From: Tom de Vries <tdevries@suse.de>
To: Andrew Burgess <aburgess@redhat.com>, gdb-patches@sourceware.org
Subject: Re: [PATCH] gdb: improve error reporting for 'save gdb-index'
Date: Tue, 5 Dec 2023 09:25:13 +0100 [thread overview]
Message-ID: <f485a5bf-9a53-47e5-a66d-c7dc76fdcd6e@suse.de> (raw)
In-Reply-To: <3182b7a3eb3764cf62a8bbff1318808188fdf261.1701718143.git.aburgess@redhat.com>
On 12/4/23 20:29, Andrew Burgess wrote:
> While making recent changes to 'save gdb-index' command I triggered
> some errors -- of the kind a user might be expected to trigger if they
> do something wrong -- and I didn't find GDB's output as helpful as it
> might be.
>
> For example:
>
> $ gdb -q /tmp/hello.x
> ...
> (gdb) save gdb-index /non_existing_dir
> Error while writing index for `/tmp/hello': mkstemp: No such file or directory.
>
> That the error message mentions '/tmp/hello', which does exit, but
exit -> exist
> doesn't mention '/non_existing_dir', which doesn't is, I think,
> confusing.
>
Agreed.
> Also, I find the 'mkstemp' in the error message confusing for a user
> facing error. A user might not know what mkstemp means, and even if
> they do, that it appears in the error message is an internal GDB
> detail. The user doesn't care what function failed, why want to know
why want -> but wants ?
> what was wrong with their input, and what they should do to fix
> things.
>
> Similarly, for a directory that does exist, but can't be written to:
>
> (gdb) save gdb-index /no_access_dir
> Error while writing index for `/tmp/hello': mkstemp: Permission denied.
>
> In this case, the 'Permission denied' might make the user thing there
> is a permissions issue with '/tmp/hello', which is not the case.
>
> After this patch, the new errors are:
>
> (gdb) save gdb-index /non_existing_dir
> Error while writing index for `/tmp/hello': `/non_existing_dir': No such file or directory.
>
> and:
>
> (gdb) save gdb-index /no_access_dir
> Error while writing index for `/tmp/hello': `/no_access_dir': Permission denied.
>
> we also have:
>
> (gdb) save gdb-index /tmp/not_a_directory
> Error while writing index for `/tmp/hello': `/tmp/not_a_directory': Is not a directory.
>
> I think these do a better job of guiding the user towards fixing the
> problem.
>
Agreed, I think this is a nice improvement.
> I've added a new test that exercises all of these cases, and also
> checks the case where a user tries to use an executable that already
> contains an index in order to generate an index.
>
> Nothing that worked correctly before this patch should give an error
> after this patch; I've only changed the output when the user was going
> to get an error anyway.
> ---
> gdb/dwarf2/index-write.c | 10 +-
> gdb/testsuite/gdb.base/gdb-index-err.c | 22 +++++
> gdb/testsuite/gdb.base/gdb-index-err.exp | 118 +++++++++++++++++++++++
> 3 files changed, 149 insertions(+), 1 deletion(-)
> create mode 100644 gdb/testsuite/gdb.base/gdb-index-err.c
> create mode 100644 gdb/testsuite/gdb.base/gdb-index-err.exp
>
> diff --git a/gdb/dwarf2/index-write.c b/gdb/dwarf2/index-write.c
> index b4a0117330e..4e5490e734e 100644
> --- a/gdb/dwarf2/index-write.c
> +++ b/gdb/dwarf2/index-write.c
> @@ -1569,6 +1569,13 @@ struct index_wip_file
> index_wip_file (const char *dir, const char *basename,
> const char *suffix)
> {
> + /* Validate DIR is a valid directory. */
double space.
> + struct stat buf;
> + if (stat (dir, &buf) == -1)
> + perror_with_name (string_printf (_("`%s'"), dir).c_str ());
> + if ((buf.st_mode & S_IFDIR) != S_IFDIR)
> + error (_("`%s': Is not a directory."), dir);
> +
> filename = (std::string (dir) + SLASH_STRING + basename
> + suffix);
>
> @@ -1577,7 +1584,8 @@ struct index_wip_file
> scoped_fd out_file_fd = gdb_mkostemp_cloexec (filename_temp.data (),
> O_BINARY);
> if (out_file_fd.get () == -1)
> - perror_with_name (("mkstemp"));
> + perror_with_name (string_printf (_("couldn't open `%s'"),
> + filename_temp.data ()).c_str ());
>
> out_file = out_file_fd.to_file ("wb");
>
> diff --git a/gdb/testsuite/gdb.base/gdb-index-err.c b/gdb/testsuite/gdb.base/gdb-index-err.c
> new file mode 100644
> index 00000000000..3a264f239ed
> --- /dev/null
> +++ b/gdb/testsuite/gdb.base/gdb-index-err.c
> @@ -0,0 +1,22 @@
> +/* This testcase is part of GDB, the GNU debugger.
> +
> + Copyright 2023 Free Software Foundation, Inc.
> +
> + This program is free software; you can redistribute it and/or modify
> + it under the terms of the GNU General Public License as published by
> + the Free Software Foundation; either version 3 of the License, or
> + (at your option) any later version.
> +
> + This program is distributed in the hope that it will be useful,
> + but WITHOUT ANY WARRANTY; without even the implied warranty of
> + MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> + GNU General Public License for more details.
> +
> + You should have received a copy of the GNU General Public License
> + along with this program. If not, see <http://www.gnu.org/licenses/>. */
> +
> +int
> +main ()
> +{
> + return 0;
> +}
> diff --git a/gdb/testsuite/gdb.base/gdb-index-err.exp b/gdb/testsuite/gdb.base/gdb-index-err.exp
> new file mode 100644
> index 00000000000..9493c3aafd2
> --- /dev/null
> +++ b/gdb/testsuite/gdb.base/gdb-index-err.exp
> @@ -0,0 +1,118 @@
> +# Copyright 2023 Free Software Foundation, Inc.
> +#
> +# This program is free software; you can redistribute it and/or modify
> +# it under the terms of the GNU General Public License as published by
> +# the Free Software Foundation; either version 3 of the License, or
> +# (at your option) any later version.
> +#
> +# This program is distributed in the hope that it will be useful,
> +# but WITHOUT ANY WARRANTY; without even the implied warranty of
> +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
> +# GNU General Public License for more details.
> +#
> +# You should have received a copy of the GNU General Public License
> +# along with this program. If not, see <http://www.gnu.org/licenses/>.
> +
> +# Test some error messages that can arise from the 'save gdb-index'
> +# command.
> +
> +standard_testfile
> +
> +if {[build_executable "build executable" $testfile $srcfile] == -1} {
> + return -1
> +}
> +
> +clean_restart $binfile
> +
I think the usual shorthand for this is prepare_for_testing.
> +# This test isn't going to work when the board file automatically adds
> +# an index section, or if the debug information is split into a
> +# separate objfile.
> +set has_cooked_index false
> +gdb_test_multiple "maint print objfiles ${binfile}" "check debug style" -lbl {
> + -re "\r\n\\.gdb_index: version ${decimal}(?=\r\n)" {
> + gdb_test_lines "" $gdb_test_name ".*"
> + }
> + -re "\r\n\\.debug_names: exists(?=\r\n)" {
> + gdb_test_lines "" $gdb_test_name ".*"
> + }
> + -re "\r\n(Cooked index in use:|Psymtabs)(?=\r\n)" {
> + set has_cooked_index true
> + gdb_test_lines "" $gdb_test_name ".*"
> + }
> + -re ".gdb_index: faked for \"readnow\"" {
> + gdb_test_lines "" $gdb_test_name ".*"
> + }
> + -re -wrap "" {
> + # Have no debug information!
> + }
> +}
> +
I think ideally we do this kind of checks using a factored out function
in gdb.exp. There may already be one, I haven't checked. If not, maybe
consider moving it there.
Anyway, LGTM.
Reviewed-By: Tom de Vries <tdevries@suse.de>
Thanks,
- Tom
> +if { !$has_cooked_index } {
> + unsupported "cannot test without a cooked index"
> + return -1
> +}
> +
> +
> +# The name of a directory that doesn't exist.
> +set bad_dir [standard_output_file "non-existent"]
> +
> +# Try to write the index into a non-existent directory.
> +foreach_with_prefix flag { "" "-dwarf-5" } {
> + gdb_test "save gdb-index ${flag} ${bad_dir}" \
> + "Error while writing index for `[string_to_regexp $binfile]': `[string_to_regexp $bad_dir]': No such file or directory\\." \
> + "try to write index to non-existent directory"
> +}
> +
> +# Create a text-file.
> +set text_file [standard_output_file "text-file"]
> +set fd [open $text_file w]
> +puts $fd "A line of text.\n"
> +close $fd
> +
> +# Try to write the index into something that is not a directory.
> +foreach_with_prefix flag { "" "-dwarf-5" } {
> + gdb_test "save gdb-index ${flag} ${text_file}" \
> + "Error while writing index for `[string_to_regexp $binfile]': `[string_to_regexp $text_file]': Is not a directory\\." \
> + "try to write index to something that's not a directory"
> +}
> +
> +# Create a directory which can't be written too.
> +set non_writable_dir [standard_output_file "private"]
> +remote_exec host "mkdir -p ${non_writable_dir}"
> +remote_exec host "chmod u-w,g-w ${non_writable_dir}"
> +
> +# Try to write the index into a non-writable directory.
> +foreach_with_prefix flag { "" "-dwarf-5" } {
> + gdb_test "save gdb-index ${flag} ${non_writable_dir}" \
> + "Error while writing index for `[string_to_regexp $binfile]': couldn't open `[string_to_regexp $non_writable_dir]/${gdb_test_file_name}.*': Permission denied\\." \
> + "try to write index to a non-writable directory"
> +}
> +
> +# Create copies of the executable, we will add an index section to
> +# each of these.
> +remote_exec host "cp $binfile ${binfile}.gdb_index"
> +remote_exec host "cp $binfile ${binfile}.dwarf_5"
> +
> +# Create a directory in which we can try to generate the index files.
> +set already_indexed_dir [standard_output_file "already_indexed"]
> +remote_exec host "mkdir -p $already_indexed_dir"
> +
> +foreach_with_prefix flag { "" "-dwarf-5" } {
> + if { $flag eq "" } {
> + set extension "gdb_index"
> + } else {
> + set extension "dwarf_5"
> + }
> +
> + # Add the index section to the executable.
> + clean_restart ${binfile}.${extension}
> + gdb_assert {[ensure_gdb_index ${binfile}.${extension} ${flag}] == 1} \
> + "add index to executable"
> +
> + # Reload the executable (which now has an index), and try to
> + # generate and index from it. This will fail.
> + clean_restart ${binfile}.${extension}
> + gdb_test "save gdb-index ${flag} $already_indexed_dir" \
> + "Error while writing index for `[string_to_regexp $binfile.$extension]': Cannot use an index to create the index" \
> + "try to generate an index from a binary with an index"
> +}
>
> base-commit: 2c4caca9873ef4e97a51a538dfbecde8fe334082
next prev parent reply other threads:[~2023-12-05 8:25 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-12-04 19:29 Andrew Burgess
2023-12-05 8:25 ` Tom de Vries [this message]
2023-12-08 16:06 ` [PATCHv2] " Andrew Burgess
2023-12-08 16:26 ` Tom Tromey
2023-12-12 14:56 ` [PATCHv3] " Andrew Burgess
2023-12-12 20:21 ` Tom Tromey
2023-12-13 8:56 ` Andrew Burgess
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=f485a5bf-9a53-47e5-a66d-c7dc76fdcd6e@suse.de \
--to=tdevries@suse.de \
--cc=aburgess@redhat.com \
--cc=gdb-patches@sourceware.org \
/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).