Skip to contentSkip to conversation

Add Micro-Star as a PCI VENDOR

PATCH In review Gordon Stanton 3 participants 4 messages Jan 25, 2007 – Jan 29, 2007

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 Add Micro-Star as a PCI VENDOR

Hi,
I noticed that a common motherboard manufacturer Micro-Star was missing it's PCI_VENDOR_ID and so I have sent this patch in. Later on I will send a patch to pci/quirks.c to unhide the SMBus on the M510C laptop once I have investigated it a bit more. It hides the SMBus the same as ASUS (and a couple of others) laptops. This is definately the correct ID, see
http://www.pcidatabase.com/vendor_details.php?idp5

I would also be happy with PCI_VENDOR_ID_MICRO_STAR instead, what do you think?

Signed-off-by Gordon Stanton <fixacpi at gmail.com>

---

include/linux/pci_ids.h +2 -0
@@ -1875,6 +1875,8 @@
 
 #define PCI_VENDOR_ID_SAMSUNG		0x144d
 
+#define PCI_VENDOR_ID_MICROSTAR		0x1462
+
 #define PCI_VENDOR_ID_MYRICOM		0x14c1
 
 #define PCI_VENDOR_ID_TITAN		0x14D2
#1 lore raw
Quoted 11 lines · Gordon Stanton

Hi,
I noticed that a common motherboard manufacturer Micro-Star was missing it's PCI_VENDOR_ID and so I have sent this patch in. Later on I will send a patch to pci/quirks.c to unhide the SMBus on the M510C laptop once I have investigated it a bit more. It hides the SMBus the same as ASUS (and a couple of others) laptops. This is definately the correct ID, see
http://www.pcidatabase.com/vendor_details.php?idp5

I would also be happy with PCI_VENDOR_ID_MICRO_STAR instead, what do you think?

I suggest sending this patch in when you need to use it, i.e. at the same time you send the quirk patch in.

thanks,

greg k-h

#2 lore raw

PATCH 1/1 Add Micro-Star as a PCI VENDOR

In reply to Gordon Stanton

Hi Gordon,

Posting to 6 mailing-lists at once, are you mad?
I kept only the relevant ones.

Quoted 10 lines · Greg KH
Quoted 9 lines · Gordon Stanton

Hi,
I noticed that a common motherboard manufacturer Micro-Star was missing it's PCI_VENDOR_ID and so I have sent this patch in. Later on I will send a patch to pci/quirks.c to unhide the SMBus on the M510C laptop once I have investigated it a bit more. It hides the SMBus the same as ASUS (and a couple of others) laptops. This is definately the correct ID, see
http://www.pcidatabase.com/vendor_details.php?idp5

Quoted 2 lines
Quoted 2 lines

I would also be happy with PCI_VENDOR_ID_MICRO_STAR instead, what do you think?

I would suggest PCI_VENDOR_ID_MSI, that's how people know them.

Quoted 2 lines

I suggest sending this patch in when you need to use it, i.e. at the same time you send the quirk patch in.

I second this suggestion.

Thanks,

Signature · 1 line
Jean Delvare
#3 lore raw

PATCH 1/1 Add Micro-Star as a PCI VENDOR

In reply to Gordon Stanton

Hi Gordon,

Quoted 2 lines · Gordon Stanton

Hi Jean,
Firstly, sorry for the personal reply. I thought it might be better though.

For what reason? First you send your message to 6 mailing lists, then replies go to just one person? I'm adding the relevant lists and persons to Cc again.

Quoted 5 lines · Jean Delvare
Quoted 2 lines

Posting to 6 mailing-lists at once, are you mad?
I kept only the relevant ones.

SMBus on PCI with ACPI crosses many areas.

I fail to see how it is related to ACPI at all.

Quoted 2 lines

I'ts fixes a PCI quirk.(linux-pci)
The fix will unhide the SMBus like ASUS needs.(acpi4asus)

Irrelevant. Asus were the first ones to do it, but half a dozen vendors do it by now. And the Asus system users certainly couldn't care less that yet another system needs it.

Quoted 2 lines

Hopefully enabling some more sensors (like fans) to be seen.(linux-acpi,lm-sensors)

"Hopefully?" You'd better make sure first. We're not going to add an unhiding quirk if there's nothing useful on the SMBus in question.

Quoted 4 lines
Quoted 1 line

I would suggest PCI_VENDOR_ID_MSI, that's how people know them.

I thought of that but MSI in relation to PCI means "Message Signaled Interrupt" so I would rather go with PCI_VENDOR_MICRO_STAR which also

Come on, the symbol is named PCI_VENDOR_ID_<something>, it's pretty clear what it stands for. If someone really thinks that PCI_VENDOR_ID_MSI could be related to message-signaled interrupts, that person better doesn't touch the kernel code at all.

We already have 62 two- or three-letter PCI vendor ID symbols in pci_ids.h, one more really isn't a problem.

Quoted 1 line

tabs in the source code nicely. Although in truth it probably should

"Tabs in the source code nicely"? What do you mean?

Quoted 1 line

be PCI_VENDOR_MICRO_STAR_INTL but that is just getting too long.

That I couldn't agree more.

Signature · 1 line
Jean Delvare