On 05-Jan-2004 Shizuka Kudo wrote: > --- John Baldwin <jhb_at_FreeBSD.org> wrote: >> >> On 23-Dec-2003 Brian Black wrote: >> > hi, >> > I have been trying to post to different lists looking for a >> > solution. though i have found some advice i have not found the >> > solution. i was recently informed that i should post this question to >> > the -current list(i should of done this earlier). I have a "Soyo Dragon >> > Light" mother board which has this award bios. I have verified that my >> > bios has been flashed with the most up-to date version(Well whatever the >> > motherboard manufacture says). >> > Now am certainly not a kernel hacker but i do understand a little >> > bit of C code. The PR goes on to talk about a "Fix", which involves >> > editing the "/urs/sys/dev/acpica/acpi_pci_link.c". After making the >> > changes i had no luck :-(. Now i have been told that some people with >> > this bios are not affected. Well would'nt that mean the the bios are >> > not the same? I have contacted "Soyo" on this matter. They state on >> > their web site that freebsd is supported, but have not returned my >> > e-mail. >> > How would i be able to determine weather this problem can be fixed >> > by FreeBsd or the manufacture of the mother board? Is This problem >> > being looked at for 5.2R i have not been able to really sink into using >> > the new 5.x series because of this problem(So what i like GUI's a >> > little). ;-) >> >> Hmm, yes. >> >> Can you try the patch at >> http://www.FreeBSD.org/~jhb/patches/acpi_link_sta.patch > > I have tried your patch with no luck. I still need to revert to test on > ACPI_ENABLE as described on 55473 This patch does test ENABLE. Do you have a working patch that you can show me? >> Index: acpi_pci_link.c >> =================================================================== >> RCS file: /usr/cvs/src/sys/dev/acpica/acpi_pci_link.c,v >> retrieving revision 1.11 >> diff -u -r1.11 acpi_pci_link.c >> --- acpi_pci_link.c 18 Dec 2003 03:25:22 -0000 1.11 >> +++ acpi_pci_link.c 23 Dec 2003 16:29:35 -0000 >> _at__at_ -489,7 +489,16 _at__at_ >> return_ACPI_STATUS (error); >> } >> >> - if (!(sta & (ACPI_STA_PRESENT | ACPI_STA_FUNCTIONAL))) { >> + /* >> + * We really should not use link devices that aren't enabled, >> + * but if they are marked as either functional or enabled, we >> + * assume they are ok. >> + * >> + * XXX: The PCI bridge code requires that both PRESENT and >> + * ENABLE are set in a separate check. >> + */ >> + if ((sta & ACPI_STA_PRESENT) == 0 || >> + (sta & (ACPI_STA_ENABLE | ACPI_STA_FUNCTIONAL)) == 0) { >> ACPI_DEBUG_PRINT((ACPI_DB_ERROR, >> "PCI interrupt link is not functional - %s\n", >> acpi_name(handle))); >> -- John Baldwin <jhb_at_FreeBSD.org> <>< http://www.FreeBSD.org/~jhb/ "Power Users Use the Power to Serve!" - http://www.FreeBSD.org/Received on Mon Jan 05 2004 - 13:30:26 UTC
This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:36 UTC