[Fwd: RE: cciss patches for 2.4.21-rc1, 4 of 4]
PATCH 1/1 [Fwd: RE: cciss patches for 2.4.21-rc1, 4 of 4]
-----Forwarded Message-----
Quoted 6 lines
From: Alan Cox <[email protected]>
To: "Miller, Mike (OS Dev)" <[email protected]>
Cc: "Cameron, Steve" <[email protected]>, Linux Kernel Mailing List <[email protected]>
Subject: RE: cciss patches for 2.4.21-rc1, 4 of 4
Date: 25 Apr 2003 14:09:53 +0100
Quoted 8 lines
Quoted 1 line · Mike (OS Dev)
I haven't seen any issues (yet) on ia64. I'm running with 5GB RAM.
That doesn't make it correct. This same problem occurs in other drivers and the usual trick is to set the pci mask to 32bit, allocate those command buffers ready, then flip back to 64bit. Just be sure one thread doesn't change it to 64bit while another is allocating commands. The reverse is fine since an accidental odd bounce is no big deal
Alan,
except that pci_alloc_consistent is guaranteed to return a 32 bit address
what you're thinking about is the mapping interface instead.