Messages in this thread | | | Date | Wed, 12 May 1999 10:29:02 -0700 (PDT) | From | Linux Lists <> | Subject | Re: Weird PCI problem |
| |
On Tue, 11 May 1999, Martin Mares wrote: > > > Using Cyctest, a diagnostics software from Cyclades. > > In NT, it was tested using the NT driver. > > > > Interestingly enough, the NT Diagnostics indicates that the _same_ > > resource Linux tries to allocate (i.e., f9e00000-f9efffff) are being > > allocated by NT, just in the right way. > > It's quite possible both the DOS diag program and the NT driver > completely ignore the region type bit.
Well, even if I try to ignore this bit in Linux, I can't access that memory area.
I've tried changing cyzload (a userspace program that loads the Cyclades-Z FPGA and its firmware by accessing its memory-mapped space) to read the address in "I/O at" instead of in "memory at" from /proc/pci (as this address already has the bit 0 masked, and it's supposely the correct one, as NT allocates the same address), but I'm still _NOT_ able to write to that memory area (I get an error on the program saying that it was not able to configure the Cyclades-Ze FPGA).
> If it turns out it's a real hardware bug, I can add it to the work-around > table in the kernel to trick the kernel that it's really a memory region ;)
In the light of the aforementioned fact, I don't this would help.
Do you have any other suggestions ?! I'm starting to run out of options ... :(
Regards, Ivan
- 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/
| |