linux-firmware: bump version to latest d877533

Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
---
 package/linux-firmware/linux-firmware.hash | 6 +++---
 package/linux-firmware/linux-firmware.mk   | 2 +-
 2 files changed, 4 insertions(+), 4 deletions(-)

Hello Marcin,

Thanks for your patch, I have some issues/questions below.

On Wed, 24 Oct 2018 12:38:45 +0200, Marcin Niestroj wrote:
> Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
> ---
>  package/linux-firmware/linux-firmware.hash | 6 +++---
>  package/linux-firmware/linux-firmware.mk   | 2 +-
>  2 files changed, 4 insertions(+), 4 deletions(-)
> 
> diff --git a/package/linux-firmware/linux-firmware.hash b/package/linux-firmware/linux-firmware.hash
> index ca6ad8f59d..91645d6262 100644
> --- a/package/linux-firmware/linux-firmware.hash
> +++ b/package/linux-firmware/linux-firmware.hash
> @@ -1,11 +1,11 @@
>  # Locally calculated
> -sha256 b279ca4d086887c2efab13e28a7ca36e409410d3df38a62d7c7b5799ee3de916 linux-firmware-44d4fca9922a252a0bd81f6307bcc072a78da54a.tar.gz
> +sha256 48f02db163bd1320cafa0c878f0c38d2184afa726f02622309be69416e420032 linux-firmware-d87753369b82c5f362250c197d04a1e1ef5bf698.tar.gz

I don't get the same hash here:

ERROR: linux-firmware-d87753369b82c5f362250c197d04a1e1ef5bf698.tar.gz has wrong sha256 hash:
ERROR: expected: 48f02db163bd1320cafa0c878f0c38d2184afa726f02622309be69416e420032
ERROR: got     : 3c5ba0e3b282e44d5cef3d34a547caf3633c90b83807469357e92cca1fff8a38

And apparently Yann E. Morin also calculated the same hash as me after
applying your version bump. Could you check what is happening ?

>  sha256 8116433f4004fc0c24d72b3d9e497808b724aa0e5e1cd63fc1bf66b715b1e2e9 LICENCE.Abilis
>  sha256 4b3ea5d5a03c0db81bee0bcb14b30d75b30ef568597bb5be7d4dee57f434265f LICENSE.amdgpu
>  sha256 38f2037aa14631b4b29826d7a99379613c41a97064d1defdee30a7a022138b20 LICENCE.Marvell
>  sha256 802b7014b26c606cf6248ae8b0ab1ce6d2d1b0db236d38dd269e676cd70710f2 LICENCE.atheros_firmware
>  sha256 3b5eb392b2d9d8c46d6aae26d06c187e5ea3029b12d13bc2b8deb8b3ce6bfa53 ath10k/QCA6174/hw3.0/notice_ath10k_firmware-4.txt
> -sha256 c565861ff7c42f5df98e15239241f1f42614e5e15f362094a2d3e8da724dc842 ath10k/QCA6174/hw3.0/notice_ath10k_firmware-6.txt
> +sha256 8ce5c6ea0542bf4aac31fc3ae16a39792ad22d0eae4543063fac56fb3380f021 ath10k/QCA6174/hw3.0/notice_ath10k_firmware-6.txt

What are the changes in this file? They should be briefly explained in
the commit log.

>  sha256 b16056fc91b82a0e3e8de8f86c2dac98201aa9dc3cbd33e8d38f1b087fcec30d LICENCE.broadcom_bcm43xx
>  sha256 a5777f9e80aca0603b0648454de996168b1c530322550ccda94d6d78bcf6c061 LICENCE.chelsio_firmware
>  sha256 60fbc9cccb455e1a3306c97db942d6f24fa93664be61d54c497637e6d0e2ae83 LICENCE.fw_sst_0f28
> @@ -27,5 +27,5 @@ sha256 8542aeabf2761935122d693561e16766ce1bcc2b0d003204f9040b7d6d929f2e LICENSE.
>  sha256 be904cd28cb292b80cdb6cf412ab0d9159d431671e987ad433c1f62e0988a9bc LICENSE.qcom
>  sha256 fc6223d4bfe9f2f9e2eddc44b9fe5721d0caf49f01cb08d602906add686d8c6f LICENSE.radeon
>  sha256 2bdd2e716f05d9737d3f9a20f9a3a3c0caee0e866100ddb0673f1178e42f92b9 LICENSE.sdma_firmware
> -sha256 ef38a9a8bb4b0f72b369d337426eea63ef8fc9d48453f127028d935f7dbc5820 WHENCE
> +sha256 a16234520f37d1fe170fb96f470897a12f74cc4a83bc6ab28eacfdde5bea1917 WHENCE

For this one, there's no real need to explain in the commit log: the
WHENCE file is updated basically everytime a firmware is added/removed.

Thanks!

Thomas

Marcin, Thomas, All,

On 2018-11-01 23:00 +0100, Thomas Petazzoni spake thusly:
> On Wed, 24 Oct 2018 12:38:45 +0200, Marcin Niestroj wrote:
> > Signed-off-by: Marcin Niestroj <m.niestroj@grinn-global.com>
> > ---
> >  package/linux-firmware/linux-firmware.hash | 6 +++---
> >  package/linux-firmware/linux-firmware.mk   | 2 +-
> >  2 files changed, 4 insertions(+), 4 deletions(-)
> > 
> > diff --git a/package/linux-firmware/linux-firmware.hash b/package/linux-firmware/linux-firmware.hash
> > index ca6ad8f59d..91645d6262 100644
> > --- a/package/linux-firmware/linux-firmware.hash
> > +++ b/package/linux-firmware/linux-firmware.hash
> > @@ -1,11 +1,11 @@
> >  # Locally calculated
> > -sha256 b279ca4d086887c2efab13e28a7ca36e409410d3df38a62d7c7b5799ee3de916 linux-firmware-44d4fca9922a252a0bd81f6307bcc072a78da54a.tar.gz
> > +sha256 48f02db163bd1320cafa0c878f0c38d2184afa726f02622309be69416e420032 linux-firmware-d87753369b82c5f362250c197d04a1e1ef5bf698.tar.gz
> 
> I don't get the same hash here:
> 
> ERROR: linux-firmware-d87753369b82c5f362250c197d04a1e1ef5bf698.tar.gz has wrong sha256 hash:
> ERROR: expected: 48f02db163bd1320cafa0c878f0c38d2184afa726f02622309be69416e420032
> ERROR: got     : 3c5ba0e3b282e44d5cef3d34a547caf3633c90b83807469357e92cca1fff8a38
> 
> And apparently Yann E. Morin also calculated the same hash as me after
> applying your version bump. Could you check what is happening ?

If using a recent distribution that has tar 1.30 (e.g. Fedroa 28+,
Arch, or other rolling-release distros), you may have fallen in the
download dependencies trap, where host-tar would not be built when doing
a 'make source' from a clean tree.

Regards,
Yann E. MORIN.

Hello,

On Thu, 1 Nov 2018 23:08:48 +0100, Yann E. MORIN wrote:

> > I don't get the same hash here:
> > 
> > ERROR: linux-firmware-d87753369b82c5f362250c197d04a1e1ef5bf698.tar.gz has wrong sha256 hash:
> > ERROR: expected: 48f02db163bd1320cafa0c878f0c38d2184afa726f02622309be69416e420032
> > ERROR: got     : 3c5ba0e3b282e44d5cef3d34a547caf3633c90b83807469357e92cca1fff8a38
> > 
> > And apparently Yann E. Morin also calculated the same hash as me after
> > applying your version bump. Could you check what is happening ?  
> 
> If using a recent distribution that has tar 1.30 (e.g. Fedroa 28+,
> Arch, or other rolling-release distros), you may have fallen in the
> download dependencies trap, where host-tar would not be built when doing
> a 'make source' from a clean tree.

Which has just been fixed as of
https://git.buildroot.org/buildroot/commit/?id=5b5feea8753520320d11896a24b970c60331bcdc,
so Marcin, you may see automagically the new/correct hash if you retry
on the latest Buildroot master (I merged this host-tar fix from Yann
yesterday evening).

Thanks,

Thomas

Hi Thomas, Yann,

Thomas Petazzoni <thomas.petazzoni@bootlin.com> writes:

> Hello,
>
> On Thu, 1 Nov 2018 23:08:48 +0100, Yann E. MORIN wrote:
>
>> > I don't get the same hash here:
>> > 
>> > ERROR: linux-firmware-d87753369b82c5f362250c197d04a1e1ef5bf698.tar.gz has wrong sha256 hash:
>> > ERROR: expected: 48f02db163bd1320cafa0c878f0c38d2184afa726f02622309be69416e420032
>> > ERROR: got     : 3c5ba0e3b282e44d5cef3d34a547caf3633c90b83807469357e92cca1fff8a38
>> > 
>> > And apparently Yann E. Morin also calculated the same hash as me after
>> > applying your version bump. Could you check what is happening ?  
>> 
>> If using a recent distribution that has tar 1.30 (e.g. Fedroa 28+,
>> Arch, or other rolling-release distros), you may have fallen in the
>> download dependencies trap, where host-tar would not be built when doing
>> a 'make source' from a clean tree.
>
> Which has just been fixed as of
> https://git.buildroot.org/buildroot/commit/?id=5b5feea8753520320d11896a24b970c60331bcdc,
> so Marcin, you may see automagically the new/correct hash if you retry
> on the latest Buildroot master (I merged this host-tar fix from Yann
> yesterday evening).

I have rebased my patch on master and I still get the same hash as
previously. That is logical, because I had host-tar already built when I
tested patch before submission. Can you confirm that problem was
actually on your side (i.e. my hash is correct)? After that I will
bump linux-firmware version to newest, extend commit message and send
PATCH v2.

>
> Thanks,
>
> Thomas