Another mainline push for "go back and rewrite it more generally" [LWN.net]

Another mainline push for "go back and rewrite it more generally"

Posted Dec 3, 2009 15:30 UTC (Thu) by NAR (subscriber, #1313)
Parent article: Another mainline push for utrace

This article underlines the statement in the previous article ("In his travels, your editor has heard complaints from developers who set out to accomplish a specific task, only to be told that they must undertake a much larger cleanup to get their code merged") with a number of examples:

It was suggested that providing a real user along with utrace...
utrace has returned [...] with a rewrite of the ptrace() [...] [should] add utrace support to the two major architectures that don't have it
A patch to expose a /proc/PID/gdb interface [...] is running into more serious difficulty [...] should be combined [with kgdb]

This is clearly a development process issue. At our project most of the developers work on adding new features (after all, this is what the customer wants), but some time is set aside to do maintenance and clean-up tasks to keep the code in a healthy state. Maybe the kernel developers could try something like this. This is a lot more than janitorial work, because it might need deep understanding of the affected areas and probably shouldn't be done by the odd developer who scratched his itch to implement some feature.

(Log in to post comments)