vboot-utils fixes

vboot-utils depend on relatively recent host kernel headers, and fail to build on some of the autobuilders.

MTD_CHAR_MAJOR definition appears in linux-3.13, and ssize_t related failures probably happen with glibc from the same timeframe.

Sadly I cannot check it properly, I can only tell it does not break builds on newer systems. If possible, please try them on some older systems before applying.

If everything works well, I'd try to send 0002 upstream, but probably not 0001. It's an ugly fix and I don't think upstream cares much about older kernels.

vboot-utils: MTD_CHAR_MAJOR definition
vboot-utils: ssize_t needs unistd.h
vboot-utils: remove hash file

 package/vboot-utils/0001-mtd-char-major.patch | 17 +++++++++++++++++
 package/vboot-utils/0002-ssize-t.patch        | 14 ++++++++++++++
 package/vboot-utils/vboot-utils.hash          |  2 --
 3 files changed, 31 insertions(+), 2 deletions(-)
 create mode 100644 package/vboot-utils/0001-mtd-char-major.patch
 create mode 100644 package/vboot-utils/0002-ssize-t.patch
 delete mode 100644 package/vboot-utils/vboot-utils.hash

2.6.4

Fixes
http://autobuild.buildroot.net/results/a6360fd21314899321071c34fcb4892c5eed9bcc/
http://autobuild.buildroot.net/results/0af3aa4a429c45d3a2609913ec1040f32378f3c8/
http://autobuild.buildroot.net/results/cc4940fd6f51b31d71014f5aedee5223180eb36e/

---
 package/vboot-utils/0001-mtd-char-major.patch | 17 +++++++++++++++++
 1 file changed, 17 insertions(+)
 create mode 100644 package/vboot-utils/0001-mtd-char-major.patch

new file mode 100644
index 0000000..9db90b2
@@ -0,0 +1,17 @@
+This definition missing from <linux/major.h> for kernels <= 3.13.
+
+Signed-off-by: Alex Suykov <[email protected]>
+
+--- a/cgpt/cgpt_wrapper.c
++++ b/cgpt/cgpt_wrapper.c
+@@ -25,6 +25,10 @@
+ #include "cgpt_nor.h"
+ #include "cryptolib.h"
+ 
++#ifndef MTD_CHAR_MAJOR
++#define MTD_CHAR_MAJOR 90
++#endif
++
+ // Check if cmdline |argv| has "-D". "-D" signifies that GPT structs are stored
+ // off device, and hence we should not wrap around cgpt.
+ static bool has_dash_D(int argc, const char *const argv[]) {

2.6.4

Dear Alex Suykov,

Applied, thanks. It would have been nicer to have a Git formatted patch, though.

I'm not sure if it's possible, but if you could submit this patch to the upstream ChromiumOS project, it would be good.

Thanks!

Thomas

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Fixes
http://autobuild.buildroot.net/results/95c8708411fbb116f313a0fdeccacf92cf9e3d8d/
http://autobuild.buildroot.net/results/cf4aa60f9fc9c48ff13d3233994b77b7c0950afe/
http://autobuild.buildroot.net/results/6f7dcafde693f06dfc88c922c5f8dd70ae2e30b5/
http://autobuild.buildroot.net/results/a2e95784d32c8842f6947801fd0d8c48f68b84e2/
http://autobuild.buildroot.net/results/f69ae4c9eee27df43389ae76048fb4ddd5a3eca6/

---
 package/vboot-utils/0002-ssize-t.patch | 14 ++++++++++++++
 1 file changed, 14 insertions(+)
 create mode 100644 package/vboot-utils/0002-ssize-t.patch

new file mode 100644
index 0000000..9f1df45
@@ -0,0 +1,14 @@
+The code uses ssize_t which is usually defined in <unistd.h>
+
+Signed-off-by: Alex Suykov <[email protected]>
+
+--- a/host/lib/fmap.c
++++ a/host/lib/fmap.c
+@@ -6,6 +6,7 @@
+ 
+ #include <stdio.h>
+ #include <string.h>
++#include <unistd.h>
+ 
+ #include "fmap.h"
+ 

2.6.4

Dear Alex Suykov,

Also applied, with same comments: a Git formatted patch would have been better, and if you can please submit upstream.

Thanks!

Thomas

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

The package uses git to download the sources.

---
 package/vboot-utils/vboot-utils.hash | 2 --
 1 file changed, 2 deletions(-)
 delete mode 100644 package/vboot-utils/vboot-utils.hash

deleted file mode 100644
index e161423..0000000
@@ -1,2 +0,0 @@
-# Git shapshot
-none	xxx	bbdd62f9b030db7ad8eef789aaf58a7ff9a25656.tar.gz

2.6.4

Dear Alex Suykov,

I think it is good to have hash files in this case, even if they simply indicate "there's no hash". This way, we will be able in the future to make hash files mandatory.

I *think* this was part of the conclusion we had at the latest Buildroot Developers Meeting, even if the write up at
http://elinux.org/Buildroot:DeveloperDaysELCE2015 doesn't make this completely clear.

Thanks!

Thomas

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

I think we didn't get to a clear conclusion, except about the github hashes.

Regards,
Arnout

Arnout Vandecappelle                          arnout at mind be
Senior Embedded Software Architect            +32-16-286500
Essensium/Mind                                http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium           BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF

Arnout,

OK, so what do we do ?

Thomas

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

IIRC the main reason why we didn't fully go for the "add hash files for all
packages" approach was that we would need to add an empty hash file for about 300 packages. But then we discovered that github could have hash files, but we weren't entirely sure yet if that was really true.

So I think now we can safely conclude that a hash file should be added for all
packages, even if it just contains 'none' hashes. It is definitely an advantage to be able to distinguish packages that should still get a hash from packages of which we already decided that the hash should be 'none'.

Regards,
Arnout

Arnout Vandecappelle                          arnout at mind be
Senior Embedded Software Architect            +32-16-286500
Essensium/Mind                                http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium           BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint:  7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF

Arnout,

Agreed.

Thomas

Thomas Petazzoni, CTO, Free Electrons
Embedded Linux, Kernel and Android engineering
http://free-electrons.com

Ok, got it.

Does it make sense to update the file name in .hash, or is it ok to leave it as is since it's not used anyway? It's (version).tar.gz now but the git checkout gets saved as vboot-utils-(version).tar.gz.

Alex, All,

On 2016-01-01 16:52 +0200, Alex Suykov spake thusly:

Yes, you should update the file name, so that it matches the name the archive is saved as. Thanks for spotting the problem! :-)

Regards,
Yann E. MORIN.

.-----------------.--------------------.------------------.--------------------.
|  Yann E. MORIN  | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software  Designer | \ / CAMPAIGN     |  ___               |
| +33 223 225 172 `------------.-------:  X  AGAINST      |  \e/  There is no  |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL    |   v   conspiracy.  |
'------------------------------^-------^------------------^--------------------'

Hi,

Yes, makes sense to me.

Bye, Peter Korsgaard

please send all of them to us via gerrit
-mike
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 819 bytes
Desc: Digital signature
URL: <http://lists.busybox.net/pipermail/buildroot/attachments/20160102/80135f9b/attachment.asc>