Guys,
Please pull the following compiler warning fix. Sorry, I've been pretty
slow in getting this pull request sent. Multiple people have reported
hitting it and I've now received 4 patches for the same warning,
http://article.gmane.org/gmane.linux.kernel.efi/4024
http://article.gmane.org/gmane.linux.kernel.efi/4017
http://article.gmane.org/gmane.linux.kernel.efi/3957
http://article.gmane.org/gmane.linux.kernel.efi/3930
The following changes since commit 7171511eaec5bf23fb06078f59784a3a0626b38f:
Linux 3.16-rc1 (2014-06-15 17:45:28 -1000)
are available in the git repository at:
git://git.kernel.org/pub/scm/linux/kernel/git/mfleming/efi.git tags/efi-urgent
for you to fetch changes up to 6fb8cc82c096fd5ccf277678639193cae07125a0:
efi: Fix compiler warnings (unused, const, type) (2014-06-19 15:03:05 +0100)
----------------------------------------------------------------
* Fix a few compiler warnings in the arm64 EFI code that lots of people
are running into and reporting - Catalin Marinas
----------------------------------------------------------------
Catalin Marinas (1):
efi: Fix compiler warnings (unused, const, type)
drivers/firmware/efi/efi.c | 6 +++---
drivers/firmware/efi/fdt.c | 2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
--
Matt Fleming, Intel Open Source Technology Center
On Fri, Jun 20, 2014 at 10:00:08AM +0100, Matt Fleming wrote:
> Please pull the following compiler warning fix. Sorry, I've been pretty
> slow in getting this pull request sent. Multiple people have reported
> hitting it and I've now received 4 patches for the same warning,
BTW, one of them is not just a simple warning but a bug. Pointer to
"long" variable on the stack passed to a function that only takes a
pointer to "int". We are probably lucky that we don't hit the bug on
arm64.
Anyway, thanks for pushing the fix.
(and lesson learnt not to trust the ARM EFI_STUB developers, won't name
them, with properly testing their code ;))
--
Catalin
On Fri, Jun 20, 2014 at 10:28:48AM +0100, Catalin Marinas wrote:
> On Fri, Jun 20, 2014 at 10:00:08AM +0100, Matt Fleming wrote:
> > Please pull the following compiler warning fix. Sorry, I've been pretty
> > slow in getting this pull request sent. Multiple people have reported
> > hitting it and I've now received 4 patches for the same warning,
>
> BTW, one of them is not just a simple warning but a bug. Pointer to
> "long" variable on the stack passed to a function that only takes a
> pointer to "int". We are probably lucky that we don't hit the bug on
> arm64.
>
> Anyway, thanks for pushing the fix.
>
> (and lesson learnt not to trust the ARM EFI_STUB developers, won't name
> them, with properly testing their code ;))
The code was tested and showed no runtime problems - but I did manage
to miss the warnings.
I did post an alternative patch for one half of this a week before
anyone else, but that one seems to have been ignored, even by gmane:
http://www.spinics.net/lists/linux-efi/msg03924.html
That form will still be required for the 32-bit arm support, since
strncmp is not available in the zImage.
/
Leif
On Fri, 20 Jun, at 02:39:19PM, Leif Lindholm wrote:
>
> I did post an alternative patch for one half of this a week before
> anyone else, but that one seems to have been ignored, even by gmane:
> http://www.spinics.net/lists/linux-efi/msg03924.html
Sorry Leif, I don't know why I didn't pick that one up. I do see it in
my mail archive. Chalk it up to human error.
> That form will still be required for the 32-bit arm support, since
> strncmp is not available in the zImage.
Are you planning on incorporating this fix into the 32-bit arm support
patch series
--
Matt Fleming, Intel Open Source Technology Center
On Fri, Jun 20, 2014 at 02:49:38PM +0100, Matt Fleming wrote:
> On Fri, 20 Jun, at 02:39:19PM, Leif Lindholm wrote:
> >
> > I did post an alternative patch for one half of this a week before
> > anyone else, but that one seems to have been ignored, even by gmane:
> > http://www.spinics.net/lists/linux-efi/msg03924.html
>
> Sorry Leif, I don't know why I didn't pick that one up. I do see it in
> my mail archive. Chalk it up to human error.
Well, it seemed to manage to trick at least one computer too...
> > That form will still be required for the 32-bit arm support, since
> > strncmp is not available in the zImage.
>
> Are you planning on incorporating this fix into the 32-bit arm support
> patch series
Yeah, I'll do that.
Just as soon as I get back off holiday :)
/
Leif
On Fri, 20 Jun, at 10:00:08AM, Matt Fleming wrote:
> Guys,
>
> Please pull the following compiler warning fix. Sorry, I've been pretty
> slow in getting this pull request sent. Multiple people have reported
> hitting it and I've now received 4 patches for the same warning,
>
> http://article.gmane.org/gmane.linux.kernel.efi/4024
> http://article.gmane.org/gmane.linux.kernel.efi/4017
> http://article.gmane.org/gmane.linux.kernel.efi/3957
> http://article.gmane.org/gmane.linux.kernel.efi/3930
>
> The following changes since commit 7171511eaec5bf23fb06078f59784a3a0626b38f:
>
> Linux 3.16-rc1 (2014-06-15 17:45:28 -1000)
>
> are available in the git repository at:
>
> git://git.kernel.org/pub/scm/linux/kernel/git/mfleming/efi.git tags/efi-urgent
>
> for you to fetch changes up to 6fb8cc82c096fd5ccf277678639193cae07125a0:
>
> efi: Fix compiler warnings (unused, const, type) (2014-06-19 15:03:05 +0100)
>
> ----------------------------------------------------------------
> * Fix a few compiler warnings in the arm64 EFI code that lots of people
> are running into and reporting - Catalin Marinas
Ping? This doesn't seem to have been picked up yet?
--
Matt Fleming, Intel Open Source Technology Center