Last week, Linux kernel contributor Qu Wenruo berated fellow code giver, Zhen Lei, for wasting kernel maintainers’ time with unnecessary fixes.
In a publication to Zhen Lei and the rest of the Linux kernel mailing list, Wenruo said he recently found a patch removed an insufficient memory debug error message from a self test used by btrfs, a filesystem supported by the Linux kernel.
“It’s nothing special, a little cleanup on the part of a kernel novice,” Wenruo wrote. “But the email address makes me careful, ‘@ huawei.com’.”
This is not the first time that similar harmless “cleaning” fixes have been provided by Huawei, said Wenruo, who observed that the fixes were also “unnecessary.”
“It makes me wonder what’s really going on here,” he wrote, noting that a quick search found a number of fixes to “clean” out of memory error messages or to fix them. spelling mistakes.
Wenruo’s theory is that Zhen Lei submitted this patch harmless to Key Performance Indicator (KPI) credit – to do something that is recognized by an employee performance measurement system as meaningful work.
Wenruo said it was good for new developers and students to submit these kind of fixes, noting that he started his kernel contributions the same way and that he hopes these small fixes lead to contributions to the kernel. long term.
âBut what you’re doing is really grabbing the KPIs,â he said. “I’ve seen several maintainers arguing with you over such ‘cleanups’ before, and you’re still fighting back to try to merge these fixes.”
“You send the patch representing your company, by doing that you are only damaging the already broken reputation.”
Wenruo’s response may seem too harsh – we asked a Linux manager if the reprimand was in line with the Linux Code of Conduct, but we got no response. However, in April, the Linux kernel community had a bad experience with researchers at the University of Minnesota who submitted “hypocritical pledges” – shoddy code – to test whether maintainers were careful. The backlash highlighted how much work volunteers contribute and how much they want when their time is wasted.
The register Zhen Lei to comment but we have had no response.
In an email to The register, Wenruo said, âSome Chinese tech companies are pushing really too hard by assigning almost impossible KPI goals, I think that’s the root cause. “
âIt pushes their employees to do things without using common sense. And obviously a toxic corporate culture like 996 (9am to 9pm, 6 days a week) and destructive competition.â
“Hopefully there will be fewer and fewer incidents like this, but without addressing the root cause, it’s just a problematic moment to hit the next incident.”
Wenruo said that as far as he was concerned, the University of Minnesota case had not played into that.
“The Minnesota incident is at least more difficult to detect, because these ‘fixes’ look like real fixes at a glance, only by digging deeper does it show some problems,” a- he declared. “But in the Huawei incident, it’s pretty straightforward, all the ‘cleanups’ are just generated from
Wenrou said he found the author’s response boring and wanted the author to save maintainers time by making it clear that “cleanup” commits are just warnings of
Zhen Lei did reply, however, on the mailing list, defending past contributions from the kernel and promising that subsequent contributions to the Linux community will be more substantial.
And in response to this, Wenruo proposed a list of projects that would be more useful to deal with.
The involvement of Huawei, which published its own Linux distribution and has become a hotspot for IT policy in the United States and Europe due to concerns about supply chain integrity and government espionage, has made this dispute a topic of interest for software developers in China.
On the Chinese Q&A site Zhihu, a person claiming to be a Red Hat engineer dissected mailing list exchange. Since the chat can be deciphered via algorithmic translation, the person posting the post argues that the Huawei shouldn’t even be in the chat, as such engagements and concerns about them occur all the time. time. Those who responded to the post claim that if a Google employee made a similar commitment, that person would be voted in favor.
We also contacted Huawei, one of the the first two Linux v5.10 contributors alongside Intel, to ask if the telecom equipment titan counts commitments for employee KPI assessment. A spokeswoman in the United States said she would look into the matter, but warned that the need to communicate with global teams could prevent an immediate response. Â®