0 | 160 | installed the package for one architecture, use `make distclean' before |
michael@0 | 161 | reconfiguring for another architecture. |
michael@0 | 162 | |
michael@0 | 163 | Installation Names |
michael@0 | 164 | ================== |
michael@0 | 165 | |
michael@0 | 166 | By default, `make install' installs the package's commands under |
michael@0 | 167 | `/usr/local/bin', include files under `/usr/local/include', etc. You |
michael@0 | 168 | can specify an installation prefix other than `/usr/local' by giving |
michael@0 | 169 | `configure' the option `--prefix=PREFIX'. |
michael@0 | 170 | |
michael@0 | 171 | You can specify separate installation prefixes for |
michael@0 | 172 | architecture-specific files and architecture-independent files. If you |
michael@0 | 173 | pass the option `--exec-prefix=PREFIX' to `configure', the package uses |
michael@0 | 174 | PREFIX as the prefix for installing programs and libraries. |
michael@0 | 175 | Documentation and other data files still use the regular prefix. |
michael@0 | 176 | |
michael@0 | 177 | In addition, if you use an unusual directory layout you can give |
michael@0 | 178 | options like `--bindir=DIR' to specify different values for particular |
michael@0 | 179 | kinds of files. Run `configure --help' for a list of the directories |
michael@0 | 180 | you can set and what kinds of files go in them. |
michael@0 | 181 | |
michael@0 | 182 | If the package supports it, you can cause programs to be installed |
michael@0 | 183 | with an extra prefix or suffix on their names by giving `configure' the |
michael@0 | 184 | option `--program-prefix=PREFIX' or `--program-suffix=SUFFIX'. |
michael@0 | 185 | |
michael@0 | 186 | Optional Features |
michael@0 | 187 | ================= |
michael@0 | 188 | |
michael@0 | 189 | Some packages pay attention to `--enable-FEATURE' options to |
michael@0 | 190 | `configure', where FEATURE indicates an optional part of the package. |
michael@0 | 191 | They may also pay attention to `--with-PACKAGE' options, where PACKAGE |
michael@0 | 192 | is something like `gnu-as' or `x' (for the X Window System). The |
michael@0 | 193 | `README' should mention any `--enable-' and `--with-' options that the |
michael@0 | 194 | package recognizes. |
michael@0 | 195 | |
michael@0 | 196 | For packages that use the X Window System, `configure' can usually |
michael@0 | 197 | find the X include and library files automatically, but if it doesn't, |
michael@0 | 198 | you can use the `configure' options `--x-includes=DIR' and |
michael@0 | 199 | `--x-libraries=DIR' to specify their locations. |
michael@0 | 200 | |
michael@0 | 201 | Specifying the System Type |
michael@0 | 202 | ========================== |
michael@0 | 203 | |
michael@0 | 204 | There may be some features `configure' cannot figure out automatically, |
michael@0 | 205 | but needs to determine by the type of machine the package will run on. |
michael@0 | 206 | Usually, assuming the package is built to be run on the _same_ |
michael@0 | 207 | architectures, `configure' can figure that out, but if it prints a |
michael@0 | 208 | message saying it cannot guess the machine type, give it the |
michael@0 | 209 | `--build=TYPE' option. TYPE can either be a short name for the system |
michael@0 | 210 | type, such as `sun4', or a canonical name which has the form: |
michael@0 | 211 | |
michael@0 | 212 | CPU-COMPANY-SYSTEM |
michael@0 | 213 | |
michael@0 | 214 | where SYSTEM can have one of these forms: |
michael@0 | 215 | |
michael@0 | 216 | OS KERNEL-OS |
michael@0 | 217 | |
michael@0 | 218 | See the file `config.sub' for the possible values of each field. If |
michael@0 | 219 | `config.sub' isn't included in this package, then this package doesn't |
michael@0 | 220 | need to know the machine type. |
michael@0 | 221 | |
michael@0 | 222 | If you are _building_ compiler tools for cross-compiling, you should |
michael@0 | 223 | use the option `--target=TYPE' to select the type of system they will |
michael@0 | 224 | produce code for. |
michael@0 | 225 | |
michael@0 | 226 | If you want to _use_ a cross compiler, that generates code for a |
michael@0 | 227 | platform different from the build platform, you should specify the |
michael@0 | 228 | "host" platform (i.e., that on which the generated programs will |
michael@0 | 229 | eventually be run) with `--host=TYPE'. |
michael@0 | 230 | |
michael@0 | 231 | Sharing Defaults |
michael@0 | 232 | ================ |
michael@0 | 233 | |
michael@0 | 234 | If you want to set default values for `configure' scripts to share, you |
michael@0 | 235 | can create a site shell script called `config.site' that gives default |
michael@0 | 236 | values for variables like `CC', `cache_file', and `prefix'. |
michael@0 | 237 | `configure' looks for `PREFIX/share/config.site' if it exists, then |
michael@0 | 238 | `PREFIX/etc/config.site' if it exists. Or, you can set the |
michael@0 | 239 | `CONFIG_SITE' environment variable to the location of the site script. |
michael@0 | 240 | A warning: not all `configure' scripts look for a site script. |
michael@0 | 241 | |
michael@0 | 242 | Defining Variables |
michael@0 | 243 | ================== |
michael@0 | 244 | |
michael@0 | 245 | Variables not defined in a site shell script can be set in the |
michael@0 | 246 | environment passed to `configure'. However, some packages may run |
michael@0 | 247 | configure again during the build, and the customized values of these |
michael@0 | 248 | variables may be lost. In order to avoid this problem, you should set |
michael@0 | 249 | them in the `configure' command line, using `VAR=value'. For example: |
michael@0 | 250 | |
michael@0 | 251 | ./configure CC=/usr/local2/bin/gcc |
michael@0 | 252 | |
michael@0 | 253 | causes the specified `gcc' to be used as the C compiler (unless it is |
michael@0 | 254 | overridden in the site shell script). |
michael@0 | 255 | |
michael@0 | 256 | Unfortunately, this technique does not work for `CONFIG_SHELL' due to |
michael@0 | 257 | an Autoconf bug. Until the bug is fixed you can use this workaround: |
michael@0 | 258 | |
michael@0 | 259 | CONFIG_SHELL=/bin/bash /bin/bash ./configure CONFIG_SHELL=/bin/bash |
michael@0 | 260 | |
michael@0 | 261 | `configure' Invocation |
michael@0 | 262 | ====================== |
michael@0 | 263 | |
michael@0 | 264 | `configure' recognizes the following options to control how it operates. |
michael@0 | 265 | |
michael@0 | 266 | `--help' |
michael@0 | 267 | `-h' |
michael@0 | 268 | Print a summary of the options to `configure', and exit. |
michael@0 | 269 | |
michael@0 | 270 | `--version' |
michael@0 | 271 | `-V' |
michael@0 | 272 | Print the version of Autoconf used to generate the `configure' |
michael@0 | 273 | script, and exit. |
michael@0 | 274 | |
michael@0 | 275 | `--cache-file=FILE' |
michael@0 | 276 | Enable the cache: use and save the results of the tests in FILE, |
michael@0 | 277 | traditionally `config.cache'. FILE defaults to `/dev/null' to |
michael@0 | 278 | disable caching. |
michael@0 | 279 | |
michael@0 | 280 | `--config-cache' |
michael@0 | 281 | `-C' |
michael@0 | 282 | Alias for `--cache-file=config.cache'. |
michael@0 | 283 | |
michael@0 | 284 | `--quiet' |
michael@0 | 285 | `--silent' |
michael@0 | 286 | `-q' |
michael@0 | 287 | Do not print messages saying which checks are being made. To |
michael@0 | 288 | suppress all normal output, redirect it to `/dev/null' (any error |
michael@0 | 289 | messages will still be shown). |
michael@0 | 290 | |
michael@0 | 291 | `--srcdir=DIR' |
michael@0 | 292 | Look for the package's source code in directory DIR. Usually |
michael@0 | 293 | `configure' can determine that directory automatically. |
michael@0 | 294 | |
michael@0 | 295 | `configure' also accepts some other, not widely useful, options. Run |
michael@0 | 296 | `configure --help' for more details. |
michael@0 | 297 |