Skip to contentSkip to conversation

PCI mmconfig support for Intel 915 bridges

PATCH In review Jesse Barnes 4 participants 5 messages Jan 11, 2007 – Apr 25, 2007

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 PCI mmconfig support for Intel 915 bridges

This is a resend of the patch I sent earlier to Oliver. It adds support for Intel 915 bridge chips to the new PCI MMConfig detection code. Tested and works on my sole 915 based platform (a Toshiba laptop). I added register masking per Oliver's suggestion, and moved the __init qualifier to after the 'static const char' to match Ogawa-san's recent cleanup patches.

Over time we can probably associate more PCI IDs with this routine, since i915 family contains a few other chips. But since I didn't have platforms to test such additions on, they're left out for now.

Thanks,
Jesse

arch/i386/pci/mmconfig-shared.c +21 -0
diff -Napur -X /home/jbarnes/dontdiff linux-2.6.19-mmconfig.orig/arch/i386/pci/mmconfig-shared.c linux-2.6.19-mmconfig/arch/i386/pci/mmconfig-shared.c
@@ -71,6 +71,26 @@ static __init const char *pci_mmcfg_e752
 	return "Intel Corporation E7520 Memory Controller Hub";
 }
 
+static const char __init *pci_mmcfg_intel_915(void)
+{
+	u32 pciexbar, len = 0;
+
+	pci_conf1_read(0, 0, PCI_DEVFN(0,0), 0x48, 4, &pciexbar);
+
+	/* No enable bit or size field, so assume 256M range is enabled. */
+	len = 0x10000000U;
+	pci_mmcfg_config_num = 1;
+	pciexbar &= 0xe0000000; /* mask out potentially bogus bits */
+
+	pci_mmcfg_config = kzalloc(sizeof(pci_mmcfg_config[0]), GFP_KERNEL);
+	pci_mmcfg_config[0].base_address = pciexbar;
+	pci_mmcfg_config[0].pci_segment_group_number = 0;
+	pci_mmcfg_config[0].start_bus_number = 0;
+	pci_mmcfg_config[0].end_bus_number = (len >> 20) - 1;
+
+	return "Intel Corporation 915PM/GM/GMS Express Memory Controller Hub";
+}
+
 static __init const char *pci_mmcfg_intel_945(void)
 {
 	u32 pciexbar, mask = 0, len = 0;
@@ -126,6 +146,7 @@ struct pci_mmcfg_hostbridge_probe {
 
 static __initdata struct pci_mmcfg_hostbridge_probe pci_mmcfg_probes[] = {
 	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7520_MCH, pci_mmcfg_e7520 },
+	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82915GM_HB, pci_mmcfg_intel_915 },
 	{ PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82945G_HB, pci_mmcfg_intel_945 },
 };
 
#1 lore raw
Quoted 11 lines · Jesse Barnes

This is a resend of the patch I sent earlier to Oliver. It adds support for Intel 915 bridge chips to the new PCI MMConfig detection code. Tested and works on my sole 915 based platform (a Toshiba laptop). I added register masking per Oliver's suggestion, and moved the __init qualifier to after the 'static const char' to match Ogawa-san's recent cleanup patches.

Over time we can probably associate more PCI IDs with this routine, since i915 family contains a few other chips. But since I didn't have platforms to test such additions on, they're left out for now.

Andrew, you sent me a series of emails to tell me the patches had moved to another subsystem tree, can you tell me which one? There is at least an anti-regression patch to add on top (the Asus/Nvidia special case is causing oopses right now).

OG.

#2 lore raw
In reply to Jesse Barnes

I see the other mmconfig stuff is upstream now, can we get this added too? The Asus bug Olivier mentioned also appears to be fixed, so this patch should be safe.

Thanks,
Jesse

Quoted 64 lines · Jesse Barnes

This is a resend of the patch I sent earlier to Oliver. It adds support for Intel 915 bridge chips to the new PCI MMConfig detection code. Tested and works on my sole 915 based platform (a Toshiba laptop). I added register masking per Oliver's suggestion, and moved the __init qualifier to after the 'static const char' to match Ogawa-san's recent cleanup patches.

Over time we can probably associate more PCI IDs with this routine, since i915 family contains a few other chips. But since I didn't have platforms to test such additions on, they're left out for now.

Thanks,
Jesse

diff -Napur -X /home/jbarnes/dontdiff linux-2.6.19-mmconfig.orig/arch/i386/pci/mmconfig-shared.c linux-2.6.19-mmconfig/arch/i386/pci/mmconfig-shared.c ---
linux-2.6.19-mmconfig.orig/arch/i386/pci/mmconfig-shared.c 2007-01-07
10:10:29.000000000 -0800 +++
linux-2.6.19-mmconfig/arch/i386/pci/mmconfig-shared.c 2007-01-10
18:46:46.000000000 -0800 @@ -71,6 +71,26 @@ static __init const char *pci_mmcfg_e752

arch/i386/pci/mmconfig-shared.c +12 -0
 	return "Intel Corporation E7520 Memory Controller Hub";
 }
 
+static const char __init *pci_mmcfg_intel_915(void)
+{
+	u32 pciexbar, len = 0;
+
+	pci_conf1_read(0, 0, PCI_DEVFN(0,0), 0x48, 4, &pciexbar);
+
+	/* No enable bit or size field, so assume 256M range is enabled. */
+	len = 0x10000000U;
+	pci_mmcfg_config_num = 1;
+	pciexbar &= 0xe0000000; /* mask out potentially bogus bits */
+
+	pci_mmcfg_config = kzalloc(sizeof(pci_mmcfg_config[0]),

GFP_KERNEL); + pci_mmcfg_config[0].base_address = pciexbar;

arch/i386/pci/mmconfig-shared.c +5 -0
+	pci_mmcfg_config[0].pci_segment_group_number = 0;
+	pci_mmcfg_config[0].start_bus_number = 0;
+	pci_mmcfg_config[0].end_bus_number = (len >> 20) - 1;
+
+	return "Intel Corporation 915PM/GM/GMS Express Memory Controller

Hub"; +}

arch/i386/pci/mmconfig-shared.c +1 -0
+
 static __init const char *pci_mmcfg_intel_945(void)
 {
 	u32 pciexbar, mask = 0, len = 0;
@@ -126,6 +146,7 @@ struct pci_mmcfg_hostbridge_probe {
 
 static __initdata struct pci_mmcfg_hostbridge_probe
 pci_mmcfg_probes[] = { { PCI_VENDOR_ID_INTEL,
 PCI_DEVICE_ID_INTEL_E7520_MCH, pci_mmcfg_e7520 }, +	{
 PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82915GM_HB,
 pci_mmcfg_intel_915 }, { PCI_VENDOR_ID_INTEL,

PCI_DEVICE_ID_INTEL_82945G_HB, pci_mmcfg_intel_945 }, };

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

#3 lore raw
Quoted 6 lines · On Wed, 25 Apr 2007 15:35:15 -0700 Jesse Barnes <[email protected]> wrote:

I see the other mmconfig stuff is upstream now, can we get this added too? The Asus bug Olivier mentioned also appears to be fixed, so this patch should be safe.

Thanks,
Jesse

^^^^^ is an evil top-poster!

Quoted 9 lines · Jesse Barnes
Quoted 7 lines

This is a resend of the patch I sent earlier to Oliver. It adds support for Intel 915 bridge chips to the new PCI MMConfig detection code. Tested and works on my sole 915 based platform (a Toshiba laptop). I added register masking per Oliver's suggestion, and moved the __init qualifier to after the 'static const char' to match Ogawa-san's recent cleanup patches.

Quoted 6 lines
Quoted 6 lines

Over time we can probably associate more PCI IDs with this routine, since i915 family contains a few other chips. But since I didn't have platforms to test such additions on, they're left out for now.

Can we have a refresh, retest and resend please?

#4 lore raw

Keep reading...

Quoted 1 line · Andrew Morton

On Wed, 25 Apr 2007 15:35:15 -0700 Jesse Barnes

Quoted 8 lines · <[email protected]> wrote:
Quoted 6 lines

I see the other mmconfig stuff is upstream now, can we get this added too? The Asus bug Olivier mentioned also appears to be fixed, so this patch should be safe.

Thanks,
Jesse

^^^^^ is an evil top-poster!

Hey I wanted to let you see the original! :)

Quoted 8 lines
Quoted 8 lines · Jesse Barnes
Quoted 7 lines

This is a resend of the patch I sent earlier to Oliver. It adds support for Intel 915 bridge chips to the new PCI MMConfig detection code. Tested and works on my sole 915 based platform (a Toshiba laptop). I added register masking per Oliver's suggestion, and moved the __init qualifier to after the 'static const char' to match Ogawa-san's recent cleanup patches.

Quoted 8 lines
Quoted 6 lines
Quoted 6 lines

Over time we can probably associate more PCI IDs with this routine, since i915 family contains a few other chips. But since I didn't have platforms to test such additions on, they're left out for now.

Can we have a refresh, retest and resend please?

Sure. I'll try to resend tomorrow or so.

Thanks,
Jesse