rpm/xen
SHA256
1
0
Fork 0
forked from pool/xen
Code Pull Requests Activity

- bnc#658704 - SLES11 SP1 Xen boot panic in x2apic mode

Browse Source 
22707-x2apic-preenabled-check.patch
- bnc#641419 - L3: Xen: qemu-dm reports "xc_map_foreign_batch: mmap failed:
  Cannot allocate memory"
  7434-qemu-rlimit-as.patch
- Additional or upstream patches from Jan
  22693-fam10-mmio-conf-base-protect.patch
  22694-x86_64-no-weak.patch
  22708-xenctx-misc.patch
  21432-4.0-cpu-boot-failure.patch
  22645-amd-flush-filter.patch
  qemu-fix-7433.patch

- Maintain compatibility with the extid flag even though it is
  deprecated for both legacy and sxp config files.
  hv_extid_compatibility.patch 

- bnc#649209-improve suspend eventchn lock
  suspend_evtchn_lock.patch

- Removed the hyper-v shim patches in favor of using the upstream 
  version. 

- bnc#641419 - L3: Xen: qemu-dm reports "xc_map_foreign_batch: mmap
  failed: Cannot allocate memory" 
  qemu-rlimit-as.patch

- Upstream c/s 7433 to replace qemu_altgr_more.patch
  7433-qemu-altgr.patch

OBS-URL: https://build.opensuse.org/package/show/Virtualization/xen?expand=0&rev=90
This commit is contained in:
Charles Arnold 2011-01-14 18:24:51 +00:00 committed by Git OBS Bridge
parent bfba6bf4b3
commit 9a05aa7fc4
90 changed files with 3973 additions and 2611 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

55
20158-revert.patch Normal file
View File

@ -0,0 +1,55 @@
# HG changeset patch
# User jfehlig@localhost
# Date 1293498049 25200
# Node ID 7584dc7bc3de64340f193a0cab188e85bd50d594
# Parent f9ce5858f7eae84bec34aa10d3585c7e6f6ca6c9
Revert 20158:f9ce5858f7ea xend: Support "bootloader" mode for "drbd:" devices
block-drbd (provided by drbd project) handles all the details that
c/s 20158 introduces within xend. Subtleties of configuring storage
subsystems should not be handled by xend, but instead delegated to the block
script mechansim provided by the xen toool stack.
Note: Patch submitted upstream but not (yet) accepted since xend is on
the way out.
http://lists.xensource.com/archives/html/xen-devel/2010-12/msg01160.html
Index: xen-4.0.1-testing/tools/python/xen/util/blkif.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/util/blkif.py
+++ xen-4.0.1-testing/tools/python/xen/util/blkif.py
@@ -79,13 +79,6 @@ def parse_uname(uname):
if typ == "phy" and not fn.startswith("/"):
fn = "/dev/%s" %(fn,)
- if typ == "drbd":
- if not fn.startswith("drbd"):
- (drbdadmstdin, drbdadmstdout) = os.popen2("/sbin/drbdadm "+"sh-dev "+fn)
- fn = drbdadmstdout.readline().strip()
- else:
- fn = "/dev/%s" %(fn,)
-
if typ in ("tap", "tap2"):
(taptype, fn) = fn.split(":", 1)
if taptype in ("tapdisk", "ioemu"):
Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/XendDomainInfo.py
+++ xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
@@ -3283,15 +3283,6 @@ class XendDomainInfo:
(fn, types) = parse_uname(disk)
- # If this is a drbd volume, check if we need to activate it
- if disk.find(":") != -1:
- (disktype, diskname) = disk.split(':', 1)
- if disktype == 'drbd':
- (drbdadmstdin, drbdadmstdout) = os.popen2(["/sbin/drbdadm", "state", diskname])
- (state, junk) = drbdadmstdout.readline().split('/', 1)
- if state == 'Secondary':
- os.system('/sbin/drbdadm primary ' + diskname)
-
def _shouldMount(types):
if types[0] in ('file', 'phy'):
return False

25
21344-4.0-testing-xenstore-fd-leak.patch Normal file
View File

@ -0,0 +1,25 @@
# HG changeset patch
# User Keir Fraser <keir.fraser@citrix.com>
# Date 1284535198 -3600
# Node ID 86b99b9ad6ef9492b66c98e2377105f3ea584cbf
# Parent 89605b79f56529d34d963d0d91c67c46294a7b7c
Fix fd leak in xenstore
Missing from commit 'libxl: Backported stuff from unstable'
Without this change, xs_daemon_open/xs_daemon_close will leak
file descriptors.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
diff -r 89605b79f565 -r 86b99b9ad6ef tools/xenstore/xs.c
--- a/tools/xenstore/xs.c Mon Sep 13 17:51:50 2010 +0100
+++ b/tools/xenstore/xs.c Wed Sep 15 08:19:58 2010 +0100
@@ -285,6 +285,8 @@
mutex_unlock(&h->request_mutex);
mutex_unlock(&h->reply_mutex);
mutex_unlock(&h->watch_mutex);
+
+ close_fds_free(h);
}
static bool read_all(int fd, void *data, unsigned int len)

35
21432-4.0-cpu-boot-failure.patch Normal file
View File

@ -0,0 +1,35 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1294742461 0
# Node ID 7a203c60d588a7a15a28a6fb16c69feafd157e0d
# Parent 64bb7d9904a64cc4561c6d541d857acf46e42180
x86: don't crash when a CPU cannot be brought online during boot
References: bnc#656369, bnc#658704
x86_cpu_to_apicid[] gets set to BAD_APICID when bringup of a secondary
CPU fails, yet srat_detect_node() wants to use this as array index.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Index: xen-4.0.1-testing/xen/arch/x86/setup.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/setup.c
+++ xen-4.0.1-testing/xen/arch/x86/setup.c
@@ -1103,10 +1103,13 @@ void __init __start_xen(unsigned long mb
__cpu_up(i);
}
- /* Set up cpu_to_node[]. */
- srat_detect_node(i);
- /* Set up node_to_cpumask based on cpu_to_node[]. */
- numa_add_cpu(i);
+ if ( cpu_online(i) )
+ {
+ /* Set up cpu_to_node[]. */
+ srat_detect_node(i);
+ /* Set up node_to_cpumask based on cpu_to_node[]. */
+ numa_add_cpu(i);
+ }
}
printk("Brought up %ld CPUs\n", (long)num_online_cpus());

465
21810-x2apic-acpi.patch Normal file
View File

@ -0,0 +1,465 @@
# HG changeset patch
# User Keir Fraser <keir.fraser@citrix.com>
# Date 1279284738 -3600
# Node ID fedab6367c9a0141d49853c77a23d6642ba70ff6
# Parent 1f7c2418e58c7d1d5650ea211016b30114de48f2
ACPI: add support for x2APIC ACPI extensions
References: bnc#656369, bnc#658704
All logical processors with APIC ID values of 255 and greater will
have their APIC reported through Processor X2APIC structure (type-9
entry type) and all logical processors with APIC ID less than 255 will
have their APIC reported through legacy Processor Local APIC (type-0
entry type) only. This is the same case even for NMI structure
reporting.
The Processor X2APIC Affinity structure provides the association
between the X2APIC ID of a logical processor and the proximity domain
to which the logical processor belongs.
This patch adds 2 new subtables to MADT and one new subtable to SRAT.
This patch also changes x86_acpiid_to_apicid from u8 to u32 for x2APIC
ID, and changes mp_register_lapic to accept 32-bit id. But there are
still some 8-bit apic id hardcode and assumptions in Xen code, it
needs to be fixed in future.
Signed-off-by: Weidong Han <weidong.han@intel.com>
--- a/xen/arch/x86/acpi/boot.c
+++ b/xen/arch/x86/acpi/boot.c
@@ -81,7 +81,7 @@ u8 acpi_enable_value, acpi_disable_value
#warning ACPI uses CMPXCHG, i486 and later hardware
#endif
-u8 x86_acpiid_to_apicid[MAX_MADT_ENTRIES] =
+u32 x86_acpiid_to_apicid[MAX_MADT_ENTRIES] =
{[0 ... MAX_MADT_ENTRIES - 1] = 0xff };
EXPORT_SYMBOL(x86_acpiid_to_apicid);
@@ -156,6 +156,35 @@ static int __init acpi_parse_madt(struct
}
static int __init
+acpi_parse_x2apic(struct acpi_subtable_header *header, const unsigned long end)
+{
+ struct acpi_table_x2apic *processor = NULL;
+
+ processor = (struct acpi_table_x2apic *)header;
+
+ if (BAD_MADT_ENTRY(processor, end))
+ return -EINVAL;
+
+ acpi_table_print_madt_entry(header);
+
+ /* Record local apic id only when enabled */
+ if (processor->flags.enabled)
+ x86_acpiid_to_apicid[processor->acpi_uid] = processor->id;
+
+ /*
+ * We need to register disabled CPU as well to permit
+ * counting disabled CPUs. This allows us to size
+ * cpus_possible_map more accurately, to permit
+ * to not preallocating memory for all NR_CPUS
+ * when we use CPU hotplug.
+ */
+ mp_register_lapic(processor->id, /* X2APIC ID */
+ processor->flags.enabled); /* Enabled? */
+
+ return 0;
+}
+
+static int __init
acpi_parse_lapic(struct acpi_subtable_header * header, const unsigned long end)
{
struct acpi_table_lapic *processor = NULL;
@@ -201,6 +230,25 @@ acpi_parse_lapic_addr_ovr(struct acpi_su
}
static int __init
+acpi_parse_x2apic_nmi(struct acpi_subtable_header *header,
+ const unsigned long end)
+{
+ struct acpi_table_x2apic_nmi *x2apic_nmi = NULL;
+
+ x2apic_nmi = (struct acpi_table_x2apic_nmi *)header;
+
+ if (BAD_MADT_ENTRY(x2apic_nmi, end))
+ return -EINVAL;
+
+ acpi_table_print_madt_entry(header);
+
+ if (x2apic_nmi->lint != 1)
+ printk(KERN_WARNING PREFIX "NMI not connected to LINT 1!\n");
+
+ return 0;
+}
+
+static int __init
acpi_parse_lapic_nmi(struct acpi_subtable_header * header, const unsigned long end)
{
struct acpi_table_lapic_nmi *lapic_nmi = NULL;
@@ -465,7 +513,7 @@ static int __init acpi_parse_fadt(struct
*/
static int __init acpi_parse_madt_lapic_entries(void)
{
- int count;
+ int count, x2count;
if (!cpu_has_apic)
return -ENODEV;
@@ -488,11 +536,13 @@ static int __init acpi_parse_madt_lapic_
count = acpi_table_parse_madt(ACPI_MADT_LAPIC, acpi_parse_lapic,
MAX_APICS);
- if (!count) {
+ x2count = acpi_table_parse_madt(ACPI_MADT_X2APIC, acpi_parse_x2apic,
+ MAX_APICS);
+ if (!count && !x2count) {
printk(KERN_ERR PREFIX "No LAPIC entries present\n");
/* TBD: Cleanup to allow fallback to MPS */
return -ENODEV;
- } else if (count < 0) {
+ } else if (count < 0 || x2count < 0) {
printk(KERN_ERR PREFIX "Error parsing LAPIC entry\n");
/* TBD: Cleanup to allow fallback to MPS */
return count;
@@ -500,7 +550,10 @@ static int __init acpi_parse_madt_lapic_
count =
acpi_table_parse_madt(ACPI_MADT_LAPIC_NMI, acpi_parse_lapic_nmi, 0);
- if (count < 0) {
+ x2count =
+ acpi_table_parse_madt(ACPI_MADT_X2APIC_NMI,
+ acpi_parse_x2apic_nmi, 0);
+ if (count < 0 || x2count < 0) {
printk(KERN_ERR PREFIX "Error parsing LAPIC NMI entry\n");
/* TBD: Cleanup to allow fallback to MPS */
return count;
--- a/xen/arch/x86/mpparse.c
+++ b/xen/arch/x86/mpparse.c
@@ -838,7 +838,7 @@ void __init mp_register_lapic_address (
int __devinit mp_register_lapic (
- u8 id,
+ u32 id,
u8 enabled)
{
struct mpc_config_processor processor;
--- a/xen/arch/x86/srat.c
+++ b/xen/arch/x86/srat.c
@@ -164,6 +164,36 @@ void __init acpi_numa_slit_init(struct a
}
#endif
+/* Callback for Proximity Domain -> x2APIC mapping */
+void __init
+acpi_numa_x2apic_affinity_init(struct acpi_srat_x2apic_cpu_affinity *pa)
+{
+ int pxm, node;
+ int apic_id;
+
+ if (srat_disabled())
+ return;
+ if (pa->header.length < sizeof(struct acpi_srat_x2apic_cpu_affinity)) {
+ bad_srat();
+ return;
+ }
+ if ((pa->flags & ACPI_SRAT_CPU_ENABLED) == 0)
+ return;
+ pxm = pa->proximity_domain;
+ node = setup_node(pxm);
+ if (node < 0) {
+ printk(KERN_ERR "SRAT: Too many proximity domains %x\n", pxm);
+ bad_srat();
+ return;
+ }
+
+ apic_id = pa->apic_id;
+ apicid_to_node[apic_id] = node;
+ acpi_numa = 1;
+ printk(KERN_INFO "SRAT: PXM %u -> APIC %u -> Node %u\n",
+ pxm, apic_id, node);
+}
+
/* Callback for Proximity Domain -> LAPIC mapping */
void __init
acpi_numa_processor_affinity_init(struct acpi_srat_cpu_affinity *pa)
--- a/xen/drivers/acpi/numa.c
+++ b/xen/drivers/acpi/numa.c
@@ -90,6 +90,21 @@ void __init acpi_table_print_srat_entry(
#endif /* ACPI_DEBUG_OUTPUT */
break;
+ case ACPI_SRAT_TYPE_X2APIC_CPU_AFFINITY:
+#ifdef ACPI_DEBUG_OUTPUT
+ {
+ struct acpi_srat_x2apic_cpu_affinity *p =
+ (struct acpi_srat_x2apic_cpu_affinity *)header;
+ ACPI_DEBUG_PRINT((ACPI_DB_INFO,
+ "SRAT Processor (x2apicid[0x%08x]) in"
+ " proximity domain %d %s\n",
+ p->apic_id,
+ p->proximity_domain,
+ (p->flags & ACPI_SRAT_CPU_ENABLED) ?
+ "enabled" : "disabled"));
+ }
+#endif /* ACPI_DEBUG_OUTPUT */
+ break;
default:
printk(KERN_WARNING PREFIX
"Found unsupported SRAT entry (type = 0x%x)\n",
@@ -105,6 +120,33 @@ static int __init acpi_parse_slit(struct
return 0;
}
+void __init __attribute__ ((weak))
+acpi_numa_x2apic_affinity_init(struct acpi_srat_x2apic_cpu_affinity *pa)
+{
+ printk(KERN_WARNING PREFIX
+ "Found unsupported x2apic [0x%08x] SRAT entry\n", pa->apic_id);
+ return;
+}
+
+
+static int __init
+acpi_parse_x2apic_affinity(struct acpi_subtable_header *header,
+ const unsigned long end)
+{
+ struct acpi_srat_x2apic_cpu_affinity *processor_affinity;
+
+ processor_affinity = (struct acpi_srat_x2apic_cpu_affinity *)header;
+ if (!processor_affinity)
+ return -EINVAL;
+
+ acpi_table_print_srat_entry(header);
+
+ /* let architecture-dependent part to do it */
+ acpi_numa_x2apic_affinity_init(processor_affinity);
+
+ return 0;
+}
+
static int __init
acpi_parse_processor_affinity(struct acpi_subtable_header * header,
const unsigned long end)
@@ -164,6 +206,8 @@ int __init acpi_numa_init(void)
{
/* SRAT: Static Resource Affinity Table */
if (!acpi_table_parse(ACPI_SIG_SRAT, acpi_parse_srat)) {
+ acpi_table_parse_srat(ACPI_SRAT_TYPE_X2APIC_CPU_AFFINITY,
+ acpi_parse_x2apic_affinity, NR_CPUS);
acpi_table_parse_srat(ACPI_SRAT_PROCESSOR_AFFINITY,
acpi_parse_processor_affinity,
NR_CPUS);
--- a/xen/drivers/acpi/tables.c
+++ b/xen/drivers/acpi/tables.c
@@ -63,6 +63,18 @@ void __init acpi_table_print_madt_entry(
}
break;
+ case ACPI_MADT_TYPE_LOCAL_X2APIC:
+ {
+ struct acpi_madt_local_x2apic *p =
+ (struct acpi_madt_local_x2apic *)header;
+ printk(KERN_INFO PREFIX
+ "X2APIC (apic_id[0x%02x] uid[0x%02x] %s)\n",
+ p->local_apic_id, p->uid,
+ (p->lapic_flags & ACPI_MADT_ENABLED) ?
+ "enabled" : "disabled");
+ }
+ break;
+
case ACPI_MADT_TYPE_IO_APIC:
{
struct acpi_madt_io_apic *p =
@@ -117,6 +129,24 @@ void __init acpi_table_print_madt_entry(
}
break;
+ case ACPI_MADT_TYPE_LOCAL_X2APIC_NMI:
+ {
+ u16 polarity, trigger;
+ struct acpi_madt_local_x2apic_nmi *p =
+ (struct acpi_madt_local_x2apic_nmi *)header;
+
+ polarity = p->inti_flags & ACPI_MADT_POLARITY_MASK;
+ trigger = (p->inti_flags & ACPI_MADT_TRIGGER_MASK) >> 2;
+
+ printk(KERN_INFO PREFIX
+ "X2APIC_NMI (uid[0x%02x] %s %s lint[0x%x])\n",
+ p->uid,
+ mps_inti_flags_polarity[polarity],
+ mps_inti_flags_trigger[trigger],
+ p->lint);
+ }
+ break;
+
case ACPI_MADT_TYPE_LOCAL_APIC_OVERRIDE:
{
struct acpi_madt_local_apic_override *p =
--- a/xen/include/acpi/actbl1.h
+++ b/xen/include/acpi/actbl1.h
@@ -404,7 +404,9 @@ enum acpi_madt_type {
ACPI_MADT_TYPE_IO_SAPIC = 6,
ACPI_MADT_TYPE_LOCAL_SAPIC = 7,
ACPI_MADT_TYPE_INTERRUPT_SOURCE = 8,
- ACPI_MADT_TYPE_RESERVED = 9 /* 9 and greater are reserved */
+ ACPI_MADT_TYPE_LOCAL_X2APIC = 9,
+ ACPI_MADT_TYPE_LOCAL_X2APIC_NMI = 10,
+ ACPI_MADT_TYPE_RESERVED = 11 /* 11 and greater are reserved */
};
/*
@@ -505,6 +507,26 @@ struct acpi_madt_interrupt_source {
#define ACPI_MADT_CPEI_OVERRIDE (1)
+/* 9: Processor Local X2APIC (ACPI 4.0) */
+
+struct acpi_madt_local_x2apic {
+ struct acpi_subtable_header header;
+ u16 reserved; /* Reserved - must be zero */
+ u32 local_apic_id; /* Processor X2_APIC ID */
+ u32 lapic_flags;
+ u32 uid; /* Extended X2_APIC processor ID */
+};
+
+/* 10: Local X2APIC NMI (ACPI 4.0) */
+
+struct acpi_madt_local_x2apic_nmi {
+ struct acpi_subtable_header header;
+ u16 inti_flags;
+ u32 uid; /* Processor X2_APIC ID */
+ u8 lint; /* LINTn to which NMI is connected */
+ u8 reserved[3];
+};
+
/*
* Common flags fields for MADT subtables
*/
@@ -646,11 +668,14 @@ struct acpi_table_srat {
enum acpi_srat_type {
ACPI_SRAT_TYPE_CPU_AFFINITY = 0,
ACPI_SRAT_TYPE_MEMORY_AFFINITY = 1,
- ACPI_SRAT_TYPE_RESERVED = 2
+ ACPI_SRAT_TYPE_X2APIC_CPU_AFFINITY = 2,
+ ACPI_SRAT_TYPE_RESERVED = 3 /* 3 and greater are reserved */
};
/* SRAT sub-tables */
+/* 0: Processor Local APIC/SAPIC Affinity */
+
struct acpi_srat_cpu_affinity {
struct acpi_subtable_header header;
u8 proximity_domain_lo;
@@ -661,9 +686,7 @@ struct acpi_srat_cpu_affinity {
u32 reserved; /* Reserved, must be zero */
};
-/* Flags */
-
-#define ACPI_SRAT_CPU_ENABLED (1) /* 00: Use affinity structure */
+/* 1: Memory Affinity */
struct acpi_srat_mem_affinity {
struct acpi_subtable_header header;
@@ -682,6 +705,22 @@ struct acpi_srat_mem_affinity {
#define ACPI_SRAT_MEM_HOT_PLUGGABLE (1<<1) /* 01: Memory region is hot pluggable */
#define ACPI_SRAT_MEM_NON_VOLATILE (1<<2) /* 02: Memory region is non-volatile */
+/* 2: Processor Local X2_APIC Affinity (ACPI 4.0) */
+
+struct acpi_srat_x2apic_cpu_affinity {
+ struct acpi_subtable_header header;
+ u16 reserved; /* Reserved, must be zero */
+ u32 proximity_domain;
+ u32 apic_id;
+ u32 flags;
+ u32 clock_domain;
+ u32 reserved2;
+};
+
+/* Flags for struct acpi_srat_cpu_affinity and struct acpi_srat_x2apic_cpu_affinity */
+
+#define ACPI_SRAT_CPU_ENABLED (1) /* 00: Use affinity structure */
+
/*******************************************************************************
*
* TCPA - Trusted Computing Platform Alliance table
--- a/xen/include/asm-x86/acpi.h
+++ b/xen/include/asm-x86/acpi.h
@@ -151,7 +151,7 @@ struct acpi_sleep_info {
#endif /* CONFIG_ACPI_SLEEP */
#define MAX_MADT_ENTRIES 256
-extern u8 x86_acpiid_to_apicid[];
+extern u32 x86_acpiid_to_apicid[];
#define MAX_LOCAL_APIC 256
extern u32 pmtmr_ioport;
--- a/xen/include/asm-x86/mpspec.h
+++ b/xen/include/asm-x86/mpspec.h
@@ -24,7 +24,7 @@ extern int pic_mode;
extern int using_apic_timer;
#ifdef CONFIG_ACPI
-extern int mp_register_lapic (u8 id, u8 enabled);
+extern int mp_register_lapic (u32 id, u8 enabled);
extern void mp_unregister_lapic(uint32_t apic_id, uint32_t cpu);
extern void mp_register_lapic_address (u64 address);
extern void mp_register_ioapic (u8 id, u32 address, u32 gsi_base);
--- a/xen/include/xen/acpi.h
+++ b/xen/include/xen/acpi.h
@@ -57,6 +57,8 @@ enum acpi_madt_entry_id {
ACPI_MADT_IOSAPIC,
ACPI_MADT_LSAPIC,
ACPI_MADT_PLAT_INT_SRC,
+ ACPI_MADT_X2APIC,
+ ACPI_MADT_X2APIC_NMI,
ACPI_MADT_ENTRY_COUNT
};
@@ -76,6 +78,17 @@ struct acpi_table_lapic {
} flags;
} __attribute__ ((packed));
+struct acpi_table_x2apic {
+ struct acpi_subtable_header header;
+ u16 reserved;
+ u32 id;
+ struct {
+ u32 enabled:1;
+ u32 reserved:31;
+ } flags;
+ u32 acpi_uid;
+} __attribute__ ((packed));
+
struct acpi_table_ioapic {
struct acpi_subtable_header header;
u8 id;
@@ -105,6 +118,14 @@ struct acpi_table_lapic_nmi {
u8 lint;
} __attribute__ ((packed));
+struct acpi_table_x2apic_nmi {
+ struct acpi_subtable_header header;
+ acpi_interrupt_flags flags;
+ u32 acpi_uid;
+ u8 lint;
+ u8 reserved[3];
+} __attribute__ ((packed));
+
struct acpi_table_lapic_addr_ovr {
struct acpi_subtable_header header;
u8 reserved[2];
@@ -280,6 +301,7 @@ void acpi_table_print_srat_entry (struct
/* the following four functions are architecture-dependent */
void acpi_numa_slit_init (struct acpi_table_slit *slit);
void acpi_numa_processor_affinity_init (struct acpi_srat_cpu_affinity *pa);
+void acpi_numa_x2apic_affinity_init(struct acpi_srat_x2apic_cpu_affinity *pa);
void acpi_numa_memory_affinity_init (struct acpi_srat_mem_affinity *ma);
void acpi_numa_arch_fixup(void);

168
21989-x2apic-resume.patch Normal file
View File

@ -0,0 +1,168 @@
# HG changeset patch
# User Keir Fraser <keir.fraser@citrix.com>
# Date 1281707886 -3600
# Node ID 3cee41690fa26853acb0be65065c52f6029ca599
# Parent 01d185dab39e9be399b203bec91870e04f576c23
x2APIC: Improve x2APIC suspend/resume
References: bnc#656369, bnc#658704
x2apic depends on interrupt remapping, so it should disable interrupt
remapping behind x2apic disabling. And also this patch wraps
__enable_x2apic to get rid of duplicated code.
Signed-off-by: Weidong Han <weidong.han@intel.com>
Index: xen-4.0.1-testing/xen/arch/x86/apic.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/apic.c
+++ xen-4.0.1-testing/xen/arch/x86/apic.c
@@ -496,9 +496,20 @@ static void apic_pm_activate(void)
apic_pm_state.active = 1;
}
+static void __enable_x2apic(void)
+{
+ u32 lo, hi;
+
+ rdmsr(MSR_IA32_APICBASE, lo, hi);
+ if ( !(lo & MSR_IA32_APICBASE_EXTD) )
+ {
+ lo |= MSR_IA32_APICBASE_ENABLE | MSR_IA32_APICBASE_EXTD;
+ wrmsr(MSR_IA32_APICBASE, lo, 0);
+ }
+}
+
static void resume_x2apic(void)
{
- uint64_t msr_content;
struct IO_APIC_route_entry **ioapic_entries = NULL;
ASSERT(x2apic_enabled);
@@ -520,14 +531,7 @@ static void resume_x2apic(void)
mask_IO_APIC_setup(ioapic_entries);
iommu_enable_IR();
-
- rdmsrl(MSR_IA32_APICBASE, msr_content);
- if ( !(msr_content & MSR_IA32_APICBASE_EXTD) )
- {
- msr_content |= MSR_IA32_APICBASE_ENABLE | MSR_IA32_APICBASE_EXTD;
- msr_content = (uint32_t)msr_content;
- wrmsrl(MSR_IA32_APICBASE, msr_content);
- }
+ __enable_x2apic();
restore_IO_APIC_setup(ioapic_entries);
unmask_8259A();
@@ -743,9 +747,10 @@ int lapic_suspend(void)
apic_pm_state.apic_tmict = apic_read(APIC_TMICT);
apic_pm_state.apic_tdcr = apic_read(APIC_TDCR);
apic_pm_state.apic_thmr = apic_read(APIC_LVTTHMR);
-
+
local_irq_save(flags);
disable_local_APIC();
+ iommu_disable_IR();
local_irq_restore(flags);
return 0;
}
@@ -1042,15 +1047,8 @@ static void enable_bsp_x2apic(void)
if ( !x2apic_preenabled )
{
- u32 lo, hi;
-
- rdmsr(MSR_IA32_APICBASE, lo, hi);
- if ( !(lo & MSR_IA32_APICBASE_EXTD) )
- {
- lo |= MSR_IA32_APICBASE_ENABLE | MSR_IA32_APICBASE_EXTD;
- wrmsr(MSR_IA32_APICBASE, lo, 0);
- printk("x2APIC mode enabled.\n");
- }
+ __enable_x2apic();
+ printk("x2APIC mode enabled.\n");
}
restore_out:
@@ -1064,19 +1062,12 @@ out:
static void enable_ap_x2apic(void)
{
- u32 lo, hi;
-
ASSERT(smp_processor_id() != 0);
/* APs only enable x2apic when BSP did so. */
BUG_ON(!x2apic_enabled);
- rdmsr(MSR_IA32_APICBASE, lo, hi);
- if ( !(lo & MSR_IA32_APICBASE_EXTD) )
- {
- lo |= MSR_IA32_APICBASE_ENABLE | MSR_IA32_APICBASE_EXTD;
- wrmsr(MSR_IA32_APICBASE, lo, 0);
- }
+ __enable_x2apic();
}
void enable_x2apic(void)
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/intremap.c
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/intremap.c
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/intremap.c
@@ -873,6 +873,24 @@ int iommu_enable_IR(void)
}
/*
+ * This function is used to disable Interrutp remapping when
+ * suspend local apic
+ */
+void iommu_disable_IR(void)
+{
+ struct acpi_drhd_unit *drhd;
+
+ if ( !iommu_supports_eim() )
+ return;
+
+ for_each_drhd_unit ( drhd )
+ disable_intremap(drhd->iommu);
+
+ for_each_drhd_unit ( drhd )
+ disable_qinval(drhd->iommu);
+}
+
+/*
* Check if interrupt remapping is enabled or not
* return 1: enabled
* return 0: not enabled
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/iommu.c
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/iommu.c
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/iommu.c
@@ -2126,10 +2126,11 @@ static void vtd_suspend(void)
iommu_disable_translation(iommu);
- if ( iommu_intremap )
- disable_intremap(iommu);
-
- if ( iommu_qinval )
+ /* If interrupt remapping is enabled, queued invalidation
+ * will be disabled following interupt remapping disabling
+ * in local apic suspend
+ */
+ if ( !iommu_intremap && iommu_qinval )
disable_qinval(iommu);
}
}
Index: xen-4.0.1-testing/xen/include/xen/iommu.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/xen/iommu.h
+++ xen-4.0.1-testing/xen/include/xen/iommu.h
@@ -59,6 +59,7 @@ struct iommu {
int iommu_setup(void);
int iommu_supports_eim(void);
int iommu_enable_IR(void);
+void iommu_disable_IR(void);
int intremap_enabled(void);
int iommu_add_device(struct pci_dev *pdev);

60
22223-vtd-workarounds.patch
View File

@ -41,15 +41,19 @@ enhancement.
Signed-off-by: Allen Kay <allen.m.kay@intel.com>
--- a/xen/drivers/passthrough/vtd/Makefile
+++ b/xen/drivers/passthrough/vtd/Makefile
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/Makefile
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/Makefile
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/Makefile
@@ -6,3 +6,4 @@ obj-y += dmar.o
obj-y += utils.o
obj-y += qinval.o
obj-y += intremap.o
+obj-y += quirks.o
--- a/xen/drivers/passthrough/vtd/dmar.c
+++ b/xen/drivers/passthrough/vtd/dmar.c
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/dmar.c
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/dmar.c
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/dmar.c
@@ -46,6 +46,7 @@ LIST_HEAD(acpi_rmrr_units);
LIST_HEAD(acpi_atsr_units);
LIST_HEAD(acpi_rhsa_units);
@ -86,8 +90,10 @@ Signed-off-by: Allen Kay <allen.m.kay@intel.com>
break;
case ACPI_DEV_IOAPIC:
--- a/xen/drivers/passthrough/vtd/dmar.h
+++ b/xen/drivers/passthrough/vtd/dmar.h
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/dmar.h
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/dmar.h
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/dmar.h
@@ -114,5 +114,6 @@ void *map_to_nocache_virt(int nr_iommus,
int vtd_hw_check(void);
void disable_pmr(struct iommu *iommu);
@ -95,8 +101,10 @@ Signed-off-by: Allen Kay <allen.m.kay@intel.com>
+int is_igd_drhd(struct acpi_drhd_unit *drhd);
#endif /* _DMAR_H_ */
--- a/xen/drivers/passthrough/vtd/extern.h
+++ b/xen/drivers/passthrough/vtd/extern.h
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/extern.h
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/extern.h
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/extern.h
@@ -26,6 +26,7 @@
extern int qinval_enabled;
@ -162,8 +170,10 @@ Signed-off-by: Allen Kay <allen.m.kay@intel.com>
+void me_wifi_quirk(struct domain *domain, u8 bus, u8 devfn, int map);
#endif // _VTD_EXTERN_H_
--- a/xen/drivers/passthrough/vtd/iommu.c
+++ b/xen/drivers/passthrough/vtd/iommu.c
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/iommu.c
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/iommu.c
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/iommu.c
@@ -43,7 +43,6 @@
#endif
@ -382,7 +392,7 @@ Signed-off-by: Allen Kay <allen.m.kay@intel.com>
irq_to_iommu = xmalloc_array(struct iommu*, nr_irqs);
BUG_ON(!irq_to_iommu);
@@ -2163,7 +2206,7 @@ static void vtd_resume(void)
@@ -2164,7 +2207,7 @@ static void vtd_resume(void)
(u32) iommu_state[i][DMAR_FEUADDR_REG]);
spin_unlock_irqrestore(&iommu->register_lock, flags);
@ -391,8 +401,10 @@ Signed-off-by: Allen Kay <allen.m.kay@intel.com>
}
}
--- a/xen/drivers/passthrough/vtd/iommu.h
+++ b/xen/drivers/passthrough/vtd/iommu.h
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/iommu.h
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/iommu.h
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/iommu.h
@@ -501,6 +501,7 @@ struct intel_iommu {
struct qi_ctrl qi_ctrl;
struct ir_ctrl ir_ctrl;
@ -401,8 +413,10 @@ Signed-off-by: Allen Kay <allen.m.kay@intel.com>
};
#endif
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/quirks.c
===================================================================
--- /dev/null
+++ b/xen/drivers/passthrough/vtd/quirks.c
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/quirks.c
@@ -0,0 +1,262 @@
+/*
+ * Copyright (c) 2010, Intel Corporation.
@ -666,8 +680,10 @@ Signed-off-by: Allen Kay <allen.m.kay@intel.com>
+
+ }
+}
--- a/xen/drivers/passthrough/vtd/vtd.h
+++ b/xen/drivers/passthrough/vtd/vtd.h
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/vtd.h
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/vtd.h
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/vtd.h
@@ -23,6 +23,9 @@
#include <xen/iommu.h>
@ -697,8 +713,10 @@ Signed-off-by: Allen Kay <allen.m.kay@intel.com>
-void iommu_free(struct acpi_drhd_unit *drhd);
-
#endif // _VTD_H_
--- a/xen/drivers/passthrough/vtd/x86/vtd.c
+++ b/xen/drivers/passthrough/vtd/x86/vtd.c
Index: xen-4.0.1-testing/xen/drivers/passthrough/vtd/x86/vtd.c
===================================================================
--- xen-4.0.1-testing.orig/xen/drivers/passthrough/vtd/x86/vtd.c
+++ xen-4.0.1-testing/xen/drivers/passthrough/vtd/x86/vtd.c
@@ -27,6 +27,7 @@
#include "../iommu.h"
#include "../dmar.h"
@ -707,8 +725,10 @@ Signed-off-by: Allen Kay <allen.m.kay@intel.com>
/*
* iommu_inclusive_mapping: when set, all memory below 4GB is included in dom0
--- a/xen/include/asm-x86/fixmap.h
+++ b/xen/include/asm-x86/fixmap.h
Index: xen-4.0.1-testing/xen/include/asm-x86/fixmap.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/asm-x86/fixmap.h
+++ xen-4.0.1-testing/xen/include/asm-x86/fixmap.h
@@ -52,6 +52,7 @@ enum fixed_addresses {
FIX_MSIX_IO_RESERV_BASE,
FIX_MSIX_IO_RESERV_END = FIX_MSIX_IO_RESERV_BASE + FIX_MSIX_MAX_PAGES -1,

48
22388-x2apic-panic.patch Normal file
View File

@ -0,0 +1,48 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1289813498 0
# Node ID 87f248de52304bc96a80dc093250fed0197f37e0
# Parent 8af10bd7f2d0e9300f1bc68f7e2fb137fa839365
x2apic: Remove a panic condition in enabling x2APIC
Currently Xen triggers a panic if user disables VT-d by command line
while not disable x2APIC. This requires users to specify both
"iommu=0" and "x2apic=0" to disable VT-d if the platform supports
x2APIC. It's not user friendly. This patch removes the panic
condition. That's to say, don't require user to specify "x2apic=0"
when specify "iommu=0". As long as VT-d is not enabled (disabled in
BIOS or in command line), x2APIC won't be enabled naturally (x2APIC
depends on VT-d Interrupt remapping).
Signed-off-by: Weidong Han <weidong.han@intel.com>
--- a/xen/arch/x86/apic.c
+++ b/xen/arch/x86/apic.c
@@ -1000,22 +1000,11 @@ static void enable_bsp_x2apic(void)
}
else
{
- if ( !iommu_enabled || !iommu_intremap || !iommu_qinval )
- panic("Cannot enable x2APIC due to iommu or interrupt "
- "remapping or queued invalidation is disabled "
- "by command line!\n");
- else
- {
- if ( x2apic_preenabled )
- panic("x2APIC: already enabled by BIOS, but "
- "iommu_supports_eim fails\n");
- else
- {
- printk("Not enable x2APIC due to "
- "iommu_supports_eim fails!\n");
- return;
- }
- }
+ if ( x2apic_preenabled )
+ panic("x2APIC: already enabled by BIOS, but "
+ "iommu_supports_eim failed!\n");
+ printk("Not enabling x2APIC: depends oniommu_supports_eim\n");
+ return;
}
}

57
22448-x86_64-gdt-ldt-fault-filter.patch Normal file
View File

@ -0,0 +1,57 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1291041272 0
# Node ID 5cd9612db2bba51f63ff7897aca4c72cc4b8e8c2
# Parent aba70e59a90dcb0cacc2c6834a6ce19d03d88601
x86-64: don't crash Xen upon direct pv guest access to GDT/LDT mapping area
handle_gdt_ldt_mapping_fault() is intended to deal with indirect
accesses (i.e. those caused by descriptor loads) to the GDT/LDT
mapping area only. While for 32-bit segment limits indeed prevent the
function being entered for direct accesses (i.e. a #GP fault will be
raised even before the address translation gets done, on 64-bit even
user mode accesses would lead to control reaching the BUG_ON() at the
beginning of that function.
Fortunately the fix is simple: Since the guest kernel runs in ring 3,
any guest direct access will have the "user mode" bit set, whereas
descriptor loads always do the translations to access the actual
descriptors as kernel mode ones.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Further, relax the BUG_ON() in handle_gdt_ldt_mapping_fault() to a
check-and-bail. This avoids any problems in future, if we don't
execute x86_64 guest kernels in ring 3 (e.g., because we use a
lightweight HVM container).
Signed-off-by: Keir Fraser <keir@xen.org>
--- a/xen/arch/x86/traps.c
+++ b/xen/arch/x86/traps.c
@@ -1051,8 +1051,14 @@ static int handle_gdt_ldt_mapping_fault(
unsigned int is_ldt_area = (offset >> (GDT_LDT_VCPU_VA_SHIFT-1)) & 1;
unsigned int vcpu_area = (offset >> GDT_LDT_VCPU_VA_SHIFT);
- /* Should never fault in another vcpu's area. */
- BUG_ON(vcpu_area != curr->vcpu_id);
+ /*
+ * If the fault is in another vcpu's area, it cannot be due to
+ * a GDT/LDT descriptor load. Thus we can reasonably exit immediately, and
+ * indeed we have to since map_ldt_shadow_page() works correctly only on
+ * accesses to a vcpu's own area.
+ */
+ if ( vcpu_area != curr->vcpu_id )
+ return 0;
/* Byte offset within the gdt/ldt sub-area. */
offset &= (1UL << (GDT_LDT_VCPU_VA_SHIFT-1)) - 1UL;
@@ -1223,7 +1229,7 @@ static int fixup_page_fault(unsigned lon
if ( unlikely(IN_HYPERVISOR_RANGE(addr)) )
{
- if ( !(regs->error_code & PFEC_reserved_bit) &&
+ if ( !(regs->error_code & (PFEC_user_mode | PFEC_reserved_bit)) &&
(addr >= GDT_LDT_VIRT_START) && (addr < GDT_LDT_VIRT_END) )
return handle_gdt_ldt_mapping_fault(
addr - GDT_LDT_VIRT_START, regs);

43
22451-hvm-cap-clobber.patch Normal file
View File

@ -0,0 +1,43 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1291116848 0
# Node ID 8420b82c22c2cad54860ffdbe8bcec21c65c86be
# Parent d281061e6ec06bbbf8e2b0fadbeb9d1a04bc32e2
x86 hvm: Do not overwrite boot-cpu capability data on VMX/SVM startup.
Apparently required back in the earliest days of Xen, we now properly
initialise CPU capabilities early during bootstrap. Re-writing
capability data later now causes problems if specific features have
been deliberately masked out.
Thanks to Weidong Han at Intel for finding such a bug where XSAVE
feature is masked out by default, but then erroneously written back
during VMX initialisation. This would cause memory corruption problems
during boot for XSAVE-capable systems.
Signed-off-by: Keir Fraser <keir@xen.org>
--- a/xen/arch/x86/hvm/svm/svm.c
+++ b/xen/arch/x86/hvm/svm/svm.c
@@ -915,9 +915,6 @@ void start_svm(struct cpuinfo_x86 *c)
return;
}
- /* Xen does not fill x86_capability words except 0. */
- boot_cpu_data.x86_capability[5] = cpuid_ecx(0x80000001);
-
if ( !test_bit(X86_FEATURE_SVME, &boot_cpu_data.x86_capability) )
return;
--- a/xen/arch/x86/hvm/vmx/vmx.c
+++ b/xen/arch/x86/hvm/vmx/vmx.c
@@ -1433,9 +1433,6 @@ void start_vmx(void)
return;
}
- /* Xen does not fill x86_capability words except 0. */
- boot_cpu_data.x86_capability[4] = cpuid_ecx(1);
-
if ( !test_bit(X86_FEATURE_VMXE, &boot_cpu_data.x86_capability) )
return;

62
22452-x86-irq-migrate-directed-eoi.patch Normal file
View File

@ -0,0 +1,62 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1291234227 0
# Node ID 62bf12040b0f4d802dc7e1cd61294776c8a41a89
# Parent 8420b82c22c2cad54860ffdbe8bcec21c65c86be
x86: fix IRQ migration when using directed EOI (broken with c/s 20465)
In directed-EOI mode, there is no chance to do the migration in
mask_and_ack_level_ioapic_irq(), as the remote IRR bit can't possibly
be clear after issuing the EOI to the LAPIC. Consequently, there's no
point to even try. Instead, migration must be done in
end_level_ioapic_irq(), and it requires masking the interrupt source
prior to issuing the EOI to the IO-APIC.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/arch/x86/io_apic.c
+++ b/xen/arch/x86/io_apic.c
@@ -1642,11 +1642,14 @@ static void mask_and_ack_level_ioapic_ir
ack_APIC_irq();
+ if ( directed_eoi_enabled )
+ return;
+
if ((irq_desc[irq].status & IRQ_MOVE_PENDING) &&
!io_apic_level_ack_pending(irq))
- move_native_irq(irq);
+ move_masked_irq(irq);
- if (!directed_eoi_enabled && !(v & (1 << (i & 0x1f)))) {
+ if ( !(v & (1 << (i & 0x1f))) ) {
atomic_inc(&irq_mis_count);
spin_lock(&ioapic_lock);
__edge_IO_APIC_irq(irq);
@@ -1662,12 +1665,22 @@ static void end_level_ioapic_irq (unsign
if ( !ioapic_ack_new )
{
- if ( irq_desc[irq].status & IRQ_DISABLED )
- return;
-
if ( directed_eoi_enabled )
+ {
+ if ( !(irq_desc[irq].status & (IRQ_DISABLED|IRQ_MOVE_PENDING)) )
+ {
+ eoi_IO_APIC_irq(irq);
+ return;
+ }
+
+ mask_IO_APIC_irq(irq);
eoi_IO_APIC_irq(irq);
- else
+ if ( (irq_desc[irq].status & IRQ_MOVE_PENDING) &&
+ !io_apic_level_ack_pending(irq) )
+ move_masked_irq(irq);
+ }
+
+ if ( !(irq_desc[irq].status & IRQ_DISABLED) )
unmask_IO_APIC_irq(irq);
return;

36
22466-x86-sis-apic-bug.patch Normal file
View File

@ -0,0 +1,36 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1291746724 0
# Node ID bfd13358b8bf3a6ca49702d773435a67f1ea5551
# Parent d4b35162b3a11f7d9c28501192b7e231609e561d
x86: remove BUG_ON() from QUIRK_IOAPIC_*_REGSEL handler
References: bnc#657692
Since (non-pvops, 32-bit only up to 2.6.27) Linux would report "BAD"
unconditionally on all SiS chipset versions (it only looks for a PCI
device at 0000:00:00.0 with SiS as the vendor), we must not crash if
the report on a 64-bit hypervisor doesn't match the #define (which is
zero).
While we could honor the quirk indication even on 64-bit, it doesn't
seem worthwhile, as there's no evidence that newer SiS chipsets
(supporting 64-bit CPUs) are actually affected.
This should also address bug 1687 (mis-reported, however, afaict).
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/arch/x86/platform_hypercall.c
+++ b/xen/arch/x86/platform_hypercall.c
@@ -192,7 +192,10 @@ ret_t do_platform_op(XEN_GUEST_HANDLE(xe
dprintk(XENLOG_INFO, "Domain 0 says that IO-APIC REGSEL is %s\n",
sis_apic_bug ? "bad" : "good");
#else
- BUG_ON(sis_apic_bug != (quirk_id == QUIRK_IOAPIC_BAD_REGSEL));
+ if ( sis_apic_bug != (quirk_id == QUIRK_IOAPIC_BAD_REGSEL) )
+ dprintk(XENLOG_WARNING,
+ "Domain 0 thinks that IO-APIC REGSEL is %s\n",
+ sis_apic_bug ? "good" : "bad");
#endif
break;
default:

28
22470-vlapic-tick-loss.patch Normal file
View File

@ -0,0 +1,28 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1291883699 0
# Node ID 0c97247c64d61511acff37d1903cb304b387f397
# Parent 84a72f957fe9e1a5af3dfb8f53334faa555e22f8
x86:vlapic: Fix possible guest tick losing after save/restore
Guest vcpu may totally lose all ticks if the vlapic->pt.irq was not
restored during save/restore process. Fix it.
Signed-off-by: Wei Gang <gang.wei@intel.com>
--- a/xen/arch/x86/hvm/vlapic.c
+++ b/xen/arch/x86/hvm/vlapic.c
@@ -863,12 +863,12 @@ static void lapic_rearm(struct vlapic *s
unsigned long tmict = vlapic_get_reg(s, APIC_TMICT);
uint64_t period;
+ s->pt.irq = vlapic_get_reg(s, APIC_LVTT) & APIC_VECTOR_MASK;
if ( (tmict = vlapic_get_reg(s, APIC_TMICT)) == 0 )
return;
period = ((uint64_t)APIC_BUS_CYCLE_NS *
(uint32_t)tmict * s->hw.timer_divisor);
- s->pt.irq = vlapic_get_reg(s, APIC_LVTT) & APIC_VECTOR_MASK;
create_periodic_time(vlapic_vcpu(s), &s->pt, period,
vlapic_lvtt_period(s) ? period : 0,
s->pt.irq, vlapic_pt_cb,

480
22475-x2apic-cleanup.patch Normal file
View File

@ -0,0 +1,480 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1291922374 0
# Node ID d9fc83a64a82624e24876250dd88b2cd9528d266
# Parent 49d2aa5cee4ecc2411d8d638e4ee32c10e9b2761
x86: x2apic: Large cleanup
References: bnc#656369, bnc#658704
Signed-off-by: Keir Fraser <keir@xen.org>
--- a/xen/arch/x86/apic.c
+++ b/xen/arch/x86/apic.c
@@ -67,12 +67,12 @@ static int enable_local_apic __initdata
*/
int apic_verbosity;
+static int opt_x2apic = 1;
+boolean_param("x2apic", opt_x2apic);
+
int x2apic_enabled __read_mostly = 0;
int directed_eoi_enabled __read_mostly = 0;
-/* x2APIC is enabled in BIOS */
-static int x2apic_preenabled;
-
/*
* The following vectors are part of the Linux architecture, there
* is no hardware IRQ pin equivalent for them, they are triggered
@@ -944,30 +944,24 @@ no_apic:
return -1;
}
-void check_x2apic_preenabled(void)
+void x2apic_setup(void)
{
+ struct IO_APIC_route_entry **ioapic_entries = NULL;
u32 lo, hi;
- if ( !x2apic_is_available() )
- return;
-
- rdmsr(MSR_IA32_APICBASE, lo, hi);
- if ( lo & MSR_IA32_APICBASE_EXTD )
+ if ( smp_processor_id() != 0 )
{
- printk("x2APIC mode is already enabled by BIOS.\n");
- x2apic_preenabled = 1;
- x2apic_enabled = 1;
+ if ( x2apic_enabled )
+ __enable_x2apic();
+ return;
}
-}
-static void enable_bsp_x2apic(void)
-{
- struct IO_APIC_route_entry **ioapic_entries = NULL;
- const struct genapic *x2apic_genapic = NULL;
-
- ASSERT(smp_processor_id() == 0);
+ if ( !cpu_has_x2apic )
+ return;
- if ( x2apic_preenabled )
+ /* Check whether x2apic mode was already enabled by the BIOS. */
+ rdmsr(MSR_IA32_APICBASE, lo, hi);
+ if ( lo & MSR_IA32_APICBASE_EXTD )
{
/*
* Interrupt remapping should be also enabled by BIOS when
@@ -977,39 +971,33 @@ static void enable_bsp_x2apic(void)
if ( !intremap_enabled() )
panic("Interrupt remapping is not enabled by BIOS while "
"x2APIC is already enabled by BIOS!\n");
+
+ printk("x2APIC mode is already enabled by BIOS.\n");
+ x2apic_enabled = 1;
}
- x2apic_genapic = apic_x2apic_probe();
- if ( x2apic_genapic )
- genapic = x2apic_genapic;
- else
+ if ( !opt_x2apic )
{
- if ( x2apic_cmdline_disable() )
+ if ( !x2apic_enabled )
{
- if ( x2apic_preenabled )
- {
- /* Ignore x2apic=0, and set default x2apic mode */
- genapic = &apic_x2apic_cluster;
- printk("x2APIC: already enabled by BIOS, ignore x2apic=0.\n");
- }
- else
- {
- printk("Not enable x2APIC due to x2apic=0 is set.\n");
- return;
- }
+ printk("Not enabling x2APIC: disabled by cmdline.\n");
+ return;
}
- else
+ printk("x2APIC: Already enabled by BIOS: Ignoring cmdline disable.\n");
+ }
+
+ if ( !iommu_supports_eim() )
+ {
+ if ( !x2apic_enabled )
{
- if ( x2apic_preenabled )
- panic("x2APIC: already enabled by BIOS, but "
- "iommu_supports_eim failed!\n");
- printk("Not enabling x2APIC: depends oniommu_supports_eim\n");
+ printk("Not enabling x2APIC: depends on iommu_supports_eim.\n");
return;
}
+ panic("x2APIC: already enabled by BIOS, but "
+ "iommu_supports_eim failed!\n");
}
- ioapic_entries = alloc_ioapic_entries();
- if ( !ioapic_entries )
+ if ( (ioapic_entries = alloc_ioapic_entries()) == NULL )
{
printk("Allocate ioapic_entries failed\n");
goto out;
@@ -1031,13 +1019,13 @@ static void enable_bsp_x2apic(void)
goto restore_out;
}
- x2apic_enabled = 1;
+ genapic = apic_x2apic_probe();
printk("Switched to APIC driver %s.\n", genapic->name);
- if ( !x2apic_preenabled )
+ if ( !x2apic_enabled )
{
+ x2apic_enabled = 1;
__enable_x2apic();
- printk("x2APIC mode enabled.\n");
}
restore_out:
@@ -1049,24 +1037,6 @@ out:
free_ioapic_entries(ioapic_entries);
}
-static void enable_ap_x2apic(void)
-{
- ASSERT(smp_processor_id() != 0);
-
- /* APs only enable x2apic when BSP did so. */
- BUG_ON(!x2apic_enabled);
-
- __enable_x2apic();
-}
-
-void enable_x2apic(void)
-{
- if ( smp_processor_id() == 0 )
- enable_bsp_x2apic();
- else
- enable_ap_x2apic();
-}
-
void __init init_apic_mappings(void)
{
unsigned long apic_phys;
--- a/xen/arch/x86/cpu/common.c
+++ b/xen/arch/x86/cpu/common.c
@@ -250,8 +250,8 @@ static void __init early_cpu_detect(void
c->x86 = 4;
if (c->cpuid_level >= 0x00000001) {
- u32 junk, tfms, cap0, misc;
- cpuid(0x00000001, &tfms, &misc, &junk, &cap0);
+ u32 cap4, tfms, cap0, misc;
+ cpuid(0x00000001, &tfms, &misc, &cap4, &cap0);
c->x86 = (tfms >> 8) & 15;
c->x86_model = (tfms >> 4) & 15;
if (c->x86 == 0xf)
@@ -260,9 +260,12 @@ static void __init early_cpu_detect(void
c->x86_model += ((tfms >> 16) & 0xF) << 4;
c->x86_mask = tfms & 15;
cap0 &= ~cleared_caps[0];
+ cap4 &= ~cleared_caps[4];
if (cap0 & (1<<19))
c->x86_cache_alignment = ((misc >> 8) & 0xff) * 8;
- c->x86_capability[0] = cap0; /* Added for Xen bootstrap */
+ /* Leaf 0x1 capabilities filled in early for Xen. */
+ c->x86_capability[0] = cap0;
+ c->x86_capability[4] = cap4;
}
}
--- a/xen/arch/x86/genapic/x2apic.c
+++ b/xen/arch/x86/genapic/x2apic.c
@@ -23,89 +23,44 @@
#include <xen/smp.h>
#include <asm/mach-default/mach_mpparse.h>
-static int x2apic = 1;
-boolean_param("x2apic", x2apic);
-
-static int x2apic_phys; /* By default we use logical cluster mode. */
+static int x2apic_phys; /* By default we use logical cluster mode. */
boolean_param("x2apic_phys", x2apic_phys);
-int x2apic_cmdline_disable(void)
-{
- return (x2apic == 0);
-}
-
-static int probe_x2apic_phys(void)
+static void init_apic_ldr_x2apic_phys(void)
{
- return x2apic && x2apic_phys && x2apic_is_available() &&
- iommu_supports_eim();
}
-static int probe_x2apic_cluster(void)
-{
- return x2apic && !x2apic_phys && x2apic_is_available() &&
- iommu_supports_eim();
-}
-
-const struct genapic apic_x2apic_phys = {
- APIC_INIT("x2apic_phys", probe_x2apic_phys),
- GENAPIC_X2APIC_PHYS
-};
-
-const struct genapic apic_x2apic_cluster = {
- APIC_INIT("x2apic_cluster", probe_x2apic_cluster),
- GENAPIC_X2APIC_CLUSTER
-};
-
-const struct genapic *apic_x2apic_probe(void)
-{
- if ( !x2apic || !x2apic_is_available() )
- return NULL;
-
- if ( !iommu_supports_eim() )
- return NULL;
-
- if ( x2apic_phys )
- return &apic_x2apic_phys;
- else
- return &apic_x2apic_cluster;
-}
-
-void init_apic_ldr_x2apic_phys(void)
-{
- return;
-}
-
-void init_apic_ldr_x2apic_cluster(void)
+static void init_apic_ldr_x2apic_cluster(void)
{
int cpu = smp_processor_id();
cpu_2_logical_apicid[cpu] = apic_read(APIC_LDR);
}
-void clustered_apic_check_x2apic(void)
+
+static void clustered_apic_check_x2apic(void)
{
- return;
}
-cpumask_t target_cpus_x2apic(void)
+static cpumask_t target_cpus_x2apic(void)
{
return cpu_online_map;
}
-cpumask_t vector_allocation_domain_x2apic(int cpu)
+static cpumask_t vector_allocation_domain_x2apic(int cpu)
{
return cpumask_of_cpu(cpu);
}
-unsigned int cpu_mask_to_apicid_x2apic_phys(cpumask_t cpumask)
+static unsigned int cpu_mask_to_apicid_x2apic_phys(cpumask_t cpumask)
{
return cpu_physical_id(first_cpu(cpumask));
}
-unsigned int cpu_mask_to_apicid_x2apic_cluster(cpumask_t cpumask)
+static unsigned int cpu_mask_to_apicid_x2apic_cluster(cpumask_t cpumask)
{
return cpu_2_logical_apicid[first_cpu(cpumask)];
}
-void send_IPI_mask_x2apic_phys(const cpumask_t *cpumask, int vector)
+static void send_IPI_mask_x2apic_phys(const cpumask_t *cpumask, int vector)
{
unsigned int cpu, cfg;
unsigned long flags;
@@ -132,7 +87,7 @@ void send_IPI_mask_x2apic_phys(const cpu
local_irq_restore(flags);
}
-void send_IPI_mask_x2apic_cluster(const cpumask_t *cpumask, int vector)
+static void send_IPI_mask_x2apic_cluster(const cpumask_t *cpumask, int vector)
{
unsigned int cpu, cfg;
unsigned long flags;
@@ -148,3 +103,34 @@ void send_IPI_mask_x2apic_cluster(const
local_irq_restore(flags);
}
+
+static const struct genapic apic_x2apic_phys = {
+ APIC_INIT("x2apic_phys", NULL),
+ .int_delivery_mode = dest_Fixed,
+ .int_dest_mode = 0 /* physical delivery */,
+ .init_apic_ldr = init_apic_ldr_x2apic_phys,
+ .clustered_apic_check = clustered_apic_check_x2apic,
+ .target_cpus = target_cpus_x2apic,
+ .vector_allocation_domain = vector_allocation_domain_x2apic,
+ .cpu_mask_to_apicid = cpu_mask_to_apicid_x2apic_phys,
+ .send_IPI_mask = send_IPI_mask_x2apic_phys,
+ .send_IPI_self = send_IPI_self_x2apic
+};
+
+static const struct genapic apic_x2apic_cluster = {
+ APIC_INIT("x2apic_cluster", NULL),
+ .int_delivery_mode = dest_LowestPrio,
+ .int_dest_mode = 1 /* logical delivery */,
+ .init_apic_ldr = init_apic_ldr_x2apic_cluster,
+ .clustered_apic_check = clustered_apic_check_x2apic,
+ .target_cpus = target_cpus_x2apic,
+ .vector_allocation_domain = vector_allocation_domain_x2apic,
+ .cpu_mask_to_apicid = cpu_mask_to_apicid_x2apic_cluster,
+ .send_IPI_mask = send_IPI_mask_x2apic_cluster,
+ .send_IPI_self = send_IPI_self_x2apic
+};
+
+const struct genapic *apic_x2apic_probe(void)
+{
+ return x2apic_phys ? &apic_x2apic_phys : &apic_x2apic_cluster;
+}
--- a/xen/arch/x86/setup.c
+++ b/xen/arch/x86/setup.c
@@ -1006,9 +1006,6 @@ void __init __start_xen(unsigned long mb
tboot_probe();
- /* Check if x2APIC is already enabled in BIOS */
- check_x2apic_preenabled();
-
/* Unmap the first page of CPU0's stack. */
memguard_guard_stack(cpu0_stack);
@@ -1039,8 +1036,7 @@ void __init __start_xen(unsigned long mb
init_apic_mappings();
- if ( x2apic_is_available() )
- enable_x2apic();
+ x2apic_setup();
percpu_free_unused_areas();
--- a/xen/arch/x86/smpboot.c
+++ b/xen/arch/x86/smpboot.c
@@ -325,8 +325,7 @@ void __devinit smp_callin(void)
*/
wait_for_init_deassert(&init_deasserted);
- if ( x2apic_enabled )
- enable_x2apic();
+ x2apic_setup();
/*
* (This works even if the APIC is not enabled.)
--- a/xen/drivers/passthrough/vtd/intremap.c
+++ b/xen/drivers/passthrough/vtd/intremap.c
@@ -129,15 +129,10 @@ int iommu_supports_eim(void)
struct acpi_drhd_unit *drhd;
int apic;
- if ( !iommu_enabled || !iommu_qinval || !iommu_intremap )
+ if ( !iommu_enabled || !iommu_qinval || !iommu_intremap ||
+ list_empty(&acpi_drhd_units) )
return 0;
- if ( list_empty(&acpi_drhd_units) )
- {
- dprintk(XENLOG_WARNING VTDPREFIX, "VT-d is not supported\n");
- return 0;
- }
-
/* We MUST have a DRHD unit for each IOAPIC. */
for ( apic = 0; apic < nr_ioapics; apic++ )
if ( !ioapic_to_drhd(IO_APIC_ID(apic)) )
--- a/xen/include/asm-x86/apic.h
+++ b/xen/include/asm-x86/apic.h
@@ -25,21 +25,8 @@ extern int apic_verbosity;
extern int x2apic_enabled;
extern int directed_eoi_enabled;
-extern void check_x2apic_preenabled(void);
-extern int x2apic_cmdline_disable(void);
-extern void enable_x2apic(void);
-
-static __inline int x2apic_is_available(void)
-{
- unsigned int op = 1, eax, ecx;
-
- asm ( "cpuid"
- : "=a" (eax), "=c" (ecx)
- : "0" (op)
- : "bx", "dx" );
-
- return (ecx & (1U << 21));
-}
+void x2apic_setup(void);
+const struct genapic *apic_x2apic_probe(void);
/*
* Define the default level of output to be very little
--- a/xen/include/asm-x86/genapic.h
+++ b/xen/include/asm-x86/genapic.h
@@ -49,8 +49,6 @@ struct genapic {
APICFUNC(acpi_madt_oem_check)
extern const struct genapic *genapic;
-extern const struct genapic apic_x2apic_phys;
-extern const struct genapic apic_x2apic_cluster;
void init_apic_ldr_flat(void);
void clustered_apic_check_flat(void);
@@ -70,39 +68,6 @@ cpumask_t vector_allocation_domain_flat(
.send_IPI_mask = send_IPI_mask_flat, \
.send_IPI_self = send_IPI_self_flat
-const struct genapic *apic_x2apic_probe(void);
-void init_apic_ldr_x2apic_phys(void);
-void init_apic_ldr_x2apic_cluster(void);
-void clustered_apic_check_x2apic(void);
-cpumask_t target_cpus_x2apic(void);
-unsigned int cpu_mask_to_apicid_x2apic_phys(cpumask_t cpumask);
-unsigned int cpu_mask_to_apicid_x2apic_cluster(cpumask_t cpumask);
-void send_IPI_mask_x2apic_phys(const cpumask_t *mask, int vector);
-void send_IPI_mask_x2apic_cluster(const cpumask_t *mask, int vector);
-void send_IPI_self_x2apic(int vector);
-cpumask_t vector_allocation_domain_x2apic(int cpu);
-#define GENAPIC_X2APIC_PHYS \
- .int_delivery_mode = dest_Fixed, \
- .int_dest_mode = 0 /* physical delivery */, \
- .init_apic_ldr = init_apic_ldr_x2apic_phys, \
- .clustered_apic_check = clustered_apic_check_x2apic, \
- .target_cpus = target_cpus_x2apic, \
- .vector_allocation_domain = vector_allocation_domain_x2apic, \
- .cpu_mask_to_apicid = cpu_mask_to_apicid_x2apic_phys, \
- .send_IPI_mask = send_IPI_mask_x2apic_phys, \
- .send_IPI_self = send_IPI_self_x2apic
-
-#define GENAPIC_X2APIC_CLUSTER \
- .int_delivery_mode = dest_LowestPrio, \
- .int_dest_mode = 1 /* logical delivery */, \
- .init_apic_ldr = init_apic_ldr_x2apic_cluster, \
- .clustered_apic_check = clustered_apic_check_x2apic, \
- .target_cpus = target_cpus_x2apic, \
- .vector_allocation_domain = vector_allocation_domain_x2apic, \
- .cpu_mask_to_apicid = cpu_mask_to_apicid_x2apic_cluster, \
- .send_IPI_mask = send_IPI_mask_x2apic_cluster, \
- .send_IPI_self = send_IPI_self_x2apic
-
void init_apic_ldr_phys(void);
void clustered_apic_check_phys(void);
cpumask_t target_cpus_phys(void);
@@ -121,4 +86,6 @@ cpumask_t vector_allocation_domain_phys(
.send_IPI_mask = send_IPI_mask_phys, \
.send_IPI_self = send_IPI_self_phys
+void send_IPI_self_x2apic(int vector);
+
#endif

34
22484-vlapic-tmcct-periodic.patch Normal file
View File

@ -0,0 +1,34 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1291999745 0
# Node ID 3b3fce9050b9b14f3a11123f6943998413ea7e7e
# Parent 901c118b363e978566775f9d118b4f2f15598212
hvm vlapic: Fix tmcct read logic when in periodic mode.
Signed-off-by: Keir Fraser <keir@xen.org>
--- a/xen/arch/x86/hvm/vlapic.c
+++ b/xen/arch/x86/hvm/vlapic.c
@@ -428,12 +428,19 @@ int vlapic_ipi(
static uint32_t vlapic_get_tmcct(struct vlapic *vlapic)
{
struct vcpu *v = current;
- uint32_t tmcct, tmict = vlapic_get_reg(vlapic, APIC_TMICT);
+ uint32_t tmcct = 0, tmict = vlapic_get_reg(vlapic, APIC_TMICT);
uint64_t counter_passed;
counter_passed = ((hvm_get_guest_time(v) - vlapic->timer_last_update)
- / APIC_BUS_CYCLE_NS / vlapic->hw.timer_divisor);
- tmcct = tmict - counter_passed;
+ / (APIC_BUS_CYCLE_NS * vlapic->hw.timer_divisor));
+
+ if ( tmict != 0 )
+ {
+ if ( vlapic_lvtt_period(vlapic) )
+ counter_passed %= tmict;
+ if ( counter_passed < tmict )
+ tmcct = tmict - counter_passed;
+ }
HVM_DBG_LOG(DBG_LEVEL_VLAPIC_TIMER,
"timer initial count %d, timer current count %d, "

91
22504-iommu-dom0-holes.patch Normal file
View File

@ -0,0 +1,91 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292320377 0
# Node ID fd4cbfbbd83e6091a343844eae1da1468f54b72b
# Parent ab785e37499c8cdadd1fd5e4ab1bfbbacebf358b
x86/iommu: don't map RAM holes above 4G
References: bnc#658163
Matching the comment in iommu_set_dom0_mapping(), map only actual RAM
from the address range starting at 4G. It's not clear though whether
that comment is actually correct (which is why I'm sending this as
RFC), but it is certain that on systems with sparse physical memory
map we're currently wasting a potentially significant amount of memory
for setting up IOMMU page tables that will never be used.
The main question is what happens for MMIO ranges living above 4G. Of
course, the same issue would currently exist for any such ranges
sitting beyond the end of RAM.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/drivers/passthrough/amd/pci_amd_iommu.c
+++ b/xen/drivers/passthrough/amd/pci_amd_iommu.c
@@ -239,8 +239,16 @@ static void amd_iommu_dom0_init(struct d
if ( !iommu_passthrough && !need_iommu(d) )
{
/* Set up 1:1 page table for dom0 */
- for ( i = 0; i < max_page; i++ )
- amd_iommu_map_page(d, i, i, IOMMUF_readable|IOMMUF_writable);
+ for ( i = 0; i < max_pdx; i++ )
+ {
+ unsigned long pfn = pdx_to_pfn(i);
+
+ /*
+ * XXX Should we really map all non-RAM (above 4G)? Minimally
+ * a pfn_valid() check would seem desirable here.
+ */
+ amd_iommu_map_page(d, pfn, pfn, IOMMUF_readable|IOMMUF_writable);
+ }
}
amd_iommu_setup_dom0_devices(d);
--- a/xen/drivers/passthrough/vtd/x86/vtd.c
+++ b/xen/drivers/passthrough/vtd/x86/vtd.c
@@ -128,14 +128,14 @@ void hvm_dpci_isairq_eoi(struct domain *
void iommu_set_dom0_mapping(struct domain *d)
{
- u64 i, j, tmp, max_pfn;
+ unsigned long i, j, tmp, top;
extern int xen_in_range(unsigned long mfn);
BUG_ON(d->domain_id != 0);
- max_pfn = max_t(u64, max_page, 0x100000000ull >> PAGE_SHIFT);
+ top = max(max_pdx, pfn_to_pdx(0xffffffffUL >> PAGE_SHIFT) + 1);
- for ( i = 0; i < max_pfn; i++ )
+ for ( i = 0; i < top; i++ )
{
/*
* Set up 1:1 mapping for dom0. Default to use only conventional RAM
@@ -143,18 +143,23 @@ void iommu_set_dom0_mapping(struct domai
* inclusive mapping maps in everything below 4GB except unusable
* ranges.
*/
- if ( !page_is_ram_type(i, RAM_TYPE_CONVENTIONAL) &&
- (!iommu_inclusive_mapping ||
- page_is_ram_type(i, RAM_TYPE_UNUSABLE)) )
+ unsigned long pfn = pdx_to_pfn(i);
+
+ if ( pfn > (0xffffffffUL >> PAGE_SHIFT) ?
+ (!mfn_valid(pfn) ||
+ !page_is_ram_type(pfn, RAM_TYPE_CONVENTIONAL)) :
+ iommu_inclusive_mapping ?
+ page_is_ram_type(pfn, RAM_TYPE_UNUSABLE) :
+ !page_is_ram_type(pfn, RAM_TYPE_CONVENTIONAL) )
continue;
/* Exclude Xen bits */
- if ( xen_in_range(i) )
+ if ( xen_in_range(pfn) )
continue;
tmp = 1 << (PAGE_SHIFT - PAGE_SHIFT_4K);
for ( j = 0; j < tmp; j++ )
- iommu_map_page(d, (i*tmp+j), (i*tmp+j),
+ iommu_map_page(d, pfn * tmp + j, pfn * tmp + j,
IOMMUF_readable|IOMMUF_writable);
if (!(i & (0xfffff >> (PAGE_SHIFT - PAGE_SHIFT_4K))))

40
22506-x86-iommu-dom0-estimate.patch Normal file
View File

@ -0,0 +1,40 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292320450 0
# Node ID 618ba64260faf45e6ec391582099d7388f013e81
# Parent 72326371ae8106b91da0ca6b0436dd2b6478b7a7
x86/iommu: account for necessary allocations when calculating Dom0's initial allocation size
References: bnc#658163
As of c/s 21812:e382656e4dcc, IOMMU related allocations for Dom0
happen only after it got all of its memory allocated, and hence the
reserve (mainly for setting up its swiotlb) may get exhausted without
accounting for the necessary allocations up front.
While not precise, the estimate has been found to be within a couple
of pages for the systems it got tested on.
For the calculation to be reasonably correct, this depends on the
patch titled "x86/iommu: don't map RAM holes above 4G" sent out
yesterday.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/arch/x86/domain_build.c
+++ b/xen/arch/x86/domain_build.c
@@ -188,6 +188,15 @@ static unsigned long __init compute_dom0
if ( is_pv_32on64_domain(d) )
avail -= opt_dom0_max_vcpus - 1;
+ /* Reserve memory for iommu_dom0_init() (rough estimate). */
+ if ( iommu_enabled )
+ {
+ unsigned int s;
+
+ for ( s = 9; s < BITS_PER_LONG; s += 9 )
+ avail -= max_pdx >> s;
+ }
+
/*
* If domain 0 allocation isn't specified, reserve 1/16th of available
* memory for things like DMA buffers. This reservation is clamped to

123
22526-ept-access-once.patch Normal file
View File

@ -0,0 +1,123 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292410025 0
# Node ID 7a5ee380041707177ca9c78e800095d1f5f3d373
# Parent 01f3b350902385627d1fa9e8cd1c231953e7610c
ept: Remove lock in ept_get_entry, replace with access-once semantics.
This mirrors the RVI/shadow situation, where p2m read access is
lockless because it's done in the hardware (linear map of the p2m
table).
This fixes the original bug (call it bug A) without introducing bug B
(a deadlock).
Bug A was caused by a race when updating p2m entries: between testing
if it's valid, and testing if it's populate-on-demand, it may have
been changed from populate-on-demand to valid.
My original patch simply introduced a lock into ept_get_entry, but
that caused bug B, caused by circular locking order: p2m_change_type
[grabs p2m lock] -> set_p2m_entry -> ept_set_entry ->
ept_set_middle_level -> p2m_alloc [grabs hap lock] write cr4 ->
hap_update_paging_modes [grabes hap lock] -> hap_update_cr3 ->
gfn_to_mfn -> ept_get_entry -> [grabs p2m lock]
Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
--- a/xen/arch/x86/mm/hap/p2m-ept.c
+++ b/xen/arch/x86/mm/hap/p2m-ept.c
@@ -137,7 +137,7 @@ static int ept_next_level(struct domain
ept_entry_t **table, unsigned long *gfn_remainder,
u32 shift)
{
- ept_entry_t *ept_entry;
+ ept_entry_t *ept_entry, e;
ept_entry_t *next;
u32 index;
@@ -145,9 +145,11 @@ static int ept_next_level(struct domain
ept_entry = (*table) + index;
- if ( !is_epte_present(ept_entry) )
+ e=*ept_entry;
+
+ if ( !is_epte_present(&e) )
{
- if ( ept_entry->avail1 == p2m_populate_on_demand )
+ if ( e.avail1 == p2m_populate_on_demand )
return GUEST_TABLE_POD_PAGE;
if ( read_only )
@@ -155,15 +157,17 @@ static int ept_next_level(struct domain
if ( !ept_set_middle_entry(d, ept_entry) )
return GUEST_TABLE_MAP_FAILED;
+ else
+ e=*ept_entry;
}
/* The only time sp would be set here is if we had hit a superpage */
- if ( is_epte_superpage(ept_entry) )
+ if ( is_epte_superpage(&e) )
return GUEST_TABLE_SUPER_PAGE;
else
{
*gfn_remainder &= (1UL << shift) - 1;
- next = map_domain_page(ept_entry->mfn);
+ next = map_domain_page(e.mfn);
unmap_domain_page(*table);
*table = next;
return GUEST_TABLE_NORMAL_PAGE;
@@ -235,35 +239,39 @@ ept_set_entry(struct domain *d, unsigned
if ( mfn_valid(mfn_x(mfn)) || direct_mmio || p2m_is_paged(p2mt) ||
(p2mt == p2m_ram_paging_in_start) )
{
- ept_entry->emt = epte_get_entry_emt(d, gfn, mfn, &ipat,
+ ept_entry_t new_entry;
+
+ new_entry.emt = epte_get_entry_emt(d, gfn, mfn, &ipat,
direct_mmio);
- ept_entry->ipat = ipat;
- ept_entry->sp = order ? 1 : 0;
+ new_entry.ipat = ipat;
+ new_entry.sp = order ? 1 : 0;
if ( ret == GUEST_TABLE_SUPER_PAGE )
{
- if ( ept_entry->mfn == (mfn_x(mfn) - offset) )
+ if ( new_entry.mfn == (mfn_x(mfn) - offset) )
need_modify_vtd_table = 0;
else
- ept_entry->mfn = mfn_x(mfn) - offset;
+ new_entry.mfn = mfn_x(mfn) - offset;
- if ( (ept_entry->avail1 == p2m_ram_logdirty)
+ if ( (new_entry.avail1 == p2m_ram_logdirty)
&& (p2mt == p2m_ram_rw) )
for ( i = 0; i < 512; i++ )
paging_mark_dirty(d, mfn_x(mfn) - offset + i);
}
else
{
- if ( ept_entry->mfn == mfn_x(mfn) )
+ if ( new_entry.mfn == mfn_x(mfn) )
need_modify_vtd_table = 0;
else
- ept_entry->mfn = mfn_x(mfn);
+ new_entry.mfn = mfn_x(mfn);
}
- ept_entry->avail1 = p2mt;
- ept_entry->avail2 = 0;
+ new_entry.avail1 = p2mt;
+ new_entry.avail2 = 0;
+
+ ept_p2m_type_to_flags(&new_entry, p2mt);
- ept_p2m_type_to_flags(ept_entry, p2mt);
+ ept_entry->epte = new_entry.epte;
}
else
ept_entry->epte = 0;

165
22533-x86-32bit-apicid.patch Normal file
View File

@ -0,0 +1,165 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292413934 0
# Node ID 548c808be2a6f166657ce7cfb86df7a296a9526a
# Parent c0662cb08260d831ae8d211ba16fc7038cb889b6
x86: mpparse and cstate need to use 32bit apic id
References: bnc#656369, bnc#658704
Instead of going with mpc_config_processor struct.
that field ony have 8 bits.
We should not change that struct, because it is shared with mptable.
Also need to increase MAX_APICS.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Rather than using a fixed value of 512, make this scale with NR_CPUS
(which obviously still doesn't cover all theoretically possible
systems, but at least allows some build time control).
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/arch/x86/acpi/cpu_idle.c
+++ b/xen/arch/x86/acpi/cpu_idle.c
@@ -742,7 +742,7 @@ static void set_cx(
int get_cpu_id(u8 acpi_id)
{
int i;
- u8 apic_id;
+ u32 apic_id;
apic_id = x86_acpiid_to_apicid[acpi_id];
if ( apic_id == 0xff )
--- a/xen/arch/x86/mpparse.c
+++ b/xen/arch/x86/mpparse.c
@@ -99,7 +99,8 @@ static int mpc_record;
static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __initdata;
/* Return xen's logical cpu_id of the new added cpu or <0 if error */
-static int __devinit MP_processor_info (struct mpc_config_processor *m)
+static int __devinit MP_processor_info_x(struct mpc_config_processor *m,
+ u32 apicidx)
{
int ver, apicid, cpu = 0;
physid_mask_t phys_cpu;
@@ -107,7 +108,7 @@ static int __devinit MP_processor_info (
if (!(m->mpc_cpuflag & CPU_ENABLED))
return -EINVAL;
- apicid = mpc_apic_id(m, translation_table[mpc_record]);
+ apicid = mpc_apic_id(m, apicidx, translation_table[mpc_record]);
if (m->mpc_featureflag&(1<<0))
Dprintk(" Floating point unit present.\n");
@@ -159,7 +160,7 @@ static int __devinit MP_processor_info (
if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
Dprintk(" Bootup CPU\n");
- boot_cpu_physical_apicid = m->mpc_apicid;
+ boot_cpu_physical_apicid = apicid;
}
ver = m->mpc_apicver;
@@ -170,10 +171,10 @@ static int __devinit MP_processor_info (
if (ver == 0x0) {
printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! "
"fixing up to 0x10. (tell your hw vendor)\n",
- m->mpc_apicid);
+ apicid);
ver = 0x10;
}
- apic_version[m->mpc_apicid] = ver;
+ apic_version[apicid] = ver;
phys_cpu = apicid_to_cpu_present(apicid);
physids_or(phys_cpu_present_map, phys_cpu_present_map, phys_cpu);
@@ -216,6 +217,11 @@ static int __devinit MP_processor_info (
return cpu;
}
+static int __devinit MP_processor_info(struct mpc_config_processor *m)
+{
+ return MP_processor_info_x(m, m->mpc_apicid);
+}
+
static void __init MP_bus_info (struct mpc_config_bus *m)
{
char str[7];
@@ -844,7 +850,7 @@ int __devinit mp_register_lapic (
struct mpc_config_processor processor;
int boot_cpu = 0;
- if (MAX_APICS - id <= 0) {
+ if (MAX_APICS <= id) {
printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
id, MAX_APICS);
return -EINVAL;
@@ -864,7 +870,7 @@ int __devinit mp_register_lapic (
processor.mpc_reserved[0] = 0;
processor.mpc_reserved[1] = 0;
- return MP_processor_info(&processor);
+ return MP_processor_info_x(&processor, id);
}
void mp_unregister_lapic(uint32_t apic_id, uint32_t cpu)
--- a/xen/common/sched_sedf.c
+++ b/xen/common/sched_sedf.c
@@ -124,7 +124,6 @@ struct sedf_cpu_info {
#define PERIOD_BEGIN(inf) ((inf)->deadl_abs - (inf)->period)
-#define MIN(x,y) (((x)<(y))?(x):(y))
#define DIV_UP(x,y) (((x) + (y) - 1) / y)
#define extra_runs(inf) ((inf->status) & 6)
--- a/xen/include/asm-x86/mach-generic/mach_apic.h
+++ b/xen/include/asm-x86/mach-generic/mach_apic.h
@@ -26,15 +26,15 @@ static inline void enable_apic_mode(void
extern u32 bios_cpu_apicid[];
-static inline int mpc_apic_id(struct mpc_config_processor *m,
+static inline int mpc_apic_id(struct mpc_config_processor *m, u32 apicid,
struct mpc_config_translation *translation_record)
{
printk("Processor #%d %d:%d APIC version %d\n",
- m->mpc_apicid,
+ apicid,
(m->mpc_cpufeature & CPU_FAMILY_MASK) >> 8,
(m->mpc_cpufeature & CPU_MODEL_MASK) >> 4,
m->mpc_apicver);
- return (m->mpc_apicid);
+ return apicid;
}
static inline void setup_portio_remap(void)
--- a/xen/include/asm-x86/mpspec_def.h
+++ b/xen/include/asm-x86/mpspec_def.h
@@ -14,7 +14,7 @@
#define SMP_MAGIC_IDENT (('_'<<24)|('P'<<16)|('M'<<8)|'_')
#define MAX_MPC_ENTRY 1024
-#define MAX_APICS 256
+#define MAX_APICS MAX(256, 4 * NR_CPUS)
struct intel_mp_floating
{
--- a/xen/include/xen/kernel.h
+++ b/xen/include/xen/kernel.h
@@ -33,6 +33,13 @@
#define max_t(type,x,y) \
({ type __x = (x); type __y = (y); __x > __y ? __x: __y; })
+/*
+ * pre-processor, array size, and bit field width suitable variants;
+ * please don't use in "normal" expressions.
+ */
+#define MIN(x,y) ((x) < (y) ? (x) : (y))
+#define MAX(x,y) ((x) > (y) ? (x) : (y))
+
/**
* container_of - cast a member of a structure out to the containing structure
*

153
22534-x86-max-local-apic.patch Normal file
View File

@ -0,0 +1,153 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292414148 0
# Node ID d4b373ec948bed71b8683e09e41c6afc99f1edb6
# Parent 548c808be2a6f166657ce7cfb86df7a296a9526a
x86: increase MAX_LOCAL_APIC
References: bnc#656369, bnc#658704
otherwise apicid_to_node[MAX_LOCAL_APIC] will be overrun if apicid >
255. After patch, the mapping get right.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Make this and also MAX_MADT_ENTRIES loosely depend on NR_CPUS. Tie
MAX_APICS to MAX_LOCAL_APIC. Fix initializer of x86_acpiid_to_apicid[]
to match the array member type of u32, as well as all checks in
readers of this array and x86_cpu_to_apicid[].
While the adjustment to xen_vcpu_physid_to_x86_{acpi,apic}id() is not
backward compatible, I think it should still be done this way as the
former reserving of values beyond 0xff should never have been part of
the interface. If considered impossible, a second best solution would
appear to be to make the macros depend on __XEN_INTERFACE_VERSION__.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/arch/x86/acpi/boot.c
+++ b/xen/arch/x86/acpi/boot.c
@@ -82,7 +82,7 @@ u8 acpi_enable_value, acpi_disable_value
#endif
u32 x86_acpiid_to_apicid[MAX_MADT_ENTRIES] =
- {[0 ... MAX_MADT_ENTRIES - 1] = 0xff };
+ {[0 ... MAX_MADT_ENTRIES - 1] = BAD_APICID };
EXPORT_SYMBOL(x86_acpiid_to_apicid);
/* --------------------------------------------------------------------------
@@ -534,6 +534,7 @@ static int __init acpi_parse_madt_lapic_
mp_register_lapic_address(acpi_lapic_addr);
+ BUILD_BUG_ON(MAX_APICS != MAX_LOCAL_APIC);
count = acpi_table_parse_madt(ACPI_MADT_LAPIC, acpi_parse_lapic,
MAX_APICS);
x2count = acpi_table_parse_madt(ACPI_MADT_X2APIC, acpi_parse_x2apic,
@@ -998,12 +999,12 @@ unsigned int acpi_get_processor_id(unsig
{
unsigned int acpiid, apicid;
- if ((apicid = x86_cpu_to_apicid[cpu]) == 0xff)
- return 0xff;
+ if ((apicid = x86_cpu_to_apicid[cpu]) == BAD_APICID)
+ return INVALID_ACPIID;
for (acpiid = 0; acpiid < ARRAY_SIZE(x86_acpiid_to_apicid); acpiid++)
if (x86_acpiid_to_apicid[acpiid] == apicid)
return acpiid;
- return 0xff;
+ return INVALID_ACPIID;
}
--- a/xen/arch/x86/acpi/cpu_idle.c
+++ b/xen/arch/x86/acpi/cpu_idle.c
@@ -745,7 +745,7 @@ int get_cpu_id(u8 acpi_id)
u32 apic_id;
apic_id = x86_acpiid_to_apicid[acpi_id];
- if ( apic_id == 0xff )
+ if ( apic_id == BAD_APICID )
return -1;
for ( i = 0; i < NR_CPUS; i++ )
--- a/xen/arch/x86/smpboot.c
+++ b/xen/arch/x86/smpboot.c
@@ -96,7 +96,7 @@ struct cpuinfo_x86 cpu_data[NR_CPUS];
EXPORT_SYMBOL(cpu_data);
u32 x86_cpu_to_apicid[NR_CPUS] __read_mostly =
- { [0 ... NR_CPUS-1] = -1U };
+ { [0 ... NR_CPUS-1] = BAD_APICID };
EXPORT_SYMBOL(x86_cpu_to_apicid);
static void map_cpu_to_logical_apicid(void);
@@ -1479,7 +1479,7 @@ int cpu_add(uint32_t apic_id, uint32_t a
return -EINVAL;
/* Detect if the cpu has been added before */
- if ( x86_acpiid_to_apicid[acpi_id] != 0xff)
+ if ( x86_acpiid_to_apicid[acpi_id] != BAD_APICID )
{
if (x86_acpiid_to_apicid[acpi_id] != apic_id)
return -EINVAL;
@@ -1513,7 +1513,7 @@ int cpu_add(uint32_t apic_id, uint32_t a
{
dprintk(XENLOG_WARNING,
"Setup node failed for pxm %x\n", pxm);
- x86_acpiid_to_apicid[acpi_id] = 0xff;
+ x86_acpiid_to_apicid[acpi_id] = BAD_APICID;
mp_unregister_lapic(apic_id, cpu);
spin_unlock(&cpu_add_remove_lock);
return node;
--- a/xen/include/asm-x86/acpi.h
+++ b/xen/include/asm-x86/acpi.h
@@ -150,9 +150,11 @@ struct acpi_sleep_info {
#endif /* CONFIG_ACPI_SLEEP */
-#define MAX_MADT_ENTRIES 256
+#define MAX_MADT_ENTRIES MAX(256, 2 * NR_CPUS)
extern u32 x86_acpiid_to_apicid[];
-#define MAX_LOCAL_APIC 256
+#define MAX_LOCAL_APIC MAX(256, 4 * NR_CPUS)
+
+#define INVALID_ACPIID (-1U)
extern u32 pmtmr_ioport;
--- a/xen/include/asm-x86/numa.h
+++ b/xen/include/asm-x86/numa.h
@@ -39,7 +39,7 @@ extern int setup_node(int pxm);
extern void srat_detect_node(int cpu);
extern void setup_node_bootmem(int nodeid, u64 start, u64 end);
-extern unsigned char apicid_to_node[256];
+extern unsigned char apicid_to_node[];
#ifdef CONFIG_NUMA
extern void __init init_cpu_to_node(void);
--- a/xen/include/public/vcpu.h
+++ b/xen/include/public/vcpu.h
@@ -187,8 +187,7 @@ DEFINE_XEN_GUEST_HANDLE(vcpu_register_vc
/*
* Get the physical ID information for a pinned vcpu's underlying physical
* processor. The physical ID informmation is architecture-specific.
- * On x86: id[31:0]=apic_id, id[63:32]=acpi_id, and all values 0xff and
- * greater are reserved.
+ * On x86: id[31:0]=apic_id, id[63:32]=acpi_id.
* This command returns -EINVAL if it is not a valid operation for this VCPU.
*/
#define VCPUOP_get_physid 12 /* arg == vcpu_get_physid_t */
@@ -197,10 +196,8 @@ struct vcpu_get_physid {
};
typedef struct vcpu_get_physid vcpu_get_physid_t;
DEFINE_XEN_GUEST_HANDLE(vcpu_get_physid_t);
-#define xen_vcpu_physid_to_x86_apicid(physid) \
- ((((uint32_t)(physid)) >= 0xff) ? 0xff : ((uint8_t)(physid)))
-#define xen_vcpu_physid_to_x86_acpiid(physid) \
- ((((uint32_t)((physid)>>32)) >= 0xff) ? 0xff : ((uint8_t)((physid)>>32)))
+#define xen_vcpu_physid_to_x86_apicid(physid) ((uint32_t)(physid))
+#define xen_vcpu_physid_to_x86_acpiid(physid) ((uint32_t)((physid) >> 32))
/*
* Register a memory location to get a secondary copy of the vcpu time

82
22535-x2apic-preenabled.patch Normal file
View File

@ -0,0 +1,82 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292414185 0
# Node ID fcc91cb3efcb83823b7e9c74ee3c97c4a9c0b7d5
# Parent d4b373ec948bed71b8683e09e41c6afc99f1edb6
x86: x2apic pre-enabled but intr-remapping is not enabled
References: bnc#656369, bnc#658704
Make it aligned with Linux kernel.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
Removed unnecessary bits from the original patch, and removed
intremap_enabled() with its only caller gone.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/arch/x86/apic.c
+++ b/xen/arch/x86/apic.c
@@ -963,15 +963,6 @@ void x2apic_setup(void)
rdmsr(MSR_IA32_APICBASE, lo, hi);
if ( lo & MSR_IA32_APICBASE_EXTD )
{
- /*
- * Interrupt remapping should be also enabled by BIOS when
- * x2APIC is already enabled by BIOS, otherwise it's a BIOS
- * bug
- */
- if ( !intremap_enabled() )
- panic("Interrupt remapping is not enabled by BIOS while "
- "x2APIC is already enabled by BIOS!\n");
-
printk("x2APIC mode is already enabled by BIOS.\n");
x2apic_enabled = 1;
}
@@ -1014,6 +1005,10 @@ void x2apic_setup(void)
if ( iommu_enable_IR() )
{
+ if ( x2apic_enabled )
+ panic("Interrupt remapping could not be enabled while "
+ "x2APIC is already enabled by BIOS!\n");
+
printk("Would not enable x2APIC due to interrupt remapping "
"cannot be enabled.\n");
goto restore_out;
--- a/xen/drivers/passthrough/vtd/intremap.c
+++ b/xen/drivers/passthrough/vtd/intremap.c
@@ -884,23 +884,3 @@ void iommu_disable_IR(void)
for_each_drhd_unit ( drhd )
disable_qinval(drhd->iommu);
}
-
-/*
- * Check if interrupt remapping is enabled or not
- * return 1: enabled
- * return 0: not enabled
- */
-int intremap_enabled(void)
-{
- struct acpi_drhd_unit *drhd;
- u32 sts;
-
- for_each_drhd_unit ( drhd )
- {
- sts = dmar_readl(drhd->iommu->reg, DMAR_GSTS_REG);
- if ( !(sts & DMA_GSTS_IRES) )
- return 0;
- }
-
- return 1;
-}
--- a/xen/include/xen/iommu.h
+++ b/xen/include/xen/iommu.h
@@ -62,7 +62,6 @@ int iommu_setup(void);
int iommu_supports_eim(void);
int iommu_enable_IR(void);
void iommu_disable_IR(void);
-int intremap_enabled(void);
int iommu_add_device(struct pci_dev *pdev);
int iommu_remove_device(struct pci_dev *pdev);

101
22538-keyhandler-relax.patch Normal file
View File

@ -0,0 +1,101 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292414674 0
# Node ID a3a29e67aa7e75a094e1b4237b10a68cf829b542
# Parent 16673224c1cc2ca3bb9730f1d7c84fe4d96e5323
Reduce side effects of handling '*' debug key
References: bnc#659284
NMI watchdog should be suppressed when dumping IRQ handlers. Softirqs
should be handled periodically while processing non-IRQ handlers.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1293185578 0
# Node ID dca1b7cf2e2c27cd160bd1d1d284e3f810d4936c
# Parent e8acb9753ff1a5cd3d6a45eda5f4c6f0059c281a
re-add calls accidentally deleted from run_all_nonirq_keyhandlers()
c/s 22538:a3a29e67aa7e, having got applied in a form different from
the one submitted, resulted in the calls to
console_{start,end}_log_everything() getting removed without
replacement. Add them back since, other than run_all_keyhandlers(),
this doesn't run with log-everything already in effect.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/common/keyhandler.c
+++ b/xen/common/keyhandler.c
@@ -407,14 +407,17 @@ static void run_all_nonirq_keyhandlers(u
int k;
console_start_log_everything();
+
for ( k = 0; k < ARRAY_SIZE(key_table); k++ )
{
+ process_pending_softirqs_nested();
h = key_table[k];
if ( (h == NULL) || !h->diagnostic || h->irq_callback )
continue;
printk("[%c: %s]\n", k, h->desc);
(*h->u.fn)(k);
}
+
console_end_log_everything();
}
@@ -426,10 +429,11 @@ static void run_all_keyhandlers(unsigned
struct keyhandler *h;
int k;
+ watchdog_disable();
+
printk("'%c' pressed -> firing all diagnostic keyhandlers\n", key);
/* Fire all the IRQ-context diangostic keyhandlers now */
- console_start_log_everything();
for ( k = 0; k < ARRAY_SIZE(key_table); k++ )
{
h = key_table[k];
@@ -438,7 +442,8 @@ static void run_all_keyhandlers(unsigned
printk("[%c: %s]\n", k, h->desc);
(*h->u.irq_fn)(k, regs);
}
- console_end_log_everything();
+
+ watchdog_enable();
/* Trigger the others from a tasklet in non-IRQ context */
tasklet_schedule(&run_all_keyhandlers_tasklet);
--- a/xen/common/softirq.c
+++ b/xen/common/softirq.c
@@ -54,6 +54,16 @@ void process_pending_softirqs(void)
__do_softirq(1ul<<SCHEDULE_SOFTIRQ);
}
+void process_pending_softirqs_nested(void)
+{
+ ASSERT(!in_irq() && local_irq_is_enabled());
+ /*
+ * Do not enter scheduler as it can preempt the calling context,
+ * and do not run tasklets as we're running one currently.
+ */
+ __do_softirq((1ul<<SCHEDULE_SOFTIRQ) | (1ul<<TASKLET_SOFTIRQ));
+}
+
asmlinkage void do_softirq(void)
{
__do_softirq(0);
--- a/xen/include/xen/softirq.h
+++ b/xen/include/xen/softirq.h
@@ -39,6 +39,8 @@ void raise_softirq(unsigned int nr);
* Use this instead of do_softirq() when you do not want to be preempted.
*/
void process_pending_softirqs(void);
+/* ... and use this instead when running inside a tasklet. */
+void process_pending_softirqs_nested(void);
/*
* TASKLETS -- dynamically-allocatable tasks run in softirq context

21
22540-32on64-hypercall-debug.patch Normal file
View File

@ -0,0 +1,21 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292415031 0
# Node ID d8279118b4bbb3bb3689b9c20abb25d0c09e2b69
# Parent 20c65aa19075b1a696adb56325827412596ac0d4
x86/32on64: zero-extend hypercall index before use in memory access (debug mode only)
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/arch/x86/x86_64/compat/entry.S
+++ b/xen/arch/x86/x86_64/compat/entry.S
@@ -36,7 +36,8 @@ ENTRY(compat_hypercall)
pushq UREGS_rbx(%rsp); pushq %rcx; pushq %rdx; pushq %rsi; pushq %rdi
pushq UREGS_rbp+5*8(%rsp)
leaq compat_hypercall_args_table(%rip),%r10
- movq $6,%rcx
+ movl %eax,%eax
+ movl $6,%ecx
subb (%r10,%rax,1),%cl
movq %rsp,%rdi
movl $0xDEADBEEF,%eax

28
22549-vtd-map-page-leak.patch Normal file
View File

@ -0,0 +1,28 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292513937 0
# Node ID aa18b8ddaf05084aef3f8cff11d92248a8b6fac8
# Parent f46f46bbb69a788037f866025c88743d55dde416
vtd: Require unmap_vtd_domain_page() on a couple of early exit paths.
From: Jan Beulich <JBeulich@novell.com>
Signed-off-by: Keir Fraser <keir@xen.org>
--- a/xen/drivers/passthrough/vtd/iommu.c
+++ b/xen/drivers/passthrough/vtd/iommu.c
@@ -1330,6 +1330,7 @@ int domain_context_mapping_one(
if ( context_set_domain_id(context, domain, iommu) )
{
spin_unlock(&iommu->lock);
+ unmap_vtd_domain_page(context_entries);
return -EFAULT;
}
@@ -1671,6 +1672,7 @@ static int intel_iommu_map_page(
if ( old.val == new.val )
{
spin_unlock(&hd->mapping_lock);
+ unmap_vtd_domain_page(page);
return 0;
}
*pte = new;

29
22574-ept-skip-validation.patch Normal file
View File

@ -0,0 +1,29 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1292954974 0
# Node ID b5418c045d029e8bf5eadfd4e1ef0b1822da7186
# Parent ff1b80ccecd9ed049cc694ab117100e83eab179f
x86 hvm ept: Remove EPT guest linear address validation
For EPT violation resulting from an attempt to load the guest PDPTEs
as part of the execution of the MOV CR instruction, the EPT_GLA_VALID
is not valid. This situation should not happen in most situation,
since we always populate guest memory. But this is not ture for PAE
guest under the PoD/Page sharing situation. In that situation, a page
pointed by CR3 may be un-populated, and we need handle it in such
situation.
Signed-off-by: Jiang, Yunhong <yunhong.jiang@intel.com>
--- a/xen/arch/x86/hvm/vmx/vmx.c
+++ b/xen/arch/x86/hvm/vmx/vmx.c
@@ -2129,8 +2129,7 @@ static void ept_handle_violation(unsigne
__trace_var(TRC_HVM_NPF, 0, sizeof(_d), (unsigned char *)&_d);
}
- if ( (qualification & EPT_GLA_VALID) &&
- hvm_hap_nested_page_fault(gfn) )
+ if ( hvm_hap_nested_page_fault(gfn) )
return;
/* Everything else is an error. */

121
22632-vtd-print-entries.patch Normal file
View File

@ -0,0 +1,121 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1293185641 0
# Node ID 7cc87dcf30a13bcf600077aa661694caeb0c290c
# Parent dca1b7cf2e2c27cd160bd1d1d284e3f810d4936c
VT-d: fix and improve print_vtd_entries()
Fix leaking of mapped domain pages (root_entry and ctxt_entry when
falling out of the level traversing loop). Do this by re-arranging
things slightly so that a mapping is retained only as long as it
really is needed.
Fix the failure to use map_domain_page() in the level traversing loop
of the function.
Add a mssing return statement in one of the error paths.
Also I wonder whether not being able to call print_vtd_entries() from
iommu_page_fault_do_one() in ix86 is still correct, now that
map_domain_page() is IRQ safe.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/drivers/passthrough/vtd/utils.c
+++ b/xen/drivers/passthrough/vtd/utils.c
@@ -99,7 +99,7 @@ void print_vtd_entries(struct iommu *iom
struct context_entry *ctxt_entry;
struct root_entry *root_entry;
struct dma_pte pte;
- u64 *l;
+ u64 *l, val;
u32 l_index, level;
printk("print_vtd_entries: iommu = %p bdf = %x:%x.%x gmfn = %"PRIx64"\n",
@@ -112,6 +112,11 @@ void print_vtd_entries(struct iommu *iom
}
root_entry = (struct root_entry *)map_vtd_domain_page(iommu->root_maddr);
+ if ( root_entry == NULL )
+ {
+ printk(" root_entry == NULL\n");
+ return;
+ }
printk(" root_entry = %p\n", root_entry);
printk(" root_entry[%x] = %"PRIx64"\n", bus, root_entry[bus].val);
@@ -122,61 +127,57 @@ void print_vtd_entries(struct iommu *iom
return;
}
- ctxt_entry =
- (struct context_entry *)map_vtd_domain_page(root_entry[bus].val);
+ val = root_entry[bus].val;
+ unmap_vtd_domain_page(root_entry);
+ ctxt_entry = map_vtd_domain_page(val);
if ( ctxt_entry == NULL )
{
- unmap_vtd_domain_page(root_entry);
printk(" ctxt_entry == NULL\n");
return;
}
printk(" context = %p\n", ctxt_entry);
+ val = ctxt_entry[devfn].lo;
printk(" context[%x] = %"PRIx64"_%"PRIx64"\n",
- devfn, ctxt_entry[devfn].hi, ctxt_entry[devfn].lo);
+ devfn, ctxt_entry[devfn].hi, val);
if ( !context_present(ctxt_entry[devfn]) )
{
unmap_vtd_domain_page(ctxt_entry);
- unmap_vtd_domain_page(root_entry);
printk(" ctxt_entry[%x] not present\n", devfn);
return;
}
level = agaw_to_level(context_address_width(ctxt_entry[devfn]));
+ unmap_vtd_domain_page(ctxt_entry);
if ( level != VTD_PAGE_TABLE_LEVEL_3 &&
level != VTD_PAGE_TABLE_LEVEL_4)
{
- unmap_vtd_domain_page(ctxt_entry);
- unmap_vtd_domain_page(root_entry);
printk("Unsupported VTD page table level (%d)!\n", level);
+ return;
}
- l = maddr_to_virt(ctxt_entry[devfn].lo);
do
{
- l = (u64*)(((unsigned long)l >> PAGE_SHIFT_4K) << PAGE_SHIFT_4K);
+ l = map_vtd_domain_page(val);
printk(" l%d = %p\n", level, l);
if ( l == NULL )
{
- unmap_vtd_domain_page(ctxt_entry);
- unmap_vtd_domain_page(root_entry);
printk(" l%d == NULL\n", level);
break;
}
l_index = get_level_index(gmfn, level);
printk(" l%d_index = %x\n", level, l_index);
- printk(" l%d[%x] = %"PRIx64"\n", level, l_index, l[l_index]);
- pte.val = l[l_index];
+ pte.val = val = l[l_index];
+ unmap_vtd_domain_page(l);
+ printk(" l%d[%x] = %"PRIx64"\n", level, l_index, val);
+
+ pte.val = val;
if ( !dma_pte_present(pte) )
{
- unmap_vtd_domain_page(ctxt_entry);
- unmap_vtd_domain_page(root_entry);
printk(" l%d[%x] not present\n", level, l_index);
break;
}
-
- l = maddr_to_virt(l[l_index]);
} while ( --level );
}

58
22645-amd-flush-filter.patch Normal file
View File

@ -0,0 +1,58 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1294220923 0
# Node ID 76d897a06b316bf2278220b006d578faf31ce3fb
# Parent fece73d4d30985ce40ef87dcd379ce6beb8aadf3
x86 amd: Revert 6382:b74c15e4dd4f (AMD flush filter configuration)
Flush filter is not reliably supported by any processor, we already
have code to unconditionally disable the filter, so we don't need the
command-line config option. Remove it.
Signed-off-by: Keir Fraser <keir@xen.org>
--- a/xen/arch/x86/cpu/amd.c
+++ b/xen/arch/x86/cpu/amd.c
@@ -237,20 +237,6 @@ int cpu_has_amd_erratum(const struct cpu
return 0;
}
-/*
- * amd_flush_filter={on,off}. Forcibly Enable or disable the TLB flush
- * filter on AMD 64-bit processors.
- */
-static int flush_filter_force;
-static void flush_filter(char *s)
-{
- if (!strcmp(s, "off"))
- flush_filter_force = -1;
- if (!strcmp(s, "on"))
- flush_filter_force = 1;
-}
-custom_param("amd_flush_filter", flush_filter);
-
#define num_physpages 0
/*
@@ -545,21 +531,6 @@ static void __devinit init_amd(struct cp
break;
}
- if (c->x86 == 15) {
- rdmsr(MSR_K7_HWCR, l, h);
- printk(KERN_INFO "CPU%d: AMD Flush Filter %sabled",
- smp_processor_id(), (l & (1<<6)) ? "dis" : "en");
- if ((flush_filter_force > 0) && (l & (1<<6))) {
- l &= ~(1<<6);
- printk(" -> Forcibly enabled");
- } else if ((flush_filter_force < 0) && !(l & (1<<6))) {
- l |= 1<<6;
- printk(" -> Forcibly disabled");
- }
- wrmsr(MSR_K7_HWCR, l, h);
- printk("\n");
- }
-
display_cacheinfo(c);
if (cpuid_eax(0x80000000) >= 0x80000008) {

71
22693-fam10-mmio-conf-base-protect.patch Normal file
View File

@ -0,0 +1,71 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1294648952 0
# Node ID 08bb0eefe87107c30a9bdc6d72f26b20c55ae687
# Parent 66e8062894641e5622aa6adc54297d5700b1cea3
x86-64: don't allow wrmsr to MSR_FAM10H_MMIO_CONF_BASE when Xen itself is using it
Signed-off-by: Jan Beulich <jbeulich@novell.com>
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1294741846 0
# Node ID a2dcebb88bc4571d64b3f30c1fbb33cfe4f6e2c5
# Parent 1ae74f060a39ae1305b3a434d0faa1a19853fdab
x86-64: refine access permission check for wrmsr to MSR_FAM10H_MMIO_CONF_BASE
We really don't want the mmconf window to move/disappear whenever we
use is ourselves, not only when we enabled it.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/arch/x86/traps.c
+++ b/xen/arch/x86/traps.c
@@ -1704,6 +1704,10 @@ static int is_cpufreq_controller(struct
(d->domain_id == 0));
}
+#ifdef CONFIG_X86_64
+#include "x86_64/mmconfig.h"
+#endif
+
static int emulate_privileged_op(struct cpu_user_regs *regs)
{
struct vcpu *v = current;
@@ -2263,7 +2267,13 @@ static int emulate_privileged_op(struct
goto fail;
if ( !IS_PRIV(v->domain) )
break;
- if ( (rdmsr_safe(MSR_FAM10H_MMIO_CONF_BASE, l, h) != 0) ||
+ if ( (rdmsr_safe(MSR_FAM10H_MMIO_CONF_BASE, l, h) != 0) )
+ goto fail;
+ if (
+#ifdef CONFIG_X86_64
+ (pci_probe & PCI_PROBE_MASK) == PCI_PROBE_MMCONF ?
+ eax != l || edx != h :
+#endif
(((((u64)h << 32) | l) ^ val) &
~( FAM10H_MMIO_CONF_ENABLE |
(FAM10H_MMIO_CONF_BUSRANGE_MASK <<
--- a/xen/arch/x86/x86_64/mmconfig.h
+++ b/xen/arch/x86/x86_64/mmconfig.h
@@ -34,6 +34,8 @@
#define PCI_VENDOR_ID_NVIDIA 0x10de
+extern unsigned int pci_probe;
+
/*
* AMD Fam10h CPUs are buggy, and cannot access MMIO config space
* on their northbrige except through the * %eax register. As such, you MUST
--- a/xen/arch/x86/x86_64/mmconfig-shared.c
+++ b/xen/arch/x86/x86_64/mmconfig-shared.c
@@ -24,7 +24,7 @@
#include "mmconfig.h"
static int __initdata known_bridge;
-static unsigned int pci_probe = PCI_PROBE_CONF1 | PCI_PROBE_MMCONF;
+unsigned int pci_probe = PCI_PROBE_CONF1 | PCI_PROBE_MMCONF;
static void __init parse_mmcfg(char *s)
{

44
22694-x86_64-no-weak.patch Normal file
View File

@ -0,0 +1,44 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1294649119 0
# Node ID 9b5d121c8805b40a4338248c346303e1e18d0c4e
# Parent 08bb0eefe87107c30a9bdc6d72f26b20c55ae687
x86_64: don't use weak symbols on x86-64
References: bnc#656369, bnc#658704
Various gcc versions inline functions that are both weak and hidden,
without even giving a warning.
Certainly the risk exists that we'll see the problem again when
another weak function gets introduced, but I don't see a way to
protect us from that.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Just remove the weak attribute altogether. It's the only one in
non-ia64-specific code. We can get teh same effect with ifdefs which
although a bit unsightly is better than using compiler/linker features
we cannot trust.
Signed-off-by: Keir Fraser <keir@xen.org>
--- a/xen/drivers/acpi/numa.c
+++ b/xen/drivers/acpi/numa.c
@@ -120,14 +120,15 @@ static int __init acpi_parse_slit(struct
return 0;
}
-void __init __attribute__ ((weak))
+#ifndef CONFIG_X86
+void __init
acpi_numa_x2apic_affinity_init(struct acpi_srat_x2apic_cpu_affinity *pa)
{
printk(KERN_WARNING PREFIX
"Found unsupported x2apic [0x%08x] SRAT entry\n", pa->apic_id);
return;
}
-
+#endif
static int __init
acpi_parse_x2apic_affinity(struct acpi_subtable_header *header,

101
22707-x2apic-preenabled-check.patch Normal file
View File

@ -0,0 +1,101 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1294746050 0
# Node ID 2ff199e2842b7e4f08ea99558afc32536a77280c
# Parent ca10302ac2859b43a41afe425d79ae0df29f2a9c
x86: restore x2apic pre-enabled check logic
References: bnc#656369, bnc#658704
c/s 22475 removed the early checking without replacement, neglecting
the fact that x2apic_enabled must be set early for APIC register
accesses done during second stage ACPI table parsing (rooted at
acpi_boot_init()) to work correctly. Without this, particularly
determination of the boot CPU won't work, resulting in an attempt to
bring up that CPU again as a secondary one (which fails).
Restore the functionality, now calling it from generic_apic_probe().
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/xen/arch/x86/apic.c
+++ b/xen/arch/x86/apic.c
@@ -947,7 +947,6 @@ no_apic:
void x2apic_setup(void)
{
struct IO_APIC_route_entry **ioapic_entries = NULL;
- u32 lo, hi;
if ( smp_processor_id() != 0 )
{
@@ -959,14 +958,6 @@ void x2apic_setup(void)
if ( !cpu_has_x2apic )
return;
- /* Check whether x2apic mode was already enabled by the BIOS. */
- rdmsr(MSR_IA32_APICBASE, lo, hi);
- if ( lo & MSR_IA32_APICBASE_EXTD )
- {
- printk("x2APIC mode is already enabled by BIOS.\n");
- x2apic_enabled = 1;
- }
-
if ( !opt_x2apic )
{
if ( !x2apic_enabled )
--- a/xen/arch/x86/genapic/probe.c
+++ b/xen/arch/x86/genapic/probe.c
@@ -59,8 +59,10 @@ custom_param("apic", genapic_apic_force)
void __init generic_apic_probe(void)
{
- int i;
- int changed = cmdline_apic = (genapic != NULL);
+ int i, changed;
+
+ check_x2apic_preenabled();
+ cmdline_apic = changed = (genapic != NULL);
for (i = 0; !changed && apic_probe[i]; i++) {
if (apic_probe[i]->probe()) {
--- a/xen/arch/x86/genapic/x2apic.c
+++ b/xen/arch/x86/genapic/x2apic.c
@@ -20,6 +20,8 @@
#include <xen/cpumask.h>
#include <asm/apicdef.h>
#include <asm/genapic.h>
+#include <asm/msr.h>
+#include <asm/processor.h>
#include <xen/smp.h>
#include <asm/mach-default/mach_mpparse.h>
@@ -134,3 +136,20 @@ const struct genapic *apic_x2apic_probe(
{
return x2apic_phys ? &apic_x2apic_phys : &apic_x2apic_cluster;
}
+
+void __init check_x2apic_preenabled(void)
+{
+ u32 lo, hi;
+
+ if ( !cpu_has_x2apic )
+ return;
+
+ /* Check whether x2apic mode was already enabled by the BIOS. */
+ rdmsr(MSR_IA32_APICBASE, lo, hi);
+ if ( lo & MSR_IA32_APICBASE_EXTD )
+ {
+ printk("x2APIC mode is already enabled by BIOS.\n");
+ x2apic_enabled = 1;
+ genapic = apic_x2apic_probe();
+ }
+}
--- a/xen/include/asm-x86/apic.h
+++ b/xen/include/asm-x86/apic.h
@@ -25,6 +25,7 @@ extern int apic_verbosity;
extern int x2apic_enabled;
extern int directed_eoi_enabled;
+void check_x2apic_preenabled(void);
void x2apic_setup(void);
const struct genapic *apic_x2apic_probe(void);

112
22708-xenctx-misc.patch Normal file
View File

@ -0,0 +1,112 @@
# HG changeset patch
# User Keir Fraser <keir@xen.org>
# Date 1294746099 0
# Node ID 7926538a633297d65a5d6324bf9bd0eb158a6aff
# Parent 2ff199e2842b7e4f08ea99558afc32536a77280c
xenctx: misc adjustments
- fix off-by-one errors during symbol insertion and lookup
- don't store the symbol type, as it wasn't needed at all so far and
is only needed now at parsing time
- don't insert certain kinds of symbols
Signed-off-by: Jan Beulich <jbeulich@novell.com>
--- a/tools/xentrace/xenctx.c
+++ b/tools/xentrace/xenctx.c
@@ -19,6 +19,7 @@
#include <unistd.h>
#include <errno.h>
#include <signal.h>
+#include <ctype.h>
#include <string.h>
#include <inttypes.h>
#include <getopt.h>
@@ -58,7 +59,6 @@ int disp_tlb;
struct symbol {
guest_word_t address;
- char type;
char *name;
struct symbol *next;
} *symbol_table = NULL;
@@ -112,12 +112,12 @@ static void insert_symbol(struct symbol
/* The System.map is usually already sorted... */
if (prev
- && prev->address < symbol->address
+ && prev->address <= symbol->address
&& (!prev->next || prev->next->address > symbol->address)) {
s = prev;
} else {
/* ... otherwise do crappy/slow search for the correct place */
- while(s && s->next && s->next->address < symbol->address)
+ while (s->next && s->next->address <= symbol->address)
s = s->next;
}
@@ -130,13 +130,13 @@ static struct symbol *lookup_symbol(gues
{
struct symbol *s = symbol_table;
- while(s && s->next && s->next->address < address)
- s = s->next;
+ if (!s)
+ return NULL;
- if (s && s->address < address)
- return s;
+ while (s->next && s->next->address < address)
+ s = s->next;
- return NULL;
+ return s->next && s->next->address <= address ? s->next : s;
}
static void print_symbol(guest_word_t addr)
@@ -159,7 +159,7 @@ static void print_symbol(guest_word_t ad
static void read_symbol_table(const char *symtab)
{
- char line[256];
+ char type, line[256];
char *p;
struct symbol *symbol;
FILE *f;
@@ -178,9 +178,13 @@ static void read_symbol_table(const char
/* need more checks for syntax here... */
symbol->address = strtoull(line, &p, 16);
- p++;
- symbol->type = *p++;
- p++;
+ if (!isspace(*p++))
+ continue;
+ type = *p++;
+ if (!isalpha(type) && type != '?')
+ continue;
+ if (!isspace(*p++))
+ continue;
/* in the future we should handle the module name
* being appended here, this would allow us to use
@@ -190,7 +194,18 @@ static void read_symbol_table(const char
p[strlen(p)-1] = '\0';
symbol->name = strdup(p);
- insert_symbol(symbol);
+ switch (type) {
+ case 'A': /* global absolute */
+ case 'a': /* local absolute */
+ break;
+ case 'U': /* undefined */
+ case 'v': /* undefined weak object */
+ case 'w': /* undefined weak function */
+ continue;
+ default:
+ insert_symbol(symbol);
+ break;
+ }
if (strcmp(symbol->name, "_stext") == 0)
kernel_stext = symbol->address;

2
32on64-extra-mem.patch
View File

@ -2,7 +2,7 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/XendDomainInfo.py
+++ xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
@@ -2920,7 +2920,7 @@ class XendDomainInfo:
@@ -2919,7 +2919,7 @@ class XendDomainInfo:
self.guest_bitsize = self.image.getBitSize()
# Make sure there's enough RAM available for the domain

50
7426-xenfb-depth.patch Normal file
View File

@ -0,0 +1,50 @@
Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xenfb.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xenfb.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xenfb.c
@@ -622,6 +622,18 @@ static void xenfb_guest_copy(struct XenF
oops = 1;
}
break;
+ case 16:
+ if (bpp == 16) {
+ for (line = y; line < (y+h); line++) {
+ memcpy (data + (line * linesize) + (x * bpp / 8), xenfb->pixels + xenfb->offset
+ + (line * xenfb->row_stride) + (x * xenfb->depth / 8), w * xenfb->depth / 8);
+ }
+ } else if (bpp == 32) {
+ BLT(uint16_t, uint32_t, 5, 6, 5, 8, 8, 8);
+ } else {
+ oops = 1;
+ }
+ break;
case 24:
if (bpp == 16) {
BLT(uint32_t, uint16_t, 8, 8, 8, 5, 6, 5);
@@ -631,6 +643,18 @@ static void xenfb_guest_copy(struct XenF
oops = 1;
}
break;
+ case 32:
+ if (bpp == 16) {
+ BLT(uint32_t, uint16_t, 8, 8, 8, 5, 6, 5);
+ } else if (bpp == 32) {
+ for (line = y; line < (y+h); line++) {
+ memcpy (data + (line * linesize) + (x * bpp / 8), xenfb->pixels + xenfb->offset
+ + (line * xenfb->row_stride) + (x * xenfb->depth / 8), w * xenfb->depth / 8);
+ }
+ } else {
+ oops = 1;
+ }
+ break;
default:
oops = 1;
}
@@ -784,6 +808,7 @@ static void xenfb_update(void *opaque)
static void xenfb_invalidate(void *opaque)
{
struct XenFB *xenfb = opaque;
+ xenfb->do_resize = 1;
xenfb->up_fullscreen = 1;
}

46
7433-qemu-altgr.patch Normal file
View File

@ -0,0 +1,46 @@
--- a/tools/ioemu-qemu-xen/keymaps.c Wed Jan 05 23:16:54 2011 +0000
+++ b/tools/ioemu-qemu-xen/keymaps.c Wed Jan 05 23:48:36 2011 +0000
@@ -53,6 +53,20 @@ typedef struct {
struct key_range *localstate_range;
struct key_range *altgr_range;
} kbd_layout_t;
+
+static void del_key_range(struct key_range **krp, int code) {
+ struct key_range *kr;
+ struct key_range *kr_pr;
+ for (kr = *krp; kr; kr_pr = kr, kr = kr->next) {
+ if (code >= kr->start && code <= kr->end) {
+ if (kr == *krp)
+ *krp = kr->next;
+ else
+ kr_pr->next = kr->next;
+ qemu_free(kr);
+ }
+ }
+}
static void add_to_key_range(struct key_range **krp, int code) {
struct key_range *kr;
@@ -137,6 +151,8 @@ static kbd_layout_t *parse_keyboard_layo
if (rest && strstr(rest, "altgr")) {
add_to_key_range(&k->altgr_range, keysym);
//fprintf(stderr, "altgr keysym %04x keycode %d\n", keysym, keycode);
+ } else {
+ del_key_range(&k->altgr_range, keysym);
}
/* if(keycode&0x80)
--- a/tools/ioemu-qemu-xen/vnc.c Wed Jan 05 23:16:54 2011 +0000
+++ b/tools/ioemu-qemu-xen/vnc.c Wed Jan 05 23:48:36 2011 +0000
@@ -1279,11 +1279,9 @@ static void press_key_altgr_down(VncStat
kbd_put_keycode(0xe0);
if (down){
kbd_put_keycode(0xb8 & 0x7f);
- vs->modifiers_state[0xb8] = 1;
}
else {
kbd_put_keycode(0xb8 | 0x80);
- vs->modifiers_state[0xb8] = 0;
}
}

34
7434-qemu-rlimit-as.patch Normal file
View File

@ -0,0 +1,34 @@
# HG changeset patch
# User Jan Beulich <jbeulich@novell.com>
# Date 1294771190 0
# Node ID d224cef9d82c5c1ba9a6d73ca2628f288ce6f434
# Parent a283996796c91dd29ecff444b78798e0ce902047
bump RLIMIT_AS if restricted
References: bnc#641419
Mostly the same as is already being done for several other RLIMIT_*
values.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Acked-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
committer: Ian Jackson <Ian.Jackson@eu.citrix.com>
git-commit-id: 2aa36d470e97f4baa219f78df82e2d3fe3d9f96d
--- a/tools/ioemu-qemu-xen/vl.c
+++ b/tools/ioemu-qemu-xen/vl.c
@@ -4845,6 +4845,13 @@ int main(int argc, char **argv, char **e
rl.rlim_cur = rl.rlim_max;
if (setrlimit(RLIMIT_DATA, &rl) != 0)
perror("setrlimit(RLIMIT_DATA)");
+ if (getrlimit(RLIMIT_AS, &rl) == 0) {
+ rl.rlim_cur = rl.rlim_max;
+ if (setrlimit(RLIMIT_AS, &rl) != 0)
+ perror("setrlimit(RLIMIT_AS)");
+ } else {
+ perror("getrlimit(RLIMIT_AS)");
+ }
rl.rlim_cur = RLIM_INFINITY;
rl.rlim_max = RLIM_INFINITY;
if (setrlimit(RLIMIT_RSS, &rl) != 0)

21
altgr_2.patch
View File

@ -29,22 +29,11 @@ keycodes up and down events and make sure the intentionally added altgr keycode
Signed-off by Chunyan Liu (cyliu@novell.com)
diff -r a108300bd904 tools/ioemu-qemu-xen/vnc.c
--- a/tools/ioemu-qemu-xen/vnc.c Mon Sep 27 21:20:36 2010 +0800
+++ b/tools/ioemu-qemu-xen/vnc.c Wed Sep 29 01:55:55 2010 +0800
@@ -1279,11 +1279,9 @@
kbd_put_keycode(0xe0);
if (down){
kbd_put_keycode(0xb8 & 0x7f);
- vs->modifiers_state[0xb8] = 1;
}
else {
kbd_put_keycode(0xb8 | 0x80);
- vs->modifiers_state[0xb8] = 0;
}
}
@@ -1310,6 +1308,9 @@
Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/vnc.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/vnc.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/vnc.c
@@ -1308,6 +1308,9 @@ static void do_key_event(VncState *vs, i
shift_keys = vs->modifiers_state[0x2a] | vs->modifiers_state[0x36];
altgr_keys = vs->modifiers_state[0xb8];

4
bdrv_open2_flags_2.patch
View File

@ -2,7 +2,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xen_blktap.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
@@ -235,6 +235,7 @@ static int open_disk(struct td_state *s,
@@ -237,6 +237,7 @@ static int open_disk(struct td_state *s,
BlockDriver* drv;
char* devname;
static int devnumber = 0;
@ -10,7 +10,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
int i;
DPRINTF("Opening %s as blktap%d\n", path, devnumber);
@@ -257,7 +258,7 @@ static int open_disk(struct td_state *s,
@@ -259,7 +260,7 @@ static int open_disk(struct td_state *s,
DPRINTF("%s driver specified\n", drv ? drv->format_name : "No");
/* Open the image */

4
blktap-pv-cdrom.patch
View File

@ -741,7 +741,7 @@ Index: xen-4.0.1-testing/tools/blktap/lib/blktaplib.h
===================================================================
--- xen-4.0.1-testing.orig/tools/blktap/lib/blktaplib.h
+++ xen-4.0.1-testing/tools/blktap/lib/blktaplib.h
@@ -224,6 +224,7 @@ typedef struct msg_pid {
@@ -225,6 +225,7 @@ typedef struct msg_pid {
#define DISK_TYPE_RAM 3
#define DISK_TYPE_QCOW 4
#define DISK_TYPE_QCOW2 5
@ -795,7 +795,7 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/server/BlktapController.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/server/BlktapController.py
+++ xen-4.0.1-testing/tools/python/xen/xend/server/BlktapController.py
@@ -20,6 +20,7 @@ blktap1_disk_types = [
@@ -21,6 +21,7 @@ blktap1_disk_types = [
'ram',
'qcow',
'qcow2',

2
capslock_enable.patch
View File

@ -2,7 +2,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/vnc.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/vnc.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/vnc.c
@@ -1344,6 +1344,11 @@ static void do_key_event(VncState *vs, i
@@ -1342,6 +1342,11 @@ static void do_key_event(VncState *vs, i
}
break;
case 0x3a: /* CapsLock */

2
change_home_server.patch
View File

@ -2,7 +2,7 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/XendDomainInfo.py
+++ xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
@@ -3136,6 +3136,11 @@ class XendDomainInfo:
@@ -3135,6 +3135,11 @@ class XendDomainInfo:
self._cleanup_phantom_devs(paths)
self._cleanupVm()

8
cpupools-core-fixup.patch
View File

@ -62,7 +62,7 @@ Index: xen-4.0.1-testing/xen/common/softirq.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/softirq.c
+++ xen-4.0.1-testing/xen/common/softirq.c
@@ -104,12 +104,15 @@ static void tasklet_schedule_list(struct
@@ -114,12 +114,15 @@ static void tasklet_schedule_list(struct
{
BUG_ON(!list_empty(&t->list));
list_add_tail(&t->list, tlist);
@ -79,7 +79,7 @@ Index: xen-4.0.1-testing/xen/common/softirq.c
}
spin_unlock_irqrestore(&tasklet_lock, flags);
@@ -120,7 +123,7 @@ void tasklet_schedule(struct tasklet *t)
@@ -130,7 +133,7 @@ void tasklet_schedule(struct tasklet *t)
tasklet_schedule_list(t, &tasklet_list, smp_processor_id());
}
@ -88,7 +88,7 @@ Index: xen-4.0.1-testing/xen/common/softirq.c
{
tasklet_schedule_list(t, &per_cpu(tasklet_list_pcpu, cpu), cpu);
}
@@ -156,7 +159,15 @@ static void tasklet_action(void)
@@ -166,7 +169,15 @@ static void tasklet_action(void)
if ( t->is_scheduled )
{
BUG_ON(t->is_dead || !list_empty(&t->list));
@ -122,7 +122,7 @@ Index: xen-4.0.1-testing/xen/include/xen/softirq.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/xen/softirq.h
+++ xen-4.0.1-testing/xen/include/xen/softirq.h
@@ -50,15 +50,17 @@ struct tasklet
@@ -52,15 +52,17 @@ struct tasklet
bool_t is_scheduled;
bool_t is_running;
bool_t is_dead;

212
cpupools-core.patch
View File

@ -1,7 +1,9 @@
From: Juergen Gross <juergen.gross@ts.fujitsu.com>
--- a/xen/arch/x86/acpi/power.c
+++ b/xen/arch/x86/acpi/power.c
Index: xen-4.0.1-testing/xen/arch/x86/acpi/power.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/acpi/power.c
+++ xen-4.0.1-testing/xen/arch/x86/acpi/power.c
@@ -234,7 +234,7 @@ static int enter_state(u32 state)
return error;
}
@ -20,8 +22,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
}
static int acpi_get_wake_status(void)
--- a/xen/arch/x86/domain.c
+++ b/xen/arch/x86/domain.c
Index: xen-4.0.1-testing/xen/arch/x86/domain.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/domain.c
+++ xen-4.0.1-testing/xen/arch/x86/domain.c
@@ -1522,42 +1522,52 @@ void sync_vcpu_execstate(struct vcpu *v)
}
@ -135,8 +139,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
return 0;
}
--- a/xen/arch/x86/domain_build.c
+++ b/xen/arch/x86/domain_build.c
Index: xen-4.0.1-testing/xen/arch/x86/domain_build.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/domain_build.c
+++ xen-4.0.1-testing/xen/arch/x86/domain_build.c
@@ -9,6 +9,7 @@
#include <xen/lib.h>
#include <xen/ctype.h>
@ -154,7 +160,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
if ( opt_dom0_max_vcpus > MAX_VIRT_CPUS )
opt_dom0_max_vcpus = MAX_VIRT_CPUS;
@@ -287,7 +288,7 @@ int __init construct_dom0(
@@ -296,7 +297,7 @@ int __init construct_dom0(
unsigned long _initrd_start, unsigned long initrd_len,
char *cmdline)
{
@ -163,7 +169,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
struct cpu_user_regs *regs;
unsigned long pfn, mfn;
unsigned long nr_pages;
@@ -786,8 +787,12 @@ int __init construct_dom0(
@@ -795,8 +796,12 @@ int __init construct_dom0(
printk("Dom0 has maximum %u VCPUs\n", opt_dom0_max_vcpus);
@ -177,8 +183,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
/* Set up CR3 value for write_ptbase */
if ( paging_mode_enabled(d) )
--- a/xen/arch/x86/microcode.c
+++ b/xen/arch/x86/microcode.c
Index: xen-4.0.1-testing/xen/arch/x86/microcode.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/microcode.c
+++ xen-4.0.1-testing/xen/arch/x86/microcode.c
@@ -114,7 +114,7 @@ static int microcode_update_cpu(const vo
return err;
}
@ -206,8 +214,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
+ return continue_hypercall_on_cpu(info->cpu, NULL,
+ do_microcode_update, info);
}
--- a/xen/arch/x86/mm.c
+++ b/xen/arch/x86/mm.c
Index: xen-4.0.1-testing/xen/arch/x86/mm.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/mm.c
+++ xen-4.0.1-testing/xen/arch/x86/mm.c
@@ -243,7 +243,7 @@ void __init arch_init_memory(void)
* Any Xen-heap pages that we will allow to be mapped will have
* their domain field set to dom_xen.
@ -234,8 +244,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
BUG_ON(dom_cow == NULL);
/* First 1MB of RAM is historically marked as I/O. */
--- a/xen/arch/x86/platform_hypercall.c
+++ b/xen/arch/x86/platform_hypercall.c
Index: xen-4.0.1-testing/xen/arch/x86/platform_hypercall.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/platform_hypercall.c
+++ xen-4.0.1-testing/xen/arch/x86/platform_hypercall.c
@@ -48,12 +48,12 @@ static DEFINE_PER_CPU(uint64_t, freq);
extern int set_px_pminfo(uint32_t cpu, struct xen_processor_performance *perf);
extern long set_cx_pminfo(uint32_t cpu, struct xen_processor_power *power);
@ -251,7 +263,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
{
int cpu = (unsigned long)data;
return cpu_down(cpu);
@@ -314,7 +314,7 @@ ret_t do_platform_op(XEN_GUEST_HANDLE(xe
@@ -317,7 +317,7 @@ ret_t do_platform_op(XEN_GUEST_HANDLE(xe
if ( op->u.change_freq.flags || !cpu_online(op->u.change_freq.cpu) )
break;
per_cpu(freq, op->u.change_freq.cpu) = op->u.change_freq.freq;
@ -260,7 +272,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
cpu_frequency_change_helper,
NULL);
break;
@@ -470,7 +470,7 @@ ret_t do_platform_op(XEN_GUEST_HANDLE(xe
@@ -473,7 +473,7 @@ ret_t do_platform_op(XEN_GUEST_HANDLE(xe
break;
}
ret = continue_hypercall_on_cpu(
@ -269,8 +281,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
break;
}
break;
--- a/xen/arch/x86/setup.c
+++ b/xen/arch/x86/setup.c
Index: xen-4.0.1-testing/xen/arch/x86/setup.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/setup.c
+++ xen-4.0.1-testing/xen/arch/x86/setup.c
@@ -2,6 +2,7 @@
#include <xen/init.h>
#include <xen/lib.h>
@ -288,7 +302,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
if ( idle_domain == NULL )
BUG();
idle_domain->vcpu = idle_vcpu;
@@ -1122,8 +1123,13 @@ void __init __start_xen(unsigned long mb
@@ -1121,8 +1122,13 @@ void __init __start_xen(unsigned long mb
if ( !tboot_protect_mem_regions() )
panic("Could not protect TXT memory regions\n");
@ -303,8 +317,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
if ( (dom0 == NULL) || (alloc_dom0_vcpu0() == NULL) )
panic("Error creating domain 0\n");
--- a/xen/arch/x86/smpboot.c
+++ b/xen/arch/x86/smpboot.c
Index: xen-4.0.1-testing/xen/arch/x86/smpboot.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/smpboot.c
+++ xen-4.0.1-testing/xen/arch/x86/smpboot.c
@@ -39,6 +39,7 @@
#include <xen/mm.h>
#include <xen/domain.h>
@ -313,7 +329,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
#include <xen/irq.h>
#include <xen/delay.h>
#include <xen/softirq.h>
@@ -1306,10 +1307,11 @@ int __cpu_disable(void)
@@ -1305,10 +1306,11 @@ int __cpu_disable(void)
__sync_lazy_execstate();
/* It's now safe to remove this processor from the online map */
@ -326,7 +342,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
return 0;
}
@@ -1341,16 +1343,12 @@ static int take_cpu_down(void *unused)
@@ -1340,16 +1342,12 @@ static int take_cpu_down(void *unused)
int cpu_down(unsigned int cpu)
{
int err = 0;
@ -344,7 +360,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
/* Can not offline BSP */
if (cpu == 0) {
err = -EINVAL;
@@ -1364,6 +1362,11 @@ int cpu_down(unsigned int cpu)
@@ -1363,6 +1361,11 @@ int cpu_down(unsigned int cpu)
printk("Prepare to bring CPU%d down...\n", cpu);
@ -356,7 +372,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
cpufreq_del_cpu(cpu);
err = stop_machine_run(take_cpu_down, NULL, cpu);
@@ -1379,6 +1382,8 @@ int cpu_down(unsigned int cpu)
@@ -1378,6 +1381,8 @@ int cpu_down(unsigned int cpu)
out:
if (!err)
send_guest_global_virq(dom0, VIRQ_PCPU_STATE);
@ -365,7 +381,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
spin_unlock(&cpu_add_remove_lock);
return err;
}
@@ -1568,6 +1573,7 @@ int __devinit __cpu_up(unsigned int cpu)
@@ -1567,6 +1572,7 @@ int __devinit __cpu_up(unsigned int cpu)
process_pending_softirqs();
}
@ -373,8 +389,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
cpufreq_add_cpu(cpu);
return 0;
}
--- a/xen/arch/x86/sysctl.c
+++ b/xen/arch/x86/sysctl.c
Index: xen-4.0.1-testing/xen/arch/x86/sysctl.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/sysctl.c
+++ xen-4.0.1-testing/xen/arch/x86/sysctl.c
@@ -29,7 +29,7 @@
#define get_xen_guest_handle(val, hnd) do { val = (hnd).p; } while (0)
@ -393,8 +411,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
break;
case XEN_SYSCTL_CPU_HOTPLUG_STATUS:
ret = 0;
--- a/xen/common/Makefile
+++ b/xen/common/Makefile
Index: xen-4.0.1-testing/xen/common/Makefile
===================================================================
--- xen-4.0.1-testing.orig/xen/common/Makefile
+++ xen-4.0.1-testing/xen/common/Makefile
@@ -1,5 +1,6 @@
obj-y += bitmap.o
obj-y += cpu.o
@ -402,8 +422,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
obj-y += domctl.o
obj-y += domain.o
obj-y += event_channel.o
Index: xen-4.0.1-testing/xen/common/cpupool.c
===================================================================
--- /dev/null
+++ b/xen/common/cpupool.c
+++ xen-4.0.1-testing/xen/common/cpupool.c
@@ -0,0 +1,609 @@
+/******************************************************************************
+ * cpupool.c
@ -1014,8 +1036,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
+ * indent-tabs-mode: nil
+ * End:
+ */
--- a/xen/common/domain.c
+++ b/xen/common/domain.c
Index: xen-4.0.1-testing/xen/common/domain.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/domain.c
+++ xen-4.0.1-testing/xen/common/domain.c
@@ -209,7 +209,7 @@ static void __init parse_extra_guest_irq
custom_param("extra_guest_irqs", parse_extra_guest_irqs);
@ -1044,8 +1068,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
sched_destroy_domain(d);
/* Free page used by xen oprofile buffer. */
--- a/xen/common/domctl.c
+++ b/xen/common/domctl.c
Index: xen-4.0.1-testing/xen/common/domctl.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/domctl.c
+++ xen-4.0.1-testing/xen/common/domctl.c
@@ -11,6 +11,7 @@
#include <xen/lib.h>
#include <xen/mm.h>
@ -1121,8 +1147,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
if ( alloc_vcpu(d, i, cpu) == NULL )
goto maxvcpu_out;
--- a/xen/common/kexec.c
+++ b/xen/common/kexec.c
Index: xen-4.0.1-testing/xen/common/kexec.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/kexec.c
+++ xen-4.0.1-testing/xen/common/kexec.c
@@ -235,7 +235,7 @@ void kexec_crash(void)
BUG();
}
@ -1141,8 +1169,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
break;
case KEXEC_TYPE_CRASH:
kexec_crash(); /* Does not return */
--- a/xen/common/sched_credit.c
+++ b/xen/common/sched_credit.c
Index: xen-4.0.1-testing/xen/common/sched_credit.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/sched_credit.c
+++ xen-4.0.1-testing/xen/common/sched_credit.c
@@ -70,11 +70,15 @@
/*
* Useful macros
@ -2031,7 +2061,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
.destroy_vcpu = csched_vcpu_destroy,
.sleep = csched_vcpu_sleep,
@@ -1411,6 +1540,13 @@ const struct scheduler sched_credit_def
@@ -1411,6 +1540,13 @@ const struct scheduler sched_credit_def
.dump_cpu_state = csched_dump_pcpu,
.dump_settings = csched_dump,
.init = csched_init,
@ -2045,8 +2075,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
.tick_suspend = csched_tick_suspend,
.tick_resume = csched_tick_resume,
--- a/xen/common/sched_sedf.c
+++ b/xen/common/sched_sedf.c
Index: xen-4.0.1-testing/xen/common/sched_sedf.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/sched_sedf.c
+++ xen-4.0.1-testing/xen/common/sched_sedf.c
@@ -21,6 +21,9 @@
printk(_a ); \
} while ( 0 )
@ -2057,7 +2089,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
#ifndef NDEBUG
#define SEDF_STATS
#define CHECK(_p) \
@@ -132,7 +135,7 @@ struct sedf_cpu_info {
@@ -131,7 +134,7 @@ struct sedf_cpu_info {
#define sedf_runnable(edom) (!(EDOM_INFO(edom)->status & SEDF_ASLEEP))
@ -2066,7 +2098,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
static inline int extraq_on(struct vcpu *d, int i)
{
@@ -329,30 +332,17 @@ static inline void __add_to_runqueue_sor
@@ -328,30 +331,17 @@ static inline void __add_to_runqueue_sor
}
@ -2103,7 +2135,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
/* Every VCPU gets an equal share of extratime by default. */
inf->deadl_abs = 0;
inf->latency = 0;
@@ -383,39 +373,88 @@ static int sedf_init_vcpu(struct vcpu *v
@@ -382,39 +372,88 @@ static int sedf_init_vcpu(struct vcpu *v
}
else
{
@ -2205,7 +2237,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
return first_cpu(online_affinity);
}
@@ -751,7 +790,7 @@ static struct task_slice sedf_do_extra_s
@@ -750,7 +789,7 @@ static struct task_slice sedf_do_extra_s
-timeslice for the current period used up
-domain on waitqueue has started it's period
-and various others ;) in general: determine which domain to run next*/
@ -2214,7 +2246,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
{
int cpu = smp_processor_id();
struct list_head *runq = RUNQ(cpu);
@@ -786,6 +825,13 @@ static struct task_slice sedf_do_schedul
@@ -785,6 +824,13 @@ static struct task_slice sedf_do_schedul
}
check_waitq:
update_queues(now, runq, waitq);
@ -2228,7 +2260,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
/*now simply pick the first domain from the runqueue, which has the
earliest deadline, because the list is sorted*/
@@ -848,7 +894,7 @@ static struct task_slice sedf_do_schedul
@@ -847,7 +893,7 @@ static struct task_slice sedf_do_schedul
}
@ -2237,7 +2269,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
{
PRINT(2,"sedf_sleep was called, domain-id %i.%i\n",
d->domain->domain_id, d->vcpu_id);
@@ -1067,7 +1113,7 @@ static inline int should_switch(struct v
@@ -1066,7 +1112,7 @@ static inline int should_switch(struct v
return 1;
}
@ -2246,7 +2278,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
{
s_time_t now = NOW();
struct sedf_vcpu_info* inf = EDOM_INFO(d);
@@ -1220,8 +1266,8 @@ static void sedf_dump_domain(struct vcpu
@@ -1219,8 +1265,8 @@ static void sedf_dump_domain(struct vcpu
}
@ -2257,7 +2289,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
{
struct list_head *list, *queue, *tmp;
struct sedf_vcpu_info *d_inf;
@@ -1294,7 +1340,7 @@ static void sedf_dump_cpu_state(int i)
@@ -1293,7 +1339,7 @@ static void sedf_dump_cpu_state(int i)
/* Adjusts periods and slices of the domains accordingly to their weights. */
@ -2266,7 +2298,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
{
struct vcpu *p;
struct domain *d;
@@ -1315,6 +1361,8 @@ static int sedf_adjust_weights(struct xe
@@ -1314,6 +1360,8 @@ static int sedf_adjust_weights(struct xe
rcu_read_lock(&domlist_read_lock);
for_each_domain( d )
{
@ -2275,7 +2307,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
for_each_vcpu( d, p )
{
if ( EDOM_INFO(p)->weight )
@@ -1366,7 +1414,7 @@ static int sedf_adjust_weights(struct xe
@@ -1365,7 +1413,7 @@ static int sedf_adjust_weights(struct xe
/* set or fetch domain scheduling parameters */
@ -2284,7 +2316,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
{
struct vcpu *v;
int rc;
@@ -1376,9 +1424,6 @@ static int sedf_adjust(struct domain *p,
@@ -1375,9 +1423,6 @@ static int sedf_adjust(struct domain *p,
p->domain_id, op->u.sedf.period, op->u.sedf.slice,
op->u.sedf.latency, (op->u.sedf.extratime)?"yes":"no");
@ -2294,7 +2326,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
if ( op->cmd == XEN_DOMCTL_SCHEDOP_putinfo )
{
/* Check for sane parameters. */
@@ -1428,7 +1473,7 @@ static int sedf_adjust(struct domain *p,
@@ -1427,7 +1472,7 @@ static int sedf_adjust(struct domain *p,
}
}
@ -2303,7 +2335,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
if ( rc )
return rc;
@@ -1456,7 +1501,7 @@ static int sedf_adjust(struct domain *p,
@@ -1455,7 +1500,7 @@ static int sedf_adjust(struct domain *p,
return 0;
}
@ -2312,7 +2344,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
.name = "Simple EDF Scheduler",
.opt_name = "sedf",
.sched_id = XEN_SCHEDULER_SEDF,
@@ -1464,9 +1509,15 @@ const struct scheduler sched_sedf_def =
@@ -1463,9 +1508,15 @@ const struct scheduler sched_sedf_def =
.init_domain = sedf_init_domain,
.destroy_domain = sedf_destroy_domain,
@ -2329,8 +2361,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
.do_schedule = sedf_do_schedule,
.pick_cpu = sedf_pick_cpu,
.dump_cpu_state = sedf_dump_cpu_state,
--- a/xen/common/schedule.c
+++ b/xen/common/schedule.c
Index: xen-4.0.1-testing/xen/common/schedule.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/schedule.c
+++ xen-4.0.1-testing/xen/common/schedule.c
@@ -53,10 +53,11 @@ static void poll_timer_fn(void *data);
/* This is global for now so that private implementations can reach it */
@ -2849,9 +2883,11 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
}
#ifdef CONFIG_COMPAT
--- a/xen/common/softirq.c
+++ b/xen/common/softirq.c
@@ -88,9 +88,11 @@ void raise_softirq(unsigned int nr)
Index: xen-4.0.1-testing/xen/common/softirq.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/softirq.c
+++ xen-4.0.1-testing/xen/common/softirq.c
@@ -98,9 +98,11 @@ void raise_softirq(unsigned int nr)
}
static LIST_HEAD(tasklet_list);
@ -2864,7 +2900,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
{
unsigned long flags;
@@ -101,28 +103,44 @@ void tasklet_schedule(struct tasklet *t)
@@ -111,28 +113,44 @@ void tasklet_schedule(struct tasklet *t)
if ( !t->is_scheduled && !t->is_running )
{
BUG_ON(!list_empty(&t->list));
@ -2913,7 +2949,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
list_del_init(&t->list);
BUG_ON(t->is_dead || t->is_running || !t->is_scheduled);
@@ -138,14 +156,15 @@ static void tasklet_action(void)
@@ -148,14 +166,15 @@ static void tasklet_action(void)
if ( t->is_scheduled )
{
BUG_ON(t->is_dead || !list_empty(&t->list));
@ -2931,7 +2967,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
raise_softirq(TASKLET_SOFTIRQ);
spin_unlock_irq(&tasklet_lock);
@@ -186,6 +205,12 @@ void tasklet_init(
@@ -196,6 +215,12 @@ void tasklet_init(
void __init softirq_init(void)
{
@ -2944,8 +2980,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
open_softirq(TASKLET_SOFTIRQ, tasklet_action);
}
--- a/xen/common/sysctl.c
+++ b/xen/common/sysctl.c
Index: xen-4.0.1-testing/xen/common/sysctl.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/sysctl.c
+++ xen-4.0.1-testing/xen/common/sysctl.c
@@ -314,6 +314,14 @@ long do_sysctl(XEN_GUEST_HANDLE(xen_sysc
}
break;
@ -2961,8 +2999,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
default:
ret = arch_do_sysctl(op, u_sysctl);
break;
--- a/xen/include/asm-x86/domain.h
+++ b/xen/include/asm-x86/domain.h
Index: xen-4.0.1-testing/xen/include/asm-x86/domain.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/asm-x86/domain.h
+++ xen-4.0.1-testing/xen/include/asm-x86/domain.h
@@ -458,7 +458,8 @@ struct arch_vcpu
#define hvm_svm hvm_vcpu.u.svm
@ -2973,8 +3013,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
void vcpu_show_execution_state(struct vcpu *);
void vcpu_show_registers(const struct vcpu *);
--- a/xen/include/public/domctl.h
+++ b/xen/include/public/domctl.h
Index: xen-4.0.1-testing/xen/include/public/domctl.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/public/domctl.h
+++ xen-4.0.1-testing/xen/include/public/domctl.h
@@ -60,10 +60,10 @@ struct xen_domctl_createdomain {
/* Should domain memory integrity be verifed by tboot during Sx? */
#define _XEN_DOMCTL_CDF_s3_integrity 2
@ -3003,8 +3045,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
struct xen_domctl {
uint32_t cmd;
#define XEN_DOMCTL_createdomain 1
--- a/xen/include/public/sysctl.h
+++ b/xen/include/public/sysctl.h
Index: xen-4.0.1-testing/xen/include/public/sysctl.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/public/sysctl.h
+++ xen-4.0.1-testing/xen/include/public/sysctl.h
@@ -491,6 +491,28 @@ struct xen_sysctl_lockprof_op {
typedef struct xen_sysctl_lockprof_op xen_sysctl_lockprof_op_t;
DEFINE_XEN_GUEST_HANDLE(xen_sysctl_lockprof_op_t);
@ -3042,8 +3086,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
uint8_t pad[128];
} u;
};
--- a/xen/include/xen/sched-if.h
+++ b/xen/include/xen/sched-if.h
Index: xen-4.0.1-testing/xen/include/xen/sched-if.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/xen/sched-if.h
+++ xen-4.0.1-testing/xen/include/xen/sched-if.h
@@ -10,16 +10,26 @@
#include <xen/percpu.h>
@ -3135,8 +3181,10 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
+struct scheduler *scheduler_get_by_id(unsigned int id);
+
#endif /* __XEN_SCHED_IF_H__ */
--- a/xen/include/xen/sched.h
+++ b/xen/include/xen/sched.h
Index: xen-4.0.1-testing/xen/include/xen/sched.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/xen/sched.h
+++ xen-4.0.1-testing/xen/include/xen/sched.h
@@ -9,6 +9,7 @@
#include <xen/shared.h>
#include <public/xen.h>
@ -3145,7 +3193,7 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
#include <public/vcpu.h>
#include <public/xsm/acm.h>
#include <xen/time.h>
@@ -132,8 +133,6 @@ struct vcpu
@@ -132,8 +133,6 @@ struct vcpu
bool_t defer_shutdown;
/* VCPU is paused following shutdown request (d->is_shutting_down)? */
bool_t paused_for_shutdown;
@ -3218,9 +3266,11 @@ From: Juergen Gross <juergen.gross@ts.fujitsu.com>
#endif /* __SCHED_H__ */
/*
--- a/xen/include/xen/softirq.h
+++ b/xen/include/xen/softirq.h
@@ -58,6 +58,7 @@ struct tasklet
Index: xen-4.0.1-testing/xen/include/xen/softirq.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/xen/softirq.h
+++ xen-4.0.1-testing/xen/include/xen/softirq.h
@@ -60,6 +60,7 @@ struct tasklet
struct tasklet name = { LIST_HEAD_INIT(name.list), 0, 0, 0, func, data }
void tasklet_schedule(struct tasklet *t);

2
del_usb_xend_entry.patch
View File

@ -2,7 +2,7 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/XendDomainInfo.py
+++ xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
@@ -1313,8 +1313,15 @@ class XendDomainInfo:
@@ -1312,8 +1312,15 @@ class XendDomainInfo:
frontpath = self.getDeviceController(deviceClass).frontendPath(dev)
backpath = xstransact.Read(frontpath, "backend")
thread.start_new_thread(self.getDeviceController(deviceClass).finishDeviceCleanup, (backpath, path))

15
domUloader.py
View File

@ -50,6 +50,18 @@ kpartx_args = '-p -part'
# Helper functions
def kpartx_has_opt(opt):
""" Return True if kpartx supports option opt, otherwise False"""
have_opt = True
kpartx_cmd = 'kpartx -' + opt + ' 2>&1'
p = os.popen(kpartx_cmd)
for line in p.readlines():
if line.find('invalid option') >= 0:
have_opt = False
break
p.close()
return have_opt
def error(s):
print >> sys.stderr, "domUloader error: %s" % s
@ -75,6 +87,9 @@ def getWholedisk(part):
# """Determines whether dev is a wholedisk dev"""
# return not domUname[-1:].isdigit()
# If available, add '-f' option (bnc#613584)
if kpartx_has_opt('f'):
kpartx_args += ' -f'
class Wholedisk:
"Class representing a whole disk that may have partitions"

149
hv_apic.patch
View File

@ -1,149 +0,0 @@
Cleanup some APIC handling code in the HyperV shim.
Signed-off-by: K. Y. Srinivasan <ksrinivasan@novell.com>
Index: xen-4.0.1-testing/xen/arch/x86/hvm/hyperv/hv_intercept.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/hvm/hyperv/hv_intercept.c 2010-10-04 14:04:46.000000000 -0600
+++ xen-4.0.1-testing/xen/arch/x86/hvm/hyperv/hv_intercept.c 2010-10-04 18:30:42.000000000 -0600
@@ -252,71 +252,6 @@ hv_get_max_vcpus_supported(void)
static inline void
-hv_read_icr(u64 *icr_content)
-{
- unsigned long icr_low, icr_high;
-
- icr_low = vlapic_mmio_handler.read_handler(current,
- (vlapic_base_address(vcpu_vlapic(current)) + APIC_ICR), 4, &icr_low);
- icr_high = vlapic_mmio_handler.read_handler(current,
- (vlapic_base_address(vcpu_vlapic(current)) + APIC_ICR2), 4, &icr_high);
- *icr_content = (((u64)icr_high<< 32) | icr_low);
-
-}
-
-static inline void
-hv_read_tpr(u64 *tpr_content)
-{
-
- vlapic_mmio_handler.read_handler(current,
- (vlapic_base_address(vcpu_vlapic(current)) + APIC_TASKPRI),
- 4, (unsigned long *)&tpr_content);
-}
-
-static inline void
-hv_write_eoi(u64 msr_content)
-{
- u32 eoi = (u32)msr_content;
-
- vlapic_mmio_handler.write_handler(current,
- (vlapic_base_address(vcpu_vlapic(current)) + APIC_EOI), 4, eoi);
-
-}
-
-static inline void
-hv_write_icr(u64 msr_content)
-{
- u32 icr_low, icr_high;
- icr_low = (u32)msr_content;
- icr_high = (u32)(msr_content >> 32);
-
- if (icr_high != 0)
- {
- vlapic_mmio_handler.write_handler(current,
- (vlapic_base_address(vcpu_vlapic(current)) + APIC_ICR2), 4,
- icr_high);
- }
- if (icr_low != 0)
- {
- vlapic_mmio_handler.write_handler(current,
- (vlapic_base_address(vcpu_vlapic(current)) + APIC_ICR), 4,
- icr_low);
- }
-
-}
-
-static inline void
-hv_write_tpr(u64 msr_content)
-{
- u32 tpr = (u32)msr_content;
-
-
- vlapic_mmio_handler.write_handler(current,
- (vlapic_base_address(vcpu_vlapic(current)) + APIC_TASKPRI), 4, tpr);
-
-}
-
-static inline void
hv_hypercall_page_initialize(void *hypercall_page)
{
char *p;
@@ -810,21 +745,14 @@ hyperv_do_rd_msr(uint32_t idx, struct cp
regs->edx = (u32)(0x0);
break;
case HV_MSR_ICR:
- if (!hv_privilege_check(curp, HV_ACCESS_APIC_MSRS)) {
- goto msr_read_error;
- }
- hv_read_icr(&msr_content);
+ regs->eax = vlapic_get_reg(vcpu_vlapic(current), APIC_ICR);
+ regs->edx = vlapic_get_reg(vcpu_vlapic(current), APIC_ICR2);
#ifdef HV_STATS
cur_vcpu->stats.num_icr_reads++;
#endif
- regs->eax = (u32)(msr_content & 0xFFFFFFFF);
- regs->edx = (u32)(msr_content >> 32);
break;
case HV_MSR_TPR:
- if (!hv_privilege_check(curp, HV_ACCESS_APIC_MSRS)) {
- goto msr_read_error;
- }
- hv_read_tpr(&msr_content);
+ msr_content = vlapic_get_reg(vcpu_vlapic(current), APIC_TASKPRI);
#ifdef HV_STATS
cur_vcpu->stats.num_tpr_reads++;
#endif
@@ -922,28 +850,30 @@ hyperv_do_wr_msr(uint32_t idx, struct cp
goto msr_write_error;
case HV_MSR_EOI:
- if (!hv_privilege_check(curp, HV_ACCESS_APIC_MSRS)) {
- goto msr_write_error;
- }
- hv_write_eoi(msr_content);
+ vlapic_EOI_set(vcpu_vlapic(current));
#ifdef HV_STATS
cur_vcpu->stats.num_eoi_writes++;
#endif
break;
- case HV_MSR_ICR:
- if (!hv_privilege_check(curp, HV_ACCESS_APIC_MSRS)) {
- goto msr_write_error;
- }
- hv_write_icr(msr_content);
+ case HV_MSR_ICR: {
+ u32 eax = (u32)msr_content;
+ u32 edx = (u32)(msr_content >> 32);
+ struct vlapic *vlapic = vcpu_vlapic(current);
+ eax &= ~(1 << 12);
+ edx &= 0xff000000;
+ vlapic_set_reg(vlapic, APIC_ICR2, edx);
+ if ( vlapic_ipi(vlapic, eax, edx) == X86EMUL_OKAY )
+ vlapic_set_reg(vlapic, APIC_ICR, eax);
+ break;
+ }
+
#ifdef HV_STATS
cur_vcpu->stats.num_icr_writes++;
#endif
break;
case HV_MSR_TPR:
- if (!hv_privilege_check(curp, HV_ACCESS_APIC_MSRS)) {
- goto msr_write_error;
- }
- hv_write_tpr(msr_content);
+ vlapic_set_reg(vcpu_vlapic(current), APIC_TASKPRI,
+ (uint8_t)msr_content);
#ifdef HV_STATS
cur_vcpu->stats.num_tpr_writes++;
#endif

56
hv_tools.patch → hv_extid_compatibility.patch
View File

@ -1,37 +1,3 @@
Index: xen-4.0.1-testing/tools/python/xen/lowlevel/xc/xc.c
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/lowlevel/xc/xc.c
+++ xen-4.0.1-testing/tools/python/xen/lowlevel/xc/xc.c
@@ -944,16 +944,16 @@ static PyObject *pyxc_hvm_build(XcObject
#endif
int i;
char *image;
- int memsize, target=-1, vcpus = 1, acpi = 0, apic = 1;
+ int memsize, target=-1, vcpus = 1, acpi = 0, apic = 1, extid = 0;
PyObject *vcpu_avail_handle = NULL;
uint8_t vcpu_avail[(HVM_MAX_VCPUS + 7)/8];
static char *kwd_list[] = { "domid",
- "memsize", "image", "target", "vcpus",
- "vcpu_avail", "acpi", "apic", NULL };
- if ( !PyArg_ParseTupleAndKeywords(args, kwds, "iis|iiOii", kwd_list,
+ "memsize", "image", "target", "vcpus",
+ "vcpu_avail", "extid", "acpi", "apic", NULL };
+ if ( !PyArg_ParseTupleAndKeywords(args, kwds, "iis|iiOiii", kwd_list,
&dom, &memsize, &image, &target, &vcpus,
- &vcpu_avail_handle, &acpi, &apic) )
+ &vcpu_avail_handle, &extid, &acpi, &apic) )
return NULL;
memset(vcpu_avail, 0, sizeof(vcpu_avail));
@@ -1005,6 +1005,7 @@ static PyObject *pyxc_hvm_build(XcObject
va_hvm->checksum -= sum;
munmap(va_map, XC_PAGE_SIZE);
#endif
+ xc_set_hvm_param(self->xc_handle, dom, HVM_PARAM_EXTEND_HYPERVISOR, extid);
return Py_BuildValue("{}");
}
Index: xen-4.0.1-testing/tools/python/xen/xend/XendConfig.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/XendConfig.py
@ -44,6 +10,15 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendConfig.py
'rtc_timeoffset': int,
'parallel': str,
'serial': str,
@@ -509,6 +510,8 @@ class XendConfig(dict):
if self.is_hvm():
if 'timer_mode' not in self['platform']:
self['platform']['timer_mode'] = 1
+ if 'extid' in self['platform'] and int(self['platform']['extid']) == 1:
+ self['platform']['viridian'] = 1
if 'viridian' not in self['platform']:
self['platform']['viridian'] = 0
if 'rtc_timeoffset' not in self['platform']:
Index: xen-4.0.1-testing/tools/python/xen/xend/image.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/image.py
@ -92,7 +67,18 @@ Index: xen-4.0.1-testing/tools/python/xen/xm/create.py
'usb', 'usbdevice',
'vcpus', 'vnc', 'vncconsole', 'vncdisplay', 'vnclisten',
- 'vncunused', 'viridian', 'vpt_align',
+ 'vncunused', 'viridian', 'extid', 'vpt_align',
+ 'vncunused', 'vpt_align',
'xauthority', 'xen_extended_power_mgmt', 'xen_platform_pci',
'memory_sharing' ]
@@ -1100,6 +1104,10 @@ def configure_hvm(config_image, vals):
config_image.append([a, vals.__dict__[a]])
if vals.vncpasswd is not None:
config_image.append(['vncpasswd', vals.vncpasswd])
+ if vals.extid and vals.extid == 1:
+ config_image.append(['viridian', vals.extid])
+ elif vals.viridian:
+ config_image.append(['viridian', vals.viridian])
def make_config(vals):

29
hv_win7_eoi_bug.patch
View File

@ -1,29 +0,0 @@
Index: xen-4.0.1-testing/xen/arch/x86/hvm/hyperv/hv_intercept.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/hvm/hyperv/hv_intercept.c
+++ xen-4.0.1-testing/xen/arch/x86/hvm/hyperv/hv_intercept.c
@@ -33,6 +33,7 @@
#include <asm/config.h>
#include <asm/hvm/io.h>
+#include <asm/hvm/support.h>
#include <asm/processor.h>
#include <asm/page.h>
#include <asm/apicdef.h>
@@ -988,8 +989,15 @@ hyperv_do_wr_msr(uint32_t idx, struct cp
break;
case HV_MSR_APIC_ASSIST_PAGE:
/*
- * For now ignore this.
+ * We don't support the APIC assist page; windows7 appears to
+ * use this; set it up so that os uses the APIC MSR to indicate EOI.
*/
+ if (msr_content & 0x1) {
+ uint32_t data = 0;
+ paddr_t assist_page = msr_content & ~1UL;
+ (void)hvm_copy_to_guest_phys(assist_page, &data, sizeof(data));
+ }
+
break;
default:

203
hv_xen_base.patch
View File

@ -1,203 +0,0 @@
Index: xen-4.0.1-testing/xen/include/asm-x86/hvm/domain.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/asm-x86/hvm/domain.h
+++ xen-4.0.1-testing/xen/include/asm-x86/hvm/domain.h
@@ -96,6 +96,7 @@ struct hvm_domain {
struct vmx_domain vmx;
struct svm_domain svm;
};
+ void *hyperv_handle; /* will be NULL on creation*/
};
#endif /* __ASM_X86_HVM_DOMAIN_H__ */
Index: xen-4.0.1-testing/xen/arch/x86/hvm/Makefile
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/hvm/Makefile
+++ xen-4.0.1-testing/xen/arch/x86/hvm/Makefile
@@ -1,5 +1,6 @@
subdir-y += svm
subdir-y += vmx
+subdir-$(x86_64) += hyperv
obj-y += asid.o
obj-y += emulate.o
Index: xen-4.0.1-testing/xen/arch/x86/hvm/hvm.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/hvm/hvm.c
+++ xen-4.0.1-testing/xen/arch/x86/hvm/hvm.c
@@ -48,6 +48,7 @@
#include <asm/mc146818rtc.h>
#include <asm/spinlock.h>
#include <asm/hvm/hvm.h>
+#include <asm/hvm/hvm_extensions.h>
#include <asm/hvm/vpt.h>
#include <asm/hvm/support.h>
#include <asm/hvm/cacheattr.h>
@@ -435,6 +436,7 @@ void hvm_domain_relinquish_resources(str
void hvm_domain_destroy(struct domain *d)
{
+ hyperx_intercept_domain_destroy(d);
hvm_funcs.domain_destroy(d);
rtc_deinit(d);
stdvga_deinit(d);
@@ -756,8 +758,14 @@ int hvm_vcpu_initialise(struct vcpu *v)
v->arch.hvm_vcpu.xfeature_mask = XSTATE_FP_SSE;
}
+ if ((rc = hyperx_intercept_vcpu_initialize(v)) != 0)
+ goto fail1;
+
if ( (rc = vlapic_init(v)) != 0 )
+ {
+ hyperx_intercept_vcpu_destroy(v);
goto fail1;
+ }
if ( (rc = hvm_funcs.vcpu_initialise(v)) != 0 )
goto fail2;
@@ -808,12 +816,14 @@ int hvm_vcpu_initialise(struct vcpu *v)
hvm_funcs.vcpu_destroy(v);
fail2:
vlapic_destroy(v);
+ hyperx_intercept_vcpu_destroy(v);
fail1:
return rc;
}
void hvm_vcpu_destroy(struct vcpu *v)
{
+ hyperx_intercept_vcpu_destroy(v);
tasklet_kill(&v->arch.hvm_vcpu.assert_evtchn_irq_tasklet);
hvm_vcpu_cacheattr_destroy(v);
vlapic_destroy(v);
@@ -1873,7 +1883,7 @@ void hvm_cpuid(unsigned int input, unsig
return;
if ( cpuid_hypervisor_leaves(input, count, eax, ebx, ecx, edx) )
- return;
+ goto hvm_cpuid_done;
domain_cpuid(v->domain, input, *ecx, eax, ebx, ecx, edx);
@@ -1940,6 +1950,8 @@ void hvm_cpuid(unsigned int input, unsig
*edx &= ~bitmaskof(X86_FEATURE_RDTSCP);
break;
}
+hvm_cpuid_done:
+ hyperx_intercept_do_cpuid(input, eax, ebx, ecx, edx);
}
void hvm_rdtsc_intercept(struct cpu_user_regs *regs)
@@ -2040,6 +2052,8 @@ int hvm_msr_read_intercept(struct cpu_us
break;
/* ret == 0, This is not an MCE MSR, see other MSRs */
else if (!ret)
+ if (hyperx_intercept_do_msr_read(ecx, regs))
+ return X86EMUL_OKAY;
return hvm_funcs.msr_read_intercept(regs);
}
@@ -2138,6 +2152,8 @@ int hvm_msr_write_intercept(struct cpu_u
else if ( ret )
break;
else if (!ret)
+ if (hyperx_intercept_do_msr_write(ecx, regs))
+ return X86EMUL_OKAY;
return hvm_funcs.msr_write_intercept(regs);
}
@@ -2344,6 +2360,10 @@ int hvm_do_hypercall(struct cpu_user_reg
case 0:
break;
}
+ if (hyperx_intercept_do_hypercall(regs))
+ {
+ return HVM_HCALL_completed;
+ }
if ( (eax & 0x80000000) && is_viridian_domain(curr->domain) )
return viridian_hypercall(regs);
@@ -2878,6 +2898,18 @@ long do_hvm_op(unsigned long op, XEN_GUE
rc = -EINVAL;
break;
+ case HVM_PARAM_EXTEND_HYPERVISOR:
+#ifdef __x86_64__
+ if (a.value != 1)
+ rc = -EINVAL;
+ else if (hyperv_initialize(d))
+ rc = -ENOMEM;
+ else
+ break;
+#else
+ rc = -EINVAL;
+#endif
+ goto param_fail;
}
if ( rc == 0 )
Index: xen-4.0.1-testing/xen/include/public/arch-x86/hvm/save.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/public/arch-x86/hvm/save.h
+++ xen-4.0.1-testing/xen/include/public/arch-x86/hvm/save.h
@@ -431,9 +431,24 @@ struct hvm_viridian_context {
DECLARE_HVM_SAVE_TYPE(VIRIDIAN, 15, struct hvm_viridian_context);
+struct hvm_hyperv_dom {
+ uint64_t guestid_msr;
+ uint64_t hypercall_msr;
+ uint32_t long_mode;
+ uint32_t ext_id;
+};
+DECLARE_HVM_SAVE_TYPE(HYPERV_DOM, 16, struct hvm_hyperv_dom);
+
+struct hvm_hyperv_cpu {
+ uint64_t control_msr;
+ uint64_t version_msr;
+ uint64_t pad[27]; //KYS: sles10 sp2 compatibility
+};
+DECLARE_HVM_SAVE_TYPE(HYPERV_CPU, 17, struct hvm_hyperv_cpu);
+
/*
* Largest type-code in use
*/
-#define HVM_SAVE_CODE_MAX 15
+#define HVM_SAVE_CODE_MAX 17
#endif /* __XEN_PUBLIC_HVM_SAVE_X86_H__ */
Index: xen-4.0.1-testing/xen/arch/x86/hvm/vlapic.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/hvm/vlapic.c
+++ xen-4.0.1-testing/xen/arch/x86/hvm/vlapic.c
@@ -34,6 +34,7 @@
#include <asm/hvm/hvm.h>
#include <asm/hvm/io.h>
#include <asm/hvm/support.h>
+#include <asm/hvm/hvm_extensions.h>
#include <asm/hvm/vmx/vmx.h>
#include <public/hvm/ioreq.h>
#include <public/hvm/params.h>
@@ -291,6 +292,7 @@ static int vlapic_accept_sipi(struct vcp
hvm_vcpu_reset_state(v, trampoline_vector << 8, 0);
vcpu_unpause(v);
+ hyperx_intercept_vcpu_up(v);
return X86EMUL_OKAY;
}
Index: xen-4.0.1-testing/xen/include/public/hvm/params.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/public/hvm/params.h
+++ xen-4.0.1-testing/xen/include/public/hvm/params.h
@@ -109,6 +109,8 @@
/* Boolean: Enable aligning all periodic vpts to reduce interrupts */
#define HVM_PARAM_VPT_ALIGN 16
-#define HVM_NR_PARAMS 17
+#define HVM_PARAM_EXTEND_HYPERVISOR 17
+
+#define HVM_NR_PARAMS 18
#endif /* __XEN_PUBLIC_HVM_PARAMS_H__ */

1776
hv_xen_extension.patch
View File

File diff suppressed because it is too large Load Diff

2
ioemu-bdrv-open-CACHE_WB.patch
View File

@ -2,7 +2,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xen_blktap.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
@@ -257,8 +257,11 @@ static int open_disk(struct td_state *s,
@@ -259,8 +259,11 @@ static int open_disk(struct td_state *s,
drv = blktap_drivers[i].drv;
DPRINTF("%s driver specified\n", drv ? drv->format_name : "No");

12
ioemu-blktap-barriers.patch
View File

@ -2,7 +2,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xen_blktap.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
@@ -358,6 +358,15 @@ static void qemu_send_responses(void* op
@@ -360,6 +360,15 @@ static void qemu_send_responses(void* op
}
/**
@ -18,7 +18,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
* Callback function for the IO message pipe. Reads requests from the ring
* and processes them (call qemu read/write functions).
*
@@ -376,6 +385,7 @@ static void handle_blktap_iomsg(void* pr
@@ -378,6 +387,7 @@ static void handle_blktap_iomsg(void* pr
blkif_t *blkif = s->blkif;
tapdev_info_t *info = s->ring_info;
int page_size = getpagesize();
@ -26,7 +26,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
struct aiocb_info *aiocb_info;
@@ -408,7 +418,7 @@ static void handle_blktap_iomsg(void* pr
@@ -410,7 +420,7 @@ static void handle_blktap_iomsg(void* pr
/* Don't allow writes on readonly devices */
if ((s->flags & TD_RDONLY) &&
@ -35,7 +35,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
blkif->pending_list[idx].status = BLKIF_RSP_ERROR;
goto send_response;
}
@@ -429,7 +439,7 @@ static void handle_blktap_iomsg(void* pr
@@ -431,7 +441,7 @@ static void handle_blktap_iomsg(void* pr
DPRINTF("Sector request failed:\n");
DPRINTF("%s request, idx [%d,%d] size [%llu], "
"sector [%llu,%llu]\n",
@ -44,7 +44,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
"WRITE" : "READ"),
idx,i,
(long long unsigned)
@@ -442,8 +452,14 @@ static void handle_blktap_iomsg(void* pr
@@ -444,8 +454,14 @@ static void handle_blktap_iomsg(void* pr
blkif->pending_list[idx].secs_pending += nsects;
@ -60,7 +60,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
case BLKIF_OP_WRITE:
aiocb_info = malloc(sizeof(*aiocb_info));
@@ -463,6 +479,10 @@ static void handle_blktap_iomsg(void* pr
@@ -465,6 +481,10 @@ static void handle_blktap_iomsg(void* pr
DPRINTF("ERROR: bdrv_write() == NULL\n");
goto send_response;
}

8
ioemu-blktap-image-format.patch
View File

@ -19,7 +19,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xen_blktap.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
@@ -228,9 +228,10 @@ static int map_new_dev(struct td_state *
@@ -230,9 +230,10 @@ static int map_new_dev(struct td_state *
return -1;
}
@ -31,7 +31,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
char* devname;
static int devnumber = 0;
int i;
@@ -240,7 +241,22 @@ static int open_disk(struct td_state *s,
@@ -242,7 +243,22 @@ static int open_disk(struct td_state *s,
bs = bdrv_new(devname);
free(devname);
@ -55,7 +55,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
fprintf(stderr, "Could not open image file %s\n", path);
return -ENOMEM;
}
@@ -519,7 +535,7 @@ static void handle_blktap_ctrlmsg(void*
@@ -521,7 +537,7 @@ static void handle_blktap_ctrlmsg(void*
s = state_init();
/*Open file*/
@ -68,7 +68,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.h
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xen_blktap.h
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.h
@@ -52,4 +52,18 @@ typedef struct fd_list_entry {
@@ -53,4 +53,18 @@ typedef struct fd_list_entry {
int init_blktap(void);

2
ioemu-blktap-zero-size.patch
View File

@ -16,7 +16,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xen_blktap.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
@@ -266,6 +266,12 @@ static int open_disk(struct td_state *s,
@@ -268,6 +268,12 @@ static int open_disk(struct td_state *s,
s->size = bs->total_sectors;
s->sector_size = 512;

78
ioemu-disable-emulated-ide-if-pv.patch Normal file
View File

@ -0,0 +1,78 @@
Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/qemu-xen.h
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/qemu-xen.h
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/qemu-xen.h
@@ -1,6 +1,8 @@
#ifndef QEMU_XEN_H
#define QEMU_XEN_H
+#include "hw/boards.h"
+
/* vl.c */
extern int restore;
extern int vga_ram_size;
@@ -73,7 +75,7 @@ void handle_buffered_pio(void);
#endif
/* xenstore.c */
-void xenstore_parse_domain_config(int domid);
+void xenstore_parse_domain_config(int domid, QEMUMachine *machine);
int xenstore_parse_disable_pf_config(void);
int xenstore_fd(void);
void xenstore_process_event(void *opaque);
Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/vl.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/vl.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/vl.c
@@ -5827,10 +5827,10 @@ int main(int argc, char **argv, char **e
if ((msg = xenbus_read(XBT_NIL, "domid", &domid_s)))
fprintf(stderr,"Can not read our own domid: %s\n", msg);
else
- xenstore_parse_domain_config(atoi(domid_s));
+ xenstore_parse_domain_config(atoi(domid_s), machine);
}
#else
- xenstore_parse_domain_config(domid);
+ xenstore_parse_domain_config(domid, machine);
#endif /* CONFIG_STUBDOM */
/* we always create the cdrom drive, even if no disk is there */
Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/xenstore.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/xenstore.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/xenstore.c
@@ -397,7 +397,7 @@ static const char *xenstore_get_guest_uu
#define PT_PCI_POWER_MANAGEMENT_DEFAULT 0
int direct_pci_msitranslate;
int direct_pci_power_mgmt;
-void xenstore_parse_domain_config(int hvm_domid)
+void xenstore_parse_domain_config(int hvm_domid, QEMUMachine *machine)
{
char **e_danger = NULL;
char *buf = NULL;
@@ -712,15 +712,19 @@ void xenstore_parse_domain_config(int hv
#endif
- drives_table[nb_drives].bdrv = bs;
- drives_table[nb_drives].used = 1;
+ if (machine == &xenfv_machine) {
+ drives_table[nb_drives].bdrv = bs;
+ drives_table[nb_drives].used = 1;
#ifdef CONFIG_STUBDOM
- media_filename[nb_drives] = strdup(danger_buf);
+ media_filename[nb_drives] = strdup(danger_buf);
#else
- media_filename[nb_drives] = strdup(bs->filename);
+ media_filename[nb_drives] = strdup(bs->filename);
#endif
- nb_drives++;
-
+ nb_drives++;
+ } else {
+ qemu_aio_flush();
+ bdrv_close(bs);
+ }
}
#ifdef CONFIG_STUBDOM

2
ioemu-vnc-resize.patch
View File

@ -2,7 +2,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/vnc.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/vnc.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/vnc.c
@@ -1736,6 +1736,25 @@ static int protocol_client_msg(VncState
@@ -1734,6 +1734,25 @@ static int protocol_client_msg(VncState
}
set_encodings(vs, (int32_t *)(data + 4), limit);

4
multi-xvdp.patch
View File

@ -34,7 +34,7 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
xc = xen.lowlevel.xc.xc()
xoptions = XendOptions.instance()
@@ -3314,20 +3314,27 @@ class XendDomainInfo:
@@ -3304,20 +3304,27 @@ class XendDomainInfo:
# This is a file, not a device. pygrub can cope with a
# file if it's raw, but if it's QCOW or other such formats
# used through blktap, then we need to mount it first.
@ -76,7 +76,7 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
try:
blcfg = bootloader(blexec, fn, self, False,
@@ -3335,11 +3342,11 @@ class XendDomainInfo:
@@ -3325,11 +3332,11 @@ class XendDomainInfo:
finally:
if mounted:
log.info("Unmounting %s from %s." %

22
qemu-fix-7433.patch Normal file
View File

@ -0,0 +1,22 @@
--- a/tools/ioemu-qemu-xen/keymaps.c
+++ b/tools/ioemu-qemu-xen/keymaps.c
@@ -56,15 +56,12 @@ typedef struct {
static void del_key_range(struct key_range **krp, int code) {
struct key_range *kr;
- struct key_range *kr_pr;
- for (kr = *krp; kr; kr_pr = kr, kr = kr->next) {
+ while ((kr = *krp) != NULL) {
if (code >= kr->start && code <= kr->end) {
- if (kr == *krp)
- *krp = kr->next;
- else
- kr_pr->next = kr->next;
+ *krp = kr->next;
qemu_free(kr);
- }
+ } else
+ krp = &kr->next;
}
}

47
snapshot-without-pv-fix.patch
View File

@ -8,16 +8,6 @@ ready, so the watch will tell qemu to add the disk entry to
drives_table[], otherwise the disk in qemu will just stay opened,not
showing up in drives_table[].
Signed-off-by: Li Dongyang <lidongyang@novell.com>
---
tools/blktap/drivers/blktapctrl.c | 81 +++++++++++++++++++++++++++++++++-
tools/blktap/lib/blkif.c | 23 ++++++++++
tools/blktap/lib/blktaplib.h | 5 ++
tools/blktap/lib/xenbus.c | 69 +++++++++++++++++++++++++++++
tools/ioemu-qemu-xen/hw/xen_blktap.c | 49 +++++++++++++++-----
5 files changed, 213 insertions(+), 14 deletions(-)
Index: xen-4.0.1-testing/tools/blktap/drivers/blktapctrl.c
===================================================================
--- xen-4.0.1-testing.orig/tools/blktap/drivers/blktapctrl.c
@ -198,7 +188,7 @@ Index: xen-4.0.1-testing/tools/blktap/lib/blktaplib.h
void __init_blkif(void);
typedef struct busy_state {
@@ -209,6 +212,8 @@ typedef struct msg_pid {
@@ -210,6 +213,8 @@ typedef struct msg_pid {
#define CTLMSG_CLOSE_RSP 8
#define CTLMSG_PID 9
#define CTLMSG_PID_RSP 10
@ -298,7 +288,16 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xen_blktap.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
@@ -81,8 +81,18 @@ static void unmap_disk(struct td_state *
@@ -35,6 +35,8 @@
#ifndef QEMU_TOOL
#include "qemu-common.h"
#include "sysemu.h"
+#include "irq.h"
+#include "hw.h"
#endif
#include "xen_blktap.h"
@@ -81,8 +83,18 @@ static void unmap_disk(struct td_state *
{
tapdev_info_t *info = s->ring_info;
fd_list_entry_t *entry;
@ -317,7 +316,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
if (info != NULL && info->mem > 0)
munmap(info->mem, getpagesize() * BLKTAP_MMAP_REGION_SIZE);
@@ -242,18 +252,6 @@ static int open_disk(struct td_state *s,
@@ -242,18 +254,6 @@ static int open_disk(struct td_state *s,
s->info = ((s->flags & TD_RDONLY) ? VDISK_READONLY : 0);
@ -336,7 +335,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
return 0;
}
@@ -494,7 +492,7 @@ static void handle_blktap_ctrlmsg(void*
@@ -494,7 +494,7 @@ static void handle_blktap_ctrlmsg(void*
msg_hdr_t *msg;
msg_newdev_t *msg_dev;
msg_pid_t *msg_pid;
@ -345,14 +344,15 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
struct td_state *s = NULL;
fd_list_entry_t *entry;
@@ -584,6 +582,31 @@ static void handle_blktap_ctrlmsg(void*
@@ -584,6 +584,33 @@ static void handle_blktap_ctrlmsg(void*
len = write(write_fd, buf, msglen);
break;
+ case CTLMSG_ADDDEV:
+ s = get_state(msg->cookie);
+ if (s) {
+ if (s && !s->added) {
+#ifndef QEMU_TOOL
+ ide_unplug_harddisks();
+ for (i = 0; i < MAX_DRIVES + 1; i++) {
+ if (drives_table[i].bdrv == NULL) {
+ drives_table[i].bdrv = s->bs;
@ -361,6 +361,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
+ drives_table[i].unit = 0;
+ drives_table[i].used = 1;
+ nb_drives++;
+ s->added = 1;
+ break;
+ }
+ }
@ -377,3 +378,17 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
default:
break;
}
Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.h
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xen_blktap.h
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.h
@@ -40,7 +40,8 @@ struct td_state {
void *fd_entry;
uint64_t sector_size;
uint64_t size;
- unsigned int info;
+ unsigned int info;
+ int added;
};
typedef struct fd_list_entry {

62
snapshot-xend.patch
View File

@ -47,12 +47,13 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/server/blkif.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/server/blkif.py
+++ xen-4.0.1-testing/tools/python/xen/xend/server/blkif.py
@@ -88,6 +88,9 @@ class BlkifController(DevController):
@@ -88,6 +88,10 @@ class BlkifController(DevController):
if bootable != None:
back['bootable'] = str(bootable)
+ if 'snapshotname' in self.vm.info:
+ back['snapshot'] = self.vm.info['snapshotname']
+ self.vm.info.pop('snapshotname')
+
if security.on() == xsconstants.XS_POLICY_USE:
self.do_access_control(config, uname)
@ -354,7 +355,25 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomain.py
xc = xen.lowlevel.xc.xc()
xoptions = XendOptions.instance()
@@ -1576,6 +1577,187 @@ class XendDomain:
@@ -319,11 +320,16 @@ class XendDomain:
fd, fn = tempfile.mkstemp()
f = os.fdopen(fd, 'w+b')
try:
+ snapshotname = ''
+ if dominfo.info.has_key('snapshotname'):
+ snapshotname = dominfo.info.pop('snapshotname')
+
prettyprint(dominfo.sxpr(legacy_only = False), f,
width = 78)
finally:
f.close()
-
+ if snapshotname:
+ dominfo.info['snapshotname'] = snapshotname
try:
shutil.move(fn, self._managed_config_path(dom_uuid))
except:
@@ -1576,6 +1582,187 @@ class XendDomain:
else:
log.debug("error: Domain is not running!")
@ -667,3 +686,42 @@ Index: xen-4.0.1-testing/tools/python/xen/xm/main.py
"shutdown": xm_shutdown,
"start": xm_start,
"sysrq": xm_sysrq,
Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/XendDomainInfo.py
+++ xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
@@ -507,7 +507,6 @@ class XendDomainInfo:
self._setSchedParams()
self._storeVmDetails()
self._createChannels()
- self._createDevices()
self._storeDomDetails()
self._endRestore()
except:
@@ -2383,7 +2382,7 @@ class XendDomainInfo:
return self.getDeviceController(deviceClass).reconfigureDevice(
devid, devconfig)
- def _createDevices(self):
+ def _createDevices(self, resume = False):
"""Create the devices for a vm.
@raise: VmError for invalid devices
@@ -2432,7 +2431,7 @@ class XendDomainInfo:
if self.image:
- self.image.createDeviceModel()
+ self.image.createDeviceModel(resume)
#if have pass-through devs, need the virtual pci slots info from qemu
self.pci_device_configure_boot()
@@ -3048,7 +3047,7 @@ class XendDomainInfo:
self._introduceDomain()
self.image = image.create(self, self.info)
if self.image:
- self.image.createDeviceModel(True)
+ self._createDevices(True)
self._storeDomDetails()
self._registerWatches()
self.refreshShutdown()

87
suspend_evtchn_lock.patch Normal file
View File

@ -0,0 +1,87 @@
Improve suspend eventchn lock, use flock instead of exclusive lock
file to avoid that sometimes the lock file is not removed cleanly
and affacts later getting lock.
http://lists.xensource.com/archives/html/xen-devel/2010-11/msg01559.html
Signed-off-by cyliu@novell.com
Index: xen-4.0.1-testing/tools/libxc/xc_suspend.c
===================================================================
--- xen-4.0.1-testing.orig/tools/libxc/xc_suspend.c
+++ xen-4.0.1-testing/tools/libxc/xc_suspend.c
@@ -6,25 +6,25 @@
#include "xc_private.h"
#include "xenguest.h"
+#include <sys/file.h>
+#include <sys/fcntl.h>
#define SUSPEND_LOCK_FILE "/var/lib/xen/suspend_evtchn_lock.d"
static int lock_suspend_event(void)
{
int fd, rc;
mode_t mask;
- char buf[128];
mask = umask(022);
- fd = open(SUSPEND_LOCK_FILE, O_CREAT | O_EXCL | O_RDWR, 0666);
+ fd = open(SUSPEND_LOCK_FILE, O_CREAT | O_RDWR, 0666);
if (fd < 0)
{
ERROR("Can't create lock file for suspend event channel\n");
return -EINVAL;
}
umask(mask);
- snprintf(buf, sizeof(buf), "%10ld", (long)getpid());
+ rc = flock(fd, LOCK_EX | LOCK_NB);
- rc = write_exact(fd, buf, strlen(buf));
close(fd);
return rc;
@@ -32,30 +32,21 @@ static int lock_suspend_event(void)
static int unlock_suspend_event(void)
{
- int fd, pid, n;
- char buf[128];
+ int fd, rc;
fd = open(SUSPEND_LOCK_FILE, O_RDWR);
if (fd < 0)
return -EINVAL;
- n = read(fd, buf, 127);
+ rc = flock(fd, LOCK_UN | LOCK_NB);
close(fd);
- if (n > 0)
- {
- sscanf(buf, "%d", &pid);
- /* We are the owner, so we can simply delete the file */
- if (pid == getpid())
- {
- unlink(SUSPEND_LOCK_FILE);
- return 0;
- }
- }
+ if(!rc)
+ unlink(SUSPEND_LOCK_FILE);
- return -EPERM;
+ return rc;
}
int xc_await_suspend(int xce, int suspend_evtchn)
@@ -110,8 +101,7 @@ int xc_suspend_evtchn_init(int xc, int x
return suspend_evtchn;
cleanup:
- if (suspend_evtchn != -1)
- xc_suspend_evtchn_release(xce, suspend_evtchn);
+ xc_suspend_evtchn_release(xce, suspend_evtchn);
return -1;
}

6
tapdisk-ioemu-shutdown-fix.patch
View File

@ -20,7 +20,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/hw/xen_blktap.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
@@ -65,6 +65,7 @@ int read_fd;
@@ -67,6 +67,7 @@ int read_fd;
int write_fd;
static pid_t process;
@ -28,7 +28,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
fd_list_entry_t *fd_start = NULL;
static void handle_blktap_iomsg(void* private);
@@ -539,6 +540,7 @@ static void handle_blktap_ctrlmsg(void*
@@ -541,6 +542,7 @@ static void handle_blktap_ctrlmsg(void*
/* Allocate the disk structs */
s = state_init();
@ -36,7 +36,7 @@ Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/hw/xen_blktap.c
/*Open file*/
if (s == NULL || open_disk(s, path, msg->drivertype, msg->readonly)) {
@@ -589,7 +591,8 @@ static void handle_blktap_ctrlmsg(void*
@@ -591,7 +593,8 @@ static void handle_blktap_ctrlmsg(void*
case CTLMSG_CLOSE:
s = get_state(msg->cookie);
if (s) unmap_disk(s);

14
x86-cpufreq-report.patch
View File

@ -1,5 +1,7 @@
--- a/xen/arch/x86/platform_hypercall.c
+++ b/xen/arch/x86/platform_hypercall.c
Index: xen-4.0.1-testing/xen/arch/x86/platform_hypercall.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/platform_hypercall.c
+++ xen-4.0.1-testing/xen/arch/x86/platform_hypercall.c
@@ -21,7 +21,7 @@
#include <xen/acpi.h>
#include <asm/current.h>
@ -17,7 +19,7 @@
struct xen_platform_op curop, *op = &curop;
if ( !IS_PRIV(current->domain) )
@@ -486,6 +487,24 @@ ret_t do_platform_op(XEN_GUEST_HANDLE(xe
@@ -489,6 +490,24 @@ ret_t do_platform_op(XEN_GUEST_HANDLE(xe
op->u.mem_add.epfn,
op->u.mem_add.pxm);
break;
@ -42,8 +44,10 @@
default:
ret = -ENOSYS;
break;
--- a/xen/include/public/platform.h
+++ b/xen/include/public/platform.h
Index: xen-4.0.1-testing/xen/include/public/platform.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/public/platform.h
+++ xen-4.0.1-testing/xen/include/public/platform.h
@@ -355,6 +355,14 @@ struct xenpf_mem_hotadd
uint32_t flags;
};

8
x86-ioapic-ack-default.patch
View File

@ -1,7 +1,9 @@
Change default IO-APIC ack mode for single IO-APIC systems to old-style.
--- a/xen/arch/x86/io_apic.c
+++ b/xen/arch/x86/io_apic.c
Index: xen-4.0.1-testing/xen/arch/x86/io_apic.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/io_apic.c
+++ xen-4.0.1-testing/xen/arch/x86/io_apic.c
@@ -1559,7 +1559,7 @@ static unsigned int startup_level_ioapic
return 0; /* don't check for pending */
}
@ -11,7 +13,7 @@ Change default IO-APIC ack mode for single IO-APIC systems to old-style.
static void setup_ioapic_ack(char *s)
{
if ( !strcmp(s, "old") )
@@ -2062,6 +2062,8 @@ void __init setup_IO_APIC(void)
@@ -2075,6 +2075,8 @@ void __init setup_IO_APIC(void)
else
io_apic_irqs = ~PIC_IRQS;

2
x86-show-page-walk-early.patch
View File

@ -15,7 +15,7 @@ Index: xen-4.0.1-testing/xen/arch/x86/traps.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/traps.c
+++ xen-4.0.1-testing/xen/arch/x86/traps.c
@@ -1336,6 +1336,7 @@ asmlinkage void do_early_page_fault(stru
@@ -1342,6 +1342,7 @@ asmlinkage void do_early_page_fault(stru
unsigned long *stk = (unsigned long *)regs;
printk("Early fatal page fault at %04x:%p (cr2=%p, ec=%04x)\n",
regs->cs, _p(regs->eip), _p(cr2), regs->error_code);

13
x86_64-gdt-ldt-fault-filter.patch
View File

@ -1,13 +0,0 @@
References: bnc#656245
--- a/xen/arch/x86/traps.c
+++ b/xen/arch/x86/traps.c
@@ -1223,7 +1223,7 @@ static int fixup_page_fault(unsigned lon
if ( unlikely(IN_HYPERVISOR_RANGE(addr)) )
{
- if ( !(regs->error_code & PFEC_reserved_bit) &&
+ if ( !(regs->error_code & (PFEC_user_mode | PFEC_reserved_bit)) &&
(addr >= GDT_LDT_VIRT_START) && (addr < GDT_LDT_VIRT_END) )
return handle_gdt_ldt_mapping_fault(
addr - GDT_LDT_VIRT_START, regs);

15
xen-minimum-restart-time.patch Normal file
View File

@ -0,0 +1,15 @@
References: bnc#661298
Index: xen-4.0.1-testing/tools/python/xen/xend/XendConstants.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/XendConstants.py
+++ xen-4.0.1-testing/tools/python/xen/xend/XendConstants.py
@@ -92,7 +92,7 @@ DOM_STATES_OLD = [
SHUTDOWN_TIMEOUT = (60.0 * 5)
"""Minimum time between domain restarts in seconds."""
-MINIMUM_RESTART_TIME = 60
+MINIMUM_RESTART_TIME = 10
RESTART_IN_PROGRESS = 'xend/restart_in_progress'
DUMPCORE_IN_PROGRESS = 'xend/dumpcore_in_progress'

143
xen.changes
View File

@ -1,3 +1,146 @@
-------------------------------------------------------------------
Fri Jan 14 08:29:15 MST 2011 - carnold@novell.com
- bnc#658704 - SLES11 SP1 Xen boot panic in x2apic mode
22707-x2apic-preenabled-check.patch
- bnc#641419 - L3: Xen: qemu-dm reports "xc_map_foreign_batch: mmap failed:
Cannot allocate memory"
7434-qemu-rlimit-as.patch
- Additional or upstream patches from Jan
22693-fam10-mmio-conf-base-protect.patch
22694-x86_64-no-weak.patch
22708-xenctx-misc.patch
21432-4.0-cpu-boot-failure.patch
22645-amd-flush-filter.patch
qemu-fix-7433.patch
-------------------------------------------------------------------
Wed Jan 12 21:28:18 MST 2011 - carnold@novell.com
- Maintain compatibility with the extid flag even though it is
deprecated for both legacy and sxp config files.
hv_extid_compatibility.patch
-------------------------------------------------------------------
Wed Jan 12 16:08:10 CST 2011 - cyliu@novell.com
- bnc#649209-improve suspend eventchn lock
suspend_evtchn_lock.patch
-------------------------------------------------------------------
Tue Jan 11 11:48:10 MST 2011 - carnold@novell.com
- Removed the hyper-v shim patches in favor of using the upstream
version.
-------------------------------------------------------------------
Mon Jan 10 21:15:43 MST 2011 - carnold@novell.com
- bnc#641419 - L3: Xen: qemu-dm reports "xc_map_foreign_batch: mmap
failed: Cannot allocate memory"
qemu-rlimit-as.patch
-------------------------------------------------------------------
Mon Jan 10 15:23:18 CST 2011 - cyliu@novell.com
- Upstream c/s 7433 to replace qemu_altgr_more.patch
7433-qemu-altgr.patch
-------------------------------------------------------------------
Fri Jan 7 16:13:24 MST 2011 - jfehlig@novell.com
- bnc#661931 - Fix fd leak in xenstore library
21344-4.0-testing-xenstore-fd-leak.patch
-------------------------------------------------------------------
Tue Jan 4 18:06:31 MST 2011 - carnold@novell.com
- bnc#656369 - g5plus: sles11sp1 xen crash with 8 socket x2apic
preenabled
21989-x2apic-resume.patch
22475-x2apic-cleanup.patch
22535-x2apic-preenabled.patch
- bnc#658163 - maintenance release - Nehalem system cannot boot
into xen with maintenance release installed
22504-iommu-dom0-holes.patch
22506-x86-iommu-dom0-estimate.patch
- bnc#658704 - SLES11 SP1 Xen boot panic in x2apic mode
21810-x2apic-acpi.patch
- Upstream patches from Jan
22470-vlapic-tick-loss.patch
22484-vlapic-tmcct-periodic.patch
22526-ept-access-once.patch
22533-x86-32bit-apicid.patch
22534-x86-max-local-apic.patch
22538-keyhandler-relax.patch
22540-32on64-hypercall-debug.patch
22549-vtd-map-page-leak.patch
22574-ept-skip-validation.patch
22632-vtd-print-entries.patch
-------------------------------------------------------------------
Tue Jan 4 16:53:54 MST 2011 - carnold@novell.com
- bnc#661298 - maintenance release candidate - Windows VMs reboot
too fast, triggering failsafe
xen-minimum-restart-time.patch
-------------------------------------------------------------------
Tue Jan 4 10:23:18 CST 2011 - cyliu@novell.com
- bnc#659070 - Fail to input '|' in en-us keyboard
qemu_altgr_more.patch
-------------------------------------------------------------------
Tue Dec 28 11:23:18 MST 2010 - jfehlig@novell.com
- bnc#659466 - XEN drbd block device type not working on SLES 11 SP1
20158-revert.patch
-------------------------------------------------------------------
Mon Dec 27 14:43:13 CST 2010 - lidongyang@novell.com
- bnc#654543 - PV guest won't unplug the IDE disk created by
qemu-dm
a dirty hack, only add the device to drives_table[] if we are FV
domU, that will be unplugged anyway if a PV driver is loaded
later.
ioemu-disable-emulated-ide-if-pv.patch
-------------------------------------------------------------------
Wed Dec 22 11:56:38 CST 2010 - cyliu@novell.com
- Upstream patch to replace xenfb_32bpp.patch
7426-xenfb-depth.patch
-------------------------------------------------------------------
Tue Dec 21 11:31:52 CST 2010 - lidongyang@novell.com
- bnc#651822 make sure we only apply the snapshot once, and the
changes made after snapshot-apply hit the disk.
snapshot-xend.patch
snapshot-without-pv-fix.patch
-------------------------------------------------------------------
Fri Dec 17 14:34:18 CET 2010 - ohering@suse.de
- fate#310510 - fix xenpaging
xenpaging.HVMCOPY_gfn_paged_out.patch
- remove incorrect and unneeded cleanup from do_memory_op
subfunctions
add mainline tag to merged patches
-------------------------------------------------------------------
Thu Dec 16 17:33:52 MST 2010 - jfehlig@novell.com
- bnc#613584 - If available, use kpartx '-f' option in domUloader
-------------------------------------------------------------------
Thu Dec 16 10:21:13 MST 2010 - jfehlig@novell.com
- bnc#659872 - xend: Do no release domain lock on checkpoint
operation.
-------------------------------------------------------------------
Tue Dec 7 15:38:58 CET 2010 - ohering@suse.de

313
xen.spec
View File

@ -1,5 +1,5 @@
#
# spec file for package xen (Version 4.0.1_02)
# spec file for package xen (Version 4.0.1_04)
#
# Copyright (c) 2010 SUSE LINUX Products GmbH, Nuernberg, Germany.
#
@ -26,35 +26,35 @@ ExclusiveArch: %ix86 x86_64
%define xen_build_dir xen-4.0.1-testing
%define with_kmp 1
%define with_stubdom 1
BuildRequires: LibVNCServer-devel
BuildRequires: SDL-devel
BuildRequires: automake
BuildRequires: bin86
BuildRequires: curl-devel
BuildRequires: dev86
BuildRequires: graphviz
BuildRequires: latex2html
BuildRequires: libjpeg-devel
BuildRequires: libxml2-devel
BuildRequires: ncurses-devel
BuildRequires: openssl
BuildRequires: openssl-devel
BuildRequires: pciutils-devel
BuildRequires: python-devel
BuildRequires: texinfo
BuildRequires: transfig
BuildRequires: LibVNCServer-devel
BuildRequires: SDL-devel
BuildRequires: automake
BuildRequires: bin86
BuildRequires: curl-devel
BuildRequires: dev86
BuildRequires: graphviz
BuildRequires: latex2html
BuildRequires: libjpeg-devel
BuildRequires: libxml2-devel
BuildRequires: ncurses-devel
BuildRequires: openssl
BuildRequires: openssl-devel
BuildRequires: pciutils-devel
BuildRequires: python-devel
BuildRequires: texinfo
BuildRequires: transfig
%if %suse_version <= 1110
BuildRequires: pmtools
BuildRequires: pmtools
%else
BuildRequires: acpica
BuildRequires: acpica
%endif
%if %suse_version >= 1030
BuildRequires: texlive
BuildRequires: texlive-latex
BuildRequires: texlive
BuildRequires: texlive-latex
%else
BuildRequires: te_ams
BuildRequires: te_latex
BuildRequires: tetex
BuildRequires: te_ams
BuildRequires: te_latex
BuildRequires: tetex
%endif
%ifarch x86_64
BuildRequires: glibc-32bit glibc-devel-32bit
@ -70,7 +70,7 @@ BuildRequires: glibc-devel
%if %{?with_kmp}0
BuildRequires: kernel-source kernel-syms module-init-tools xorg-x11
%endif
Version: 4.0.1_02
Version: 4.0.1_04
Release: 1
License: GPLv2+
Group: System/Kernel
@ -115,51 +115,84 @@ Patch1: 21235-crashkernel-advanced.patch
Patch2: 21271-x86-cache-flush-global.patch
Patch3: 21301-svm-lmsl.patch
Patch4: 21304-keyhandler-alternative.patch
Patch5: 21406-x86-microcode-quiet.patch
Patch6: 21421-vts-ats-enabling.patch
Patch7: 21526-x86-nehalem-cpuid-mask.patch
Patch8: 21542-amd-erratum-411.patch
Patch9: 21615-dont-save-xen-heap-pages.patch
Patch10: 21627-cpuidle-wrap.patch
Patch11: 21653-xend-mac-addr.patch
Patch12: 21678-xend-mac-fix.patch
Patch13: 21683-vtd-kill-timer-conditional.patch
Patch14: 21723-get-domu-state.patch
Patch15: 21847-pscsi.patch
Patch16: 21866-xenapi.patch
Patch17: 21894-intel-unmask-cpuid.patch
Patch18: 22019-x86-cpuidle-online-check.patch
Patch19: 22045-python27-compat.patch
Patch20: 22051-x86-forced-EOI.patch
Patch21: 22067-x86-irq-domain.patch
Patch22: 22068-vtd-irte-RH-bit.patch
Patch23: 22071-ept-get-entry-lock.patch
Patch24: 22084-x86-xsave-off.patch
Patch25: 7410-qemu-alt-gr.patch
Patch26: 22135-heap-lock.patch
Patch27: 22148-serial-irq-dest.patch
Patch28: 22157-x86-debug-key-i.patch
Patch29: 22159-notify-evtchn-dying.patch
Patch30: 22160-Intel-C6-EOI.patch
Patch31: 22174-x86-pmtimer-accuracy.patch
Patch32: 22175-x86-irq-enter-exit.patch
Patch33: 22177-i386-irq-safe-map_domain_page.patch
Patch34: 22194-tmem-check-pv-mfn.patch
Patch35: 22213-x86-xsave-cpuid-check.patch
Patch36: 22214-x86-msr-misc-enable.patch
Patch37: 22222-x86-timer-extint.patch
Patch38: 22223-vtd-workarounds.patch
Patch39: 22231-x86-pv-ucode-msr-intel.patch
Patch40: 22232-x86-64-lahf-lm-bios-workaround.patch
Patch41: 22235-lxml-validator.patch
Patch42: 22280-kexec.patch
Patch43: 22337-vtd-scan-single-func.patch
Patch44: 22348-vtd-check-secbus-devfn.patch
Patch45: 22369-xend-pci-passthru-fix.patch
Patch46: 22389-amd-iommu-decls.patch
Patch47: 22416-acpi-check-mwait.patch
Patch48: 22417-vpmu-nehalem.patch
Patch49: 22431-p2m-remove-bug-check.patch
Patch5: 21344-4.0-testing-xenstore-fd-leak.patch
Patch6: 21406-x86-microcode-quiet.patch
Patch7: 21421-vts-ats-enabling.patch
Patch8: 21526-x86-nehalem-cpuid-mask.patch
Patch9: 21432-4.0-cpu-boot-failure.patch
Patch10: 21542-amd-erratum-411.patch
Patch11: 21615-dont-save-xen-heap-pages.patch
Patch12: 21627-cpuidle-wrap.patch
Patch13: 21653-xend-mac-addr.patch
Patch14: 21678-xend-mac-fix.patch
Patch15: 21683-vtd-kill-timer-conditional.patch
Patch16: 21723-get-domu-state.patch
Patch17: 21847-pscsi.patch
Patch18: 21866-xenapi.patch
Patch19: 21894-intel-unmask-cpuid.patch
Patch20: 22019-x86-cpuidle-online-check.patch
Patch21: 22045-python27-compat.patch
Patch22: 22051-x86-forced-EOI.patch
Patch23: 22067-x86-irq-domain.patch
Patch24: 22068-vtd-irte-RH-bit.patch
Patch25: 22071-ept-get-entry-lock.patch
Patch26: 22084-x86-xsave-off.patch
Patch27: 22135-heap-lock.patch
Patch28: 22148-serial-irq-dest.patch
Patch29: 22157-x86-debug-key-i.patch
Patch30: 22159-notify-evtchn-dying.patch
Patch31: 22160-Intel-C6-EOI.patch
Patch32: 22174-x86-pmtimer-accuracy.patch
Patch33: 22175-x86-irq-enter-exit.patch
Patch34: 22177-i386-irq-safe-map_domain_page.patch
Patch35: 22194-tmem-check-pv-mfn.patch
Patch36: 21810-x2apic-acpi.patch
Patch37: 21989-x2apic-resume.patch
Patch38: 22213-x86-xsave-cpuid-check.patch
Patch39: 22214-x86-msr-misc-enable.patch
Patch40: 22222-x86-timer-extint.patch
Patch41: 22223-vtd-workarounds.patch
Patch42: 22231-x86-pv-ucode-msr-intel.patch
Patch43: 22232-x86-64-lahf-lm-bios-workaround.patch
Patch44: 22235-lxml-validator.patch
Patch45: 22280-kexec.patch
Patch46: 22337-vtd-scan-single-func.patch
Patch47: 22348-vtd-check-secbus-devfn.patch
Patch48: 22369-xend-pci-passthru-fix.patch
Patch49: 22388-x2apic-panic.patch
Patch50: 22389-amd-iommu-decls.patch
Patch51: 22416-acpi-check-mwait.patch
Patch52: 22417-vpmu-nehalem.patch
Patch53: 22431-p2m-remove-bug-check.patch
Patch54: 22448-x86_64-gdt-ldt-fault-filter.patch
Patch55: 22451-hvm-cap-clobber.patch
Patch56: 22452-x86-irq-migrate-directed-eoi.patch
Patch57: 22466-x86-sis-apic-bug.patch
Patch58: 22470-vlapic-tick-loss.patch
Patch59: 22475-x2apic-cleanup.patch
Patch60: 22484-vlapic-tmcct-periodic.patch
Patch61: 22504-iommu-dom0-holes.patch
Patch62: 22506-x86-iommu-dom0-estimate.patch
Patch63: 22526-ept-access-once.patch
Patch64: 22533-x86-32bit-apicid.patch
Patch65: 22534-x86-max-local-apic.patch
Patch66: 22535-x2apic-preenabled.patch
Patch67: 22538-keyhandler-relax.patch
Patch68: 22540-32on64-hypercall-debug.patch
Patch69: 22549-vtd-map-page-leak.patch
Patch70: 22574-ept-skip-validation.patch
Patch71: 22632-vtd-print-entries.patch
Patch72: 22645-amd-flush-filter.patch
Patch73: 22693-fam10-mmio-conf-base-protect.patch
Patch74: 22694-x86_64-no-weak.patch
Patch75: 22707-x2apic-preenabled-check.patch
Patch76: 22708-xenctx-misc.patch
# Upstream qemu patches
Patch200: 7410-qemu-alt-gr.patch
Patch201: 7426-xenfb-depth.patch
Patch202: 7433-qemu-altgr.patch
Patch203: qemu-fix-7433.patch
Patch204: 7434-qemu-rlimit-as.patch
# Our patches
Patch300: xen-config.diff
Patch301: xend-config.diff
@ -221,6 +254,8 @@ Patch372: popen2-argument-fix.patch
Patch373: usb-list.patch
Patch374: xend-devid-or-name.patch
Patch375: 22326-cpu-pools-numa-placement.patch
Patch376: 20158-revert.patch
Patch377: suspend_evtchn_lock.patch
# Patches for snapshot support
Patch400: snapshot-ioemu-save.patch
Patch401: snapshot-ioemu-restore.patch
@ -251,69 +286,63 @@ Patch433: multi-xvdp.patch
Patch434: check_device_status.patch
Patch435: change_home_server.patch
Patch436: altgr_2.patch
Patch437: xenfb_32bpp.patch
Patch438: stdvga-cache.patch
Patch439: minios-fixups.patch
Patch440: bdrv_default_rwflag.patch
Patch441: blktap2.patch
Patch442: xen-minimum-restart-time.patch
# Jim's domain lock patch
Patch450: xend-domain-lock.patch
# Hypervisor and PV driver Patches
Patch500: 32on64-extra-mem.patch
Patch501: x86_64-gdt-ldt-fault-filter.patch
Patch502: x86-ioapic-ack-default.patch
Patch503: x86-cpufreq-report.patch
Patch504: dump-exec-state.patch
Patch505: dom-print.patch
Patch506: pvdrv-import-shared-info.patch
Patch507: x86-show-page-walk-early.patch
Patch508: x86-extra-trap-info.patch
Patch509: pvdrv_emulation_control.patch
Patch510: blktap-pv-cdrom.patch
Patch511: pv-driver-build.patch
Patch512: supported_module.diff
Patch513: magic_ioport_compat.patch
Patch501: x86-ioapic-ack-default.patch
Patch502: x86-cpufreq-report.patch
Patch503: dump-exec-state.patch
Patch504: dom-print.patch
Patch505: pvdrv-import-shared-info.patch
Patch506: x86-show-page-walk-early.patch
Patch507: x86-extra-trap-info.patch
Patch508: pvdrv_emulation_control.patch
Patch509: blktap-pv-cdrom.patch
Patch510: pv-driver-build.patch
Patch511: supported_module.diff
Patch512: magic_ioport_compat.patch
Patch650: disable_emulated_device.diff
Patch651: ioemu-disable-scsi.patch
# novell_shim patches
Patch700: hv_tools.patch
Patch701: hv_xen_base.patch
Patch702: hv_xen_extension.patch
Patch703: hv_win7_eoi_bug.patch
Patch704: hv_apic.patch
Patch652: ioemu-disable-emulated-ide-if-pv.patch
Patch700: hv_extid_compatibility.patch
# Build patch
Patch999: tmp_build.patch
# FATE 310510
Patch902: hotplug-block-losetup-a.patch
Patch10001: xenpaging.tools_xenpaging_cleanup.patch
Patch10002: xenpaging.pageout_policy.patch
Patch10003: xenpaging.xs_daemon_close.patch
Patch10004: xenpaging.get_paged_frame.patch
Patch10005: xenpaging.makefile.patch
Patch10010: xenpaging.policy_linear.patch
Patch10011: xenpaging.pagefile.patch
Patch10012: xenpaging.xenpaging_init.patch
Patch10013: xenpaging.mem_paging_tool_qemu_flush_cache.patch
Patch10014: xenpaging.machine_to_phys_mapping.patch
Patch10015: xenpaging.populate_only_if_paged.patch
Patch10017: xenpaging.autostart.patch
Patch10018: xenpaging.signal_handling.patch
Patch10019: xenpaging.MRU_SIZE.patch
Patch10020: xenpaging.guest_remove_page.patch
Patch10021: xenpaging.mem_event_check_ring-free_requests.patch
Patch10022: xenpaging.blacklist.patch
Patch10023: xenpaging.autostart_delay.patch
Patch10024: xenpaging.page_already_populated.patch
Patch10025: xenpaging.notify_policy_only_once.patch
Patch10026: xenpaging.num_pages_equal_max_pages.patch
Patch10027: xenpaging.p2m_mem_paging_populate_if_p2m_ram_paged.patch
Patch10028: xenpaging.HVMCOPY_gfn_paged_out.patch
Patch10029: xenpaging.optimize_p2m_mem_paging_populate.patch
Patch10030: xenpaging.paging_prep_enomem.patch
Patch10031: xenpaging.print-arguments.patch
Patch10040: xenpaging.doc.patch
Patch902: hotplug-block-losetup-a.patch
Patch10001: xenpaging.tools_xenpaging_cleanup.patch
Patch10002: xenpaging.pageout_policy.patch
Patch10003: xenpaging.get_paged_frame.patch
Patch10004: xenpaging.makefile.patch
Patch10010: xenpaging.policy_linear.patch
Patch10011: xenpaging.pagefile.patch
Patch10012: xenpaging.xenpaging_init.patch
Patch10013: xenpaging.mem_paging_tool_qemu_flush_cache.patch
Patch10014: xenpaging.machine_to_phys_mapping.patch
Patch10015: xenpaging.populate_only_if_paged.patch
Patch10017: xenpaging.autostart.patch
Patch10018: xenpaging.signal_handling.patch
Patch10019: xenpaging.MRU_SIZE.patch
Patch10020: xenpaging.guest_remove_page.patch
Patch10021: xenpaging.mem_event_check_ring-free_requests.patch
Patch10022: xenpaging.blacklist.patch
Patch10023: xenpaging.autostart_delay.patch
Patch10024: xenpaging.page_already_populated.patch
Patch10025: xenpaging.notify_policy_only_once.patch
Patch10026: xenpaging.num_pages_equal_max_pages.patch
Patch10027: xenpaging.p2m_mem_paging_populate_if_p2m_ram_paged.patch
Patch10028: xenpaging.HVMCOPY_gfn_paged_out.patch
Patch10029: xenpaging.optimize_p2m_mem_paging_populate.patch
Patch10030: xenpaging.paging_prep_enomem.patch
Patch10031: xenpaging.print-arguments.patch
Patch10040: xenpaging.doc.patch
# xenalyze
Patch20000: xenalyze.64bit.patch
Patch20000: xenalyze.64bit.patch
Url: http://www.cl.cam.ac.uk/Research/SRG/netos/xen/
BuildRoot: %{_tmppath}/%{name}-%{version}-build
#%define pysite %(python -c "import distutils.sysconfig; print distutils.sysconfig.get_python_lib()")
@ -697,6 +726,38 @@ tar xfj %{SOURCE2} -C $RPM_BUILD_DIR/%{xen_build_dir}/tools
%patch47 -p1
%patch48 -p1
%patch49 -p1
%patch50 -p1
%patch51 -p1
%patch52 -p1
%patch53 -p1
%patch54 -p1
%patch55 -p1
%patch56 -p1
%patch57 -p1
%patch58 -p1
%patch59 -p1
%patch60 -p1
%patch61 -p1
%patch62 -p1
%patch63 -p1
%patch64 -p1
%patch65 -p1
%patch66 -p1
%patch67 -p1
%patch68 -p1
%patch69 -p1
%patch70 -p1
%patch71 -p1
%patch72 -p1
%patch73 -p1
%patch74 -p1
%patch75 -p1
%patch76 -p1
%patch200 -p1
%patch201 -p1
%patch202 -p1
%patch203 -p1
%patch204 -p1
%patch300 -p1
%patch301 -p1
%patch302 -p1
@ -756,6 +817,8 @@ tar xfj %{SOURCE2} -C $RPM_BUILD_DIR/%{xen_build_dir}/tools
%patch373 -p1
%patch374 -p1
%patch375 -p1
%patch376 -p1
#%patch377 -p1
%patch400 -p1
%patch401 -p1
%patch402 -p1
@ -783,11 +846,11 @@ tar xfj %{SOURCE2} -C $RPM_BUILD_DIR/%{xen_build_dir}/tools
%patch434 -p1
%patch435 -p1
%patch436 -p1
%patch437 -p1
%patch438 -p1
%patch439 -p1
%patch440 -p1
%patch441 -p1
%patch442 -p1
%patch450 -p1
%patch500 -p1
%patch501 -p1
@ -802,21 +865,16 @@ tar xfj %{SOURCE2} -C $RPM_BUILD_DIR/%{xen_build_dir}/tools
%patch510 -p1
%patch511 -p1
%patch512 -p1
%patch513 -p1
%patch650 -p1
%patch651 -p1
%patch652 -p1
%patch700 -p1
%patch701 -p1
%patch702 -p1
%patch703 -p1
%patch704 -p1
%patch999 -p1
%patch902 -p1
%patch10001 -p1
%patch10002 -p1
%patch10003 -p1
%patch10004 -p1
%patch10005 -p1
%patch10010 -p1
%patch10011 -p1
%patch10012 -p1
@ -842,6 +900,7 @@ tar xfj %{SOURCE2} -C $RPM_BUILD_DIR/%{xen_build_dir}/tools
#
%patch20000 -p1
%build
XEN_EXTRAVERSION=%version-%release
XEN_EXTRAVERSION=${XEN_EXTRAVERSION#%{xvers}}

11
xend-domain-lock.patch
View File

@ -94,7 +94,7 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
XendTask.log_progress(0, 30, self._constructDomain)
XendTask.log_progress(31, 60, self._initDomain)
@@ -2990,6 +2992,11 @@ class XendDomainInfo:
@@ -2989,6 +2991,11 @@ class XendDomainInfo:
self._stateSet(DOM_STATE_HALTED)
self.domid = None # Do not push into _stateSet()!
@ -106,7 +106,7 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
finally:
self.refresh_shutdown_lock.release()
@@ -4508,6 +4515,74 @@ class XendDomainInfo:
@@ -4498,6 +4505,74 @@ class XendDomainInfo:
def has_device(self, dev_class, dev_uuid):
return (dev_uuid in self.info['%s_refs' % dev_class.lower()])
@ -232,15 +232,16 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendCheckpoint.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/XendCheckpoint.py
+++ xen-4.0.1-testing/tools/python/xen/xend/XendCheckpoint.py
@@ -133,6 +133,7 @@ def save(fd, dominfo, network, live, dst
@@ -133,6 +133,8 @@ def save(fd, dominfo, network, live, dst
dominfo.shutdown('suspend')
dominfo.waitForSuspend()
if line in ('suspend', 'suspended'):
+ dominfo.release_running_lock(domain_name)
+ if checkpoint == False:
+ dominfo.release_running_lock(domain_name)
dominfo.migrateDevices(network, dst, DEV_MIGRATE_STEP2,
domain_name)
log.info("Domain %d suspended.", dominfo.getDomid())
@@ -410,6 +411,7 @@ def restore(xd, fd, dominfo = None, paus
@@ -410,6 +412,7 @@ def restore(xd, fd, dominfo = None, paus
if not paused:
dominfo.unpause()

41
xenfb_32bpp.patch
View File

@ -1,41 +0,0 @@
Fix problem that press CTRL+ALT+2 and then CTRL+ALT+1 back to VM graphic console, screen cannot be restored.
xenfb_guest_copy only handles xenfb->depth=8 and 24 cases, because it assumes in xenfb->depth=16 or 32 cases,
buffer is shared, like in xenfb_update ==> qemu_create_displaysurface_from() does. But as the mentioned bug,
it didn't enter the xenfb_update do_resize hunk, buffer is not shared.
To solve that problem, either make it enter the xenfb_update do_resize hunk, or change xenfb_guest_copy(), let
it handle all cases. This patch is to let xenfb_guest_copy handle xenfb->depth=32 case.
diff -r e4f337bb97f7 tools/ioemu-qemu-xen/hw/xenfb.c
--- a/tools/ioemu-qemu-xen/hw/xenfb.c Wed Oct 20 19:39:28 2010 +0800
+++ b/tools/ioemu-qemu-xen/hw/xenfb.c Wed Oct 20 21:42:37 2010 +0800
@@ -612,6 +612,12 @@
uint8_t *data = ds_get_data(xenfb->c.ds);
if (!is_buffer_shared(xenfb->c.ds->surface)) {
+ if (xenfb->depth == bpp) {
+ for (line = y; line < (y+h); line++) {
+ memcpy (data + (line * linesize) + (x * bpp / 8), xenfb->pixels + xenfb->offset + (line * xenfb->row_stride) + (x * xenfb->depth / 8), w * xenfb->depth / 8);
+ }
+ }
+ else{
switch (xenfb->depth) {
case 8:
if (bpp == 16) {
@@ -631,9 +637,17 @@
oops = 1;
}
break;
+ case 32:
+ if (bpp == 16) {
+ BLT(uint32_t, uint16_t, 8, 8, 8, 5, 6, 5);
+ } else {
+ oops = 1;
+ }
+ break;
default:
oops = 1;
}
+ }
}
if (oops) /* should not happen */
xen_be_printf(&xenfb->c.xendev, 0, "%s: oops: convert %d -> %d bpp?\n",

33
xenpaging.HVMCOPY_gfn_paged_out.patch
View File

@ -13,12 +13,14 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
---
xen/arch/x86/hvm/hvm.c | 4 ++++
xen/common/memory.c | 44 +++++++++++++++++++++++++++++++++++++++-----
2 files changed, 43 insertions(+), 5 deletions(-)
xen/common/memory.c | 39 ++++++++++++++++++++++++++++++++++-----
2 files changed, 38 insertions(+), 5 deletions(-)
Index: xen-4.0.1-testing/xen/arch/x86/hvm/hvm.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/hvm/hvm.c
+++ xen-4.0.1-testing/xen/arch/x86/hvm/hvm.c
@@ -1853,6 +1853,8 @@ unsigned long copy_to_user_hvm(void *to,
@@ -1843,6 +1843,8 @@ unsigned long copy_to_user_hvm(void *to,
rc = hvm_copy_to_guest_virt_nofault((unsigned long)to, (void *)from,
len, 0);
@ -27,7 +29,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
return rc ? len : 0; /* fake a copy_to_user() return code */
}
@@ -1869,6 +1871,8 @@ unsigned long copy_from_user_hvm(void *t
@@ -1859,6 +1861,8 @@ unsigned long copy_from_user_hvm(void *t
#endif
rc = hvm_copy_from_guest_virt_nofault(to, (unsigned long)from, len, 0);
@ -36,6 +38,8 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
return rc ? len : 0; /* fake a copy_from_user() return code */
}
Index: xen-4.0.1-testing/xen/common/memory.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/memory.c
+++ xen-4.0.1-testing/xen/common/memory.c
@@ -47,6 +47,7 @@ static void increase_reservation(struct
@ -46,7 +50,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
xen_pfn_t mfn;
struct domain *d = a->domain;
@@ -80,8 +81,14 @@ static void increase_reservation(struct
@@ -80,8 +81,13 @@ static void increase_reservation(struct
if ( !guest_handle_is_null(a->extent_list) )
{
mfn = page_to_mfn(page);
@ -54,7 +58,6 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
+ ctg_ret = __copy_to_guest_offset(a->extent_list, i, &mfn, 1);
+ if ( unlikely(ctg_ret) )
+ {
+ free_domheap_pages(page, a->extent_order);
+ if ( (long)ctg_ret == -EAGAIN )
+ a->preempted = 1;
goto out;
@ -62,7 +65,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
}
}
@@ -93,6 +100,7 @@ static void populate_physmap(struct memo
@@ -93,6 +99,7 @@ static void populate_physmap(struct memo
{
struct page_info *page;
unsigned long i, j;
@ -70,7 +73,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
xen_pfn_t gpfn, mfn;
struct domain *d = a->domain;
@@ -111,8 +119,13 @@ static void populate_physmap(struct memo
@@ -111,8 +118,13 @@ static void populate_physmap(struct memo
goto out;
}
@ -85,7 +88,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
if ( a->memflags & MEMF_populate_on_demand )
{
@@ -142,8 +155,17 @@ static void populate_physmap(struct memo
@@ -142,8 +154,13 @@ static void populate_physmap(struct memo
set_gpfn_from_mfn(mfn + j, gpfn + j);
/* Inform the domain of the new page's machine address. */
@ -93,10 +96,6 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
+ cftg_ret = __copy_to_guest_offset(a->extent_list, i, &mfn, 1);
+ if ( unlikely(cftg_ret) )
+ {
+ for ( j = 0; j < (1 << a->extent_order); j++ )
+ set_gpfn_from_mfn(mfn + j, INVALID_M2P_ENTRY);
+ guest_physmap_remove_page(d, gpfn, mfn, a->extent_order);
+ free_domheap_pages(page, a->extent_order);
+ if ( (long)cftg_ret == -EAGAIN )
+ a->preempted = 1;
goto out;
@ -104,7 +103,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
}
}
}
@@ -212,6 +234,7 @@ int guest_remove_page(struct domain *d,
@@ -212,6 +229,7 @@ int guest_remove_page(struct domain *d,
static void decrease_reservation(struct memop_args *a)
{
unsigned long i, j;
@ -112,7 +111,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
xen_pfn_t gmfn;
if ( !guest_handle_subrange_okay(a->extent_list, a->nr_done,
@@ -226,8 +249,13 @@ static void decrease_reservation(struct
@@ -226,8 +244,13 @@ static void decrease_reservation(struct
goto out;
}
@ -127,7 +126,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
if ( tb_init_done )
{
@@ -511,6 +539,7 @@ long do_memory_op(unsigned long cmd, XEN
@@ -511,6 +534,7 @@ long do_memory_op(unsigned long cmd, XEN
int rc, op;
unsigned int address_bits;
unsigned long start_extent;
@ -135,7 +134,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
struct xen_memory_reservation reservation;
struct memop_args args;
domid_t domid;
@@ -524,8 +553,13 @@ long do_memory_op(unsigned long cmd, XEN
@@ -524,8 +548,13 @@ long do_memory_op(unsigned long cmd, XEN
case XENMEM_populate_physmap:
start_extent = cmd >> MEMOP_EXTENT_SHIFT;

2
xenpaging.MRU_SIZE.patch
View File

@ -4,6 +4,8 @@ Increase recently used pages from 4MB to 64MB.
Keeping more pages in memory allows the guest to make more progress if the
paging file spans the entire guest memory.
(xen-unstable changeset: 22447:aba70e59a90d)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---

8
xenpaging.autostart.patch
View File

@ -81,20 +81,20 @@ Index: xen-4.0.1-testing/tools/python/xen/xend/XendConfig.py
+ self['platform']['xenpaging'] = None
if 'timer_mode' not in self['platform']:
self['platform']['timer_mode'] = 1
if 'viridian' not in self['platform']:
if 'extid' in self['platform'] and int(self['platform']['extid']) == 1:
Index: xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
===================================================================
--- xen-4.0.1-testing.orig/tools/python/xen/xend/XendDomainInfo.py
+++ xen-4.0.1-testing/tools/python/xen/xend/XendDomainInfo.py
@@ -2442,6 +2442,7 @@ class XendDomainInfo:
@@ -2441,6 +2441,7 @@ class XendDomainInfo:
if self.image:
self.image.createDeviceModel()
self.image.createDeviceModel(resume)
+ self.image.createXenPaging()
#if have pass-through devs, need the virtual pci slots info from qemu
self.pci_device_configure_boot()
@@ -2454,6 +2455,11 @@ class XendDomainInfo:
@@ -2453,6 +2454,11 @@ class XendDomainInfo:
self.image.destroyDeviceModel()
except Exception, e:
log.exception("Device model destroy failed %s" % str(e))

14
xenpaging.guest_remove_page.patch
View File

@ -15,6 +15,8 @@ v2:
xen/include/public/mem_event.h | 1
5 files changed, 73 insertions(+), 20 deletions(-)
Index: xen-4.0.1-testing/tools/xenpaging/xenpaging.c
===================================================================
--- xen-4.0.1-testing.orig/tools/xenpaging/xenpaging.c
+++ xen-4.0.1-testing/tools/xenpaging/xenpaging.c
@@ -600,12 +600,19 @@ int main(int argc, char *argv[])
@ -42,6 +44,8 @@ v2:
}
/* Prepare the response */
Index: xen-4.0.1-testing/xen/arch/x86/mm/p2m.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/mm/p2m.c
+++ xen-4.0.1-testing/xen/arch/x86/mm/p2m.c
@@ -2000,12 +2000,15 @@ p2m_remove_page(struct domain *d, unsign
@ -65,7 +69,7 @@ v2:
}
set_p2m_entry(d, gfn, _mfn(INVALID_MFN), page_order, p2m_invalid);
}
@@ -2532,6 +2535,35 @@ int p2m_mem_paging_evict(struct domain *
@@ -2529,6 +2532,35 @@ int p2m_mem_paging_evict(struct domain *
return 0;
}
@ -101,7 +105,7 @@ v2:
void p2m_mem_paging_populate(struct domain *d, unsigned long gfn)
{
struct vcpu *v = current;
@@ -2596,17 +2628,20 @@ void p2m_mem_paging_resume(struct domain
@@ -2593,17 +2625,20 @@ void p2m_mem_paging_resume(struct domain
/* Pull the response off the ring */
mem_event_get_response(d, &rsp);
@ -132,6 +136,8 @@ v2:
}
/* Unpause domain */
Index: xen-4.0.1-testing/xen/common/memory.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/memory.c
+++ xen-4.0.1-testing/xen/common/memory.c
@@ -162,6 +162,12 @@ int guest_remove_page(struct domain *d,
@ -147,6 +153,8 @@ v2:
#else
mfn = gmfn_to_mfn(d, gmfn);
#endif
Index: xen-4.0.1-testing/xen/include/asm-x86/p2m.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/asm-x86/p2m.h
+++ xen-4.0.1-testing/xen/include/asm-x86/p2m.h
@@ -441,6 +441,8 @@ int set_shared_p2m_entry(struct domain *
@ -167,6 +175,8 @@ v2:
static inline void p2m_mem_paging_populate(struct domain *d, unsigned long gfn)
{ }
#endif
Index: xen-4.0.1-testing/xen/include/public/mem_event.h
===================================================================
--- xen-4.0.1-testing.orig/xen/include/public/mem_event.h
+++ xen-4.0.1-testing/xen/include/public/mem_event.h
@@ -37,6 +37,7 @@

6
xenpaging.machine_to_phys_mapping.patch
View File

@ -37,9 +37,11 @@ v2:
xen/common/page_alloc.c | 9 +++++++++
2 files changed, 19 insertions(+), 3 deletions(-)
Index: xen-4.0.1-testing/xen/arch/x86/mm/p2m.c
===================================================================
--- xen-4.0.1-testing.orig/xen/arch/x86/mm/p2m.c
+++ xen-4.0.1-testing/xen/arch/x86/mm/p2m.c
@@ -2598,9 +2598,16 @@ void p2m_mem_paging_resume(struct domain
@@ -2595,9 +2595,16 @@ void p2m_mem_paging_resume(struct domain
/* Fix p2m entry */
mfn = gfn_to_mfn(d, rsp.gfn, &p2mt);
@ -59,6 +61,8 @@ v2:
/* Unpause domain */
if ( rsp.flags & MEM_EVENT_FLAG_VCPU_PAUSED )
Index: xen-4.0.1-testing/xen/common/page_alloc.c
===================================================================
--- xen-4.0.1-testing.orig/xen/common/page_alloc.c
+++ xen-4.0.1-testing/xen/common/page_alloc.c
@@ -1178,9 +1178,18 @@ void free_domheap_pages(struct page_info

2
xenpaging.mem_event_check_ring-free_requests.patch
View File

@ -3,6 +3,8 @@ Subject: xenpaging: print info when free request slots drop below 2
Add debugging aid to free request slots in the ring buffer.
It should not happen that the ring gets full, print info anyway if it happens.
(xen-unstable changeset: 22439:3bbb3969236c)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---

4
xenpaging.mem_paging_tool_qemu_flush_cache.patch
View File

@ -15,9 +15,11 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
tools/ioemu-qemu-xen/xenstore.c | 3 +++
1 file changed, 3 insertions(+)
Index: xen-4.0.1-testing/tools/ioemu-qemu-xen/xenstore.c
===================================================================
--- xen-4.0.1-testing.orig/tools/ioemu-qemu-xen/xenstore.c
+++ xen-4.0.1-testing/tools/ioemu-qemu-xen/xenstore.c
@@ -1021,6 +1021,9 @@ static void xenstore_process_dm_command_
@@ -1025,6 +1025,9 @@ static void xenstore_process_dm_command_
do_pci_add(par);
free(par);
#endif

10
xenpaging.notify_policy_only_once.patch
View File

@ -8,6 +8,8 @@ other already resumed pages too early. In the worst case, a page that
was just resumed can be evicted right away, causing a deadlock in the
guest.
(xen-unstable changeset: 22441:7d2c013727d7)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---
@ -16,7 +18,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
--- xen-4.0.1-testing.orig/tools/xenpaging/xenpaging.c
+++ xen-4.0.1-testing/tools/xenpaging/xenpaging.c
@@ -377,7 +377,7 @@ int xenpaging_evict_page(xenpaging_t *pa
@@ -380,7 +380,7 @@ int xenpaging_evict_page(xenpaging_t *pa
return ret;
}
@ -25,7 +27,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
{
int ret;
@@ -387,7 +387,8 @@ static int xenpaging_resume_page(xenpagi
@@ -390,7 +390,8 @@ static int xenpaging_resume_page(xenpagi
goto out;
/* Notify policy of page being paged in */
@ -35,7 +37,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
/* Tell Xen page is ready */
ret = xc_mem_paging_resume(paging->xc_handle, paging->mem_event.domain_id,
@@ -619,7 +620,7 @@ int main(int argc, char *argv[])
@@ -621,7 +622,7 @@ int main(int argc, char *argv[])
rsp.vcpu_id = req.vcpu_id;
rsp.flags = req.flags;
@ -44,7 +46,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
if ( rc != 0 )
{
ERROR("Error resuming page");
@@ -648,7 +649,7 @@ int main(int argc, char *argv[])
@@ -650,7 +651,7 @@ int main(int argc, char *argv[])
rsp.vcpu_id = req.vcpu_id;
rsp.flags = req.flags;

6
xenpaging.num_pages_equal_max_pages.patch
View File

@ -4,6 +4,8 @@ Simplify paging size argument. If a negative number is specified, it
means the entire guest memory should be paged out. This is useful for
debugging. Also limit num_pages to the guests max_pages.
(xen-unstable changeset: 22442:1793318b35e6)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---
@ -12,7 +14,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
--- xen-4.0.1-testing.orig/tools/xenpaging/xenpaging.c
+++ xen-4.0.1-testing/tools/xenpaging/xenpaging.c
@@ -512,8 +512,6 @@ int main(int argc, char *argv[])
@@ -514,8 +514,6 @@ int main(int argc, char *argv[])
domain_id = atoi(argv[1]);
num_pages = atoi(argv[2]);
@ -21,7 +23,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
/* Seed random-number generator */
srand(time(NULL));
@@ -534,6 +532,13 @@ int main(int argc, char *argv[])
@@ -536,6 +534,13 @@ int main(int argc, char *argv[])
return 2;
}

6
xenpaging.optimize_p2m_mem_paging_populate.patch
View File

@ -7,6 +7,8 @@ does not need to be paused, p2m_mem_paging_resume will do nothing with
the request. And also xenpaging will drop the request if the vcpu does
not need a wakeup.
(xen-unstable changeset: 22444:7fe9cad00e15)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---
@ -15,7 +17,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
--- xen-4.0.1-testing.orig/xen/arch/x86/mm/p2m.c
+++ xen-4.0.1-testing/xen/arch/x86/mm/p2m.c
@@ -2571,12 +2571,12 @@ void p2m_mem_paging_populate(struct doma
@@ -2567,12 +2567,12 @@ void p2m_mem_paging_populate(struct doma
mem_event_request_t req;
p2m_type_t p2mt;
@ -30,7 +32,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
/* Fix p2m mapping */
/* XXX: It seems inefficient to have this here, as it's only needed
* in one case (ept guest accessing paging out page) */
@@ -2594,6 +2594,11 @@ void p2m_mem_paging_populate(struct doma
@@ -2590,6 +2590,11 @@ void p2m_mem_paging_populate(struct doma
vcpu_pause_nosync(v);
req.flags |= MEM_EVENT_FLAG_VCPU_PAUSED;
}

4
xenpaging.p2m_mem_paging_populate_if_p2m_ram_paged.patch
View File

@ -10,6 +10,8 @@ process of being still paged-out or already paged-in.
In fact, p2m state p2m_ram_paged is the only state where the mfn type
can be invalidated.
(xen-unstable changeset: 22443:48b10f9a436e)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---
@ -18,7 +20,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
--- xen-4.0.1-testing.orig/xen/arch/x86/mm/p2m.c
+++ xen-4.0.1-testing/xen/arch/x86/mm/p2m.c
@@ -2581,7 +2581,7 @@ void p2m_mem_paging_populate(struct doma
@@ -2577,7 +2577,7 @@ void p2m_mem_paging_populate(struct doma
/* XXX: It seems inefficient to have this here, as it's only needed
* in one case (ept guest accessing paging out page) */
gfn_to_mfn(d, gfn, &p2mt);

4
xenpaging.page_already_populated.patch
View File

@ -2,6 +2,8 @@ Subject: xenpaging: print p2mt for already paged-in pages
Add more debug output, print p2mt for pages which were requested more than once.
(xen-unstable changeset: 22440:1b87fbd10f43)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---
@ -10,7 +12,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
--- xen-4.0.1-testing.orig/tools/xenpaging/xenpaging.c
+++ xen-4.0.1-testing/tools/xenpaging/xenpaging.c
@@ -632,8 +632,10 @@ int main(int argc, char *argv[])
@@ -634,8 +634,10 @@ int main(int argc, char *argv[])
else
{
DPRINTF("page already populated (domain = %d; vcpu = %d;"

2
xenpaging.pagefile.patch
View File

@ -4,6 +4,8 @@ Open paging file only if xenpaging_init() succeeds. It can fail if the host
does not support the required virtualization features such as EPT or if
xenpaging was already started for this domain_id.
(xen-unstable changeset: 22435:7818ea9045fd)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Already-Acked-by: Patrick Colp <pjcolp@cs.ubc.ca>
Already-Acked-by: Keir Fraser <keir.fraser@citrix.com>

4
xenpaging.paging_prep_enomem.patch
View File

@ -6,6 +6,8 @@ now such condition would stall the guest because the requested page will
not come back, xenpaging simply exits. So xenpaging could very well
retry the allocation forever to rescue the guest.
(xen-unstable changeset: 22446:08158f001f19)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---
@ -64,7 +66,7 @@ Signed-off-by: Olaf Hering <olaf@aepfle.de>
*gfn = _gfn;
--- xen-4.0.1-testing.orig/xen/arch/x86/mm/p2m.c
+++ xen-4.0.1-testing/xen/arch/x86/mm/p2m.c
@@ -2614,7 +2614,7 @@ int p2m_mem_paging_prep(struct domain *d
@@ -2611,7 +2611,7 @@ int p2m_mem_paging_prep(struct domain *d
/* Get a free page */
page = alloc_domheap_page(d, 0);
if ( unlikely(page == NULL) )

2
xenpaging.policy_linear.patch
View File

@ -8,6 +8,8 @@ While doing that, xenpaging got into an endless loop because some pages
cant be paged out right away. Now the policy reports an error if the gfn
number wraps.
(xen-unstable changeset: 22434:0d0a18cd416f)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Already-Acked-by: Patrick Colp <pjcolp@cs.ubc.ca>
Already-Acked-by: Keir Fraser <keir.fraser@citrix.com>

2
xenpaging.populate_only_if_paged.patch
View File

@ -6,6 +6,8 @@ will be stopped and later resumed once the page content is usable again.
This matches other p2m_mem_paging_populate usage in the source tree.
(xen-unstable changeset: 22437:9a9bcf399856)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---

2
xenpaging.print-arguments.patch
View File

@ -3,6 +3,8 @@ Subject: xenpaging: print xenpaging cmdline options
Print xenpaging arguments to simplify domain_id mapping from xenpaging
logfile to other logfiles and Xen console output.
(xen-unstable changeset: 22445:c7b08fc1cb8d)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---

2
xenpaging.signal_handling.patch
View File

@ -3,6 +3,8 @@ Subject: xenpaging: add signal handling
Leave paging loop if xenpaging gets a signal.
Remove paging file on exit.
(xen-unstable changeset: 22438:d622cf8c372e)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---

1
xenpaging.tools_xenpaging_cleanup.patch
View File

@ -4,6 +4,7 @@ This isn't directly related to EPT checking, but does some general fix-ups
to the xenpaging code (adds some extra frees, etc.)
(xen-unstable changeset: 21890:3263d0ff9476)
Already-Signed-off-by: Patrick Colp <pjcolp@cs.ubc.ca>
Signed-off-by: Olaf Hering <olaf@aepfle.de>

5
xenpaging.xenpaging_init.patch
View File

@ -3,6 +3,9 @@ Subject: xenpaging: allow only one xenpaging binary per guest
Make sure only one xenpaging binary is active per domain.
Print info when the host lacks the required features for xenpaging.
(xen-unstable changeset: 22436:b153d3b71504)
(xen-unstable changeset: 22511:fdf95845c2f8)
Signed-off-by: Olaf Hering <olaf@aepfle.de>
Already-Acked-by: Patrick Colp <pjcolp@cs.ubc.ca>
Already-Acked-by: Keir Fraser <keir.fraser@citrix.com>
@ -29,7 +32,7 @@ v2: use perror for default case
+ ERROR("EPT not supported for this guest");
+ break;
+ default:
+ perror("Error initialising shared page");
+ ERROR("Error initialising shared page: %s", strerror(errno));
+ break;
+ }
goto err;

22
xenpaging.xs_daemon_close.patch
View File

@ -1,22 +0,0 @@
Subject: xenpaging/xenstore: fix fd leak in xenstore
Missing from commit 'libxl: Backported stuff from unstable'
Without this change, xs_daemon_open/xs_daemon_close will leak filedescriptors.
Signed-off-by: Olaf Hering <olaf@aepfle.de>
---
tools/xenstore/xs.c | 2 ++
1 file changed, 2 insertions(+)
--- xen-4.0.1-testing.orig/tools/xenstore/xs.c
+++ xen-4.0.1-testing/tools/xenstore/xs.c
@@ -285,6 +285,8 @@ void xs_daemon_close(struct xs_handle *h
mutex_unlock(&h->request_mutex);
mutex_unlock(&h->reply_mutex);
mutex_unlock(&h->watch_mutex);
+
+ close_fds_free(h);
}
static bool read_all(int fd, void *data, unsigned int len)