public inbox for u-boot@lists.denx.de
 help / color / mirror / Atom feed
From: AKASHI Takahiro <takahiro.akashi@linaro.org>
To: Ilias Apalodimas <ilias.apalodimas@linaro.org>
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de>,
	trini@konsulko.com, u-boot@lists.denx.de
Subject: Re: [PATCH v2] efi_loader: Fix warnings for unaligned accesses
Date: Fri, 12 May 2023 09:05:05 +0900	[thread overview]
Message-ID: <20230512000505.GA8438@laputa> (raw)
In-Reply-To: <CAC_iWjJki2iMZ9BZ1W6q837Zz552weMSEYn4oZTs9Vq+6Sw7ww@mail.gmail.com>

Hi Ilias,

On Thu, May 11, 2023 at 08:00:32PM +0300, Ilias Apalodimas wrote:
> Hi Heinrich,
> 
> On Thu, 11 May 2023 at 19:56, Heinrich Schuchardt <xypron.glpk@gmx.de> wrote:
> >
> > On 5/11/23 18:40, Ilias Apalodimas wrote:
> > > Tom reports that when building with clang we see this warning:
> > > field guid within 'struct efi_hii_keyboard_layout' is less aligned than 'efi_guid_t' and is usually due to 'struct efi_hii_keyboard_layout' being packed, which can lead to unaligned accesses [-Wunaligned-access]
> > >
> > > This happens because 'struct efi_hii_keyboard_layout' is defined as
> > > packed and thus has 1-byte alignment but efi_guid_t is a type that
> > > requires greater alignment than that.
> > >
> > > However the EFI spec describes the EFI_GUID as
> > > "128-bit buffer containing a unique identifier value.
> > > Unless otherwise specified"
> > >
> > > So convert the efi_guid_t -> u8 b[16] here and skip the alignment
> > > requirements.  Since the struct is packed to begin with, it makes no
> > > difference on the final memory layout.
> > >
> > > Suggested-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
> > > Reported-by: Tom Rini <trini@konsulko.com>
> > > Signed-off-by: Ilias Apalodimas <ilias.apalodimas@linaro.org>
> > > ---
> > > Changes since v1:
> > > - Adjust the commit message and add a comment on why this happens
> > >
> > >   include/efi_api.h | 28 +++++++++++++++++++++++++++-
> > >   1 file changed, 27 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/include/efi_api.h b/include/efi_api.h
> > > index 2fd0221c1c77..55a4c989fc7c 100644
> > > --- a/include/efi_api.h
> > > +++ b/include/efi_api.h
> > > @@ -1170,7 +1170,33 @@ struct efi_key_descriptor {
> > >
> > >   struct efi_hii_keyboard_layout {
> > >       u16 layout_length;
> > > -     efi_guid_t guid;
> > > +     /*
> > > +      * The EFI spec defines this as efi_guid_t.
> > > +      * clang and gcc both report alignment problems here.
> > > +      * clang with -Wunaligned-access
> > > +      * warning: field guid within 'struct efi_hii_keyboard_layout' is less
> > > +      * aligned than 'efi_guid_t' and is usually due to
> > > +      * 'struct efi_hii_keyboard_layout' being packed, which can lead to
> > > +      * unaligned accesses
> > > +      *
> > > +      * GCC with -Wpacked-not-aligned -Waddress-of-packed-member
> > > +      * 'efi_guid_t' offset 2 in 'struct efi_hii_keyboard_layout'
> > > +      * isn't aligned to 4
> > > +      *
> > > +      * Removing the alignment from efi_guid_t is not an option, since
> > > +      * it is also used in non-packed structs and that would break
> > > +      * calculations with offsetof
> > > +      *
> > > +      * This is the only place we get a report for. That happens because
> > > +      * all other declarations of efi_guid_t within a packed struct happens
> > > +      * to be 4-byte aligned.  i.e a u32, a u64 a 2 * u16 or any combination
> > > +      * that ends up landing efi_guid_t on a 4byte boundary precedes.
> > > +      *
> > > +      * Replace this with a 1-byte aligned counterpart of b[16].  This is a
> > > +      * packed struct so the memory  placement of efi_guid_t should not change
> > > +      *
> > > +      */
> > > +     u8 guid[16];

I thought that you have agreed with my comment, saying keep "efi_guid_t" here.
https://lists.denx.de/pipermail/u-boot/2023-April/515831.html

-Takahiro Akashi


> > >       u32 layout_descriptor_string_offset;
> > >       u8 descriptor_count;
> > >       /* struct efi_key_descriptor descriptors[]; follows here */
> > > --
> > > 2.39.2
> > >
> >
> > Thank you for investigating this in depth.
> 
> yw :)
> 
> >
> > Commit messages should preferably limited to 75 characters per line, see
> > scripts/checkpatch.pl. No need to resubmit.
> 
> Yea I know, I just prefered to keep the clang warning intact in the
> commit message
> 
> Thanks
> /Ilias
> >
> > Reviewed-by: Heinrich Schuchardt <xypron.glpk@gmx.de>

  reply	other threads:[~2023-05-12  0:05 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-05-11 16:40 [PATCH v2] efi_loader: Fix warnings for unaligned accesses Ilias Apalodimas
2023-05-11 16:55 ` Heinrich Schuchardt
2023-05-11 17:00   ` Ilias Apalodimas
2023-05-12  0:05     ` AKASHI Takahiro [this message]
2023-05-12  5:28       ` Ilias Apalodimas

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=20230512000505.GA8438@laputa \
    --to=takahiro.akashi@linaro.org \
    --cc=ilias.apalodimas@linaro.org \
    --cc=trini@konsulko.com \
    --cc=u-boot@lists.denx.de \
    --cc=xypron.glpk@gmx.de \
    /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