| To: | Atsushi Nemoto <nemoto@toshiba-tops.co.jp> |
|---|---|
| Subject: | Re: new asm-mips/io.h |
| From: | Ralf Baechle <ralf@oss.sgi.com> |
| Date: | Wed, 28 Nov 2001 01:19:07 +1100 |
| Cc: | linux-mips@oss.sgi.com |
| In-reply-to: | <20011127.191022.27957874.nemoto@toshiba-tops.co.jp>; from nemoto@toshiba-tops.co.jp on Tue, Nov 27, 2001 at 07:10:22PM +0900 |
| References: | <20011126200946.A8408@dea.linux-mips.net> <20011127.130406.104026562.nemoto@toshiba-tops.co.jp> <20011127180648.H29424@dea.linux-mips.net> <20011127.191022.27957874.nemoto@toshiba-tops.co.jp> |
| Sender: | owner-linux-mips@oss.sgi.com |
| User-agent: | Mutt/1.2.5i |
On Tue, Nov 27, 2001 at 07:10:22PM +0900, Atsushi Nemoto wrote: > ralf> Well, talk to it's developers before it's too late. Or as it > ralf> has already happened for some hardware I think we should simply > ralf> go with your suggestion and make all those functions vectors. > > For me, currently it happens only in big endian and I can live happy > in a little endian world. I will create new patch when I REALLY need > it. Thank you. Right now the Linux philosophy is that we don't provide any swapping facility for I/O port accesses. Some hardware supports byteswapping for I/O port and memory access, others must do that in software't, so big endian systems can set CONFIG_SWAP_IO_SPACE to enable swapping for I/O ports and memory. In the past there has been quite some confusion about this and how to use this properly in drivers ... Ralf |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: failed installation debian-mipsel (Decstation 5000/150), Jan-Benedict Glaw |
|---|---|
| Next by Date: | Re: failed installation debian-mipsel (Decstation 5000/150), Maciej W. Rozycki |
| Previous by Thread: | Re: new asm-mips/io.h, Atsushi Nemoto |
| Next by Thread: | Re: new asm-mips/io.h, Jun Sun |
| Indexes: | [Date] [Thread] [Top] [All Lists] |