MDEV-30281 MTR usability improvements (config file and misc features) - #5467
MDEV-30281 MTR usability improvements (config file and misc features)#5467midenok wants to merge 18 commits into
Conversation
Document each parameter as in, out or in-out, and note that *argv is replaced and must be freed with free_defaults(). Also fix a couple of typos.
Failure reports could dump the entire per-test server error log
(get_log_from_proc) and every suspicious line collected during shutdown
(check_warnings_post_shutdown) with no limit, which is very noisy on
tests that produce large logs.
Add seven options:
--head-log=N keep the first N lines of the server error log in a
crash report
--tail-log=N keep the last N lines of the server error log in a
crash report; combine with --head-log to keep both ends
with the middle snipped
--head-warnings=N keep the first N suspicious lines of the shutdown-
warnings report
--tail-warnings=N keep the last N suspicious lines of the shutdown-warnings
report; combine with --head-warnings to keep both ends
with the middle snipped
--head=N shortcut: set the head-* options to N and the tail-*
options to 0
--tail=N shortcut: set the tail-* options to N and the head-*
options to 0
For each option N=0 keeps nothing and a negative value includes
everything; trimmed output is prefixed with a "< snip N lines >" marker.
For both the server error log and the shutdown-warnings report the whole
text is kept unless the corresponding --head-*/--tail-* option is given,
and giving one end alone drops the opposite end. A shared splice_lines()
helper implements the trimming for both.
The generic "negative values aren't meaningful on integer options" guard
in command_line_setup() is updated to exempt the --tail*/--head-* options,
so their negative "everything" default is accepted instead of aborting
startup. --tail-lines now also accepts a negative value, mapped to the
mysqltest maximum of 10000, since the mysqltest binary itself rejects
negatives.
Add module-level POD to My::ConfigFactory explaining how a test run's server configuration is built from a single .cnf template: - template selection precedence (<testname>.cnf, suite my.cnf, defaults) done in mtr_cases.pm, and --defaults-file/--defaults-extra-file - my.cnf/INI parsing and recursive !include/!includedir merging by My::Config, with last-value-wins override semantics - generated settings applied on top via the rule sets - related .opt and .combinations files handled elsewhere Every subroutine in the module is documented too: the public new_config entry point, the rule engine, the pre-rules, the fix_* value generators and the post_* checks. The documentation is standard POD, so it can be rendered with the usual Perl tools. To read it in the terminal: perldoc mysql-test/lib/My/ConfigFactory.pm To produce a man page, plain text or HTML respectively: pod2man mysql-test/lib/My/ConfigFactory.pm | man -l - pod2text mysql-test/lib/My/ConfigFactory.pm pod2html mysql-test/lib/My/ConfigFactory.pm No functional change.
|
|
There was a problem hiding this comment.
Pull request overview
This PR implements a set of usability and quality-of-life improvements to the MariaDB Test Runner (MTR) harness, primarily in mysql-test/mariadb-test-run.pl and supporting modules, to make runs more configurable, output less noisy, suite/combination selection more flexible, and test collection faster.
Changes:
- Add support for reading MTR options from standard MariaDB option files (
[mtr]group) and improve invalid-option attribution between config vs command line. - Add crash-log and shutdown-warning trimming/stripping options, plus
--list-combinationsand suite negation (--suite=!NAME). - Add
--exec-<debugger>wrapping for mysqltest--execcommands, and speed up result-file discovery via directory indexing.
Reviewed changes
Copilot reviewed 14 out of 14 changed files in this pull request and generated 2 comments.
Show a summary per file
| File | Description |
|---|---|
mysys/my_default.c |
Improves parameter documentation for option-file loading API. |
mysql-test/suite/mtr/t/feat.test |
New integration test exercising several new MTR features via child MTR runs. |
mysql-test/suite/mtr/t/feat.result |
Expected output for the new feat.test. |
mysql-test/suite/mtr/misc/warn.test |
New helper test to generate real shutdown warnings for trimming tests. |
mysql-test/suite/mtr/misc/empty.test |
New placeholder test used for combination listing. |
mysql-test/suite/mtr/misc/empty.combinations |
New combinations file used by --list-combinations testing. |
mysql-test/suite/mtr/misc/crash.test |
New helper test that intentionally crashes the server to exercise crash-report trimming/stripping. |
mysql-test/mariadb-test-run.pl |
Core implementation: config-file option merging, suite negation, combination selection/listing, trimming/stripping logic, and improved error reporting. |
mysql-test/lib/My/Debugger.pm |
Adds exec debugger templates and exports MYSQLTEST_EXEC_WRAP for wrapping mysqltest --exec. |
mysql-test/lib/My/CoreDump.pm |
Makes “Output from gdb/lldb follows…” hint text optional (controlled by --strip-hints). |
mysql-test/lib/My/ConfigFactory.pm |
Adds substantial internal POD documentation clarifying config templating and rule generation. |
mysql-test/lib/My/Config.pm |
Enhances config parsing: !includedir, whitespace-only lines, unknown ! directives ignored, and cleaner parse error messages. |
mysql-test/lib/mtr_cases.pm |
Adds --combination-select support and speeds up result file discovery by indexing result directories once. |
client/mysqltest.cc |
Injects optional MYSQLTEST_EXEC_WRAP wrapper into --exec commands after leading NAME=VALUE assignments. |
💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.
5e60e3c to
5200f82
Compare
5200f82 to
2a1d651
Compare
MTR options are read from standard MariaDB conf file from [mtr] section. Command-line options take precedence over configuration file. There are two variants of config file location: 1. Default behaviour to look for global, server and user configs as described in https://mariadb.com/kb/en/configuring-mariadb-with-option-files/ 2. Custom config location specified by MTR_CONFIG environment variable. None of other config files are read in this case. If there is wrong setting in configuration file MTR fails and displays the error message. It is clear from the error message from what source the wrong option comes: config file or command-line. The config file name and the line number is not printed which is the subject for further improvement. Implementation reuses the in-tree My::Config parser instead of adding a third-party module. A new load_defaults() in mariadb-test-run.pl parses the config files with My::Config and merges the [mtr] options into the @argv array for further validation in Getopt::Long: - Config options are merged into the beginning of the array so that command-line options take precedence. They are separated from the command-line options by ---end-of-config--- which is used as a marker for the source of failed options after Getopt::Long; - a value-less [mtr] option that actually requires a value would let Getopt::Long swallow that marker (or the following option) as its value; this is detected and reported instead of failing obscurely, and the invalid-option message prints the bare option name; - Global, server and user config files are searched; only the first existing among the global and server locations is used, then the user's ~/.my.cnf. A config file can also be specified explicitly via MTR_CONFIG, in which case no other file is read; - Argument-less options are passed as is (My::Config::Option::option() renders them without a value); - My::Config already honours !include; it was extended to also support !includedir, to skip whitespace-only lines and to ignore any other unknown ! directive instead of failing. A line beginning with '#' is treated as a comment, as in any my.cnf, not as an option named '#...'. (An earlier version of this patch was based on the CPAN MySQL::Config module, which ignored all ! directives.)
…roak Make every wrong setting in a server config file fail with a human-friendly message and no Perl stack trace or source-code line. My::Config now treats parse errors as data errors instead of caller bugs: each parse failure dies with a clean "file:line: reason" message naming the config file and the offending line number, and without the "at FILE line N." Perl location. This also delivers the file name and line number that were previously missing. The internal misuse/invariant assertions keep using croak, which is the right tool there. load_defaults() (in mariadb-test-run.pl) catches that die and re-throws it through mtr_error(), which flushes output, prints the uniform mtr error format and exits in a controlled way. This replaces the earlier fragile approach of stripping the location suffix off a croak with a regexp. mariadb-test-run.pl: validate the parallel option value before the numeric comparison, so a non-numeric or negative value reports a clean error instead of an "Argument isn't numeric" warning with a code line. This path is shared with the command line, so it is fixed there too.
Wire the standard MariaDB defaults options into MTR's own option loading,
mirroring libmariadb's get_defaults_options()/my_load_defaults():
--no-defaults do not read any option file
--defaults-file read [mtr] from this file only
--defaults-extra-file read this file in addition (extra-file slot)
--defaults-group-suffix also read [mtr<suffix>]
--print-defaults print the [mtr] options and exit
--defaults-file and --defaults-extra-file keep their existing MTR meaning
as the server config *template* too: the two consumers read different
groups of the same file - [mysqld]/[client]/... for the template (via
collect_option), [mtr] here. A command-line --defaults-file /
--defaults-extra-file overrides the MTR_CONFIG / MTR_CONFIG_EXTRA
environment variable; --defaults-group-suffix overrides
MARIADB_GROUP_SUFFIX / MYSQL_GROUP_SUFFIX.
An explicitly named file (--defaults-file or $MTR_CONFIG) that does not
exist is an error, matching libmariadb; a missing file in the standard
search order is still skipped silently.
Order of execution in command_line_setup():
1. get_defaults_options() runs first (before the main GetOptions). It
consumes the MTR-only options (--no-defaults, --defaults-group-suffix,
--print-defaults) out of @argv, and *peeks* --defaults-file /
--defaults-extra-file off a copy of @argv so they stay in @argv for
the second consumer.
2. load_defaults() reads the [mtr] group from the selected files and
prepends the resulting options into @argv, ahead of the
---end-of-config--- marker.
3. My::Debugger::fix_options() adjusts optional-argument options.
4. The main GetOptions(%options) parses everything - the config options
(before the marker) and the command line (after it). Command-line
options come later and therefore take precedence. --defaults-file is
handed to collect_option for the template. Options still unrecognized
afterwards are validated in a manual loop that uses the marker to
report whether a bad option came from the config file or the command
line.
Getopt::Long is configured "pass_through", so a parse leaves any option it
does not declare in @argv instead of erroring. Because pass_through is on,
step 1 skims off only the defaults options without touching the rest, and
step 4 does the real parse; neither errors on options it does not declare.
--mtr-config-only (-M) makes --defaults-file / --defaults-extra-file be read for MTR's own [mtr] options only, without also being applied as the server config template. Normally those two options serve two consumers of the same file: the [mtr] group configures MTR itself (load_defaults), while the remaining groups - [mysqld]/[client]/... - become the test servers' config template (collect_option). -M opts out of the second use: once load_defaults has read [mtr], it removes --defaults-file / --defaults-extra-file from @argv, so the main GetOptions/collect_option never turns them into a template. The switch may be given on the command line, or set inside the [mtr] section of the config file itself. In the latter case load_defaults notices mtr-config-only while reading [mtr] and drops the file options on the fly, so [mtr] mtr-config-only turns any file passed via --defaults-file into an MTR-only config. get_defaults_options() consumes -M out of @argv like the other MTR-only defaults options; it is a directive, not passed through to GetOptions. -M is only relevant for a command-line --defaults-file / --defaults-extra-file (the dual-use options). The MTR_CONFIG / MTR_CONFIG_EXTRA environment variables are read only for [mtr] and never reach collect_option, so they are mtr-config-only already and -M is a no-op for them.
Extend the --suite[s] option so a suite name prefixed with '!' excludes
that suite instead of adding it. This makes "run everything except X"
easy and lets a config-file suite list be trimmed on the command line.
Semantics:
--suites=A,B run suites A and B (as before)
--suites=!A run the default set minus A
--suites=A,B,!B run A (a '!' name is removed from the positive names)
[mtr] suites=A,B
+ --suites=!B run A (command-line exclusion trims the file set)
Rules:
- If any plain (positive) names are given, they are the base set;
otherwise the base is the default suite set.
- Every '!'-prefixed name is then removed from the base.
- Names accumulate across the [mtr] config file and the command line,
so a '!' exclusion on the command line applies to the set configured
in the file (see below).
To make the cross-source case work, --suites now accumulates its values
into @opt_suites instead of a last-wins scalar: the option handler pushes
each comma-separated value, so the [mtr] config file value and the
command-line value are both collected. The positive/negative resolution
then runs once in main(), producing the final suite list.
Exclusion matching ignores the "-<overlay>" suffix used in the default
list (e.g. "rpl-"), so --suites=!rpl removes "rpl-", and slashes are
preserved (--suites=!compat/oracle works).
Add --combination-select=N (short alias -c) to run only a single combination from a .combinations file instead of every one. N is the position of the combination in the file, counting the [] sections in file order: 1 first combination -1 last combination 2 second combination -2 last but one ... ... So --combination-select=-1 runs only the last combination, -c 2 runs the second one, etc. N must be a non-zero integer; 0, a non-integer, or a value outside the range of a given file is an error. The integer format is validated up front in command_line_setup(), so a bad value is rejected even for tests that have no .combinations file. The range and the selection itself are applied in combinations_from_file(), which is the single point through which every .combinations file is read (both the suite-level "combinations" file and per-test "<test>.combinations"), so it always follows the [] section order of the file. When a test draws combinations from more than one file, N selects within each file. It is ignored when --combination is given (that already replaces the .combinations files with command-line combinations).
Add --list-combinations (alias --lc) which, for the specified test(s), prints the available combinations in the selectable "test,combination" form and exits without running anything. For each collected test it prints a summary line followed by one line per combination, e.g.: $ mtr --list-combinations versioning.foreign Combinations for versioning.foreign: timestamp,trx_id versioning.foreign,timestamp versioning.foreign,trx_id Each printed line is exactly what you pass back to mtr to run that one combination (it round-trips). When a test draws combinations from more than one .combinations file (several dimensions), the lines list the full product, e.g. "test,dim1_value,dim2_value".
A crash report contains, besides the actual server log and backtrace,
several explanatory "hint" paragraphs - bug-reporting boilerplate,
instructions for producing a better stack trace, "Attempting backtrace",
"Output from gdb follows", etc. They are useful once but noise for
someone who reads these reports all day.
Add --strip-hints (off by default). When given, these hints are omitted
and only the real content is kept.
Three sources are handled:
- The paragraphs the server writes to its error log (via the crash
signal handler) are dropped from the "Server log from this test"
excerpt by strip_crash_hints(). Each hint is a paragraph running from
a recognizable opener line to the next blank line, so trimming is
robust across server-version wording changes; the surrounding data
(CURRENT_TEST, the signal line, "Server version", the backtrace,
thread pointer, ...) is kept.
- My::CoreDump prints its own "Output from gdb/lldb follows ..." header;
it is suppressed too, leaving just the debugger output.
- mariadb-test-run.pl's own end-of-run boilerplate in mtr_report_stats()
("The log files in var/log may give you some hint ..." and the
MariaDB bug-tracker URL) is skipped as well.
$opt_strip_hints is declared with "our" so My::CoreDump and
mtr_report.pm can read it as $::opt_strip_hints.
On a crash the server writes a "Resource Limits" table (read from
/proc/self/limits) to its error log, which MTR echoes in the crash
report. Like the bug-reporting hints it is rarely useful when reading
these reports routinely.
Add --strip-limits (off by default). When given, the table is dropped
from the "Server log from this test" excerpt by strip_resource_limits().
Unlike a hint paragraph the table has no trailing blank line - it runs
straight into the next section ("Core pattern:", ...). So the filter
removes the opener line (matching both "Resource Limits:" and the newer
"Resource Limits (excludes unlimited resources):"), the
"Limit ... Soft Limit ..." header and the "Max ..." rows, and stops at
the first line that is neither, keeping the surrounding data.
This is independent of --strip-hints.
On a crash the server writes its own stack backtrace (my_print_stacktrace)
to the error log, which MTR echoes in the crash report. When a core file
is available the gdb/lldb backtrace from My::CoreDump is more useful, so
this server-side report is often just noise.
Add --strip-backtrace (off by default). When given, it is dropped from
the "Server log from this test" excerpt by strip_backtrace(). The
gdb/lldb backtrace from a core file is a separate thing and is not
affected.
strip_backtrace() treats the backtrace as a bounded block rather than
matching lines everywhere, so a line outside a backtrace that merely ends
in an [0x...] address is kept:
- a "Thread pointer:" or "Attempting backtrace" line starts a block;
- inside a block every recognisable backtrace line is dropped: the
intro prose, the thread pointer, blank lines, a "stack_bottom" or
"Stack range" header, frame lines (both "...[0x...]" symbol frames
and the bare "0x..." addresses printed by the frame-pointer walker)
and the interleaved addr2line / my_addr_resolve diagnostics;
- the first line that is not recognisable backtrace content ends the
block and is kept, so pass-through resumes and a later backtrace (or
the rest of the log) survives.
Because anything unrecognised ends the block rather than being dropped, a
backtrace with no "stack_bottom" line and no bracketed frames - as the
frame-pointer-walker and aborted-backtrace builds produce - does not
swallow the remainder of the log.
Also add --strip-log, a convenience alias that turns on all three of
--strip-hints, --strip-limits and --strip-backtrace at once.
All stripping runs before the --head-log/--tail-log trimming, so those
line counts apply to the already-stripped log.
Run a child mariadb-test-run.pl on misc.crash (SIGSEGV) and misc.warn (system-versioned SYSTEM_TIME LIMIT overflow) and check that the crash and shutdown-warnings reports are trimmed by --head-log / --tail-log / --strip-log and --tail-warnings. Also exercise --list-combinations and --suite negation. The child's volatile output (pids, timestamps, addresses, an unbounded backtrace, ...) is reduced to a deterministic skeleton with include/grep.inc and normalized with --replace_regex: the two anchor backtrace frames (my_print_stacktrace, main) are kept while the variable middle frames and the per-line values are cut down to "<...cut...>". The test is gated behind --big-test.
New debugger options run every mysqltest '--exec' command line under a wrapper (rr record, or gdb --args in an xterm), so external tools invoked by tests (myisampack, myisamchk, ...) can be traced or debugged without naming them. My::Debugger: each debugger with an 'exec' template gets an --exec-<dbg> option auto-registered. pre_setup() detects the requested one, runs its one-time 'pre' hook, and exports MYSQLTEST_EXEC_WRAP (plus _RR_TRACE_DIR for rr, and an xterm wrapper for terminal debuggers like gdb). do_exec(): when MYSQLTEST_EXEC_WRAP is set it is injected into the command line, after any leading shell NAME=VALUE assignments (quotes and backslashes honoured), so the tool - not the assignment word - is what gets wrapped. The implicit /bin/sh -c that popen() supplies provides the shell layer.
collect_one_test_case() located each test's .result/.rdiff files with a
per-test glob:
for (<{$resdirglob}/$tname*.{rdiff,result}>) { ... }
Because of the "$tname*" wildcard, glob had to opendir and read the whole
result directory on every call, and the "{rdiff,result}" brace made it do
so twice (once per extension). Run once per test, this is O(tests *
dirsize) - the suite's result directory is re-read ~2x for every test in
the suite. Profiling the collection of all default suites showed this
single line accounting for 8.3s of the 13.1s phase (63%), and the phase
issuing 408k stat() calls (339k of them ENOENT).
Read each result directory once instead, indexing its .result/.rdiff
files by base test name (the leading run of characters before the first
'.' or ','), and look a test's files up in that cache. The "$tname*"
glob was only a coarse prefilter anyway - the existing regex
"^$tname((?:,combo)*)\.(rdiff|result)$" does the real matching - so
indexing by base name yields exactly the same candidate set.
Stats (collecting all default suites, 6548 tests, warm cache):
collection time 13.1s -> 4.5s (~2.9x)
directory-open syscalls 23026 -> 5284
The collected test list (result_file, base_result, skip and comment for
every test) is byte-for-byte identical before and after.
The terminal emulator for interactive debuggers (--gdb, --exec-gdb, ...)
was hard-coded to xterm. It is now configurable via the --terminal
option or the MTR_TERM environment variable (the option takes
precedence). The template understands two placeholders, {title} (window
title) and {command} (the debugger invocation, expanded as separate argv
words); the default "xterm -title {title} -e {command}" reproduces the
previous behaviour.
My::Debugger::term_argv() expands the template into an argv list for the
mysqld/client/boot debuggers; the same template drives the --exec-<dbg>
shell-string wrapper, where {command} must be last.
For example, debug the bootstrap server for main.1st under gdb in a
KDE Konsole window:
mtr --boot-gdb --terminal='konsole -p tabtitle={title} -e {command}' main.1st
--exit-line=N stops a test before the command at line N of the test file, exactly as if an --exit directive were placed there. Handy for bisecting where a test starts to misbehave, and for debugging a test without extracting a standalone test case: record the run under rr (--rr), then reverse-replay from the end of the trace straight to the SQL command of interest. mariadb-test-run.pl gains --exit-line (-l) and forwards it to mysqltest. The option is global: it applies to every test in the run, each stopping at line N of its own test file (so N is per-file). A test whose file has fewer than N lines is unaffected and runs to completion. mysqltest gains the --exit-line (-l) option: after reading each command, if we are in the top-level test file (cur_file == file_stack) and the command starts at or past the requested line, it aborts like Q_EXIT. Gating to the main file keeps line numbers of sourced includes from triggering it. suite/mtr/feat exercises it by running a child mtr with --exit-line=5 on main.1st (line 4 "show databases;", line 5 "show tables in mysql;"): the run stops before line 5, so the recorded result's tables section is missing and the test fails with a length mismatch - proving the exit landed exactly at the requested line. The child's whole output is kept (no grep); volatile banner/footer lines are cut by prefix via $NORM_RUN.
2a1d651 to
723252a
Compare
A set of independent quality-of-life improvements to the MTR harness (
mariadb-test-run.pl), covering four areas: reading options from a standard config file, trimming noisy failure reports, running test features, and debugger options.Each item is self-contained and lands as its own commit; command-line options are off by default and preserve existing behaviour unless requested.
1. Configuration from an option file
MTR now reads its own options from the standard MariaDB option files, from the
[mtr]section. Command-line options take precedence over the file.~/.my.cnfis read last.my.ini); on Windows, or any non-standard setup, use an explicitMTR_CONFIG/--defaults-file, which work on every platform.MTR_CONFIGenv var selects a custom config location; when set, no other file is read.My::Configparser (no third-party module). A newload_defaults()inmariadb-test-run.plmerges[mtr]options into@ARGV, separated from command-line options by an---end-of-config---marker so error reporting can name the source.My::Configextended to support!includedir, skip whitespace-only lines, and ignore unknown!directives instead of failing.Standard defaults options are wired in, mirroring libmariadb's
my_load_defaults():--no-defaults--defaults-file[mtr]from this file only--defaults-extra-file--defaults-group-suffix[mtr<suffix>]--print-defaults[mtr]options and exit--mtr-config-only(alias-M)[mtr]from--defaults-file/--defaults-extra-fileonly, without also using them as the server template--defaults-file/--defaults-extra-filekeep their existing MTR meaning as the server config template too; the two consumers read different groups of the same file. Command line overridesMTR_CONFIG/MTR_CONFIG_EXTRAand the group-suffix env vars.--mtr-config-only(-M) opts out of that dual use: the file is then read for[mtr]only and dropped from@ARGV, so it is not also applied as the server template. It may be given on the command line, or set inside the[mtr]section itself (applied on the fly while[mtr]is read) — so a singlemtr-config-onlyline turns any--defaults-fileinto an MTR-only config.-Mis only relevant for a command-line--defaults-file/--defaults-extra-file(the dual-use options). TheMTR_CONFIG/MTR_CONFIG_EXTRAenv vars are read only for[mtr]and never reach the template consumer, so they are mtr-config-only already and-Mis a no-op for them.Config error reporting
Every wrong setting in a config file now fails with a clean
file:line: reasonmessage, without a Perl stack trace or source-code location, and it is clear whether a bad option came from the config file or the command line.2. Trimming noisy failure reports
Crash reports and shutdown-warning dumps could emit the entire server error log and every suspicious line with no limit. New options bound them.
Head/tail trimming
--head-log=N--tail-log=N--head-logto keep both ends--head-warnings=N--tail-warnings=N--head-warnings--head=Nhead-*= N,tail-*= 0 (!)--tail=Ntail-*= N,head-*= 0 (!)N=0 keeps nothing; a negative value keeps everything; trimmed output is marked with
< snip N lines >.Stripping boilerplate
--strip-hints--strip-limits/proc/self/limits--strip-backtracemy_print_stacktracebacktrace (the gdb/lldb core backtrace is kept)--strip-logAll stripping runs before head/tail trimming, so those line counts apply to the already-stripped log.
--strip-backtracetreats the backtrace as a bounded block (state machine) rather than matching lines everywhere.3. Running test features
--suites=!NAME— exclude a suite. Positive names (or the default set) form the base; every!-name is removed from it. Accumulates across the config file and command line, so a command-line exclusion trims the file-configured set. Ignores the-<overlay>suffix (so!rplremovesrpl-).--combination-select=N(alias-c) — run only the Nth combination of a.combinationsfile (1-based; negative counts from the end,-1= last). Applied in the single point where every.combinationsfile is read. Ignored when--combinationis given.--list-combinations(alias--lc) — print a test's available combinations in the selectabletest,combinationform and exit. Each printed line round-trips back into mtr. Multi-dimension tests list the full product.--exit-line=N(alias-l) — stop the test before the command at line N of the test file, exactly as if an--exitdirective were placed there. Implemented as a mysqltest option and forwarded by mtr; gated to the top-level test file so line numbers of sourced includes never trigger it.4. Debugger options
Running external tools under a debugger
New
--exec-rr/--exec-gdboptions run every mysqltest--execcommand under a wrapper (rr record, orgdb --argsin a terminal), so external tools invoked by tests (myisampack,myisamchk, ...) can be traced or debugged without naming them.My::Debugger: any debugger with anexectemplate auto-registers an--exec-<dbg>option;pre_setup()exportsMYSQLTEST_EXEC_WRAP(plus_RR_TRACE_DIRfor rr).do_exec()injects the wrapper into the command line after any leading shellNAME=VALUEassignments, so the tool — not the assignment word — is wrapped. The implicit/bin/sh -cfrompopen()provides the shell layer.Configurable terminal
The terminal emulator for interactive debuggers (
--gdb,--exec-gdb, ...) was hard-coded toxterm; it is now configurable via the--terminaloption or theMTR_TERMenvironment variable (the option takes precedence). The template expands two placeholders —{title}(window title) and{command}(the debugger invocation) — and the defaultxterm -title {title} -e {command}preserves the previous behaviour. For example:--terminal='gnome-terminal --title={title} -- {command}'.5. Faster test collection (~3x)
Locating each test's
.result/.rdifffiles used a per-test glob keyed on the test name, which forced a full read of the suite's result directory on every test — and twice over, once for each extension — so collection cost grew with tests × directory size. Each result directory is now read once and its.result/.rdifffiles indexed by base test name (the leading run before the first.or,); every test then looks its files up in that cache. The glob was only a coarse prefilter — the existing regex still does the real matching — so the candidate set, and the resulting test list, is byte-for-byte identical.Collecting all default suites (6548 tests, warm cache): collection time 13.1s → 4.5s (~2.9×), directory-open syscalls 23026 → 5284.