Unified getcwd() implementation

[Tim Robbins]

On Sat, May 08, 2004 at 12:55:56AM +0100, Bruce M Simpson wrote:

> On Fri, May 07, 2004 at 08:01:19PM +1000, Tim Robbins wrote:
> > Why is this necessary? Emulation of the Linux getcwd() syscall belongs
> > in the Linux emulator. There is also some fairly blatantly plagiarised code
> > in this patch.
> 
> I rolled the patch before I was a committer in response to an item on
> PHK's JKH tasklist.
> 
> The getcwd() implementation in the Linux module is not inherently Linux
> specific; but it does implement the name-finding without using the name
> cache, by scanning directories. The patch merges the name cache and
> non-name-cache based implementations of getcwd() to address the PRs which
> Marc cited in his earlier message.

It's not inherently Linux-specific, but it exists solely for the benefit
of Linux emulation. I don't see why they need to be merged at the
expense of complicating the native system calls -- we have way
too many tentacles of the Linux emulator reaching in there already.

> You seem to be concerned about plagiarism. Could you clarify or explain
> your concerns?

getcwd_scandir() has clearly been copied from linux_getcwd.c. The work
needs to be attributed to the author, Bill Sommerfeld, and the NetBSD
license needs to be honoured.


Tim