bug-gnu-utils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [PATCH] [RFC] adding support for .patches and /proc/patches.gz


From: Jon Oberheide
Subject: Re: [PATCH] [RFC] adding support for .patches and /proc/patches.gz
Date: Wed, 12 May 2004 00:59:28 -0400

On Tue, 2004-05-11 at 14:37, Bill Davidsen wrote:
> Jon Oberheide wrote:
> > Greetings,
> > 
> > This feature has been brought up several times before, as can be seen
> > here:
> > http://www.ussg.iu.edu/hypermail/linux/kernel/0404.3/0798.html
> > http://www.uwsg.iu.edu/hypermail/linux/kernel/0203.1/0598.html
> > http://www.uwsg.iu.edu/hypermail/linux/kernel/9803.0/0223.html
> > 
> > For those unfamiliar, a file linux/.patches would be adding to the
> > source tree.  When applying patches to the source tree, descriptive
> > information would be written to .patches.  After compilation and running
> > of this kernel, the .patches information would be accessible through
> > /proc/patches.gz; similar to the /proc/config.gz feature.
> 
> The first question would be, patches between the current kernel and 
> what? Vendor kernel, people may not have it. Kernel.org kernal, just the 
> patches to a current vendor kernel diff would be pretty huge in some cases.

Any patches applied against the current vanilla kernel.org kernel would
be listed in .patches.  This would include vendor, third-party, and even
pre/bk/mm patches.  

Keep in mind, .patches would not contain the entire patch, as that would
be WAY to large, but just a short entry such as the name, date last
modified, and date applied of the patch file.

> Let's say it looks like a high cost/benefit ratio, would be much less 
> effective unless it were used for every patch, and feels like something 
> you might want to do within an organization rather than as a general 
> practice.

Exactly as I stated, adoption would be the hardest part.  Paul's idea of
adding an option to patch w/o breaking POSIX sounds like a way to go. 
Of course that would require widespread documentation updates and
contacting vendors but would be very possible.

> Sorry, you asked for comments...

No need to be sorry, thanks!  :)

Regards,
Jon Oberheide
address@hidden

Attachment: signature.asc
Description: This is a digitally signed message part


reply via email to

[Prev in Thread] Current Thread [Next in Thread]