Home | History | Annotate | Line # | Download | only in dist
NEWS revision 1.1.1.12
      1 Major changes between releases                  Automated Testing Framework
      2 ===========================================================================
      3 
      4 
      5 Changes in version 0.17
      6 ***********************
      7 
      8 Experimental version released on February 14th, 2013.
      9 
     10 * Added the atf_utils_cat_file, atf_utils_compare_file,
     11   atf_utils_copy_file, atf_utils_create_file, atf_utils_file_exists,
     12   atf_utils_fork, atf_utils_grep_file, atf_utils_grep_string,
     13   atf_utils_readline, atf_utils_redirect and atf_utils_wait utility
     14   functions to atf-c-api.  Documented the already-public
     15   atf_utils_free_charpp function.
     16 
     17 * Added the cat_file, compare_file, copy_file, create_file, file_exists,
     18   fork, grep_collection, grep_file, grep_string, redirect and wait
     19   functions to the atf::utils namespace of atf-c++-api.  These are
     20   wrappers around the same functions added to the atf-c-api library.
     21 
     22 * Added the ATF_CHECK_MATCH, ATF_CHECK_MATCH_MSG, ATF_REQUIRE_MATCH and
     23   ATF_REQUIRE_MATCH_MSG macros to atf-c to simplify the validation of a
     24   string against a regular expression.
     25 
     26 * Miscellaneous fixes for manpage typos and compilation problems with
     27   clang.
     28 
     29 * Added caching of the results of those configure tests that rely on
     30   executing a test program.  This should help crossbuild systems by
     31   providing a mechanism to pre-specify what the results should be.
     32 
     33 * PR bin/45690: Make atf-report convert any non-printable characters to
     34   a plain-text representation (matching their corresponding hexadecimal
     35   entities) in XML output files.  This is to prevent the output of test
     36   cases from breaking xsltproc later.
     37 
     38 
     39 Changes in version 0.16
     40 ***********************
     41 
     42 Experimental version released on July 10th, 2012.
     43 
     44 * Added a --enable-tools flag to configure to request the build of the
     45   deprecated ATF tools, whose build is now disabled by default.  In order
     46   to continue running tests, you should migrate to Kyua instead of enabling
     47   the build of the deprecated tools.  The kyua-atf-compat package provides
     48   transitional compatibility versions of atf-run and atf-report built on
     49   top of Kyua.
     50 
     51 * Tweaked the ATF_TEST_CASE macro of atf-c++ so that the compiler can
     52   detect defined but unused test cases.
     53 
     54 * PR bin/45859: Fixed some XSLT bugs that resulted in the tc-time and
     55   tp-time XML tags leaking into the generated HTML file.  Also improved
     56   the CSS file slightly to correct alignment and color issues with the
     57   timestamps column.
     58 
     59 * Optimized atf-c++/macros.hpp so that GNU G++ consumes less memory during
     60   compilation with GNU G++.
     61 
     62 * Flipped the default to building shared libraries for atf-c and atf-c++,
     63   and started versioning them.  As a side-effect, this removes the
     64   --enable-unstable-shared flag from configure that appears to not work any
     65   more (under NetBSD).  Additionally, some distributions require the use of
     66   shared libraries for proper dependency tracking (e.g. Fedora), so it is
     67   better if we do the right versioning upstream.
     68 
     69 * Project hosting moved from an adhoc solution (custom web site and
     70   Monotone repository) to Google Code (standard wiki and Git).  ATF now
     71   lives in a subcomponent of the Kyua project.
     72 
     73 
     74 Changes in version 0.15
     75 ***********************
     76 
     77 Experimental version released on January 16th, 2012.
     78 
     79 * Respect stdin in atf-check.  The previous release silenced stdin for any
     80   processes spawned by atf, not only test programs, which caused breakage
     81   in tests that pipe data through atf-check.
     82 
     83 * Performance improvements to atf-sh.
     84 
     85 * Enabled detection of unused parameters and variables in the code and
     86   fixed all warnings.
     87 
     88 * Changed the behavior of "developer mode".  Compiler warnings are now
     89   enabled unconditionally regardless of whether we are in developer mode or
     90   not; developer mode is now only used to perform strict warning checks and
     91   to enable assertions.  Additionally, developer mode is now only
     92   automatically enabled when building from the repository, not for formal
     93   releases.
     94 
     95 * Added new Autoconf M4 macros (ATF_ARG_WITH, ATF_CHECK_C and
     96   ATF_CHECK_CXX) to provide a consistent way of defining a --with-arg flag
     97   in configure scripts and detecting the presence of any of the ATF
     98   bindings.  Note that ATF_CHECK_SH was already introduced in 0.14, but it
     99   has now been modified to also honor --with-atf if instantiated.
    100 
    101 * Added timing support to atf-run / atf-report.
    102 
    103 * Added support for a 'require.memory' property, to specify the minimum
    104   amount of physical memory needed by the test case to yield valid results.
    105 
    106 * PR bin/45690: Force an ISO-8859-1 encoding in the XML files generated by
    107   atf-report so that invalid data in the output of test cases does not
    108   mangle our report.
    109 
    110 
    111 Changes in version 0.14
    112 ***********************
    113 
    114 Experimental version released on June 14th, 2011.
    115 
    116 * Added a pkg-config file for atf-sh and an aclocal file to ease the
    117   detection of atf-sh from autoconf scripts.
    118 
    119 * Made the default test case body defined by atf_sh fail.  This is to
    120   ensure that test cases are properly defined in test programs and helps
    121   in catching typos in the names of the body functions.
    122 
    123 * PR bin/44882: Made atf-run connect the stdin of test cases to /dev/zero.
    124   This provides more consistent results with "normal" execution (in
    125   particular, when tests are executed detached from a terminal).
    126 
    127 * Made atf-run hardcode TZ=UTC for test cases.  It used to undefine TZ, but
    128   that does not take into account that libc determines the current timezone
    129   from a configuration file.
    130 
    131 * All test programs will now print a warning when they are not run through
    132   atf-run(1) stating that this is unsupported and may deliver incorrect
    133   results.
    134 
    135 * Added support for the 'require.files' test-case property.  This allows
    136   test cases to specify installed files that must be present for the test
    137   case to run.
    138 
    139 
    140 Changes in version 0.13
    141 ***********************
    142 
    143 Experimental version released on March 31st, 2011.
    144 
    145 This is the first release after the creation of the Kyua project, a more
    146 modular and reliable replacement for ATF.  From now on, ATF will change to
    147 accomodate the transition to this new codebase, but ATF will still continue
    148 to see development in the short/medium term.  Check out the project page at
    149 http://code.google.com/p/kyua/ for more details.
    150 
    151 The changes in this release are:
    152 
    153 * Added support to run the tests with the Kyua runtime engine (kyua-cli), a
    154   new package that aims to replace atf-run and atf-report.  The ATF tests
    155   can be run with the new system by issuing a 'make installcheck-kyua' from
    156   the top-level directory of the project (assuming the 'kyua' binary is
    157   available during the configuration stage of ATF).
    158 
    159 * atf-run and atf-report are now in maintenance mode (but *not* deprecated
    160   yet!).  Kyua already implements a new, much more reliable runtime engine
    161   that provides similar features to these tools.  That said, it is not
    162   complete yet so all development efforts should go towards it.
    163 
    164 * If GDB is installed, atf-run dumps the stack trace of crashing test
    165   programs in an attempt to aid debugging.  Contributed by Antti Kantee.
    166 
    167 * Reverted default timeout change in previous release and reset its value
    168   to 5 minutes.  This was causing several issues, specially when running
    169   the existing NetBSD test suite in qemu.
    170 
    171 * Fixed the 'match' output checker in atf-check to properly validate the
    172   last line of a file even if it does not have a newline.
    173 
    174 * Added the ATF_REQUIRE_IN and ATF_REQUIRE_NOT_IN macros to atf-c++ to
    175   check for the presence (or lack thereof) of an element in a collection.
    176 
    177 * PR bin/44176: Fixed a race condition in atf-run that would crash atf-run
    178   when the cleanup of a test case triggered asynchronous modifications to
    179   its work directory (e.g. killing a daemon process that cleans up a pid
    180   file in the work directory).
    181 
    182 * PR bin/44301: Fixed the sample XSLT file to report bogus test programs
    183   instead of just listing them as having 0 test cases.
    184 
    185 
    186 Changes in version 0.12
    187 ***********************
    188 
    189 Experimental version released on November 7th, 2010.
    190 
    191 * Added the ATF_REQUIRE_THROW_RE to atf-c++, which is the same as
    192   ATF_REQUIRE_THROW but allows checking for the validity of the exception's
    193   error message by means of a regular expression.
    194 
    195 * Added the ATF_REQUIRE_MATCH to atf-c++, which allows checking for a
    196   regular expression match in a string.
    197 
    198 * Changed the default timeout for test cases from 5 minutes to 30 seconds.
    199   30 seconds is long enough for virtually all tests to complete, and 5
    200   minutes is a way too long pause in a test suite where a single test case
    201   stalls.
    202 
    203 * Deprecated the use.fs property.  While this seemed like a good idea in
    204   the first place to impose more control on what test cases can do, it
    205   turns out to be bad.  First, use.fs=false prevents bogus test cases
    206   from dumping core so after-the-fact debugging is harder.  Second,
    207   supporting use.fs adds a lot of unnecessary complexity.  atf-run will
    208   now ignore any value provided to use.fs and will allow test cases to
    209   freely access the file system if they wish to.
    210 
    211 * Added the atf_tc_get_config_var_as_{bool,long}{,_wd} functions to the atf-c
    212   library.  The 'text' module became private in 0.11 but was being used
    213   externally to simplify the parsing of configuration variables.
    214 
    215 * Made atf-run recognize the 'unprivileged-user' configuration variable
    216   and automatically drop root privileges when a test case sets
    217   require.user=unprivileged.  Note that this is, by no means, done for
    218   security purposes; this is just for user convenience; tests should, in
    219   general, not be blindly run as root in the first place.
    220 
    221 
    222 Changes in version 0.11
    223 ***********************
    224 
    225 Experimental version released on October 20th, 2010.
    226 
    227 * The ATF_CHECK* macros in atf-c++ were renamed to ATF_REQUIRE* to match
    228   their counterparts in atf-c.
    229 
    230 * Clearly separated the modules in atf-c that are supposed to be public
    231   from those that are implementation details.  The header files for the
    232   internal modules are not installed any more.
    233 
    234 * Made the atf-check tool private.  It is only required by atf-sh and being
    235   public has the danger of causing confusion.  Also, making it private
    236   simplifies the public API of atf.
    237 
    238 * Changed atf-sh to enable per-command error checking (set -e) by default.
    239   This catches many cases in which a test case is broken but it is not
    240   reported as such because execution continues.
    241 
    242 * Fixed the XSTL and CSS stylesheets to support expected failures.
    243 
    244 
    245 Changes in version 0.10
    246 ***********************
    247 
    248 Experimental version released on July 2nd, 2010.
    249 
    250 Miscellaneous features
    251 
    252 * Added expected failures support to test cases and atf-run.  These
    253   include, for example, expected clean exits, expected reception of fatal
    254   signals, expected timeouts and expected errors in condition checks.
    255   These statuses can be used to denote test cases that are known to fail
    256   due to a bug in the code they are testing.  atf-report reports these
    257   tests separately but they do not count towards the failed test cases
    258   amount.
    259 
    260 * Added the ATF_CHECK_ERRNO and ATF_REQUIRE_ERRNO to the C library to
    261   allow easy checking of call failures that update errno.
    262 
    263 * Added the has.cleanup meta-data property to test caes that specifies
    264   whether the test case has a cleanup routine or not; its value is
    265   automatically set.  This property is read by atf-run to know if it has to
    266   run the cleanup routine; skipping this run for every test case
    267   significantly speeds up the run time of test suites.
    268 
    269 * Reversed the order of the ATF_CHECK_THROW macro in the C++ binding to
    270   take the expected exception as the first argument and the statement to
    271   execute as the second argument.
    272 
    273 Changes in atf-check
    274 
    275 * Changed atf-check to support negating the status and output checks by
    276   prefixing them with not- and added support to specify multiple checkers
    277   for stdout and stderr, not only one.
    278 
    279 * Added the match output checker to atf-check to look for regular
    280   expressions in the stdout and stderr of commands.
    281 
    282 * Modified the exit checks in atf-check to support checking for the
    283   reception of signals.
    284 
    285 Code simplifications and cleanups
    286 
    287 * Removed usage messages from test programs to simplify the
    288   implementation of every binding by a significant amount.  They just now
    289   refer the user to the appropriate manual page and do not attempt to wrap
    290   lines on terminal boundaries.  Test programs are not supposed to be run
    291   by users directly so this minor interface regression is not important.
    292 
    293 * Removed the atf-format internal utility, which is unused after the
    294   change documented above.
    295 
    296 * Removed the atf-cleanup internal utility.  It has been unused since the
    297   test case isolation was moved to atf-run in 0.8
    298 
    299 * Splitted the Makefile.am into smaller files for easier maintenance and
    300   dropped the use of M4.  Only affects users building from the repository
    301   sources.
    302 
    303 * Intermixed tests with the source files in the source tree to provide
    304   them more visibility and easier access.  The tests directory is gone from
    305   the source tree and tests are now suffixed by _test, not prefixed by t_.
    306 
    307 * Simplifications to the atf-c library: removed the io, tcr and ui
    308   modules as they had become unnecessary after all simplifications
    309   introduced since the 0.8 release.
    310 
    311 * Removed the application/X-atf-tcr format introduced in 0.8 release.
    312   Tests now print a much simplified format that is easy to parse and nicer
    313   to read by end users.  As a side effect, the default for test cases is
    314   now to print their results to stdout unless otherwise stated by providing
    315   the -r flag.
    316 
    317 * Removed XML distribution documents and replaced them with plain-text
    318   documents.  They provided little value and introduced a lot of complexity
    319   to the build system.
    320 
    321 * Simplified the output of atf-version by not attempting to print a
    322   revision number when building form a distfile.  Makes the build system
    323   easier to maintain.
    324 
    325 
    326 Changes in version 0.9
    327 **********************
    328 
    329 Experimental version released on June 3rd, 2010.
    330 
    331 * Added atf-sh, an interpreter to process test programs written using
    332   the shell API.  This is not really a shell interpreter by itself though:
    333   it is just a wrapper around the system shell that eases the loading of
    334   the necessary ATF libraries.
    335 
    336 * Removed atf-compile in favour of atf-sh.
    337 
    338 * Added the use.fs metadata property to test case, which is used to
    339   specify which test cases require file system access.  This is to
    340   highlight dependencies on external resources more clearly and to speed up
    341   the execution of test suites by skipping the creation of many unnecessary
    342   work directories.
    343 
    344 * Fixed test programs to get a sane default value for their source
    345   directory.  This means that it should not be necessary any more to pass
    346   -s when running test programs that do not live in the current directory.
    347 
    348 * Defining test case headers became optional.  This is trivial to achieve
    349   in shell-based tests but a bit ugly in C and C++.  In C, use the new
    350   ATF_TC_WITHOUT_HEAD macro to define the test case, and in C++ use
    351   ATF_TEST_CASE_WITHOUT_HEAD.
    352 
    353 
    354 Changes in version 0.8
    355 **********************
    356 
    357 Experimental version released on May 7th, 2010.
    358 
    359 * Test programs no longer run several test cases in a row.  The execution
    360   of a test program now requires a test case name, and that single test
    361   case is executed.  To execute several test cases, use the atf-run utility
    362   as usual.
    363 
    364 * Test programs no longer fork a subprocess to isolate the execution of
    365   test cases.  They run the test case code in-process, and a crash of the
    366   test case will result in a crash of the test program.  This is to ease
    367   debugging of faulty test cases.
    368 
    369 * Test programs no longer isolate their test cases.  This means that they
    370   will not create temporary directories nor sanitize the environment any
    371   more.  Yes: running a test case that depends on system state by hand will
    372   most likely yield different results depending on where (machine,
    373   directory, user environment, etc.) it is run.  Isolation has been moved
    374   to atf-run.
    375 
    376 * Test programs no longer print a cryptic format (application/X-atf-tcs)
    377   on a special file channel.  They can now print whatever they want on the
    378   screen.  Because test programs can now only run one test case every time,
    379   providing controlled output is not necessary any more.
    380 
    381 * Test programs no longer write their status into a special file
    382   descriptor.  Instead, they create a file with the results, which is later
    383   parsed by atf-run.  This changes the semantics of the -r flag.
    384 
    385 * atf-run has been adjusted to perform the test case isolation.  As a
    386   result, there is now a single canonical place that implements the
    387   isolation of test caes.  In previous releases, the three language
    388   bindings (C, C++ and shell) had to be kept in sync with each other (read:
    389   not a nice thing to do at all).  As a side effect of this change, writing
    390   bindings for other languages will be much, much easier from now on.
    391 
    392 * atf-run forks test programs on a test case basis, instead of on a test
    393   program basis as it did before.  This is to provide the test case
    394   isolation that was before implemented by the test programs themselves.
    395 
    396 * Removed the atf-exec tool.  This was used to implement test case
    397   isolation in atf-sh, but it is now unnecessary.
    398 
    399 * It is now optional to define the descr meta-data property.  It has been
    400   proven to be mostly useless, because test cases often carry a descriptive
    401   name of their own.
    402 
    403 
    404 Changes in version 0.7
    405 **********************
    406 
    407 Experimental version released on December 22nd, 2009.
    408 
    409 * Added build-time checks to atf-c and atf-c++.  A binding for atf-sh
    410   will come later.
    411 
    412 * Migrated all build-time checks for header files to proper ATF tests.
    413   This demonstrates the use of the new feature described above.
    414 
    415 * Added an internal API for child process management.
    416 
    417 * Converted all plain-text distribution documents to a Docbook canonical
    418   version, and include pre-generated plain text and HTML copies in the
    419   distribution file.
    420 
    421 * Simplified the contents of the Makefile.am by regenerating it from a
    422   canonical Makefile.am.m4 source.  As a side-effect, some dependency
    423   specifications were fixed.
    424 
    425 * Migrated all checks from the check target to installcheck, as these
    426   require ATF to be installed.
    427 
    428 * Fixed sign comparison mismatches triggered by the now-enabled
    429   -Wsign-compare.
    430 
    431 * Fixed many memory and object leaks.
    432 
    433 
    434 Changes in version 0.6
    435 **********************
    436 
    437 Experimental version released on January 18th, 2009.
    438 
    439 * Make atf-exec be able to kill its child process after a certain period
    440   of time; this is controlled through the new -t option.
    441 
    442 * Change atf-sh to use atf-exec's -t option to control the test case's
    443   timeouts, instead of doing it internally.  Same behavior as before, but
    444   noticeably faster.
    445 
    446 * atf-exec's -g option and atf-killpg are gone due to the previous
    447   change.
    448 
    449 * Added the atf-check(1) tool, a program that executes a given command
    450   and checks its exit code against a known value and allows the management
    451   of stdout and stderr in multiple ways.  This replaces the previous
    452   atf_check function in the atf-sh library and exposes this functionality
    453   to both atf-c and atf-c++.
    454 
    455 * Added the ATF_REQUIRE family of macros to the C interface.  These help
    456   in checking for fatal test conditions.  The old ATF_CHECK macros now
    457   perform non-fatal checks only.  I.e. by using ATF_CHECK, the test case
    458   can now continue its execution and the failures will not be reported
    459   until the end of the whole run.
    460 
    461 * Extended the amount of ATF_CHECK_* C macros with new ones to provide
    462   more features to the developer.  These also have their corresponding
    463   counterparts in the ATF_REQUIRE_* family.  The new macros (listing the
    464   suffixes only) are: _EQ (replaces _EQUAL), _EQ_MSG, _STREQ and
    465   _STREQ_MSG.
    466 
    467 
    468 Changes in version 0.5
    469 **********************
    470 
    471 Experimental version released on May 1st, 2008.
    472 
    473 * Clauses 3 and 4 of the BSD license used by the project were dropped.
    474   All the code is now under a 2-clause BSD license compatible with the GNU
    475   General Public License (GPL).
    476 
    477 * Added a C-only binding so that binary test programs do not need to be
    478   tied to C++ at all.  This binding is now known as the atf-c library.
    479 
    480 * Renamed the C++ binding to atf-c++ for consistency with the new atf-c.
    481 
    482 * Renamed the POSIX shell binding to atf-sh for consistency with the new
    483   atf-c and atf-c++.
    484 
    485 * Added a -w flag to test programs through which it is possible to
    486   specify the work directory to be used.  This was possible in prior
    487   releases by defining the workdir configuration variable (-v workdir=...),
    488   but was a conceptually incorrect mechanism.
    489 
    490 * Test programs now preserve the execution order of test cases when they
    491   are given in the command line.  Even those mentioned more than once are
    492   executed multiple times to comply with the user's requests.
    493 
    494 
    495 Changes in version 0.4
    496 **********************
    497 
    498 Experimental version released on February 4th, 2008.
    499 
    500 * Added two new manual pages, atf-c++-api and atf-sh-api, describing the
    501   C++ and POSIX shell interfaces used to write test programs.
    502 
    503 * Added a pkg-config file, useful to get the flags to build against the
    504   C++ library or to easily detect the presence of ATF.
    505 
    506 * Added a way for test cases to require a specific architecture and/or
    507   machine type through the new 'require.arch' and 'require.machine'
    508   meta-data properties, respectively.
    509 
    510 * Added the 'timeout' property to test cases, useful to set an
    511   upper-bound limit for the test's run time and thus prevent global test
    512   program stalls due to the test case's misbehavior.
    513 
    514 * Added the atf-exec(1) internal utility, used to execute a command
    515   after changing the process group it belongs to.
    516 
    517 * Added the atf-killpg(1) internal utility, used to kill process groups.
    518 
    519 * Multiple portability fixes.  Of special interest, full support for
    520   SunOS (Solaris Express Developer Edition 2007/09) using the Sun Studio 12
    521   C++ compiler.
    522 
    523 * Fixed a serious bug that prevented atf-run(1) from working at all
    524   under Fedora 8 x86_64.  Due to the nature of the bug, other platforms
    525   were likely affected too.
    526 
    527 
    528 Changes in version 0.3
    529 **********************
    530 
    531 Experimental version released on November 11th, 2007.
    532 
    533 * Added XML output support to atf-report.  This is accompanied by a DTD
    534   for the format's structure and sample XSLT/CSS files to post-process this
    535   output and convert it to a plain HTML report.
    536 
    537 * Changed atf-run to add system information to the report it generates.
    538   This is currently used by atf-report's XML output only, and is later
    539   printed in the HTML reports in a nice and useful summary table.  The user
    540   and system administrator are allowed to tune this feature by means of
    541   hooks.
    542 
    543 * Removed the test cases' 'isolated' property.  This was intended to
    544   avoid touching the file system at all when running the related test case,
    545   but this has not been true for a long while: some control files are
    546   unconditionally required for several purposes, and we cannot easily get
    547   rid of them.  This way we remove several critical and delicate pieces of
    548   code.
    549 
    550 * Improved atf-report's CSV output format to include information about
    551   test programs too.
    552 
    553 * Fixed the tests that used atf-compile to not require this tool as a
    554   helper.  Avoids systems without build-time utilities to skip many tests
    555   that could otherwise be run.  (E.g. NetBSD without the comp.tgz set
    556   installed.)
    557 
    558 * Many general cleanups: Fixed many pieces of code marked as ugly and/or
    559   incomplete.
    560 
    561 
    562 Changes in version 0.2
    563 **********************
    564 
    565 Experimental version released on September 20th, 2007.
    566 
    567 * Test cases now get a known umask on entry.
    568 
    569 * atf-run now detects many unexpected failures caused by test programs and
    570   reports them as bogus tests.  atf-report is able to handle these new
    571   errors and nicely reports them to the user.
    572 
    573 * All the data formats read and written by the tools have been
    574   documented and cleaned up.  These include those grammars that define how
    575   the different components communicate with each other as well as the
    576   format of files written by the developers and users: the Atffiles and the
    577   configuration files.
    578 
    579 * Added the atf-version tool, a utility that displays information about
    580   the currently installed version of ATF.
    581 
    582 * Test cases can now define an optional cleanup routine to undo their
    583   actions regardless of their exit status.
    584 
    585 * atf-report now summarizes the list of failed (bogus) test programs
    586   when using the ticker output format.
    587 
    588 * Test programs now capture some termination signals and clean up any
    589   temporary files before exiting the program.
    590 
    591 * Multiple bug fixes and improvements all around.
    592 
    593 
    594 Changes in version 0.1
    595 **********************
    596 
    597 Experimental version released on August 20th, 2007.
    598 
    599 * First public version.  This was released coinciding with the end of the
    600   Google Summer of Code 2007 program.
    601 
    602 
    603 ===========================================================================
    604 vim: filetype=text:textwidth=75:expandtab:shiftwidth=2:softtabstop=2
    605