[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Gustavo Zacarias]

Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
---
 package/libtasn1/libtasn1.hash | 2 +-
 package/libtasn1/libtasn1.mk   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/package/libtasn1/libtasn1.hash b/package/libtasn1/libtasn1.hash
index 653e018..41e10a6 100644
--- a/package/libtasn1/libtasn1.hash
+++ b/package/libtasn1/libtasn1.hash
@@ -1,2 +1,2 @@
 # Locally calculated after checking pgp signature
-sha256	89b3b5dce119273431544ecb305081f3530911001bb12e5d76588907edb71bda	libtasn1-4.5.tar.gz
+sha256	3462fc25e2d2536878c39a8825f5e36ba2e2611b27ef535e064f4c56258e508b	libtasn1-4.6.tar.gz
diff --git a/package/libtasn1/libtasn1.mk b/package/libtasn1/libtasn1.mk
index eeff76c..6818f30 100644
--- a/package/libtasn1/libtasn1.mk
+++ b/package/libtasn1/libtasn1.mk
@@ -4,7 +4,7 @@
 #
 ################################################################################
 
-LIBTASN1_VERSION = 4.5
+LIBTASN1_VERSION = 4.6
 LIBTASN1_SITE = http://ftp.gnu.org/gnu/libtasn1
 LIBTASN1_DEPENDENCIES = host-bison
 LIBTASN1_LICENSE = GPLv3+ LGPLv2.1+
-- 
2.4.6

[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Vicente Olivert Riera]

Dear Gustavo Zacarias,

On 11/09/15 23:24, Gustavo Zacarias wrote:
> Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
Reviewed-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
Tested-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>

Build test for MIPS architecture.

$ grep ^LIBTASN1_VERSION package/libtasn1/libtasn1.mk
LIBTASN1_VERSION = 4.6

$ file output/target/usr/bin/asn1Coding
output/target/usr/bin/asn1Coding: ELF 32-bit MSB executable, MIPS, 
MIPS32 rel2 version 1 (SYSV), dynamically linked (uses shared libs), for 
GNU/Linux 2.6.32, with unknown capability 0x41000000 = 0xf676e75, with 
unknown capability 0x10000 = 0x70403, stripped

Regards,

Vincent.

> ---
>   package/libtasn1/libtasn1.hash | 2 +-
>   package/libtasn1/libtasn1.mk   | 2 +-
>   2 files changed, 2 insertions(+), 2 deletions(-)
>
> diff --git a/package/libtasn1/libtasn1.hash b/package/libtasn1/libtasn1.hash
> index 653e018..41e10a6 100644
> --- a/package/libtasn1/libtasn1.hash
> +++ b/package/libtasn1/libtasn1.hash
> @@ -1,2 +1,2 @@
>   # Locally calculated after checking pgp signature
> -sha256	89b3b5dce119273431544ecb305081f3530911001bb12e5d76588907edb71bda	libtasn1-4.5.tar.gz
> +sha256	3462fc25e2d2536878c39a8825f5e36ba2e2611b27ef535e064f4c56258e508b	libtasn1-4.6.tar.gz
> diff --git a/package/libtasn1/libtasn1.mk b/package/libtasn1/libtasn1.mk
> index eeff76c..6818f30 100644
> --- a/package/libtasn1/libtasn1.mk
> +++ b/package/libtasn1/libtasn1.mk
> @@ -4,7 +4,7 @@
>   #
>   ################################################################################
>
> -LIBTASN1_VERSION = 4.5
> +LIBTASN1_VERSION = 4.6
>   LIBTASN1_SITE = http://ftp.gnu.org/gnu/libtasn1
>   LIBTASN1_DEPENDENCIES = host-bison
>   LIBTASN1_LICENSE = GPLv3+ LGPLv2.1+
>

[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Thomas Petazzoni]

Dear Gustavo Zacarias,

On Fri, 11 Sep 2015 19:24:29 -0300, Gustavo Zacarias wrote:
> Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
> ---
>  package/libtasn1/libtasn1.hash | 2 +-
>  package/libtasn1/libtasn1.mk   | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)

Applied, thanks.

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

[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Thomas Petazzoni]

Vicente,

On Sat, 12 Sep 2015 10:26:04 +0100, Vicente Olivert Riera wrote:

> On 11/09/15 23:24, Gustavo Zacarias wrote:
> > Signed-off-by: Gustavo Zacarias <gustavo@zacarias.com.ar>
> Reviewed-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>
> Tested-by: Vicente Olivert Riera <Vincent.Riera@imgtec.com>

Thanks for the testing!

However, may I suggest that you focus your review/test time to more
complicated patches? I typically apply trivial package bumps (such as
this one) without even doing a test build. There is really nothing to
review in such trivial patches (they just change the version and the
hash), and the build testing will be done by the autobuilders.

So I believe that if you have some time to review/test patches, it
would definitely be more beneficial on the more complicated ones (and
by complicated I don't mean "long", I simply mean not as trivial as a
package bump). For example, your help/testing on the systemd /usr merge
patch was much more useful than the testing of trivial package bumps.

Of course, feel free to continue testing package bumps as well, it does
not harm at all. I'm just trying to make your time as useful as
possible :-)

> $ file output/target/usr/bin/asn1Coding
> output/target/usr/bin/asn1Coding: ELF 32-bit MSB executable, MIPS, 

libtasn1 is a library, so it's actually more important to verify that
the library has been built correctly. Though I agree that we could
assume that asn1Coding is linked against the library, so that most
likely the library is also properly built for MIPS.

BTW, the fact that you are regularly verifying that the binaries built
by Buildroot are built for MIPS, I am wondering if it wouldn't be worth
the effort to add a final check at the end of the build to verify that
all binaries are built for the proper architecture/ABI. I remember that
some time ago, the nfs-utils package was installing binaries built for
the host machine into the target filesystem.

Best regards,

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

[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Peter Korsgaard]

>>>>> "Thomas" == Thomas Petazzoni <thomas.petazzoni@free-electrons.com> writes:

Hi,

 > BTW, the fact that you are regularly verifying that the binaries built
 > by Buildroot are built for MIPS, I am wondering if it wouldn't be worth
 > the effort to add a final check at the end of the build to verify that
 > all binaries are built for the proper architecture/ABI. I remember that
 > some time ago, the nfs-utils package was installing binaries built for
 > the host machine into the target filesystem.

Yes, it could be a nice test. How would we do it? With readelf? We would
need to filter out any shell scripts and similar.

-- 
Bye, Peter Korsgaard

[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Thomas Petazzoni]

Peter,

On Sun, 13 Sep 2015 21:52:30 +0200, Peter Korsgaard wrote:

>  > BTW, the fact that you are regularly verifying that the binaries built
>  > by Buildroot are built for MIPS, I am wondering if it wouldn't be worth
>  > the effort to add a final check at the end of the build to verify that
>  > all binaries are built for the proper architecture/ABI. I remember that
>  > some time ago, the nfs-utils package was installing binaries built for
>  > the host machine into the target filesystem.
> 
> Yes, it could be a nice test. How would we do it? With readelf? We would
> need to filter out any shell scripts and similar.

Either readelf, or just "file". But maybe readelf is fine as it's
available in the toolchain. We can skip scripts by testing the first
two bytes of each file and see if they are #!, no?

Best regards,

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

[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Vicente Olivert Riera]

Dear Thomas Petazzoni,

On 09/13/2015 10:20 PM, Thomas Petazzoni wrote:
> Peter,
> 
> On Sun, 13 Sep 2015 21:52:30 +0200, Peter Korsgaard wrote:
> 
>>  > BTW, the fact that you are regularly verifying that the binaries built
>>  > by Buildroot are built for MIPS, I am wondering if it wouldn't be worth
>>  > the effort to add a final check at the end of the build to verify that
>>  > all binaries are built for the proper architecture/ABI. I remember that
>>  > some time ago, the nfs-utils package was installing binaries built for
>>  > the host machine into the target filesystem.

It's a good idea.

>> Yes, it could be a nice test. How would we do it? With readelf? We would
>> need to filter out any shell scripts and similar.
> 
> Either readelf, or just "file". But maybe readelf is fine as it's
> available in the toolchain. We can skip scripts by testing the first
> two bytes of each file and see if they are #!, no?

Or use "file -i" and grep for "charset=binary".

Also, your script to calculate the size of the rootfs is a good start
point since it already collects all the files in target :-) We could
reuse part of that code, or directly use the .txt file that it generates.

Regards,

Vincent.

> Best regards,
> 
> Thomas
> 

[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Arnout Vandecappelle]

On 13-09-15 23:20, Thomas Petazzoni wrote:
> Peter,
> 
> On Sun, 13 Sep 2015 21:52:30 +0200, Peter Korsgaard wrote:
> 
>>  > BTW, the fact that you are regularly verifying that the binaries built
>>  > by Buildroot are built for MIPS, I am wondering if it wouldn't be worth
>>  > the effort to add a final check at the end of the build to verify that
>>  > all binaries are built for the proper architecture/ABI. I remember that
>>  > some time ago, the nfs-utils package was installing binaries built for
>>  > the host machine into the target filesystem.
>>
>> Yes, it could be a nice test. How would we do it? With readelf? We would
>> need to filter out any shell scripts and similar.
> 
> Either readelf, or just "file". But maybe readelf is fine as it's
> available in the toolchain. We can skip scripts by testing the first
> two bytes of each file and see if they are #!, no?

 file has the advantage of also working for scripts, and should even work for
FLAT binaries (which readelf wouldn't like). But it doesn't show the CPU, ISA
and FP details, which are useful to check as well.

 So the perfect approach would be: file, and readelf if file says it's an ELF.

 Or, of course, python-magic and python-pyelftools :-)

 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

[Buildroot] [PATCH] libtasn1: bump to version 4.6

[Thomas Petazzoni]

Arnout,

On Tue, 15 Sep 2015 00:09:05 +0200, Arnout Vandecappelle wrote:

>  file has the advantage of also working for scripts, and should even work for
> FLAT binaries (which readelf wouldn't like). But it doesn't show the CPU, ISA
> and FP details, which are useful to check as well.
> 
>  So the perfect approach would be: file, and readelf if file says it's an ELF.

Right.

>  Or, of course, python-magic and python-pyelftools :-)

Indeed, but then we would have to make the test optional so that
Buildroot can continue to work on systems where those additional Python
modules are not available. Which isn't as good as an unconditional
test that relies on already available utilities.

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