Messages in this thread | | | Subject | Re: Article: IBM wants to "clean up the license" of Linux | Date | Sun, 03 Jan 1999 21:21:17 -0500 | From | "Brandon S. Allbery KF8NH" <> |
| |
In message <19990103205053.A4065@tiktok.cygnus.com>, Michael Meissner writes: +----- | On Sat, Jan 02, 1999 at 06:34:27PM -0500, Brandon S. Allbery KF8NH wrote: | > (config.guess rant: *why* the exact processor ID? About half of configure | > scripts fall over in ECE Linux builds because they don't expect "i686". | > This should be "x86". And if it has to be exact, where are AMD and Cyrix?) | | If you configure it for i686-unknown-linux-gnu, then the default compiler will | compile code that is tuned for an i686 (ie, Pentium-Pro, Pentium-II), but still | only use instructions common to all members of the x86 family. Similarly if +--->8
I'm not talking about the kernel or about gcc. I'm talking about random packages that *won't configure on Pentium II boxes* because config.guess coughs up "i686-...", config.sub sees it and aborts --- its patterns look for "i[345]86-" (and occasionally "i[34]86-", in older packages). I now keep around a replacement config.guess and config.sub for these packages.
Some of the packages which do this are fairly recent, too.
This is an autoconf issue; if you insist on tracking exact processor versions, get rid of the charclass in config.sub or make it accept any digit > 2. Otherwise the world breaks yet again when someone decides to detect the next IA32 increment as an "i786", just as large parts of the world break now on "i686".
-- brandon s. allbery [os/2][linux][solaris][japh] allbery@kf8nh.apk.net system administrator [WAY too many hats] allbery@ece.cmu.edu carnegie mellon / electrical and computer engineering KF8NH We are Linux. Resistance is an indication that you missed the point.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |