diff mbox series

[4/5] package/intel-mediadriver: bump version to 23.1.0

Message ID 20230130174339.36794-4-bernd.kuhls@t-online.de
State Accepted
Headers show
Series [1/5] package/libva: bump version to 2.17.0 | expand

Commit Message

Bernd Kuhls Jan. 30, 2023, 5:43 p.m. UTC
Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
---
 package/intel-mediadriver/intel-mediadriver.hash | 2 +-
 package/intel-mediadriver/intel-mediadriver.mk   | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

Comments

Thomas Petazzoni Jan. 30, 2023, 9:15 p.m. UTC | #1
On Mon, 30 Jan 2023 18:43:38 +0100
Bernd Kuhls <bernd.kuhls@t-online.de> wrote:

> Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
> ---
>  package/intel-mediadriver/intel-mediadriver.hash | 2 +-
>  package/intel-mediadriver/intel-mediadriver.mk   | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/package/intel-mediadriver/intel-mediadriver.hash b/package/intel-mediadriver/intel-mediadriver.hash
> index a5e9c64e97..e4e41e4ad3 100644
> --- a/package/intel-mediadriver/intel-mediadriver.hash
> +++ b/package/intel-mediadriver/intel-mediadriver.hash
> @@ -1,3 +1,3 @@
>  # Locally computed
> -sha256  b553290e829dfd824eb62295c9f07dbe8062ce7998f7c527cc92856d0792562d  intel-media-22.6.6.tar.gz
> +sha256  edae1bdc07b694bd6e42099feddcf875e3b5d7254894f4b3fb292128577baf82  intel-media-23.1.0.tar.gz
>  sha256  74979d5aaee78b8da82e3aafd415a216b6131dfff6d95d6930927c8a4e3bded3  LICENSE.md

Here as well, I'm getting a different hash. Did github change
something? Could you keep the tarball you have locally so that we can
compare it to the one we get now from Github?

Best regards,

Thomas
Bernd Kuhls Jan. 30, 2023, 9:30 p.m. UTC | #2
Am Mon, 30 Jan 2023 22:15:14 +0100 schrieb Thomas Petazzoni via buildroot:

> Here as well, I'm getting a different hash. Did github change something?
> Could you keep the tarball you have locally so that we can compare it to
> the one we get now from Github?

Hi Thomas,

both tarballs contain exactly the same files but the tarballs differ in 
size:

-rw-r--r-- 1 bernd bernd 28517107 Jan 30 22:24 intel-media-23.1.0.tar.gz
-rw-r--r-- 1 bernd bernd 28601938 Jan 18 19:14 intel-media-23.1.0.tar.gz

Regards, Bernd
Thomas Petazzoni Jan. 30, 2023, 9:39 p.m. UTC | #3
On Mon, 30 Jan 2023 22:30:19 +0100
Bernd Kuhls <bernd.kuhls@t-online.de> wrote:

> both tarballs contain exactly the same files but the tarballs differ in 
> size:
> 
> -rw-r--r-- 1 bernd bernd 28517107 Jan 30 22:24 intel-media-23.1.0.tar.gz
> -rw-r--r-- 1 bernd bernd 28601938 Jan 18 19:14 intel-media-23.1.0.tar.gz

Could you run diffoscope between both tarballs?

Thomas
Bernd Kuhls Jan. 30, 2023, 9:49 p.m. UTC | #4
Am Mon, 30 Jan 2023 22:39:41 +0100 schrieb Thomas Petazzoni via buildroot:

> On Mon, 30 Jan 2023 22:30:19 +0100 Bernd Kuhls <bernd.kuhls@t-online.de>
> wrote:
> 
>> both tarballs contain exactly the same files but the tarballs differ in
>> size:
>> 
>> -rw-r--r-- 1 bernd bernd 28517107 Jan 30 22:24
>> intel-media-23.1.0.tar.gz -rw-r--r-- 1 bernd bernd 28601938 Jan 18
>> 19:14 intel-media-23.1.0.tar.gz
> 
> Could you run diffoscope between both tarballs?

The result is a 235 MB log file starting with:

--- old_edae/intel-media-23.1.0.tar.gz
+++ new_f47f/intel-media-23.1.0.tar.gz
│┄ Format-specific differences are supported for Gzipped files but no 
file-specific differences were detected; falling back to a binary diff. 
file(1) reports: gzip compressed data, from Unix, original size modulo 
2^32 245893120

Regards, Bernd
Bernd Kuhls Jan. 30, 2023, 10 p.m. UTC | #5
Am Mon, 30 Jan 2023 22:15:14 +0100 schrieb Thomas Petazzoni via buildroot:

> Here as well, I'm getting a different hash. Did github change something?

Hi Thomas,

see https://github.com/keybase/client/issues/10800#issuecomment-375831096
and the following discussion :(

Regards, Bernd
Thomas Petazzoni Jan. 30, 2023, 10:04 p.m. UTC | #6
On Mon, 30 Jan 2023 22:49:04 +0100
Bernd Kuhls <bernd.kuhls@t-online.de> wrote:

> --- old_edae/intel-media-23.1.0.tar.gz
> +++ new_f47f/intel-media-23.1.0.tar.gz
> │┄ Format-specific differences are supported for Gzipped files but no 
> file-specific differences were detected; falling back to a binary diff. 
> file(1) reports: gzip compressed data, from Unix, original size modulo 
> 2^32 245893120

If you gunzip both files, are the resulting tarballs identical?

Thomas
Thomas Petazzoni Jan. 30, 2023, 10:05 p.m. UTC | #7
On Mon, 30 Jan 2023 22:49:04 +0100
Bernd Kuhls <bernd.kuhls@t-online.de> wrote:

> The result is a 235 MB log file starting with:
> 
> --- old_edae/intel-media-23.1.0.tar.gz
> +++ new_f47f/intel-media-23.1.0.tar.gz
> │┄ Format-specific differences are supported for Gzipped files but no 
> file-specific differences were detected; falling back to a binary diff. 
> file(1) reports: gzip compressed data, from Unix, original size modulo 
> 2^32 245893120

Could you put online somewhere the "old" tarball, so that we can have a
look at what's going on?

Thanks a lot!

Thomas
Peter Seiderer Jan. 31, 2023, 5:26 p.m. UTC | #8
Hello *,

On Mon, 30 Jan 2023 23:00:15 +0100, Bernd Kuhls <bernd.kuhls@t-online.de> wrote:

> Am Mon, 30 Jan 2023 22:15:14 +0100 schrieb Thomas Petazzoni via buildroot:
>
> > Here as well, I'm getting a different hash. Did github change something?
>
> Hi Thomas,
>
> see https://github.com/keybase/client/issues/10800#issuecomment-375831096
> and the following discussion :(

And https://github.blog/changelog/2023-01-30-git-archive-checksums-may-change/

Regards,
Peter


>
> Regards, Bernd
>
> _______________________________________________
> buildroot mailing list
> buildroot@buildroot.org
> https://lists.buildroot.org/mailman/listinfo/buildroot
Thomas Petazzoni Feb. 8, 2023, 8:58 a.m. UTC | #9
On Mon, 30 Jan 2023 18:43:38 +0100
Bernd Kuhls <bernd.kuhls@t-online.de> wrote:

> Signed-off-by: Bernd Kuhls <bernd.kuhls@t-online.de>
> ---
>  package/intel-mediadriver/intel-mediadriver.hash | 2 +-
>  package/intel-mediadriver/intel-mediadriver.mk   | 2 +-
>  2 files changed, 2 insertions(+), 2 deletions(-)

Github has fixed their tarball generation logic so that the hashes are
now correct again, so I've applied!

Thomas
diff mbox series

Patch

diff --git a/package/intel-mediadriver/intel-mediadriver.hash b/package/intel-mediadriver/intel-mediadriver.hash
index a5e9c64e97..e4e41e4ad3 100644
--- a/package/intel-mediadriver/intel-mediadriver.hash
+++ b/package/intel-mediadriver/intel-mediadriver.hash
@@ -1,3 +1,3 @@ 
 # Locally computed
-sha256  b553290e829dfd824eb62295c9f07dbe8062ce7998f7c527cc92856d0792562d  intel-media-22.6.6.tar.gz
+sha256  edae1bdc07b694bd6e42099feddcf875e3b5d7254894f4b3fb292128577baf82  intel-media-23.1.0.tar.gz
 sha256  74979d5aaee78b8da82e3aafd415a216b6131dfff6d95d6930927c8a4e3bded3  LICENSE.md
diff --git a/package/intel-mediadriver/intel-mediadriver.mk b/package/intel-mediadriver/intel-mediadriver.mk
index 3473c949fd..ebac7aa6f5 100644
--- a/package/intel-mediadriver/intel-mediadriver.mk
+++ b/package/intel-mediadriver/intel-mediadriver.mk
@@ -6,7 +6,7 @@ 
 
 # based on https://software.intel.com/en-us/articles/build-and-debug-open-source-media-stack
 
-INTEL_MEDIADRIVER_VERSION = 22.6.6
+INTEL_MEDIADRIVER_VERSION = 23.1.0
 INTEL_MEDIADRIVER_SITE = https://github.com/intel/media-driver/archive
 INTEL_MEDIADRIVER_SOURCE= intel-media-$(INTEL_MEDIADRIVER_VERSION).tar.gz
 INTEL_MEDIADRIVER_LICENSE = MIT, BSD-3-Clause