diff options
author | Grant Grundler <grundler@parisc-linux.org> | 2005-10-21 22:37:43 -0400 |
---|---|---|
committer | Kyle McMartin <kyle@parisc-linux.org> | 2005-10-21 22:37:43 -0400 |
commit | 86a61ee9c9f3d8b632d29e86ac6610c43ebbb4f0 (patch) | |
tree | d9289057621307b63d5f0f902ab0f37826ccb1f8 /drivers/parisc/dino.c | |
parent | 64908ad95c34f25849412d6d4735ac10f8fb6575 (diff) | |
download | op-kernel-dev-86a61ee9c9f3d8b632d29e86ac6610c43ebbb4f0.zip op-kernel-dev-86a61ee9c9f3d8b632d29e86ac6610c43ebbb4f0.tar.gz |
[PARISC] Update ccio-dma from parisc tree
revert use of %%sr0 in fdc asm.
Thanks to Joel Soete for pointing out this oversight.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
2.6.14-rc2-pa3 fdc/lci should be %r0 instead 0 for index (PA 1.1 compliance)
From: Joel Soete <soete.joel@tiscali.be>
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Explain why we need insert_resource() instead of request_resource().
Fundementally, this is more convoluted for ccio driver because of
o legacy (HP-PB) transperant bridges.
o support for MMIO behind card-mode Dino (PCI)
o support for above bridges without ccio in the box
SBA driver doesn't have to worry about those issues.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Use insert_resource instead of request_resource now that the subdevices
will already have their resources claimed
Signed-off-by: Matthew Wilcox <willy@parisc-linux.org>
re-enable use of "inline" for perf critical functions.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
2.6.12-rc4-pa5 fix sign extension of MMIO range
Fixes the problem of claiming a range that is disabled on 64-bit kernel:
ccio_init_resource() claimed CCIO bus address space (ffffffff00000000,
ffffffffffffffff)
also removes use of __FILE__.
Tested on both 32 and 64-bit systems by Joel.
From: Joel Soete <soete.joel@tiscali.be>
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
2.6.12-rc1-pa7 incorrect BUG_ON in ccio
ccio-dma.c line 1317 was preventing K-class with 4GB RAM from booting.
Any ccio machine with >=2GB of RAM would have (incorrectly) triggered this.
Signed-off-by: Grant Grundler <grundler@parisc-linux.org>
Convert to ioremap and __raw_read/write
Signed-off-by: Matthew Wilcox <willy@parisc-linux.org>
Signed-off-by: Kyle McMartin <kyle@parisc-linux.org>
Diffstat (limited to 'drivers/parisc/dino.c')
0 files changed, 0 insertions, 0 deletions