| Message ID | 20200123070913.626488-18-mst@redhat.com |
|---|---|
| State | New |
| Headers | show |
| Series | [PULL,v3,01/18] q35: implement 128K SMRAM at default SMBASE address | expand |
On Thu, 23 Jan 2020 at 07:11, Michael S. Tsirkin <mst@redhat.com> wrote: > > From: Corey Minyard <cminyard@mvista.com> > > Per the ACPI spec (version 6.1, section 6.1.5 _HID) it is not required > on enumerated buses (like PCI in this case), _ADR is required (and is > already there). And the _HID value is wrong. Linux appears to ignore > the _HID entry, but Windows 10 detects it as 'Unknown Device' and there > is no driver available. See https://bugs.launchpad.net/qemu/+bug/1856724 Is this commit in fact a fix for LP:1856724 ? If so, we could usefully add a comment to that bug noting the commit which fixes it and mark the bug 'fix committed', since it seems to affect various users who would like to know the status. thanks -- PMM
On Thu, Jan 30, 2020 at 07:05:16PM +0000, Peter Maydell wrote: > On Thu, 23 Jan 2020 at 07:11, Michael S. Tsirkin <mst@redhat.com> wrote: > > > > From: Corey Minyard <cminyard@mvista.com> > > > > Per the ACPI spec (version 6.1, section 6.1.5 _HID) it is not required > > on enumerated buses (like PCI in this case), _ADR is required (and is > > already there). And the _HID value is wrong. Linux appears to ignore > > the _HID entry, but Windows 10 detects it as 'Unknown Device' and there > > is no driver available. See https://bugs.launchpad.net/qemu/+bug/1856724 > > Is this commit in fact a fix for LP:1856724 ? If so, we could > usefully add a comment to that bug noting the commit which > fixes it and mark the bug 'fix committed', since it seems > to affect various users who would like to know the status. > > thanks > -- PMM Right. Corey could you do that pls?
I checked a few days ago and someone had already beat me to it. -corey On Mon, Feb 3, 2020, 06:34 Michael S. Tsirkin <mst@redhat.com> wrote: > On Thu, Jan 30, 2020 at 07:05:16PM +0000, Peter Maydell wrote: > > On Thu, 23 Jan 2020 at 07:11, Michael S. Tsirkin <mst@redhat.com> wrote: > > > > > > From: Corey Minyard <cminyard@mvista.com> > > > > > > Per the ACPI spec (version 6.1, section 6.1.5 _HID) it is not required > > > on enumerated buses (like PCI in this case), _ADR is required (and is > > > already there). And the _HID value is wrong. Linux appears to ignore > > > the _HID entry, but Windows 10 detects it as 'Unknown Device' and there > > > is no driver available. See > https://bugs.launchpad.net/qemu/+bug/1856724 > > > > Is this commit in fact a fix for LP:1856724 ? If so, we could > > usefully add a comment to that bug noting the commit which > > fixes it and mark the bug 'fix committed', since it seems > > to affect various users who would like to know the status. > > > > thanks > > -- PMM > > Right. Corey could you do that pls? > >
On Mon, 3 Feb 2020 at 12:03, Corey Minyard <cminyard@mvista.com> wrote: > > I checked a few days ago and someone had already beat me to it. Yeah, we had a discussion on IRC about it and came to the conclusion that this was a fix. To avoid future confusion, my suggestion would be that commit messages that fix bugs should explicitly say they fix bugs they refer to; "See $BUG" implies to me a looser connection like "this is another bug in this area" or "this partially addresses that bug" or "this bug report mentioned this issue in passing while describing a different bug". thanks -- PMM
On Mon, Feb 03, 2020 at 12:10:33PM +0000, Peter Maydell wrote: > On Mon, 3 Feb 2020 at 12:03, Corey Minyard <cminyard@mvista.com> wrote: > > > > I checked a few days ago and someone had already beat me to it. > > Yeah, we had a discussion on IRC about it and came to the > conclusion that this was a fix. To avoid future confusion, > my suggestion would be that commit messages that fix bugs > should explicitly say they fix bugs they refer to; "See $BUG" > implies to me a looser connection like "this is another bug in > this area" or "this partially addresses that bug" or > "this bug report mentioned this issue in passing while > describing a different bug". > > thanks > -- PMM Right. Probably Fixes: <> that (I think) github pioneered.
diff --git a/hw/i386/acpi-build.c b/hw/i386/acpi-build.c index e25df838f0..9c4e46fa74 100644 --- a/hw/i386/acpi-build.c +++ b/hw/i386/acpi-build.c @@ -1816,7 +1816,6 @@ static void build_smb0(Aml *table, I2CBus *smbus, int devnr, int func) Aml *scope = aml_scope("_SB.PCI0"); Aml *dev = aml_device("SMB0"); - aml_append(dev, aml_name_decl("_HID", aml_eisaid("APP0005"))); aml_append(dev, aml_name_decl("_ADR", aml_int(devnr << 16 | func))); build_acpi_ipmi_devices(dev, BUS(smbus), "\\_SB.PCI0.SMB0"); aml_append(scope, dev);