ipmitool recipe update needed?

Vernon Mauery vernon.mauery at linux.intel.com
Wed Apr 17 02:05:26 AEST 2019


On 16-Apr-2019 04:27 PM, Alexander Amelkin wrote:
>11.04.2019 22:07, William Kennington wrote:
>> The problem with the archive tarballs is that their contents are not
>> stable (they are autogenerated when needed from git), and they aren't
>> dist tarballs so they haven't been run over `./bootstrap.sh;
>> ./configure; make dist` yet.
>
>Well, they are stable enough because they are autogenerated from the same git hash each time.
>
>As to not being 'dist' tarballs, is that actually a problem (at least for op-build) ?
>Release tarballs on sourceforge haven't been run through bootstrap and configure as well, because that's useless.
>You would need to re-run it on your target build system anyway.
>
>As Vernon suggested, using git instead of tarballs would be even better. The code in the repository has been significantly improved since 1.8.18 already, yet I don't feel so far like tagging a new release. Nonetheless, I don't see why the top of trunk couldn't be used for op-build.

I have pushed a patch to meta-oe that just uses a recent hash from the 
upstream git. Ideally, it would go tag to tag or something like that. I 
am not sure about the tagging or release schedule for ipmitool.

--Vernon


More information about the openbmc mailing list