From: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
To: linux-sparse@vger.kernel.org
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
Subject: [PATCH 13/13] flex-array: warn when a flexible array member has some padding
Date: Thu, 1 Oct 2020 01:18:28 +0200 [thread overview]
Message-ID: <20200930231828.66751-14-luc.vanoostenryck@gmail.com> (raw)
In-Reply-To: <20200930231828.66751-1-luc.vanoostenryck@gmail.com>
If some padding is added because of the presence of a flexible
array member, the size of the structure will be greater than
the offset of this flexible array which can cause some
problems if the assumption is made that these 2 size must be
identical (which is easy to do since such flexible arrays are
conceptually 'after' the structure itself).
Signed-off-by: Luc Van Oostenryck <luc.vanoostenryck@gmail.com>
---
options.c | 2 ++
options.h | 1 +
sparse.1 | 6 ++++++
symbol.c | 2 ++
validation/flex-array-padding.c | 1 -
5 files changed, 11 insertions(+), 1 deletion(-)
diff --git a/options.c b/options.c
index b46900b973a6..92d64e6ab64e 100644
--- a/options.c
+++ b/options.c
@@ -102,6 +102,7 @@ int Wenum_mismatch = 1;
int Wexternal_function_has_definition = 1;
int Wflexible_array_array = 1;
int Wflexible_array_nested = 0;
+int Wflexible_array_padding = 0;
int Wflexible_array_sizeof = 0;
int Wimplicit_int = 1;
int Winit_cstring = 0;
@@ -845,6 +846,7 @@ static const struct flag warnings[] = {
{ "external-function-has-definition", &Wexternal_function_has_definition },
{ "flexible-array-array", &Wflexible_array_array },
{ "flexible-array-nested", &Wflexible_array_nested },
+ { "flexible-array-padding", &Wflexible_array_padding },
{ "flexible-array-sizeof", &Wflexible_array_sizeof },
{ "implicit-int", &Wimplicit_int },
{ "init-cstring", &Winit_cstring },
diff --git a/options.h b/options.h
index d23ed472eaac..31d9c5859977 100644
--- a/options.h
+++ b/options.h
@@ -101,6 +101,7 @@ extern int Wenum_mismatch;
extern int Wexternal_function_has_definition;
extern int Wflexible_array_array;
extern int Wflexible_array_nested;
+extern int Wflexible_array_padding;
extern int Wflexible_array_sizeof;
extern int Wimplicit_int;
extern int Winit_cstring;
diff --git a/sparse.1 b/sparse.1
index 9b1a59c6b9d4..8c61e869dc57 100644
--- a/sparse.1
+++ b/sparse.1
@@ -268,6 +268,12 @@ Sparse issues these warnings by default. To turn them off, use
Warn about structures containing a flexible array being contained into
another structure, union or array.
+Sparse does not issue these warnings by default.
+.
+.TP
+.B -Wflexible-array-padding
+Warn about padding alignments caused by the presence of a flexible array member.
+
Sparse does not issue these warnings by default.
.
.TP
diff --git a/symbol.c b/symbol.c
index a9f646eb053f..eabb2226651f 100644
--- a/symbol.c
+++ b/symbol.c
@@ -213,6 +213,8 @@ static struct symbol * examine_struct_union_type(struct symbol *sym, int advance
}
if (info.flex_array) {
info.has_flex_array = 1;
+ if (sym->offset != bits_to_bytes(bit_size) && Wflexible_array_padding)
+ warning(info.flex_array->pos, "flexible array member has padding");
}
if (info.has_flex_array)
sym->has_flex_array = 1;
diff --git a/validation/flex-array-padding.c b/validation/flex-array-padding.c
index 2ba77971266e..95b349e1199a 100644
--- a/validation/flex-array-padding.c
+++ b/validation/flex-array-padding.c
@@ -12,7 +12,6 @@ static int foo(struct s *s)
/*
* check-name: flex-array-padding
* check-command: test-linearize -Wflexible-array-padding $file
- * check-known-to-fail
*
* check-output-ignore
*
--
2.28.0
next prev parent reply other threads:[~2020-09-30 23:18 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-09-30 23:18 [PATCH 00/13] add warnings for flexible arrays Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 01/13] flex-array: add testcases Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 02/13] flex-array: factor out common part of lay_out_{struct,union}() Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 03/13] flex-array: do not lay out invalid struct members Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 04/13] flex-array: flexible array members have zero size and alignment is OK Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 05/13] flex-array: detect structures with a flexible array member Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 06/13] flex-array: warn on flexible arrays in unions Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 07/13] flex-array: warn if flexible array is not last Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 08/13] flex-array: identify structures with a flexible array member Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 09/13] flex-array: add helper has_flexible_array() Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 10/13] flex-array: warn when using sizeof() on a flexible array Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 11/13] flex-array: warn an arrays containing " Luc Van Oostenryck
2020-09-30 23:18 ` [PATCH 12/13] flex-array: warn on flexible array in nested aggregate types Luc Van Oostenryck
2020-09-30 23:18 ` Luc Van Oostenryck [this message]
2020-10-01 16:34 ` [PATCH 13/13] flex-array: warn when a flexible array member has some padding Linus Torvalds
2020-10-01 19:17 ` Luc Van Oostenryck
2020-10-01 19:27 ` Linus Torvalds
2020-10-01 19:41 ` Luc Van Oostenryck
2020-10-01 19:51 ` Luc Van Oostenryck
2020-10-01 16:36 ` [PATCH 00/13] add warnings for flexible arrays Linus Torvalds
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=20200930231828.66751-14-luc.vanoostenryck@gmail.com \
--to=luc.vanoostenryck@gmail.com \
--cc=linux-sparse@vger.kernel.org \
--cc=torvalds@linux-foundation.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 NNTP newsgroup(s).