As root, please do (_exactly_ as shown) ...
mmapw -b /dev/fbs/aperture 0x01FE01002864 0x7Cmmapr /dev/fbs/aperture 0x01FE01010000 4 > /dev/null
... and tell me if the second command (the `mmapr`) still crashes your system. The `mmapw` is very important here. Its effect is to tell your PCI bridge to ignore most PCI errors that occur on its secondary bus segment.
Marc, still the system seems to crashing for the mmapr. I have executed first mampw followed by mmapr. I confirmed this beahviour by trying twice. Is there any way that I could confirm my mmapw of 0x7c is successfull using? Will this work "mmapr -b /dev/fbs/aperture 0x01FE01002864". If so were can I
To verify that the value was correctly written, you can pipe mmapr's stdout into /usr/bin/od, i.e. ...
mmapr -b /dev/fbs/aperture 0x01FE01002864 | /usr/bin/od -t x1
There is a typo here. This should have been ...
mmapr -b /dev/fbs/aperture 0x01FE01002864 1 | /usr/bin/od -t x1
... i.e. with the length of the area specified.
I should, at some point, change mmapr to optionally pretty-print its output.
At this point, I need to ask you to install an adapter into one of the machine's PCI slots. Any adapter will do, as long as it's PCI of course. It doesn't even need to be a video adapter. The intent is to put an adapter behind the PCI bridge. If after installing this adapter, the command ...
mmapr /dev/fbs/aperture 0x01FE01010000 4 > /dev/null
... no longer crashes the system, then please send me the pci5.dat2 that results from ...
mmapr /dev/fbs/aperture 0x01FE01002800 256 > pci5.dat2
... and I'll compare that with the pci5.dat you've previously sent. I'll also need to see the updated `prtconf -Ppv` output.
Actually, come to think of it, the prtconf output would be useful whether or not installing the extra adapter crashes the system on the mmapr command.
Marc even after adding the PCI card, still the system crashing. I had Permedia card which I wanted to finally test with XFree86 in sparc. After placing the new card in PCI slot I have taken the "prtconf -Ppv" output, Please refer the attached log "logPci.txt".
Still I have doubt with the mmapw of "0x7C". Because when I read back I am getting octal value of 0. Following is the sequence of comamnd I have executed in the shell.
bash-2.03# mmapw -b /dev/fbs/aperture 0x01FE01002864 0x7C bash-2.03# bash-2.03# mmapr -b /dev/fbs/aperture 0x01FE01002864 | /usr/bin/od -t x1
mmapr [-{bwlqL}] <file> <offset> <length>
endianness flags:
-b output one byte at a time -w output up to two aligned bytes at a time -l output up to four aligned bytes at a time (default) -q output up to eight aligned bytes at a time -L same as -l
0000000 bash-2.03# mmapr /dev/fbs/aperture 0x01FE01010000 4 > /dev/null
------ System hanged here ---------
Is the second mmapr is as expected or not ?
See above.
Given the system still hangs, even after installing an adapter, I have requested access to such a machine here on campus so that I can explore this problem more thouroughly. I'll get back to you.
Marc.
+----------------------------------+-----------------------------------+ | Marc Aurele La France | work: 1-780-492-9310 | | Computing and Network Services | fax: 1-780-492-1729 | | 352 General Services Building | email: [EMAIL PROTECTED] | | University of Alberta +-----------------------------------+ | Edmonton, Alberta | | | T6G 2H1 | Standard disclaimers apply | | CANADA | | +----------------------------------+-----------------------------------+ XFree86 developer and VP. ATI driver and X server internals. _______________________________________________ XFree86 mailing list [EMAIL PROTECTED] http://XFree86.Org/mailman/listinfo/xfree86
