Hey y'all,
I've modified the board-status script to have as few external dependencies as possible, be a self-extracting, self-running "binary". and removed the time-stamp to keep the number initial commits down. Feel free to tell me how bad it is and how it murdered your children. :P
John.
On 14.11.2014 12:44, John Lewis wrote:
Hey y'all,
I've modified the board-status script to have as few external dependencies as possible, be a self-extracting, self-running "binary". and removed the time-stamp to keep the number initial commits down. Feel free to tell me how bad it is and how it murdered your children. :P
Never send binaries to the list. And about the idea: current code needs git tree. I tried to make changes to make it possible to have board-status without git tree.
Result: those change were ignored, bikeshed or vetoed.
Long story short: nobody cares.
John.
On 21/11/14 21:58, Vladimir 'φ-coder/phcoder' Serbinenko wrote:
On 14.11.2014 12:44, John Lewis wrote:
Hey y'all,
I've modified the board-status script to have as few external dependencies as possible, be a self-extracting, self-running "binary". and removed the time-stamp to keep the number initial commits down. Feel free to tell me how bad it is and how it murdered your children. :P
Never send binaries to the list.
A link could be just as malicious. What way should I do it?
And about the idea: current code needs git tree. I tried to make changes to make it possible to have board-status without git tree.
Result: those change were ignored, bikeshed or vetoed.
Long story short: nobody cares.
At some point Ron did care, otherwise we wouldn't have been talking about it, and I certainly do. I can tell you anecdotally which Chromebooks are being used and should remain live in the code-base (hint: all of them) but I thought we were trying to be more scientific about it than that.
John.
On 22.11.2014 12:22, John Lewis wrote:
On 21/11/14 21:58, Vladimir 'φ-coder/phcoder' Serbinenko wrote:
On 14.11.2014 12:44, John Lewis wrote:
Hey y'all,
I've modified the board-status script to have as few external dependencies as possible, be a self-extracting, self-running "binary". and removed the time-stamp to keep the number initial commits down. Feel free to tell me how bad it is and how it murdered your children. :P
Never send binaries to the list.
A link could be just as malicious. What way should I do it?
A link would be fine. It wouldn't make everybody uninterested in this topic but subscribed to the list download 1M file.
And about the idea: current code needs git tree. I tried to make changes to make it possible to have board-status without git tree.
Result: those change were ignored, bikeshed or vetoed.
Long story short: nobody cares.
At some point Ron did care, otherwise we wouldn't have been talking about it, and I certainly do. I can tell you anecdotally which Chromebooks are being used and should remain live in the code-base (hint: all of them) but I thought we were trying to be more scientific about it than that.
Speaking is about where it usually stops. It's been very disappointing. Usual scenario is that we get lots of discussion to do sth, everybody is dissatisfied with current status and when I spend huge chunk of my personal time to fix it, nobody is there to even review the patches. That's understandable: otherwise what will we speak about next time if the issue is solved? Still it's disappointing and I have better things to do with my personal time than making patches nobody cares about. The top of rename stack is at 7185. It's automated test and Stefan agreed to it in principle in Prague, yet it's sitting there for already 3 weeks.
John.
On 22/11/14 15:55, Vladimir 'φ-coder/phcoder' Serbinenko wrote:
On 22.11.2014 12:22, John Lewis wrote:
On 21/11/14 21:58, Vladimir 'φ-coder/phcoder' Serbinenko wrote:
On 14.11.2014 12:44, John Lewis wrote:
Hey y'all,
I've modified the board-status script to have as few external dependencies as possible, be a self-extracting, self-running "binary". and removed the time-stamp to keep the number initial commits down. Feel free to tell me how bad it is and how it murdered your children. :P
Never send binaries to the list.
A link could be just as malicious. What way should I do it?
A link would be fine. It wouldn't make everybody uninterested in this topic but subscribed to the list download 1M file.
I understand what you mean now - people perhaps stuck on low-bandwidth links with expensive data plans.
And about the idea: current code needs git tree. I tried to make changes to make it possible to have board-status without git tree.
Result: those change were ignored, bikeshed or vetoed.
Long story short: nobody cares.
At some point Ron did care, otherwise we wouldn't have been talking about it, and I certainly do. I can tell you anecdotally which Chromebooks are being used and should remain live in the code-base (hint: all of them) but I thought we were trying to be more scientific about it than that.
Speaking is about where it usually stops. It's been very disappointing. Usual scenario is that we get lots of discussion to do sth, everybody is dissatisfied with current status and when I spend huge chunk of my personal time to fix it, nobody is there to even review the patches. That's understandable: otherwise what will we speak about next time if the issue is solved? Still it's disappointing and I have better things to do with my personal time than making patches nobody cares about. The top of rename stack is at 7185. It's automated test and Stefan agreed to it in principle in Prague, yet it's sitting there for already 3 weeks.
Okay, well the binary as is will upload stuff to the current board-status repo, so maybe that's enough. I'm trying to encourage people to use it. We'll see if we can get maybe a few hundred people to upload something, given time.
John.
On 22.11.2014 17:28, John Lewis wrote:
On 22/11/14 15:55, Vladimir 'φ-coder/phcoder' Serbinenko wrote:
On 22.11.2014 12:22, John Lewis wrote:
On 21/11/14 21:58, Vladimir 'φ-coder/phcoder' Serbinenko wrote:
On 14.11.2014 12:44, John Lewis wrote:
Hey y'all,
I've modified the board-status script to have as few external dependencies as possible, be a self-extracting, self-running "binary". and removed the time-stamp to keep the number initial commits down. Feel free to tell me how bad it is and how it murdered your children. :P
Never send binaries to the list.
A link could be just as malicious. What way should I do it?
A link would be fine. It wouldn't make everybody uninterested in this topic but subscribed to the list download 1M file.
I understand what you mean now - people perhaps stuck on low-bandwidth links with expensive data plans.
And about the idea: current code needs git tree. I tried to make changes to make it possible to have board-status without git tree.
Result: those change were ignored, bikeshed or vetoed.
Long story short: nobody cares.
At some point Ron did care, otherwise we wouldn't have been talking about it, and I certainly do. I can tell you anecdotally which Chromebooks are being used and should remain live in the code-base (hint: all of them) but I thought we were trying to be more scientific about it than that.
Speaking is about where it usually stops. It's been very disappointing. Usual scenario is that we get lots of discussion to do sth, everybody is dissatisfied with current status and when I spend huge chunk of my personal time to fix it, nobody is there to even review the patches. That's understandable: otherwise what will we speak about next time if the issue is solved? Still it's disappointing and I have better things to do with my personal time than making patches nobody cares about. The top of rename stack is at 7185. It's automated test and Stefan agreed to it in principle in Prague, yet it's sitting there for already 3 weeks.
Okay, well the binary as is will upload stuff to the current board-status repo, so maybe that's enough. I'm trying to encourage people to use it. We'll see if we can get maybe a few hundred people to upload something, given time.
Determining of mainboard is wrong for clones. I was trying to fix this a while ago but I stopped carying.
John.
Dear John,
Am Freitag, den 14.11.2014, 11:44 +0000 schrieb John Lewis:
I've modified the board-status script to have as few external dependencies as possible, be a self-extracting, self-running "binary". and removed the time-stamp to keep the number initial commits down. Feel free to tell me how bad it is and how it murdered your children. :P
thanks a lot for encouraging the users of your firmware images to upload the board status results [1] to coreboot’s board-status repository [2].
Looking at the pushed commits, the `dirty` string is appended to all of them. This means, when you built the image, there were local modifications to the source tree. It’d be great if you could check what these modifications are. Sometimes it is just a change for the 3rdparty submodule. Sometimes these are bigger changes. The output `git diff` or `git status` should clear that up.
Additionally, the file `coreboot_timestamps.txt` just contains
No timestamps found in coreboot table.
meaning collecting time stamps is not enabled in Kconfig. It’d be great if you could enable that too. Soon this will be enabled by default if Kyösti’s change set #8023 will get committed.
If you could distribute new firmware images with these changes, that would be awesome!
Thanks a lot,
Paul
[1] https://johnlewis.ie/a-call-to-arms/ [2] http://review.coreboot.org/gitweb?p=board-status.git;a=summary [3] http://review.coreboot.org/8023
-
John Lewis -
Paul Menzel -
Vladimir 'φ-coder/phcoder' Serbinenko