public inbox for gcc-bugs@sourceware.org
help / color / mirror / Atom feed
* [Bug c++/114928] New: #pragma packed(push,1) should give the same warning as __attribute__((packed))
@ 2024-05-02 22:52 dhylands at gmail dot com
  2024-05-04  2:09 ` [Bug c++/114928] " egallager at gcc dot gnu.org
  0 siblings, 1 reply; 2+ messages in thread
From: dhylands at gmail dot com @ 2024-05-02 22:52 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114928

            Bug ID: 114928
           Summary: #pragma packed(push,1) should give the same warning as
                    __attribute__((packed))
           Product: gcc
           Version: 13.2.1
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c++
          Assignee: unassigned at gcc dot gnu.org
          Reporter: dhylands at gmail dot com
  Target Milestone: ---

Created attachment 58091
  --> https://gcc.gnu.org/bugzilla/attachment.cgi?id=58091&action=edit
File that compiles with no warnings

I'm using the ARM Toolchain
arm-gnu-toolchain-13.2.Rel1-x86_64-aarch64-none-linux-gnu/bin/aarch64-none-linux-gnu-g++

running on Ubuntu 20.04

If I delcare my structure using:

struct Foo {
    char y;
    int x;
} __attribute__((packed));

then the compiler correctly gives me a warning about taking the address of a
packed member:

warning: taking address of packed member of ‘Foo’ may result in an unaligned
pointer value [-Waddress-of-packed-member]

However, if I change the structure to use #pragma pack(push, 1) 

#pragma pack(push, 1)
struct Foo {
    char y;
    int x;
};
#pragma pack(pop)

then I don't get the warning (and my bug is that it should give the same
warning).

In either case the member offsets and structure size are the same.

^ permalink raw reply	[flat|nested] 2+ messages in thread

* [Bug c++/114928] #pragma packed(push,1) should give the same warning as __attribute__((packed))
  2024-05-02 22:52 [Bug c++/114928] New: #pragma packed(push,1) should give the same warning as __attribute__((packed)) dhylands at gmail dot com
@ 2024-05-04  2:09 ` egallager at gcc dot gnu.org
  0 siblings, 0 replies; 2+ messages in thread
From: egallager at gcc dot gnu.org @ 2024-05-04  2:09 UTC (permalink / raw)
  To: gcc-bugs

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=114928

Eric Gallager <egallager at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |egallager at gcc dot gnu.org
           See Also|                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=60972

--- Comment #1 from Eric Gallager <egallager at gcc dot gnu.org> ---
There are a number of other bugs open regarding inconsistencies between #pragma
pack and __attribute__((packed)); see for instance bug 60972 and related bugs
(not sure if this is a dup or even fully related, but it's at least worth
putting under "See Also"...)

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2024-05-04  2:09 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-05-02 22:52 [Bug c++/114928] New: #pragma packed(push,1) should give the same warning as __attribute__((packed)) dhylands at gmail dot com
2024-05-04  2:09 ` [Bug c++/114928] " egallager at gcc dot gnu.org

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).