50
Context
Around a year and a half ago, I’ve asked my former company for some time to
work on an issue that was impacting the debugging capabilities in our project:
gdbserver couldn’t debug multithreaded applications running on a PowerPC32
architecture. The connection to the gdbserver was broken and it couldn’t
control the debug session anymore. Multiple people have already investigated
this problem and I had a good starting point, but we still weren’t sure in
which software component the issue lied: it could have been the toolchain, the
gdbserver, the Linux kernel or the custom patches we applied on top of the
kernel tree. We were quite far away from finding the root cause.
Somebody is pretty salty for no good reason. The maintainers own patch is nicer code than the suggested patch - and the change is small enough that there just isn’t anything available to guide the reporter to a better solution without wasting everyone’s time.
I’d probably have added a thanks for debugging effort into the commit message myself - but “please accept my patch because I want to have code in the kernel” is a very stupid thing to say, and the maintainer offering a suitable problem to fix is more than I’d have done in that situation.
As someone who had a mildly unpleasant interaction with kernel folks, I can totally understand the issue.
This is one of the very few open source projects I had the feeling they don’t appreciate new contributers. There is no on boarding material available and picking the wrong subproject mailing list results in being ignored. You have to spend days without any possibility of help and if your are lucky you get mentioned as a reporter. For the next issue you start from square one as there was no guidance, so you could only learn the bare minimum.
So yeah, his patch may be underwhelming. But the help and credit he got for days or weeks of unpaid work was basically nothing. You may be okay with spending days and only getting credits for the bug report, but I suspect many aren’t and will not contribute again after such an experience. And post like this try to point out the issue they have and why many people won’t contribute to the kernel ever again.
Especially in this particular case the effort is in debugging the problem, not doing the actual fix - which is the bug report, where he got credited for. lkml is not the place for “how I debugged this problem” - that’d be what goes into his blog. And if you look around you’ll see a lot of “how I helped solving this problem” kind of blog posts.
This change is so simple that guiding him to do it in a good way would involve fixing it yourself in the explanation - and then you’d not show the code so he can do it himself? That’s just silly. If he cares about that he came out of that with quite a bit of experience on how to handle it the next time - and he mentions he even got an (assumed to be starter friendly) other issue suggested if he wants to have code in the kernel.
From the perspective of hiring people he turned this from a “nice work debugging a problem, might be a useful candidate” to “tries getting low quality code merged for vanity reasons, let’s avoid that guy”
I didn’t meant to defend the patch and I see your point. But I personally think that it’s not unreasonable to expect to land a bugfix commit after spending multiple days debugging a complex issue, that’s why understand that he feels robbed of a kernel contribution.
I don’t know what could have been a good solution for this scenario. But taking potential future contributors feelings more serious would help to keep them around and make them feel appreciated.
The shit storm he brew up in response to getting feedback on his very first pull request is way more concerning than churning out low-quality code.
Coding skills can be improved, specially from the first pull request onward. Toxic behavior such as putting up very public smear campaigns in response to getting feedback on his very first patch submission is a major red flag, and is as toxic as it gets.
That’s roughly what I meant - he should’ve come out of that experience having learned a lot (there’s even an explanation why the other code is better on the mailing list), and had the option of working on a different problem (while he didn’t say which I assume it was selected to be more beginner friendly). And instead he’s throwing a temper tantrum - that’s too risky behaviour for hiring.
We should keep in mind that this is a one-sided account on how a mundane bugfix issue was handled. Grain of salt required.
Nevertheless, the blog author said he received feedback from members of the Linux kernel security mailing list. Even though I think he could have received more credit than reporting the issue, that was basically his contribution: he pinpointed where the bug was. He also contributed a couple of patches that were faulty and unusable, and the maintainer had to step in and roll out his own fix.
I understand that fixing a nontrivial bug is a badge of honor, and getting credit for critical contributions might have more implications than a warm feeling. However, if the submitted patches were unusable then what would be the desirable outcome? I mean, should Linux users be deprived of a bug fix because a first-timr contributor is struggling with putting together a working patch?
Good point, this could just misrepresentat the situation. I also haven’t looked over the mailing list thread and comments here are very salty.
But giving him the benefit of doubt of a nice potential contributer who just debugged a very hard issue and sending in a basic concept of a potential fix. I think it would be beneficial for their community to take the wish for more credit more serious and try to make him feel welcome. But I recognize it was probably hard to do in this case.
Overall I just wanted to recognize that I do see how he feels robbed of his contribution. It reminded me that I also had an experience with the kernel developers that made me not want to contribute again.
I think they did. Apparently the maintainer trusted the first-time contributor enough to propose tackling another bug.
If the goal is to get more contributions, I think that’s exactly what should happen. I feel the kernel maintainer is being treated unfairly.
Whining about getting extra work feels like the author didn’t intended to contribute anything else and just put all this reputation chips on that one isolated ticket.
There is no trust needed when asking someone to fix a bug. It’s not like the maintainer would lose anything if the contributor failed to fix the bug.
Besides, I think it is natural to want recognition when you do a lot of work for free. Many other people wouldn’t do this unpaid work at all; recognizing their contribution is the bare minimum of good manners. Even in a company where employees are paid for their work, it is customary to give credit to co-workers who have helped you. Most people don’t like to work in places where they don’t feel appreciated, and that is also true in Open-Source.
Unfortunately I don’t have my original patch, because I only sent that to the Linux security mailing list. I don’t think it’s a stupid thing to want to have code in the kernel, especially after spending all my time debugging this issue. The fix was trivial once I’ve pointed to the exact place where the buffer overflow happened and I should have received credit for all my effort.
You did receive credit. A good bug report allows reproducing and ideally fixing the issue - which can involve considerable effort. This is the difference between your report, and the one you linked from 6 years ago.
Like I said, I’d probably have added an additional thanks for that in my commit message - but I’m unfamiliar with the kind of reports this particular subsystem typically receives, so it is quite possible your report is just something average coming in there.
I personally prefer to include code suggesting a fix in my bug reports - but I usually don’t expect it to be just merged as I’m not familiar with surrounding code. I also don’t expect that to receive an additional mention - it’s just part of the report, and is often cleaner in demonstrating the issue than a problem description.
The way that you jumped straight onto broadcasting drama when your very first Linux kernel patch stumbled on the code review stage is a major red flag.
I would hate to work with you because I would feel that I would be risking being subjected to a very public character attack each time I had to review one of your patches.
I’m not broadcasting drama, I’m sharing my side of the story on my personal blog and distribute it to other social media platforms.
The patch didn’t stumble on the code review stage, the PowerPC maintainer didn’t want to accept patches from me and implemented his own fix.
Why would you hate people who would describe their interactions with you? The only reason I see is that you would hate how you’ve dealt with them.
I’m not broadcasting drama, I’m sharing my side of the story on my personal blog and distribute it to other social media platforms.
That’s literally broadcasting drama.
I think you’ll be happier in the long run if you can forgive and move on.