Re: everyone who has VM problems/messages (writable swappage or crash or lockup)
Linus Torvalds (torvalds@transmeta.com)
9 Jul 1998 19:02:33 GMT
In article <Pine.LNX.3.96.980709142708.29927B-100000@lo-pc3035a>,
Alex Buell <alex.buell@tahallah.demon.co.uk> wrote:
>On Thu, 9 Jul 1998, Alan Cox wrote:
>
>> Its quite happy to rearrange the order of operations for performance and
>> to remove "dead code". Of course the former when not correctly marked in
>> device drivers is a disaster and the latter tends to include little things
>> like delay loops that have been handled wrongly.
>
>Ahh, now I understand what the problem is. Maybe we need an option to add
>to egcs, something like -wlobotomy or something like that. :o) And rename
>it ofotcn (One flew over the cuckoo's nest).. excuse my silliness!
>
>Oh well, at least having two compilers that works so differently helps to
>expose "holes" in the kernel sources, and improve its portability to other
>architectures, right?
Let's not be too happy about egcs. _Some_ of the reports of problems
were due to legal optimizations done by egcs.
Some were not.
Linus
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu