public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Carl Love <cel@us.ibm.com>
To: Andrew Burgess <aburgess@redhat.com>, Tom Tromey <tom@tromey.com>,
	Andrew Burgess via Gdb-patches <gdb-patches@sourceware.org>,
	cel@us.ibm.com
Subject: RE: [PATCH 2/2] gdb/python: improve formatting of help text for user defined commands
Date: Tue, 31 May 2022 16:10:47 -0700	[thread overview]
Message-ID: <73f20772cf68f5f966d82ff005b430a34f6ba1a3.camel@us.ibm.com> (raw)
In-Reply-To: <8457aaf504438e99371cfa8f2f363cfe73da6786.camel@us.ibm.com>

On Tue, 2022-05-31 at 14:54 -0700, Carl Love via Gdb-patches wrote:
> Andrew:
> 
> On Sat, 2022-05-28 at 11:11 +0100, Andrew Burgess via Gdb-patches
> wrote:
> > Tom Tromey <tom@tromey.com> writes:
> > 
> > > > > > > > "Andrew" == Andrew Burgess via Gdb-patches <
> > > > > > > > gdb-patches@sourceware.org> writes:
> > > 
> > > Andrew> Thanks to Eli and Lancelot for their feedback.
> > > 
> > > Andrew> In this revision I have:
> > > Andrew>   - Updated the NEWS entry in line with Eli's feedback,
> > > Andrew>   - Updated some comments that Lancelot identified as
> > > being
> > > out of date,
> > > Andrew>   - Added an additional comment into the test script to
> > > justify some
> > > Andrew>     'verbose -log' lines.
> > > Andrew>   - There's no functional change to the code.
> > > 
> > > This looks ok to me, thank you.
> > 
> > Thanks, pushed.
> > 
> > Andrew
> > 
> 
> I am seeing failures with the new test gdb/testsuite/gdb.python/py-
> doc-
> reformat.exp on X86-64 and Power.  The errors show up in the full

My bad, I forgot to mention that the failure is on Power 7, RHEL 7.9. 
I was digging a bit more on the "undefined source" error messages below
about the source .  I noticed that the python version was 2.7.5.  I
checked the test on a Power 10, RHEL 9.0, box and found that it worked.
The power 10 box has python version  3.9.10.  I am using the bash shell
on both systems to run the gdb tests.  Not sure if a newer version of
python is needed for the source command.  So far I haven't found
anything about the source command being version dependent.

The intel box is Ubuntu 22.04, does not have python but rather python3
version 3.10.4.   Given that I was able to get it to run on a Power 10
box but not a Power 7 box would indicate it is some sort of a system
setup issue?  Any thoughts on what it might be?  I will keep digging.


> regression run and when just this test is run.  When running just
> this
> test I see in the gdb/testsuite/gdb.log file the following issues.
> 
> There are a couple of errors about "Undefined command "source
> /home/...".  
> 
> Then later I see the a series of FAILS for numerous tests. 
> 
>    ....
>    (gdb) set height 0
>    (gdb) set width 0
>    (gdb) dir
>    Reinitialize source path to empty? (y or n) y
>    Source directories searched: $cdir:$cwd
>    (gdb) dir /home/carll/GDB/build-
> current/gdb/testsuite/../../../binutils-gdb-
> current/gdb/testsuite/gdb.python
>    Source directories searched: /home/carll/GDB/build-
> current/gdb/testsuite/../../../binutils-gdb-
> current/gdb/testsuite/gdb.python:$cdir:$cwd
>    (gdb) source /home/carll/GDB/build-
> current/gdb/testsuite/outputs/gdb.python/py-doc-reformat/py-doc-
> reformat-0.py
>    /home/carll/GDB/build-current/gdb/testsuite/outputs/gdb.python/py-
> doc-reformat/py-doc-reformat-0.py:1: Error in sourced command file:
>    Undefined command: "class".  Try "help".
>    (gdb) ERROR: Undefined command "source /home/carll/GDB/build-
> current/gdb/testsuite/outputs/gdb.python/py-doc-reformat/py-doc-
> reformat-0.py".
>    UNRESOLVED: gdb.python/py-doc-reformat.exp: test 0: source python
> command file
>    help test-cmd
>    Undefined command: "test-cmd".  Try "help".
>    (gdb) ERROR: Undefined command "help test-cmd".
>    UNRESOLVED: gdb.python/py-doc-reformat.exp: test 0: help test-cmd
>    help set test-param
>    Evaluate expression EXP and assign result to variable VAR.
>    Usage: set VAR = EXP
>    This uses assignment syntax appropriate for the current language
>    (VAR = EXP or VAR := EXP for example).
>    VAR may be a debugger "convenience" variable (names starting
>    with $), a register (a few standard names starting with $), or an
> actual
>    variable in the program being debugged.  EXP is any valid
> expression.
>    Use "set variable" for variables with names identical to set
> subcommands.
> 
>    With a subcommand, this command modifies parts of the gdb
> environment.
>    You can see these environment settings with the "show" command.
> 
>    List of set subcommands:
> 
>    set ada -- Prefix command for changing Ada-specific settings.
>    ...
>    set use-deprecated-index-sections -- Set whether to use deprecated
> gdb_index sections.
>    set variable, set var -- Evaluate expression EXP and assign result
> to variable VAR.
>    set verbose -- Set verbosity.
>    set watchdog -- Set watchdog timer.
>    set width -- Set number of characters where GDB should wrap lines
> of its output.
>    set write -- Set writing into executable and core files.
>    Type "help set" followed by set subcommand name for full
> documentation.
>    Type "apropos word" to search for commands related to "word".
>    Type "apropos -v word" for full documentation of commands related
> to "word".
>    Command name abbreviations are allowed if unambiguous.
>    (gdb) FAIL: gdb.python/py-doc-reformat.exp: test 1: help set test-
> param
>    help show test-param
>    Undefined show command: "test-param".  Try "help show".
>    (gdb) FAIL: gdb.python/py-doc-reformat.exp: test 1: help show
> test-param
>    Start of: test_cmd 2
>    ...
> 
> Note, these two failures for test 1, repeat for each of the following
> tests 10, 11, 12, 13, 14, 15, 16, 17, 18.
> 
>    ...
> 
>    Input:
>      """
>      This is the first line.
>        Indented second line.
>      This is the third line."""
>    builtin_spawn /home/carll/bin/gdb -nw -nx -iex set height 0 -iex
> set width 0
>    GNU gdb (GDB) 13.0.50.20220529-git
>    ...
>                   === gdb Summary ===
> 
>    # of unexpected failures        44
>    # of unresolved testcases       44
> 
> Then final error count across all of the tests is 44.  
> 
> I tried doing the commands manually:
> 
>    (gdb)  help set test-param                                     <-
> this should be test 1, which in the regression
>                                                                      
> run is listed as a failure but seemed to work
>    	                                                          manua
> lly.  Wondering if there is an issue in
>                                                                      
> the error reporting.
>    Evaluate expression EXP and assign result to variable VAR.
>    Usage: set VAR = EXP
>    This uses assignment syntax appropriate for the current language
>    (VAR = EXP or VAR := EXP for example).
>    VAR may be a debugger "convenience" variable (names starting
>    with $), a register (a few standard names starting with $), or an
> actual
>    variable in the program being debugged.  EXP is any valid
> expression.
>    Use "set variable" for variables with names identical to set
> subcommands.
> 
>    With a subcommand, this command modifies parts of the gdb
> environment.
>    You can see these environment settings with the "show" command.
> 
>    List of set subcommands:
> 
>    set ada -- Prefix command for changing Ada-specific settings.
>    set agent -- Set debugger's willingness to use agent as a helper.
>    set annotate -- Set annotation_level.
>    set architecture, set processor -- Set architecture of target.
>    set args -- Set argument list to give program being debugged when
> it is started.
>    set auto-connect-native-target -- Set whether GDB may
> automatically connect to the native target.
>    set auto-load -- Auto-loading specific settings.
>    set auto-solib-add -- Set autoloading of shared library symbols.
>    set backtrace -- Set backtrace specific variables.
>    set basenames-may-differ -- Set whether a source file may have
> multiple base names.
>    set breakpoint -- Breakpoint specific settings.
>    set can-use-hw-watchpoints -- Set debugger's willingness to use
> watchpoint hardware.
>    set case-sensitive -- Set case sensitivity in name search
> (on/off/auto).
>    set charset -- Set the host and target character sets.
>    set check, set ch, set c -- Set the status of the type/range
> checker.
>    set circular-trace-buffer -- Set target's use of circular trace
> buffer.
>    set code-cache -- Set cache use for code segment access.
>    set coerce-float-to-double -- Set coercion of floats to doubles
> when calling functions.
>    set compile-args -- Set compile command GCC command-line
> arguments.
>    set compile-gcc -- Set compile command GCC driver filename.
>    set complaints -- Set max number of complaints about incorrect
> symbols.
>    set confirm -- Set whether to confirm potentially dangerous
> operations.
>    set cp-abi -- Set the ABI used for inspecting C++ objects.
>    set cwd -- Set the current working directory to be used when the
> inferior is started.
>    set data-directory -- Set GDB's data directory.
>    set dcache -- Use this command to set number of lines in dcache
> and line-size.
>    set debug -- Generic command for setting gdb debugging flags.
>    set debug-file-directory -- Set the directories where separate
> debug symbols are searched for.
>    set debuginfod -- Set debuginfod options.
>    set default-collect -- Set the list of expressions to collect by
> default.
>    set demangle-style -- Set the current C++ demangling style.
>    set detach-on-fork -- Set whether gdb will detach the child of a
> fork.
>    set directories -- Set the search path for finding source files.
>    set disable-randomization -- Set disabling of debuggee's virtual
> address space randomization.
>    set disassemble-next-line -- Set whether to disassemble next
> source line or insn when execution stops.
>    --Type <RET> for more, q to quit, c to continue without paging--q
>    Quit
>    (gdb) help show test-param
>    Undefined show command: "test-param".  Try "help
> show".                             <- failure
> 
> I haven't found any specific system setup errors to explain the
> fails. 
> Please let me know if I can help further with these failures.  
> 
>                             Carl Love
> 
> 


  reply	other threads:[~2022-05-31 23:10 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-05-17 10:51 [PATCH 0/2] Improve help test for commands defined in Python Andrew Burgess
2022-05-17 10:51 ` [PATCH 1/2] gdb: use gdb::unique_xmalloc_ptr<char> for docs in cmdpy_init Andrew Burgess
2022-05-26 17:43   ` Tom Tromey
2022-05-28 10:10     ` Andrew Burgess
2022-05-17 10:51 ` [PATCH 2/2] gdb/python: improve formatting of help text for user defined commands Andrew Burgess
2022-05-17 12:19   ` Eli Zaretskii
2022-05-17 22:41   ` Lancelot SIX
2022-05-18 10:19   ` Andrew Burgess
2022-05-18 12:09     ` Eli Zaretskii
2022-05-26 17:46     ` Tom Tromey
2022-05-28 10:11       ` Andrew Burgess
2022-05-31 21:54         ` Carl Love
2022-05-31 23:10           ` Carl Love [this message]
2022-06-03 18:05             ` Carl Love
2022-06-06 11:40               ` [PUSHED] gdb/testsuite: add missing skip_python_tests call in py-doc-reformat.exp Andrew Burgess
2022-06-07 16:08                 ` Carl Love

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=73f20772cf68f5f966d82ff005b430a34f6ba1a3.camel@us.ibm.com \
    --to=cel@us.ibm.com \
    --cc=aburgess@redhat.com \
    --cc=gdb-patches@sourceware.org \
    --cc=tom@tromey.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).