From: "Martin Liška" <mliska@suse.cz>
To: Jan Hubicka <hubicka@ucw.cz>
Cc: Martin Jambor <mjambor@suse.cz>, gcc-patches@gcc.gnu.org
Subject: Re: [PATCH] ipa: silent -Wodr notes with -w
Date: Thu, 9 Feb 2023 08:44:05 +0100 [thread overview]
Message-ID: <08e947d9-2d38-974e-2a84-35a931919e75@suse.cz> (raw)
In-Reply-To: <Y+Q550rqR89C48Pb@kam.mff.cuni.cz>
On 2/9/23 01:10, Jan Hubicka wrote:
>>> On 2/1/23 15:26, Martin Jambor wrote:
>>>> Hi,
>>>>
>>>> On Fri, Dec 02 2022, Martin Liška wrote:
>>>>> If -w is used, warn_odr properly sets *warned = false and
>>>>> so it should be preserved when calling warn_types_mismatch.
>>>>>
>>>>> Noticed that during a LTO reduction where I used -w.
>>>>>
>>>>> Patch can bootstrap on x86_64-linux-gnu and survives regression tests.
>>>>>
>>>>> Ready to be installed?
>>>>> Thanks,
>>>>> Martin
>>>>>
>>>>> gcc/ChangeLog:
>>>>>
>>>>> * ipa-devirt.cc (odr_types_equivalent_p): Respect *warned
>>>>> value if set.
>>>>
>>>
>>> Hi.
>>>
>>>> Sorry for skipping this for so long, usually ODR stuff is... interesting
>>>> to the point I gladly leave it to Honza.
>>>
>>> Makes sense, however, he's not much active when it comes to patch review.
>>
>> Sorry, I was confused by the patch and delayed reply to figure out what
>> you are trying to fix. Indeed the dererence is missing here, however
>> every caller that sets warn to true should also set warned to non-NULL.
>> So indeed derefernce is missing, but I think the check for
>> warned == NULL should not be necessary.
>
> This seems to bootstrap with LTO. I am not sure what testcase you
> looked in, but unless there as a good reason to include the NULL check,
> I would rmeove it as it makes it harder to see what is going on.
Hi.
Thanks for the patch. Apparently, I noticed that during reduction of a test-case
where I used -w in order to silent all warnings.
So go ahead with your patch.
Martin
>
> honza
>
> diff --git a/gcc/ipa-devirt.cc b/gcc/ipa-devirt.cc
> index 14cf132c767..819860258d1 100644
> --- a/gcc/ipa-devirt.cc
> +++ b/gcc/ipa-devirt.cc
> @@ -1221,6 +1221,9 @@ odr_types_equivalent_p (tree t1, tree t2, bool warn, bool *warned,
> hash_set<type_pair> *visited,
> location_t loc1, location_t loc2)
> {
> + /* If we are asked to warn, we need warned to keep track if warning was
> + output. */
> + gcc_assert (!warn || warned);
> /* Check first for the obvious case of pointer identity. */
> if (t1 == t2)
> return true;
> @@ -1300,7 +1303,7 @@ odr_types_equivalent_p (tree t1, tree t2, bool warn, bool *warned,
> warn_odr (t1, t2, NULL, NULL, warn, warned,
> G_("it is defined as a pointer to different type "
> "in another translation unit"));
> - if (warn && (warned == NULL || *warned))
> + if (warn && *warned)
> warn_types_mismatch (TREE_TYPE (t1), TREE_TYPE (t2),
> loc1, loc2);
> return false;
> @@ -1315,7 +1318,7 @@ odr_types_equivalent_p (tree t1, tree t2, bool warn, bool *warned,
> warn_odr (t1, t2, NULL, NULL, warn, warned,
> G_("a different type is defined "
> "in another translation unit"));
> - if (warn && (warned == NULL || *warned))
> + if (warn && *warned)
> warn_types_mismatch (TREE_TYPE (t1), TREE_TYPE (t2), loc1, loc2);
> return false;
> }
> @@ -1333,7 +1336,7 @@ odr_types_equivalent_p (tree t1, tree t2, bool warn, bool *warned,
> warn_odr (t1, t2, NULL, NULL, warn, warned,
> G_("a different type is defined in another "
> "translation unit"));
> - if (warn && (warned == NULL || *warned))
> + if (warn && *warned)
> warn_types_mismatch (TREE_TYPE (t1), TREE_TYPE (t2), loc1, loc2);
> }
> gcc_assert (TYPE_STRING_FLAG (t1) == TYPE_STRING_FLAG (t2));
> @@ -1375,7 +1378,7 @@ odr_types_equivalent_p (tree t1, tree t2, bool warn, bool *warned,
> warn_odr (t1, t2, NULL, NULL, warn, warned,
> G_("has different return value "
> "in another translation unit"));
> - if (warn && (warned == NULL || *warned))
> + if (warn && *warned)
> warn_types_mismatch (TREE_TYPE (t1), TREE_TYPE (t2), loc1, loc2);
> return false;
> }
> @@ -1398,7 +1401,7 @@ odr_types_equivalent_p (tree t1, tree t2, bool warn, bool *warned,
> warn_odr (t1, t2, NULL, NULL, warn, warned,
> G_("has different parameters in another "
> "translation unit"));
> - if (warn && (warned == NULL || *warned))
> + if (warn && *warned)
> warn_types_mismatch (TREE_VALUE (parms1),
> TREE_VALUE (parms2), loc1, loc2);
> return false;
> @@ -1484,7 +1487,7 @@ odr_types_equivalent_p (tree t1, tree t2, bool warn, bool *warned,
> warn_odr (t1, t2, f1, f2, warn, warned,
> G_("a field of same name but different type "
> "is defined in another translation unit"));
> - if (warn && (warned == NULL || *warned))
> + if (warn && *warned)
> warn_types_mismatch (TREE_TYPE (f1), TREE_TYPE (f2), loc1, loc2);
> return false;
> }
prev parent reply other threads:[~2023-02-09 7:44 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-12-02 11:27 Martin Liška
2022-12-09 8:27 ` Martin Liška
2022-12-22 12:15 ` Martin Liška
2023-01-13 9:09 ` Martin Liška
2023-01-24 13:34 ` Martin Liška
2023-02-01 13:13 ` Martin Liška
2023-02-01 14:26 ` Martin Jambor
2023-02-01 14:44 ` Martin Liška
2023-02-07 0:35 ` Jan Hubicka
2023-02-09 0:10 ` Jan Hubicka
2023-02-09 7:44 ` Martin Liška [this message]
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=08e947d9-2d38-974e-2a84-35a931919e75@suse.cz \
--to=mliska@suse.cz \
--cc=gcc-patches@gcc.gnu.org \
--cc=hubicka@ucw.cz \
--cc=mjambor@suse.cz \
/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).