From: Matthias Maennich <maennich@google.com>
To: Giuliano Procida <gprocida@google.com>
Cc: libabigail@sourceware.org, dodji@seketeli.org, kernel-team@android.com
Subject: Re: [PATCH v2] Fix has_net_changes for --leaf-changes-only mode.
Date: Mon, 30 Mar 2020 11:17:54 +0200 [thread overview]
Message-ID: <20200330091754.GH101337@google.com> (raw)
In-Reply-To: <20200327190135.98611-1-gprocida@google.com>
On Fri, Mar 27, 2020 at 07:01:35PM +0000, Giuliano Procida wrote:
>This function was not aware of --leaf-changes-only mode.
>
> - Stats counters for changed variables and types have
> different names in the different modes.
> - Leaf type changes were not included in leaf mode.
>
>For some inputs, this resulted in abidiff producing an empty report but
>returning a non-zero exit status in --leaf-changes-only mode.
>
>For other inputs, including some existing tests, the combination of
>both issues still resulted in the correct return code.
>
>This patch makes has_net_changes mirror emit_diff_stats, modulo flags
>like --non-reachable-types which if absent can still result in
>discrepancies between output and return code.
>
> * src/abg-comparison.cc
> (corpus_diff::priv::emit_diff_stats): Code whitespace.
> (corpus_diff::has_net_changes): Reorder the logic to match
> emit_diff_stats, add a note about this; in --leaf-changes-only
> mode use the right counters for function and variable changes
> and include type changes.
>
>Signed-off-by: Giuliano Procida <gprocida@google.com>
Would feel better to have some tests covering this along with the patch.
But this still looks correct to me.
Reviewed-by: Matthias Maennich <maennich@google.com>
Cheers,
Matthias
>---
> src/abg-comparison.cc | 30 ++++++++++++++++++------------
> 1 file changed, 18 insertions(+), 12 deletions(-)
>
>diff --git a/src/abg-comparison.cc b/src/abg-comparison.cc
>index 46bf9e30..50791cb5 100644
>--- a/src/abg-comparison.cc
>+++ b/src/abg-comparison.cc
>@@ -9932,7 +9932,7 @@ corpus_diff::priv::emit_diff_stats(const diff_stats& s,
> out << " (" << s.num_changed_func_filtered_out() << " filtered out)";
> out << ", ";
>
>- out << s.net_num_func_added()<< " Added";
>+ out << s.net_num_func_added() << " Added";
> if (s.num_added_func_filtered_out())
> out << " (" << s.num_added_func_filtered_out() << " filtered out)";
> if (total_nb_function_changes <= 1)
>@@ -10560,7 +10560,7 @@ corpus_diff::has_incompatible_changes() const
> || stats.net_num_func_removed() != 0
> || (stats.num_func_with_virtual_offset_changes() != 0
> // If all reports about functions with sub-type changes
>- // have been suppressd, then even those about functions
>+ // have been suppressed, then even those about functions
> // that are virtual don't matter anymore because the
> // user willingly requested to shut them down
> && stats.net_num_func_changed() != 0)
>@@ -10602,21 +10602,27 @@ corpus_diff::has_net_changes() const
> const diff_stats& stats = const_cast<corpus_diff*>(this)->
> apply_filters_and_suppressions_before_reporting();
>
>+ // Logic here should match emit_diff_stats.
>+ // TODO: Possibly suppress things that won't be shown there.
>+ bool leaf = context()->show_leaf_changes_only();
> return (architecture_changed()
> || soname_changed()
>- || stats.net_num_func_changed()
>- || stats.net_num_vars_changed()
>- || stats.net_num_func_added()
>- || stats.net_num_added_func_syms()
> || stats.net_num_func_removed()
>- || stats.net_num_removed_func_syms()
>- || stats.net_num_vars_added()
>- || stats.net_num_added_var_syms()
>+ || (leaf && stats.num_leaf_type_changes())
>+ || (leaf ? stats.net_num_leaf_func_changes()
>+ : stats.net_num_func_changed())
>+ || stats.net_num_func_added()
> || stats.net_num_vars_removed()
>- || stats.net_num_removed_var_syms()
>- || stats.net_num_added_unreachable_types()
>+ || (leaf ? stats.net_num_leaf_var_changes()
>+ : stats.net_num_vars_changed())
>+ || stats.net_num_vars_added()
> || stats.net_num_removed_unreachable_types()
>- || stats.net_num_changed_unreachable_types());
>+ || stats.net_num_changed_unreachable_types()
>+ || stats.net_num_added_unreachable_types()
>+ || stats.net_num_removed_func_syms()
>+ || stats.net_num_added_func_syms()
>+ || stats.net_num_removed_var_syms()
>+ || stats.net_num_added_var_syms());
> }
>
> /// Apply the different filters that are registered to be applied to
>--
>2.26.0.rc2.310.g2932bb562d-goog
>
next prev parent reply other threads:[~2020-03-30 9:17 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-24 17:13 [RFC] Fix has_net_changes for leaf-changes-only mode Giuliano Procida
2020-03-24 17:17 ` Giuliano Procida
2020-03-26 10:37 ` Matthias Maennich
2020-03-26 12:01 ` Giuliano Procida
2020-03-26 17:12 ` Dodji Seketeli
2020-03-27 18:42 ` Giuliano Procida
2020-03-27 19:01 ` [PATCH v2] Fix has_net_changes for --leaf-changes-only mode Giuliano Procida
2020-03-30 9:17 ` Matthias Maennich [this message]
2020-07-01 13:19 ` [PATCH v3 0/1] Fix abidiff exit code when diffs are suppressed Giuliano Procida
2020-07-01 13:19 ` [PATCH v3 1/1] Fix has_net_changes for --leaf-changes-only mode Giuliano Procida
2020-07-17 13:54 ` Dodji Seketeli
2020-07-17 16:34 ` Giuliano Procida
2020-07-21 6:28 ` Dodji Seketeli
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=20200330091754.GH101337@google.com \
--to=maennich@google.com \
--cc=dodji@seketeli.org \
--cc=gprocida@google.com \
--cc=kernel-team@android.com \
--cc=libabigail@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).