Messages in this thread | | | Date | Tue, 11 Sep 2007 00:56:53 +0200 | From | Rene Herman <> | Subject | Re: [-mm patch] unexport sys_{open,read} |
| |
On 09/11/2007 12:41 AM, Adrian Bunk wrote:
> On Tue, Sep 11, 2007 at 12:15:56AM +0200, Rene Herman wrote: >> On 09/11/2007 12:18 AM, Adrian Bunk wrote: >> >>> On Mon, Sep 10, 2007 at 01:17:59PM -0700, Andrew Morton wrote: >>>> There is no benefit in making some rigid set of rules. >>> Is is considered beneficial to provide API stability for external modules >>> or not? >> If I may... >> >> Yes, it is. Just not at any significant cost and Andrew is saying that he >> considers the _UNUSED() thing not significant. > > But there is no API stability for external modules this way.
I agree that doing things only half is semi-regularly worse than doing them not at all, and this specific case might be the worst example of all, as I read that using sys_open/read is actively harmful, so, well...
I read the thread since I tend to keep lots of external crap around. Not in any way that would mean I'd have any grounds for complaining about anything; mostly just driver stuff in various states of completeness that I never seem to get around to cleaning up enough to submit to anyone.
But as such, I can comment on the fact that I'm much more likely to notice the warning than I am to notice a thread on LKML, say. How much more likely I'd be to then also actually do anything about it before it just breaks anyway is another matter, but again, well...
> It simply doesn't make sense to give the few sys_open() abusers even > more grace period while changes to the IRQ API affecting nearly everyone > are allowed without any requirements of ensuring API stability. > > I'm not a fan of API stability for external modules, but if API > stability was considered important it should be done consequently and > not only for some patches that have the bad fate of having to go through > Andrew to Linus.
In this case I believe it makes sense to just rip it out, but generally it doesn't need to be such a fully robotic yes/no decision, I'd say.
Rene. - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |