RealTime IT News

Linux Creator Calls Backporting 'Good Thing'

The creator of the Linux operating system, Linus Torvalds, has weighed in on the issue of backporting features from newer Linux kernels into older ones, calling the practice a good thing for the most part.

Torvalds comments, in an e-mail interview with internetnews.com, came after SUSE'S CTO, Juergen Geck, told an audience at the Real World Linux Conference in Toronto that Red Hat's practice of backporting features from the 2.6 kernel into the 2.4 Kernel is a "bad thing" because it interferes with standardization of the open source operating system.

Geck's comments were part of a keynote theme that urged industry players to avoid practices that could fragment open source standardization efforts.

The comments sparked discussions in the open source community. When asked by e-mail to comment for internetnews.com, Torvalds wrote:

"I think it makes sense from a company standpoint to basically 'cherry-pick' stuff from the development version that they feel is important to their customers. And in that sense I think the back-porting is actually a very good thing."

A vice president of Novell and co-founder of Ximian, which, like SUSE is also owned by Novell , also defended Red Hat's practice of backporting. "Very few people ship a vanilla Linus Torvalds kernel," Miguel de Icaza told internetnews.com.

He explained that developers will often bundle features that customers require, but that don't make it in the official kernel release. For example, Red Hat backported the Native Posix Threading Library (NPTL) from its 2.5.x development tree to the 2.4 kernel, which were part of larger goals of achieving scalability and stability enhancements in current enterprise releases.

"I am very happy that they decided to do this work, since the NPTL thread stack is very important to run multi-threaded applications reliably," de Icaza said.

De Icaza said his current project Mono, an open source implementation of the .NET framework, has also benefited from the backported NPTL feature.

"The Mono Debugger is an application that was vastly simplified by the use of NPTL," he told internetnews.com. "We would not have been able to remove a large portion of complexity without it."

The recently re-elected leader of the Debian Linux project said some backporting of features just make sense. "I think users benefit from backporting features from 2.6 selectively," Martin Michlmayr told internetnews.com.

"Our kernel maintainer has a clear policy that only features not in 2.4 will be accepted for inclusion as additional patches if they are in 2.6. This ensures that the feature is supported and will be presented in future kernels."

Michlmayr also noted that backporting certain features to version 2.4 of the Linux kernel can also allow for easier migration to the 2.6 version.

However, Bruce Perens, a former Debian Project Leader and author of the Open Source Definition, wasn't as quick to compliment Red Hat.

In a public post, Perens wrote, "I have a large customer who refuses to run Red Hat's kernel even when they run Red Hat's distribution. And it's just for the reason that [SUSE] talks about. The kernel is so far diverged from the main thread of Linux that it's a dead-end, and there's no hope of getting it supported from anyone but Red Hat. I don't know if they meant it as a lock-in play, but it works out that way. And my customer doesn't have patience for Red Hat's support."

Despite his comments, Perens told internetnews.com he didn't think the issue was that big a deal and hoped the community wouldn't over-react. However he noted that if the kernel wasn't under the widely used General Public License , it would be.

"I want to emphasize, though, that the GPL mitigates forks" Perens said. "Since all of the parties involved can copy any software from any of the forks into their own one, because of the GPL license, forks tend to merge."

Gael Duval, founder of the Mandrake Linux distribution, also expressed some doubts about the practice of backporting features, though he admitted that his working group does it on occasion.

"It's certainly not a good idea to do that for consumers because it won't be supported officially by the Linux community of developers," Duval told internetnews.com.

But occasionally, such as on behalf of a client's needs, the practice comes in handy. "At Mandrakesoft we happen to do it for large accounts when they need a special feature to be integrated in a kernel."

And even Torvalds' support of the practice comes with some caveats. "There are parts of it that worry me logistically," Torvalds wrote in the e-mail to internetnews.com. "What usually ends up happening is that the back-ported patches aren't being very cleanly maintained, and that ends up making it harder for people to do a good job of maintaining a coherent base for the stable kernel."

Although kernel 'coherency' is a victim of backported features, according to Torvalds, its impact is not long lived. "That lack of 'coherency' makes long-term maintenance harder (and is probably why the SuSE people aren't thrilled, because it also makes it harder to keep different trees reasonably well in sync)," Torvalds continued.

"But as long as the long-term goal ends up to drop the old stable kernel in favour of the development kernel anyway, the pain is likely to be fairly temporary."

Torvalds said he would prefer to have more users use the development version of the kernel, though he admitted that there is an upside to having others backport development features. "The fact that having part of the features back-ported does mean that those back-ported features get much wider testing. Which is a huge boon," Torvalds wrote.

"So you win some, you lose some, so far I suspect it's been mostly positive."