From: "Singapore Citizen Mr. Teo En Ming (Zhang Enming)" <singapore.mr.teo.en.ming@gmail.com>
To: "xen-users@lists.xen.org" <xen-users@lists.xen.org>,
"xen-devel@lists.xen.org" <xen-devel@lists.xen.org>,
"Teo En Ming (Zhang Enming)" <singapore.mr.teo.en.ming@gmail.com>,
"Pasi Kärkkäinen" <pasik@iki.fi>,
"Ian Campbell" <Ian.Campbell@citrix.com>,
"Konrad Rzeszutek Wilk" <konrad.wilk@oracle.com>
Subject: Result of Applying IGD VGA Passthrough Patches to Xen 4.4-unstable Changeset 27238
Date: Tue, 16 Jul 2013 12:31:28 +0800 [thread overview]
Message-ID: <51E4CCA0.8000604@gmail.com> (raw)
[-- Attachment #1: Type: text/plain, Size: 1449 bytes --]
Hi,
I have just applied IGD VGA Passthrough patches to Xen 4.4-unstable
changeset 27238. Please refer to the 2 attached patches.
1. [Xen-devel] [PATCH 2/3] V4 qemu-xen-trad: Correctly expose PCH ISA
bridge for IGD passthrough
Link:
http://lists.xenproject.org/archives/html/xen-devel/2013-06/msg01720.html
2. [Xen-devel] [PATCH 3/3] qemu-xen-trad: IGD passthrough: Expose vendor
specific pci cap on host bridge.
Link:
http://lists.xenproject.org/archives/html/xen-devel/2013-02/msg00538.html
These are the steps I took to patch, compile and install Xen
4.4-unstable changeset 27238.
cd
hg clone -r 27238 http://xenbits.xensource.com/xen-unstable.hg
xen-unstable.hg-cs27238
cd xen-unstable.hg-cs27238
sudo -s
./configure
make world
make clean
cd tools/qemu-xen-traditional-dir-remote
patch -p1 < patch2of3.txt
patch -p1 < patch3of3.txt
cd ../..
make world
make install
However, I still encountered problems installing Intel HD Graphics
display drivers. The driver installation would auto detect the display
adapter, causing the LCD monitor to blink 1-2 times. But after that, the
LCD monitor goes into Power Saving mode (blank screen) forever again. I
tried to destroy the Windows 8 HVM domU and restart it. Windows 8 HVM
domU would start up for a few seconds, then falls back to blank LCD
monitor screen forever again.
Please advise.
Thank you very much.
--
Yours sincerely,
Singapore Citizen Mr. Teo En Ming (Zhang Enming)
[-- Attachment #2: patch2of3.txt --]
[-- Type: text/plain, Size: 2523 bytes --]
Signed-off-by: Stefano Stabellini <stefano.stabellini@xxxxxxxxxxxxx>,
Rui Guo <firemeteor@xxxxxxxxxxxxxxxxxxxxx>
Tested-by: Rui Guo <firemeteor@xxxxxxxxxxxxxxxxxxxxx>
Xen-devel: http://marc.info/?l=xen-devel&m=135548433715750
---
hw/pci.c | 10 ++++++++++
hw/pci.h | 3 +++
hw/pt-graphics.c | 9 ++++++---
3 files changed, 19 insertions(+), 3 deletions(-)
diff --git a/hw/pci.c b/hw/pci.c
index f051de1..c423285 100644
--- a/hw/pci.c
+++ b/hw/pci.c
@@ -938,6 +938,16 @@ PCIBus *pci_bridge_init(PCIBus *bus, int devfn, uint16_t vid, uint16_t did,
return s->bus;
}
+PCIBus *pci_isa_bridge_init(PCIBus *bus, int devfn, uint16_t vid, uint16_t did,
+ uint8_t rid, pci_map_irq_fn map_irq, const char *name)
+{
+ PCIBus *s = pci_bridge_init(bus, devfn, vid, did, rid, map_irq, name);
+
+ pci_config_set_class(s->parent_dev->config, PCI_CLASS_BRIDGE_ISA);
+ s->parent_dev->config[PCI_HEADER_TYPE] = 0x80;
+ return s;
+}
+
int pt_chk_bar_overlap(PCIBus *bus, int devfn, uint32_t addr,
uint32_t size, uint8_t type)
{
diff --git a/hw/pci.h b/hw/pci.h
index edc58b6..cacbdd2 100644
--- a/hw/pci.h
+++ b/hw/pci.h
@@ -271,6 +271,9 @@ void pci_info(void);
PCIBus *pci_bridge_init(PCIBus *bus, int devfn, uint16_t vid, uint16_t did,
uint8_t rid, pci_map_irq_fn map_irq, const char *name);
+PCIBus *pci_isa_bridge_init(PCIBus *bus, int devfn, uint16_t vid, uint16_t did,
+ uint8_t rid, pci_map_irq_fn map_irq, const char *name);
+
#define NR_PCI_FUNC 8
#define NR_PCI_DEV 32
#define NR_PCI_DEVFN (NR_PCI_FUNC * NR_PCI_DEV)
diff --git a/hw/pt-graphics.c b/hw/pt-graphics.c
index c6f8869..7302b25 100644
--- a/hw/pt-graphics.c
+++ b/hw/pt-graphics.c
@@ -3,6 +3,7 @@
*/
#include "pass-through.h"
+#include "pci.h"
#include "pci/header.h"
#include "pci/pci.h"
@@ -40,9 +41,11 @@ void intel_pch_init(PCIBus *bus)
did = pt_pci_host_read(pci_dev_1f, PCI_DEVICE_ID, 2);
rid = pt_pci_host_read(pci_dev_1f, PCI_REVISION, 1);
- if ( vid == PCI_VENDOR_ID_INTEL )
- pci_bridge_init(bus, PCI_DEVFN(0x1f, 0), vid, did, rid,
- pch_map_irq, "intel_bridge_1f");
+ if (vid == PCI_VENDOR_ID_INTEL) {
+ pci_isa_bridge_init(bus, PCI_DEVFN(0x1f, 0), vid, did, rid,
+ pch_map_irq, "intel_bridge_1f");
+
+ }
}
uint32_t igd_read_opregion(struct pt_dev *pci_dev)
--
1.7.10.4
[-- Attachment #3: patch3of3.txt --]
[-- Type: text/plain, Size: 4450 bytes --]
Signed-off-by: Jean Guyader <jean.guyader@xxxxxxxxx>,
Rui Guo <firemeteor@xxxxxxxxxxxxxxxxxxxxx>
Tested-by: Rui Guo <firemeteor@xxxxxxxxxxxxxxxxxxxxx>
Xen-devel: http://marc.info/?l=xen-devel&m=135748187808766
---
hw/pass-through.c | 2 +-
hw/pt-graphics.c | 69 ++++++++++++++++++++++++++++++++++++++++++++++++-----
2 files changed, 64 insertions(+), 7 deletions(-)
diff --git a/hw/pass-through.c b/hw/pass-through.c
index 304c438..2e795e1 100644
--- a/hw/pass-through.c
+++ b/hw/pass-through.c
@@ -2101,7 +2101,7 @@ struct pci_dev *pt_pci_get_dev(int bus, int dev, int fn)
u32 pt_pci_host_read(struct pci_dev *pci_dev, u32 addr, int len)
{
- u32 val = -1;
+ u32 val = 0;
pci_access_init();
pci_read_block(pci_dev, addr, (u8 *) &val, len);
diff --git a/hw/pt-graphics.c b/hw/pt-graphics.c
index 5d4cf4a..269aade 100644
--- a/hw/pt-graphics.c
+++ b/hw/pt-graphics.c
@@ -144,6 +144,53 @@ write_default:
pci_default_write_config(pci_dev, config_addr, val, len);
}
+#define PCI_INTEL_VENDOR_CAP 0x34
+#define PCI_INTEL_VENDOR_CAP_TYPE 0x09
+/*
+ * This function returns 0 is the value hasn't been
+ * updated. That mean the offset doesn't anything to
+ * do with the vendor capability.
+ */
+static uint32_t igd_pci_read_vendor_cap(PCIDevice *pci_dev, uint32_t config_addr, int len,
+ uint32_t *val)
+{
+ struct pci_dev *pci_dev_host_bridge = pt_pci_get_dev(0, 0, 0);
+ uint32_t vendor_cap = 0;
+ uint32_t cap_type = 0;
+ uint32_t cap_size = 0;
+
+ vendor_cap = pt_pci_host_read(pci_dev_host_bridge, PCI_INTEL_VENDOR_CAP, 1);
+ if (!vendor_cap)
+ return 0;
+
+ cap_type = pt_pci_host_read(pci_dev_host_bridge, vendor_cap, 1);
+ if (cap_type != PCI_INTEL_VENDOR_CAP_TYPE)
+ return 0;
+
+ if (config_addr == PCI_INTEL_VENDOR_CAP)
+ {
+ *val = vendor_cap;
+ return 1;
+ }
+
+ /* Remove the next capability link */
+ if (config_addr == vendor_cap + 1)
+ {
+ *val = 0;
+ return 1;
+ }
+
+ cap_size = pt_pci_host_read(pci_dev_host_bridge, vendor_cap + 2, 1);
+ if (config_addr >= vendor_cap &&
+ config_addr + len <= vendor_cap + cap_size)
+ {
+ *val = pt_pci_host_read(pci_dev_host_bridge, config_addr, len);
+ return 1;
+ }
+
+ return 0;
+}
+
uint32_t igd_pci_read(PCIDevice *pci_dev, uint32_t config_addr, int len)
{
struct pci_dev *pci_dev_host_bridge;
@@ -151,12 +198,22 @@ uint32_t igd_pci_read(PCIDevice *pci_dev, uint32_t config_addr, int len)
assert(pci_dev->devfn == 0x00);
if ( !igd_passthru )
- goto read_default;
+ {
+ val = pci_default_read_config(pci_dev, config_addr, len);
+ goto read_return;
+ }
+ /* Exposing writable register does not lead to security risk since this
+ only apply to read. This may confuse the guest, but it works good so far.
+ Will switch to mask & merge style only if this is proved broken.
+ Note: Always expose aligned address if any byte of the dword is to be
+ exposed. This provides a consistent view, at least for read. */
switch (config_addr)
{
case 0x00: /* vendor id */
case 0x02: /* device id */
+ case 0x04: /* command */
+ case 0x06: /* status, needed for the cap list bit*/
case 0x08: /* revision id */
case 0x2c: /* sybsystem vendor id */
case 0x2e: /* sybsystem id */
@@ -169,7 +226,9 @@ uint32_t igd_pci_read(PCIDevice *pci_dev, uint32_t config_addr, int len)
case 0xa8: /* SNB: base of GTT stolen memory */
break;
default:
- goto read_default;
+ if (!(igd_passthru && igd_pci_read_vendor_cap(pci_dev, config_addr, len, &val)))
+ val = pci_default_read_config(pci_dev, config_addr, len);
+ goto read_return;
}
/* Host read */
@@ -180,15 +239,13 @@ uint32_t igd_pci_read(PCIDevice *pci_dev, uint32_t config_addr, int len)
}
val = pt_pci_host_read(pci_dev_host_bridge, config_addr, len);
+
+read_return:
#ifdef PT_DEBUG_PCI_CONFIG_ACCESS
PT_LOG_DEV(pci_dev, "addr=%x len=%x val=%x\n",
config_addr, len, val);
#endif
return val;
-
-read_default:
-
- return pci_default_read_config(pci_dev, config_addr, len);
}
/*
--
1.7.10.4
[-- Attachment #4: Type: text/plain, Size: 126 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
http://lists.xen.org/xen-devel
next reply other threads:[~2013-07-16 4:31 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-07-16 4:31 Singapore Citizen Mr. Teo En Ming (Zhang Enming) [this message]
2013-07-16 7:24 ` Result of Applying IGD VGA Passthrough Patches to Xen 4.4-unstable Changeset 27238 Pasi Kärkkäinen
2013-07-16 8:00 ` Singapore Citizen Mr. Teo En Ming (Zhang Enming)
2013-07-16 8:51 ` Pasi Kärkkäinen
2013-07-16 13:31 ` Singapore Citizen Mr. Teo En Ming (Zhang Enming)
2013-07-17 22:05 ` Pasi Kärkkäinen
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=51E4CCA0.8000604@gmail.com \
--to=singapore.mr.teo.en.ming@gmail.com \
--cc=Ian.Campbell@citrix.com \
--cc=konrad.wilk@oracle.com \
--cc=pasik@iki.fi \
--cc=xen-devel@lists.xen.org \
--cc=xen-users@lists.xen.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).