cvs commit: src/sys/dev/acpica acpi_pci.c src/sys/dev/pci pci.c

[M. Warner Losh]

In message: <200404270912.48113.jhb at FreeBSD.org>
            John Baldwin <jhb at FreeBSD.org> writes:
: On Sunday 25 April 2004 10:11 pm, Warner Losh wrote:
: > imp         2004/04/25 19:11:38 PDT
: >
: >   FreeBSD src repository
: >
: >   Modified files:
: >     sys/dev/acpica       acpi_pci.c
: >     sys/dev/pci          pci.c
: >   Log:
: >   Fix two typos from PR: 65694
: >
: >   1) In pci.c, we need to check the child device's state, not the parent
: >      device's state.
: >   2) In acpi_pci.c, we have to run the power state change after the acpi
: >      method when the old_state is > new state, not the other way around.
: 
: Umm, the ACPI method may turn the power to the device off (i.e., turn off the 
: power resource that provides power to the device), meaning that you can't 
: access it's PCI config registers anymore.  It's quite intentional to turn the 
: device off using PCI first and ACPI second and then to turn the device on via 
: ACPI first and then PCI second.
: 
: Unfortunately, the PR number is wrong (that's some ports PR) and there are no 
: PR's in the database with an originator that contains 'Remesov' or the text 
: 'acpi_pci' in the text fields at least none that query-pr-summary.cgi can 
: find.  Ugh, actually, query-pr is just hosed as it thinks we have no open 
: PRs.  *sigh*

You misunderstand.  When turning the device OFF, we run the acpi
last.  When turning the device ON, we run it first.  The code ran it
twice in one case and never in the other.

Warner