[PATCH 00/28] Drivers: hv: vmbus

[PATCH 00/28] Drivers: hv: vmbus

[K. Y. Srinivasan]

Miscellaneous fixes/improvements. The patches that were sent over the last
couple of months have been consolidated in this submission. All review
comments have been addressed.

Alex Ng (1):
  Drivers: hv: balloon: Enable dynamic memory protocol negotiation with
    Windows 10 hosts

Christopher Oo (1):
  Drivers: hv_vmbus: Fix signal to host condition

Dan Carpenter (1):
  hv: util: checking the wrong variable

Denis V. Lunev (1):
  mshyperv: fix recognition of Hyper-V guest crash MSR's

Dexuan Cui (4):
  Drivers: hv: vmbus: Further improve CPU affiliation logic
  Drivers: hv: vmbus: add a sysfs attr to show the binding of
    channel/VP
  tools: hv: add a python script lsvmbus to list VMBus devices
  Drivers: hv: vmbus: document the VMBus sysfs files

K. Y. Srinivasan (4):
  Drivers: hv: vmbus: Permit sending of packets without payload
  Drivers: hv: vmbus: Consider ND NIC in binding channels to CPUs
  Drivers: hv: vmbus: Improve the CPU affiliation for channels
  Drivers: hv: vmbus: Implement a clocksource based on the TSC page

Nik Nyby (1):
  Drivers: hv: vmbus: fix typo in hv_port_info struct

Viresh Kumar (1):
  drivers/hv: Migrate to new 'set-state' interface

Vitaly Kuznetsov (12):
  Drivers: hv: vmbus: remove hv_synic_free_cpu() call from
    hv_synic_cleanup()
  kexec: define kexec_in_progress in !CONFIG_KEXEC case
  Drivers: hv: vmbus: add special kexec handler
  Drivers: hv: don't do hypercalls when hypercall_page is NULL
  Drivers: hv: vmbus: add special crash handler
  Drivers: hv: vmbus: prefer 'die' notification chain to 'panic'
  Drivers: hv: kvp: check kzalloc return value
  Drivers: hv: fcopy: dynamically allocate smsg_out in
    fcopy_send_data()
  Drivers: hv: vmbus: don't send CHANNELMSG_UNLOAD on pre-Win2012R2
    hosts
  cpu-hotplug: convert cpu_hotplug_disabled to a counter
  cpu-hotplug: export cpu_hotplug_enable/cpu_hotplug_disable
  Drivers: hv: vmbus: use cpu_hotplug_enable/disable

jakeo@microsoft.com (2):
  drivers:hv: Modify hv_vmbus to search for all MMIO ranges available.
  drivers:hv: Move MMIO range picking from hyper_fb to hv_vmbus

 Documentation/ABI/stable/sysfs-bus-vmbus       |   29 ++
 Documentation/power/suspend-and-cpuhotplug.txt |    6 +-
 MAINTAINERS                                    |    1 +
 arch/x86/include/asm/mshyperv.h                |    5 +
 arch/x86/include/uapi/asm/hyperv.h             |    2 +
 arch/x86/kernel/cpu/mshyperv.c                 |   47 ++++
 drivers/hv/channel.c                           |    4 +-
 drivers/hv/channel_mgmt.c                      |   34 ++-
 drivers/hv/hv.c                                |  152 ++++++++---
 drivers/hv/hv_balloon.c                        |   26 ++-
 drivers/hv/hv_fcopy.c                          |   21 +-
 drivers/hv/hv_kvp.c                            |    3 +
 drivers/hv/hv_utils_transport.c                |    2 +-
 drivers/hv/hyperv_vmbus.h                      |   16 +-
 drivers/hv/ring_buffer.c                       |   14 +-
 drivers/hv/vmbus_drv.c                         |  353 +++++++++++++++++++-----
 drivers/video/fbdev/hyperv_fb.c                |   46 ++--
 include/linux/hyperv.h                         |    7 +-
 include/linux/kexec.h                          |    1 +
 kernel/cpu.c                                   |   24 +-
 tools/hv/lsvmbus                               |  101 +++++++
 21 files changed, 718 insertions(+), 176 deletions(-)
 create mode 100644 Documentation/ABI/stable/sysfs-bus-vmbus
 create mode 100644 tools/hv/lsvmbus

-- 
1.7.4.1

[PATCH 01/28] Drivers: hv: vmbus: remove hv_synic_free_cpu() call from hv_synic_cleanup()

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

We already have hv_synic_free() which frees all per-cpu pages for all
CPUs, let's remove the hv_synic_free_cpu() call from hv_synic_cleanup()
so it will be possible to do separate cleanup (writing to MSRs) and final
freeing. This is going to be used to assist kexec.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/hv.c        |    2 --
 drivers/hv/vmbus_drv.c |    1 +
 2 files changed, 1 insertions(+), 2 deletions(-)

diff --git a/drivers/hv/hv.c b/drivers/hv/hv.c
index d3943bc..5b87042 100644
--- a/drivers/hv/hv.c
+++ b/drivers/hv/hv.c
@@ -530,6 +530,4 @@ void hv_synic_cleanup(void *arg)
 	rdmsrl(HV_X64_MSR_SCONTROL, sctrl.as_uint64);
 	sctrl.enable = 0;
 	wrmsrl(HV_X64_MSR_SCONTROL, sctrl.as_uint64);
-
-	hv_synic_free_cpu(cpu);
 }
diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index cf20400..00d5158 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -1120,6 +1120,7 @@ static void __exit vmbus_exit(void)
 		tasklet_kill(hv_context.event_dpc[cpu]);
 		smp_call_function_single(cpu, hv_synic_cleanup, NULL, 1);
 	}
+	hv_synic_free();
 	acpi_bus_unregister_driver(&vmbus_acpi_driver);
 	hv_cpu_hotplug_quirk(false);
 }
-- 
1.7.4.1

[PATCH 02/28] kexec: define kexec_in_progress in !CONFIG_KEXEC case

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

If some piece of code wants to check kexec_in_progress it has to be put
in #ifdef CONFIG_KEXEC block to not break the build in !CONFIG_KEXEC
case. Overcome this limitation by defining kexec_in_progress to false.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 include/linux/kexec.h |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/include/linux/kexec.h b/include/linux/kexec.h
index e804306..b63218f 100644
--- a/include/linux/kexec.h
+++ b/include/linux/kexec.h
@@ -323,6 +323,7 @@ struct pt_regs;
 struct task_struct;
 static inline void crash_kexec(struct pt_regs *regs) { }
 static inline int kexec_should_crash(struct task_struct *p) { return 0; }
+#define kexec_in_progress false
 #endif /* CONFIG_KEXEC */
 
 #endif /* !defined(__ASSEBMLY__) */
-- 
1.7.4.1

[PATCH 03/28] Drivers: hv: vmbus: add special kexec handler

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

When general-purpose kexec (not kdump) is being performed in Hyper-V guest
the newly booted kernel fails with an MCE error coming from the host. It
is the same error which was fixed in the "Drivers: hv: vmbus: Implement
the protocol for tearing down vmbus state" commit - monitor pages remain
special and when they're being written to (as the new kernel doesn't know
these pages are special) bad things happen. We need to perform some
minimalistic cleanup before booting a new kernel on kexec. To do so we
need to register a special machine_ops.shutdown handler to be executed
before the native_machine_shutdown(). Registering a shutdown notification
handler via the register_reboot_notifier() call is not sufficient as it
happens to early for our purposes. machine_ops is not being exported to
modules (and I don't think we want to export it) so let's do this in
mshyperv.c

The minimalistic cleanup consists of cleaning up clockevents, synic MSRs,
guest os id MSR, and hypercall MSR.

Kdump doesn't require all this stuff as it lives in a separate memory
space.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 arch/x86/include/asm/mshyperv.h |    2 ++
 arch/x86/kernel/cpu/mshyperv.c  |   24 ++++++++++++++++++++++++
 drivers/hv/vmbus_drv.c          |   14 ++++++++++++++
 3 files changed, 40 insertions(+), 0 deletions(-)

diff --git a/arch/x86/include/asm/mshyperv.h b/arch/x86/include/asm/mshyperv.h
index c163215..d3db910 100644
--- a/arch/x86/include/asm/mshyperv.h
+++ b/arch/x86/include/asm/mshyperv.h
@@ -20,4 +20,6 @@ void hyperv_vector_handler(struct pt_regs *regs);
 void hv_setup_vmbus_irq(void (*handler)(void));
 void hv_remove_vmbus_irq(void);
 
+void hv_setup_kexec_handler(void (*handler)(void));
+void hv_remove_kexec_handler(void);
 #endif
diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
index aad4bd8..fa483ed 100644
--- a/arch/x86/kernel/cpu/mshyperv.c
+++ b/arch/x86/kernel/cpu/mshyperv.c
@@ -18,6 +18,7 @@
 #include <linux/efi.h>
 #include <linux/interrupt.h>
 #include <linux/irq.h>
+#include <linux/kexec.h>
 #include <asm/processor.h>
 #include <asm/hypervisor.h>
 #include <asm/hyperv.h>
@@ -28,10 +29,13 @@
 #include <asm/i8259.h>
 #include <asm/apic.h>
 #include <asm/timer.h>
+#include <asm/reboot.h>
 
 struct ms_hyperv_info ms_hyperv;
 EXPORT_SYMBOL_GPL(ms_hyperv);
 
+static void (*hv_kexec_handler)(void);
+
 #if IS_ENABLED(CONFIG_HYPERV)
 static void (*vmbus_handler)(void);
 
@@ -67,8 +71,27 @@ void hv_remove_vmbus_irq(void)
 }
 EXPORT_SYMBOL_GPL(hv_setup_vmbus_irq);
 EXPORT_SYMBOL_GPL(hv_remove_vmbus_irq);
+
+void hv_setup_kexec_handler(void (*handler)(void))
+{
+	hv_kexec_handler = handler;
+}
+EXPORT_SYMBOL_GPL(hv_setup_kexec_handler);
+
+void hv_remove_kexec_handler(void)
+{
+	hv_kexec_handler = NULL;
+}
+EXPORT_SYMBOL_GPL(hv_remove_kexec_handler);
 #endif
 
+static void hv_machine_shutdown(void)
+{
+	if (kexec_in_progress && hv_kexec_handler)
+		hv_kexec_handler();
+	native_machine_shutdown();
+}
+
 static uint32_t  __init ms_hyperv_platform(void)
 {
 	u32 eax;
@@ -141,6 +164,7 @@ static void __init ms_hyperv_init_platform(void)
 	no_timer_check = 1;
 #endif
 
+	machine_ops.shutdown = hv_machine_shutdown;
 }
 
 const __refconst struct hypervisor_x86 x86_hyper_ms_hyperv = {
diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index 00d5158..31748a2 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -1060,6 +1060,17 @@ static struct acpi_driver vmbus_acpi_driver = {
 	},
 };
 
+static void hv_kexec_handler(void)
+{
+	int cpu;
+
+	hv_synic_clockevents_cleanup();
+	vmbus_initiate_unload();
+	for_each_online_cpu(cpu)
+		smp_call_function_single(cpu, hv_synic_cleanup, NULL, 1);
+	hv_cleanup();
+};
+
 static int __init hv_acpi_init(void)
 {
 	int ret, t;
@@ -1092,6 +1103,8 @@ static int __init hv_acpi_init(void)
 	if (ret)
 		goto cleanup;
 
+	hv_setup_kexec_handler(hv_kexec_handler);
+
 	return 0;
 
 cleanup:
@@ -1104,6 +1117,7 @@ static void __exit vmbus_exit(void)
 {
 	int cpu;
 
+	hv_remove_kexec_handler();
 	vmbus_connection.conn_state = DISCONNECTED;
 	hv_synic_clockevents_cleanup();
 	vmbus_disconnect();
-- 
1.7.4.1

[PATCH 04/28] Drivers: hv: don't do hypercalls when hypercall_page is NULL

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

At the very late stage of kexec a driver (which are not being unloaded) can
try to post a message or signal an event. This will crash the kernel as we
already did hv_cleanup() and the hypercall page is NULL.

Move all common (between 32 and 64 bit code) declarations to the beginning
of the do_hypercall() function. Unfortunately we have to write the
!hypercall_page check twice to not mix declarations and code.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/hv.c |   13 ++++++++-----
 1 files changed, 8 insertions(+), 5 deletions(-)

diff --git a/drivers/hv/hv.c b/drivers/hv/hv.c
index 5b87042..41d8072 100644
--- a/drivers/hv/hv.c
+++ b/drivers/hv/hv.c
@@ -93,11 +93,14 @@ static int query_hypervisor_info(void)
  */
 static u64 do_hypercall(u64 control, void *input, void *output)
 {
-#ifdef CONFIG_X86_64
-	u64 hv_status = 0;
 	u64 input_address = (input) ? virt_to_phys(input) : 0;
 	u64 output_address = (output) ? virt_to_phys(output) : 0;
 	void *hypercall_page = hv_context.hypercall_page;
+#ifdef CONFIG_X86_64
+	u64 hv_status = 0;
+
+	if (!hypercall_page)
+		return (u64)ULLONG_MAX;
 
 	__asm__ __volatile__("mov %0, %%r8" : : "r" (output_address) : "r8");
 	__asm__ __volatile__("call *%3" : "=a" (hv_status) :
@@ -112,13 +115,13 @@ static u64 do_hypercall(u64 control, void *input, void *output)
 	u32 control_lo = control & 0xFFFFFFFF;
 	u32 hv_status_hi = 1;
 	u32 hv_status_lo = 1;
-	u64 input_address = (input) ? virt_to_phys(input) : 0;
 	u32 input_address_hi = input_address >> 32;
 	u32 input_address_lo = input_address & 0xFFFFFFFF;
-	u64 output_address = (output) ? virt_to_phys(output) : 0;
 	u32 output_address_hi = output_address >> 32;
 	u32 output_address_lo = output_address & 0xFFFFFFFF;
-	void *hypercall_page = hv_context.hypercall_page;
+
+	if (!hypercall_page)
+		return (u64)ULLONG_MAX;
 
 	__asm__ __volatile__ ("call *%8" : "=d"(hv_status_hi),
 			      "=a"(hv_status_lo) : "d" (control_hi),
-- 
1.7.4.1

[PATCH 05/28] Drivers: hv: vmbus: add special crash handler

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

Full kernel hang is observed when kdump kernel starts after a crash. This
hang happens in vmbus_negotiate_version() function on
wait_for_completion() as Hyper-V host (Win2012R2 in my testing) never
responds to CHANNELMSG_INITIATE_CONTACT as it thinks the connection is
already established. We need to perform some mandatory minimalistic
cleanup before we start new kernel.

Reported-by: K. Y. Srinivasan <kys@microsoft.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 arch/x86/include/asm/mshyperv.h |    2 ++
 arch/x86/kernel/cpu/mshyperv.c  |   22 ++++++++++++++++++++++
 drivers/hv/vmbus_drv.c          |   14 ++++++++++++++
 3 files changed, 38 insertions(+), 0 deletions(-)

diff --git a/arch/x86/include/asm/mshyperv.h b/arch/x86/include/asm/mshyperv.h
index d3db910..d02f9c9 100644
--- a/arch/x86/include/asm/mshyperv.h
+++ b/arch/x86/include/asm/mshyperv.h
@@ -22,4 +22,6 @@ void hv_remove_vmbus_irq(void);
 
 void hv_setup_kexec_handler(void (*handler)(void));
 void hv_remove_kexec_handler(void);
+void hv_setup_crash_handler(void (*handler)(struct pt_regs *regs));
+void hv_remove_crash_handler(void);
 #endif
diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
index fa483ed..5ea5bbc 100644
--- a/arch/x86/kernel/cpu/mshyperv.c
+++ b/arch/x86/kernel/cpu/mshyperv.c
@@ -35,6 +35,7 @@ struct ms_hyperv_info ms_hyperv;
 EXPORT_SYMBOL_GPL(ms_hyperv);
 
 static void (*hv_kexec_handler)(void);
+static void (*hv_crash_handler)(struct pt_regs *regs);
 
 #if IS_ENABLED(CONFIG_HYPERV)
 static void (*vmbus_handler)(void);
@@ -83,6 +84,18 @@ void hv_remove_kexec_handler(void)
 	hv_kexec_handler = NULL;
 }
 EXPORT_SYMBOL_GPL(hv_remove_kexec_handler);
+
+void hv_setup_crash_handler(void (*handler)(struct pt_regs *regs))
+{
+	hv_crash_handler = handler;
+}
+EXPORT_SYMBOL_GPL(hv_setup_crash_handler);
+
+void hv_remove_crash_handler(void)
+{
+	hv_crash_handler = NULL;
+}
+EXPORT_SYMBOL_GPL(hv_remove_crash_handler);
 #endif
 
 static void hv_machine_shutdown(void)
@@ -92,6 +105,14 @@ static void hv_machine_shutdown(void)
 	native_machine_shutdown();
 }
 
+static void hv_machine_crash_shutdown(struct pt_regs *regs)
+{
+	if (hv_crash_handler)
+		hv_crash_handler(regs);
+	native_machine_crash_shutdown(regs);
+}
+
+
 static uint32_t  __init ms_hyperv_platform(void)
 {
 	u32 eax;
@@ -165,6 +186,7 @@ static void __init ms_hyperv_init_platform(void)
 #endif
 
 	machine_ops.shutdown = hv_machine_shutdown;
+	machine_ops.crash_shutdown = hv_machine_crash_shutdown;
 }
 
 const __refconst struct hypervisor_x86 x86_hyper_ms_hyperv = {
diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index 31748a2..1ed9b32 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -1071,6 +1071,18 @@ static void hv_kexec_handler(void)
 	hv_cleanup();
 };
 
+static void hv_crash_handler(struct pt_regs *regs)
+{
+	vmbus_initiate_unload();
+	/*
+	 * In crash handler we can't schedule synic cleanup for all CPUs,
+	 * doing the cleanup for current CPU only. This should be sufficient
+	 * for kdump.
+	 */
+	hv_synic_cleanup(NULL);
+	hv_cleanup();
+};
+
 static int __init hv_acpi_init(void)
 {
 	int ret, t;
@@ -1104,6 +1116,7 @@ static int __init hv_acpi_init(void)
 		goto cleanup;
 
 	hv_setup_kexec_handler(hv_kexec_handler);
+	hv_setup_crash_handler(hv_crash_handler);
 
 	return 0;
 
@@ -1118,6 +1131,7 @@ static void __exit vmbus_exit(void)
 	int cpu;
 
 	hv_remove_kexec_handler();
+	hv_remove_crash_handler();
 	vmbus_connection.conn_state = DISCONNECTED;
 	hv_synic_clockevents_cleanup();
 	vmbus_disconnect();
-- 
1.7.4.1

[PATCH 06/28] Drivers: hv: vmbus: prefer 'die' notification chain to 'panic'

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

current_pt_regs() sometimes returns regs of the userspace process and in
case of a kernel crash this is not what we need to report. E.g. when we
trigger crash with sysrq we see the following:
...
 RIP: 0010:[<ffffffff815b8696>]  [<ffffffff815b8696>] sysrq_handle_crash+0x16/0x20
 RSP: 0018:ffff8800db0a7d88  EFLAGS: 00010246
 RAX: 000000000000000f RBX: ffffffff820a0660 RCX: 0000000000000000
...
at the same time current_pt_regs() give us:
ip=7f899ea7e9e0, ax=ffffffffffffffda, bx=26c81a0, cx=7f899ea7e9e0, ...
These registers come from the userspace process triggered the crash. As we
don't even know which process it was this information is rather useless.

When kernel crash happens through 'die' proper regs are being passed to
all receivers on the die_chain (and panic_notifier_list is being notified
with the string passed to panic() only). If panic() is called manually
(e.g. on BUG()) we won't get 'die' notification so keep the 'panic'
notification reporter as well but guard against double reporting.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/vmbus_drv.c |   40 ++++++++++++++++++++++++++++++++++++----
 1 files changed, 36 insertions(+), 4 deletions(-)

diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index 1ed9b32..b6114cc 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -39,6 +39,7 @@
 #include <asm/mshyperv.h>
 #include <linux/notifier.h>
 #include <linux/ptrace.h>
+#include <linux/kdebug.h>
 #include "hyperv_vmbus.h"
 
 static struct acpi_device  *hv_acpi_dev;
@@ -48,12 +49,18 @@ static struct completion probe_event;
 static int irq;
 
 
-static int hyperv_panic_event(struct notifier_block *nb,
-			unsigned long event, void *ptr)
+static void hyperv_report_panic(struct pt_regs *regs)
 {
-	struct pt_regs *regs;
+	static bool panic_reported;
 
-	regs = current_pt_regs();
+	/*
+	 * We prefer to report panic on 'die' chain as we have proper
+	 * registers to report, but if we miss it (e.g. on BUG()) we need
+	 * to report it on 'panic'.
+	 */
+	if (panic_reported)
+		return;
+	panic_reported = true;
 
 	wrmsrl(HV_X64_MSR_CRASH_P0, regs->ip);
 	wrmsrl(HV_X64_MSR_CRASH_P1, regs->ax);
@@ -65,9 +72,32 @@ static int hyperv_panic_event(struct notifier_block *nb,
 	 * Let Hyper-V know there is crash data available
 	 */
 	wrmsrl(HV_X64_MSR_CRASH_CTL, HV_CRASH_CTL_CRASH_NOTIFY);
+}
+
+static int hyperv_panic_event(struct notifier_block *nb, unsigned long val,
+			      void *args)
+{
+	struct pt_regs *regs;
+
+	regs = current_pt_regs();
+
+	hyperv_report_panic(regs);
 	return NOTIFY_DONE;
 }
 
+static int hyperv_die_event(struct notifier_block *nb, unsigned long val,
+			    void *args)
+{
+	struct die_args *die = (struct die_args *)args;
+	struct pt_regs *regs = die->regs;
+
+	hyperv_report_panic(regs);
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block hyperv_die_block = {
+	.notifier_call = hyperv_die_event,
+};
 static struct notifier_block hyperv_panic_block = {
 	.notifier_call = hyperv_panic_event,
 };
@@ -842,6 +872,7 @@ static int vmbus_bus_init(int irq)
 	 * Only register if the crash MSRs are available
 	 */
 	if (ms_hyperv.features & HV_FEATURE_GUEST_CRASH_MSR_AVAILABLE) {
+		register_die_notifier(&hyperv_die_block);
 		atomic_notifier_chain_register(&panic_notifier_list,
 					       &hyperv_panic_block);
 	}
@@ -1139,6 +1170,7 @@ static void __exit vmbus_exit(void)
 	tasklet_kill(&msg_dpc);
 	vmbus_free_channels();
 	if (ms_hyperv.features & HV_FEATURE_GUEST_CRASH_MSR_AVAILABLE) {
+		unregister_die_notifier(&hyperv_die_block);
 		atomic_notifier_chain_unregister(&panic_notifier_list,
 						 &hyperv_panic_block);
 	}
-- 
1.7.4.1

[PATCH 07/28] Drivers: hv: kvp: check kzalloc return value

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

kzalloc() return value check was accidentally lost in 11bc3a5fa91f:
"Drivers: hv: kvp: convert to hv_utils_transport" commit.

We don't need to reset kvp_transaction.state here as we have the
kvp_timeout_func() timeout function and in case we're in OOM situation
it is preferable to wait.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/hv_kvp.c |    3 +++
 1 files changed, 3 insertions(+), 0 deletions(-)

diff --git a/drivers/hv/hv_kvp.c b/drivers/hv/hv_kvp.c
index d85798d..74c38a9 100644
--- a/drivers/hv/hv_kvp.c
+++ b/drivers/hv/hv_kvp.c
@@ -353,6 +353,9 @@ kvp_send_key(struct work_struct *dummy)
 		return;
 
 	message = kzalloc(sizeof(*message), GFP_KERNEL);
+	if (!message)
+		return;
+
 	message->kvp_hdr.operation = operation;
 	message->kvp_hdr.pool = pool;
 	in_msg = kvp_transaction.kvp_msg;
-- 
1.7.4.1

[PATCH 08/28] Drivers: hv: fcopy: dynamically allocate smsg_out in fcopy_send_data()

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

struct hv_start_fcopy is too big to be on stack on i386, the following
warning is reported:

>> drivers/hv/hv_fcopy.c:159:1: warning: the frame size of 1088 bytes is larger than 1024 bytes [-Wframe-larger-than=]

Reported-by: kbuild test robot <fengguang.wu@intel.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/hv_fcopy.c |   21 +++++++++++++--------
 1 files changed, 13 insertions(+), 8 deletions(-)

diff --git a/drivers/hv/hv_fcopy.c b/drivers/hv/hv_fcopy.c
index b50dd33..db4b887 100644
--- a/drivers/hv/hv_fcopy.c
+++ b/drivers/hv/hv_fcopy.c
@@ -116,7 +116,7 @@ static int fcopy_handle_handshake(u32 version)
 
 static void fcopy_send_data(struct work_struct *dummy)
 {
-	struct hv_start_fcopy smsg_out;
+	struct hv_start_fcopy *smsg_out = NULL;
 	int operation = fcopy_transaction.fcopy_msg->operation;
 	struct hv_start_fcopy *smsg_in;
 	void *out_src;
@@ -136,21 +136,24 @@ static void fcopy_send_data(struct work_struct *dummy)
 	switch (operation) {
 	case START_FILE_COPY:
 		out_len = sizeof(struct hv_start_fcopy);
-		memset(&smsg_out, 0, out_len);
-		smsg_out.hdr.operation = operation;
+		smsg_out = kzalloc(sizeof(*smsg_out), GFP_KERNEL);
+		if (!smsg_out)
+			return;
+
+		smsg_out->hdr.operation = operation;
 		smsg_in = (struct hv_start_fcopy *)fcopy_transaction.fcopy_msg;
 
 		utf16s_to_utf8s((wchar_t *)smsg_in->file_name, W_MAX_PATH,
 				UTF16_LITTLE_ENDIAN,
-				(__u8 *)&smsg_out.file_name, W_MAX_PATH - 1);
+				(__u8 *)&smsg_out->file_name, W_MAX_PATH - 1);
 
 		utf16s_to_utf8s((wchar_t *)smsg_in->path_name, W_MAX_PATH,
 				UTF16_LITTLE_ENDIAN,
-				(__u8 *)&smsg_out.path_name, W_MAX_PATH - 1);
+				(__u8 *)&smsg_out->path_name, W_MAX_PATH - 1);
 
-		smsg_out.copy_flags = smsg_in->copy_flags;
-		smsg_out.file_size = smsg_in->file_size;
-		out_src = &smsg_out;
+		smsg_out->copy_flags = smsg_in->copy_flags;
+		smsg_out->file_size = smsg_in->file_size;
+		out_src = smsg_out;
 		break;
 
 	default:
@@ -168,6 +171,8 @@ static void fcopy_send_data(struct work_struct *dummy)
 			fcopy_transaction.state = HVUTIL_READY;
 		}
 	}
+	kfree(smsg_out);
+
 	return;
 }
 
-- 
1.7.4.1

[PATCH 09/28] Drivers: hv: balloon: Enable dynamic memory protocol negotiation with Windows 10 hosts

[K. Y. Srinivasan]

From: Alex Ng <alexng@microsoft.com>

Support Win10 protocol for Dynamic Memory. Thia patch allows guests on Win10 hosts
to hot-add memory even when dynamic memory is not enabled on the guest.

Signed-off-by: Alex Ng <alexng@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/hv_balloon.c |   26 ++++++++++++++++++++------
 1 files changed, 20 insertions(+), 6 deletions(-)

diff --git a/drivers/hv/hv_balloon.c b/drivers/hv/hv_balloon.c
index 8a725cd..b853b4b 100644
--- a/drivers/hv/hv_balloon.c
+++ b/drivers/hv/hv_balloon.c
@@ -62,11 +62,13 @@
 enum {
 	DYNMEM_PROTOCOL_VERSION_1 = DYNMEM_MAKE_VERSION(0, 3),
 	DYNMEM_PROTOCOL_VERSION_2 = DYNMEM_MAKE_VERSION(1, 0),
+	DYNMEM_PROTOCOL_VERSION_3 = DYNMEM_MAKE_VERSION(2, 0),
 
 	DYNMEM_PROTOCOL_VERSION_WIN7 = DYNMEM_PROTOCOL_VERSION_1,
 	DYNMEM_PROTOCOL_VERSION_WIN8 = DYNMEM_PROTOCOL_VERSION_2,
+	DYNMEM_PROTOCOL_VERSION_WIN10 = DYNMEM_PROTOCOL_VERSION_3,
 
-	DYNMEM_PROTOCOL_VERSION_CURRENT = DYNMEM_PROTOCOL_VERSION_WIN8
+	DYNMEM_PROTOCOL_VERSION_CURRENT = DYNMEM_PROTOCOL_VERSION_WIN10
 };
 
 
@@ -1296,13 +1298,25 @@ static void version_resp(struct hv_dynmem_device *dm,
 	if (dm->next_version == 0)
 		goto version_error;
 
-	dm->next_version = 0;
 	memset(&version_req, 0, sizeof(struct dm_version_request));
 	version_req.hdr.type = DM_VERSION_REQUEST;
 	version_req.hdr.size = sizeof(struct dm_version_request);
 	version_req.hdr.trans_id = atomic_inc_return(&trans_id);
-	version_req.version.version = DYNMEM_PROTOCOL_VERSION_WIN7;
-	version_req.is_last_attempt = 1;
+	version_req.version.version = dm->next_version;
+
+	/*
+	 * Set the next version to try in case current version fails.
+	 * Win7 protocol ought to be the last one to try.
+	 */
+	switch (version_req.version.version) {
+	case DYNMEM_PROTOCOL_VERSION_WIN8:
+		dm->next_version = DYNMEM_PROTOCOL_VERSION_WIN7;
+		version_req.is_last_attempt = 0;
+		break;
+	default:
+		dm->next_version = 0;
+		version_req.is_last_attempt = 1;
+	}
 
 	ret = vmbus_sendpacket(dm->dev->channel, &version_req,
 				sizeof(struct dm_version_request),
@@ -1442,7 +1456,7 @@ static int balloon_probe(struct hv_device *dev,
 
 	dm_device.dev = dev;
 	dm_device.state = DM_INITIALIZING;
-	dm_device.next_version = DYNMEM_PROTOCOL_VERSION_WIN7;
+	dm_device.next_version = DYNMEM_PROTOCOL_VERSION_WIN8;
 	init_completion(&dm_device.host_event);
 	init_completion(&dm_device.config_event);
 	INIT_LIST_HEAD(&dm_device.ha_region_list);
@@ -1474,7 +1488,7 @@ static int balloon_probe(struct hv_device *dev,
 	version_req.hdr.type = DM_VERSION_REQUEST;
 	version_req.hdr.size = sizeof(struct dm_version_request);
 	version_req.hdr.trans_id = atomic_inc_return(&trans_id);
-	version_req.version.version = DYNMEM_PROTOCOL_VERSION_WIN8;
+	version_req.version.version = DYNMEM_PROTOCOL_VERSION_WIN10;
 	version_req.is_last_attempt = 0;
 
 	ret = vmbus_sendpacket(dev->channel, &version_req,
-- 
1.7.4.1

[PATCH 10/28] Drivers: hv: vmbus: Permit sending of packets without payload

[K. Y. Srinivasan]

The guest may have to send a completion packet back to the host.
To support this usage, permit sending a packet without a payload -
we would be only sending the descriptor in this case.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/channel.c |    4 +++-
 1 files changed, 3 insertions(+), 1 deletions(-)

diff --git a/drivers/hv/channel.c b/drivers/hv/channel.c
index 603ce97..c4dcab0 100644
--- a/drivers/hv/channel.c
+++ b/drivers/hv/channel.c
@@ -601,6 +601,7 @@ int vmbus_sendpacket_ctl(struct vmbus_channel *channel, void *buffer,
 	u64 aligned_data = 0;
 	int ret;
 	bool signal = false;
+	int num_vecs = ((bufferlen != 0) ? 3 : 1);
 
 
 	/* Setup the descriptor */
@@ -618,7 +619,8 @@ int vmbus_sendpacket_ctl(struct vmbus_channel *channel, void *buffer,
 	bufferlist[2].iov_base = &aligned_data;
 	bufferlist[2].iov_len = (packetlen_aligned - packetlen);
 
-	ret = hv_ringbuffer_write(&channel->outbound, bufferlist, 3, &signal);
+	ret = hv_ringbuffer_write(&channel->outbound, bufferlist, num_vecs,
+				  &signal);
 
 	/*
 	 * Signalling the host is conditional on many factors:
-- 
1.7.4.1

[PATCH 11/28] drivers:hv: Modify hv_vmbus to search for all MMIO ranges available.

[K. Y. Srinivasan]

From: jakeo@microsoft.com <jakeo@microsoft.com>

This patch changes the logic in hv_vmbus to record all of the ranges in the
VM's firmware (BIOS or UEFI) that offer regions of memory-mapped I/O space for
use by paravirtual front-end drivers.  The old logic just found one range
above 4GB and called it good.  This logic will find any ranges above 1MB.

It would have been possible with this patch to just use existing resource
allocation functions, rather than keep track of the entire set of Hyper-V
related MMIO regions in VMBus.  This strategy, however, is not sufficient
when the resource allocator needs to be aware of the constraints of a
Hyper-V virtual machine, which is what happens in the next patch in the series.
So this first patch exists to show the first steps in reworking the MMIO
allocation paths for Hyper-V front-end drivers.

Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/vmbus_drv.c          |  116 ++++++++++++++++++++++++++++++---------
 drivers/video/fbdev/hyperv_fb.c |    2 +-
 include/linux/hyperv.h          |    2 +-
 3 files changed, 92 insertions(+), 28 deletions(-)

diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index b6114cc..21bb287 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -102,10 +102,7 @@ static struct notifier_block hyperv_panic_block = {
 	.notifier_call = hyperv_panic_event,
 };
 
-struct resource hyperv_mmio = {
-	.name  = "hyperv mmio",
-	.flags = IORESOURCE_MEM,
-};
+struct resource *hyperv_mmio;
 EXPORT_SYMBOL_GPL(hyperv_mmio);
 
 static int vmbus_exists(void)
@@ -1013,30 +1010,105 @@ void vmbus_device_unregister(struct hv_device *device_obj)
 
 
 /*
- * VMBUS is an acpi enumerated device. Get the the information we
+ * VMBUS is an acpi enumerated device. Get the information we
  * need from DSDT.
  */
-
+#define VTPM_BASE_ADDRESS 0xfed40000
 static acpi_status vmbus_walk_resources(struct acpi_resource *res, void *ctx)
 {
+	resource_size_t start = 0;
+	resource_size_t end = 0;
+	struct resource *new_res;
+	struct resource **old_res = &hyperv_mmio;
+	struct resource **prev_res = NULL;
+
 	switch (res->type) {
 	case ACPI_RESOURCE_TYPE_IRQ:
 		irq = res->data.irq.interrupts[0];
+		return AE_OK;
+
+	/*
+	 * "Address" descriptors are for bus windows. Ignore
+	 * "memory" descriptors, which are for registers on
+	 * devices.
+	 */
+	case ACPI_RESOURCE_TYPE_ADDRESS32:
+		start = res->data.address32.address.minimum;
+		end = res->data.address32.address.maximum;
 		break;
 
 	case ACPI_RESOURCE_TYPE_ADDRESS64:
-		hyperv_mmio.start = res->data.address64.address.minimum;
-		hyperv_mmio.end = res->data.address64.address.maximum;
+		start = res->data.address64.address.minimum;
+		end = res->data.address64.address.maximum;
 		break;
+
+	default:
+		/* Unused resource type */
+		return AE_OK;
+
 	}
+	/*
+	 * Ignore ranges that are below 1MB, as they're not
+	 * necessary or useful here.
+	 */
+	if (end < 0x100000)
+		return AE_OK;
+
+	new_res = kzalloc(sizeof(*new_res), GFP_ATOMIC);
+	if (!new_res)
+		return AE_NO_MEMORY;
+
+	/* If this range overlaps the virtual TPM, truncate it. */
+	if (end > VTPM_BASE_ADDRESS && start < VTPM_BASE_ADDRESS)
+		end = VTPM_BASE_ADDRESS;
+
+	new_res->name = "hyperv mmio";
+	new_res->flags = IORESOURCE_MEM;
+	new_res->start = start;
+	new_res->end = end;
+
+	do {
+		if (!*old_res) {
+			*old_res = new_res;
+			break;
+		}
+
+		if ((*old_res)->end < new_res->start) {
+			new_res->sibling = *old_res;
+			if (prev_res)
+				(*prev_res)->sibling = new_res;
+			*old_res = new_res;
+			break;
+		}
+
+		prev_res = old_res;
+		old_res = &(*old_res)->sibling;
+
+	} while (1);
 
 	return AE_OK;
 }
 
+static int vmbus_acpi_remove(struct acpi_device *device)
+{
+	struct resource *cur_res;
+	struct resource *next_res;
+
+	if (hyperv_mmio) {
+		for (cur_res = hyperv_mmio; cur_res; cur_res = next_res) {
+			next_res = cur_res->sibling;
+			kfree(cur_res);
+		}
+	}
+
+	return 0;
+}
+
 static int vmbus_acpi_add(struct acpi_device *device)
 {
 	acpi_status result;
 	int ret_val = -ENODEV;
+	struct acpi_device *ancestor;
 
 	hv_acpi_dev = device;
 
@@ -1046,35 +1118,27 @@ static int vmbus_acpi_add(struct acpi_device *device)
 	if (ACPI_FAILURE(result))
 		goto acpi_walk_err;
 	/*
-	 * The parent of the vmbus acpi device (Gen2 firmware) is the VMOD that
-	 * has the mmio ranges. Get that.
+	 * Some ancestor of the vmbus acpi device (Gen1 or Gen2
+	 * firmware) is the VMOD that has the mmio ranges. Get that.
 	 */
-	if (device->parent) {
-		result = acpi_walk_resources(device->parent->handle,
-					METHOD_NAME__CRS,
-					vmbus_walk_resources, NULL);
+	for (ancestor = device->parent; ancestor; ancestor = ancestor->parent) {
+		result = acpi_walk_resources(ancestor->handle, METHOD_NAME__CRS,
+					     vmbus_walk_resources, NULL);
 
 		if (ACPI_FAILURE(result))
-			goto acpi_walk_err;
-		if (hyperv_mmio.start && hyperv_mmio.end)
-			request_resource(&iomem_resource, &hyperv_mmio);
+			continue;
+		if (hyperv_mmio)
+			break;
 	}
 	ret_val = 0;
 
 acpi_walk_err:
 	complete(&probe_event);
+	if (ret_val)
+		vmbus_acpi_remove(device);
 	return ret_val;
 }
 
-static int vmbus_acpi_remove(struct acpi_device *device)
-{
-	int ret = 0;
-
-	if (hyperv_mmio.start && hyperv_mmio.end)
-		ret = release_resource(&hyperv_mmio);
-	return ret;
-}
-
 static const struct acpi_device_id vmbus_acpi_device_ids[] = {
 	{"VMBUS", 0},
 	{"VMBus", 0},
diff --git a/drivers/video/fbdev/hyperv_fb.c b/drivers/video/fbdev/hyperv_fb.c
index 807ee22..b54ee1c 100644
--- a/drivers/video/fbdev/hyperv_fb.c
+++ b/drivers/video/fbdev/hyperv_fb.c
@@ -688,7 +688,7 @@ static int hvfb_getmem(struct fb_info *info)
 	par->mem.name = KBUILD_MODNAME;
 	par->mem.flags = IORESOURCE_MEM | IORESOURCE_BUSY;
 	if (gen2vm) {
-		ret = allocate_resource(&hyperv_mmio, &par->mem,
+		ret = allocate_resource(hyperv_mmio, &par->mem,
 					screen_fb_size,
 					0, -1,
 					screen_fb_size,
diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h
index 30d3a1f..217e14b 100644
--- a/include/linux/hyperv.h
+++ b/include/linux/hyperv.h
@@ -1233,7 +1233,7 @@ extern bool vmbus_prep_negotiate_resp(struct icmsg_hdr *,
 
 void hv_process_channel_removal(struct vmbus_channel *channel, u32 relid);
 
-extern struct resource hyperv_mmio;
+extern struct resource *hyperv_mmio;
 
 /*
  * Negotiated version with the Host.
-- 
1.7.4.1

[PATCH 12/28] drivers:hv: Move MMIO range picking from hyper_fb to hv_vmbus

[K. Y. Srinivasan]

From: jakeo@microsoft.com <jakeo@microsoft.com>

This patch deletes the logic from hyperv_fb which picked a range of MMIO space
for the frame buffer and adds new logic to hv_vmbus which picks ranges for
child drivers.  The new logic isn't quite the same as the old, as it considers
more possible ranges.

Signed-off-by: Jake Oshins <jakeo@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/vmbus_drv.c          |   88 +++++++++++++++++++++++++++++++++++++--
 drivers/video/fbdev/hyperv_fb.c |   46 +++++++++-----------
 include/linux/hyperv.h          |    7 ++-
 3 files changed, 110 insertions(+), 31 deletions(-)

diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index 21bb287..83759c1 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -39,6 +39,7 @@
 #include <asm/mshyperv.h>
 #include <linux/notifier.h>
 #include <linux/ptrace.h>
+#include <linux/screen_info.h>
 #include <linux/kdebug.h>
 #include "hyperv_vmbus.h"
 
@@ -103,7 +104,6 @@ static struct notifier_block hyperv_panic_block = {
 };
 
 struct resource *hyperv_mmio;
-EXPORT_SYMBOL_GPL(hyperv_mmio);
 
 static int vmbus_exists(void)
 {
@@ -891,8 +891,8 @@ err_cleanup:
 }
 
 /**
- * __vmbus_child_driver_register - Register a vmbus's driver
- * @drv: Pointer to driver structure you want to register
+ * __vmbus_child_driver_register() - Register a vmbus's driver
+ * @hv_driver: Pointer to driver structure you want to register
  * @owner: owner module of the drv
  * @mod_name: module name string
  *
@@ -924,7 +924,8 @@ EXPORT_SYMBOL_GPL(__vmbus_driver_register);
 
 /**
  * vmbus_driver_unregister() - Unregister a vmbus's driver
- * @drv: Pointer to driver structure you want to un-register
+ * @hv_driver: Pointer to driver structure you want to
+ *             un-register
  *
  * Un-register the given driver that was previous registered with a call to
  * vmbus_driver_register()
@@ -1104,6 +1105,85 @@ static int vmbus_acpi_remove(struct acpi_device *device)
 	return 0;
 }
 
+/**
+ * vmbus_allocate_mmio() - Pick a memory-mapped I/O range.
+ * @new:		If successful, supplied a pointer to the
+ *			allocated MMIO space.
+ * @device_obj:		Identifies the caller
+ * @min:		Minimum guest physical address of the
+ *			allocation
+ * @max:		Maximum guest physical address
+ * @size:		Size of the range to be allocated
+ * @align:		Alignment of the range to be allocated
+ * @fb_overlap_ok:	Whether this allocation can be allowed
+ *			to overlap the video frame buffer.
+ *
+ * This function walks the resources granted to VMBus by the
+ * _CRS object in the ACPI namespace underneath the parent
+ * "bridge" whether that's a root PCI bus in the Generation 1
+ * case or a Module Device in the Generation 2 case.  It then
+ * attempts to allocate from the global MMIO pool in a way that
+ * matches the constraints supplied in these parameters and by
+ * that _CRS.
+ *
+ * Return: 0 on success, -errno on failure
+ */
+int vmbus_allocate_mmio(struct resource **new, struct hv_device *device_obj,
+			resource_size_t min, resource_size_t max,
+			resource_size_t size, resource_size_t align,
+			bool fb_overlap_ok)
+{
+	struct resource *iter;
+	resource_size_t range_min, range_max, start, local_min, local_max;
+	const char *dev_n = dev_name(&device_obj->device);
+	u32 fb_end = screen_info.lfb_base + (screen_info.lfb_size << 1);
+	int i;
+
+	for (iter = hyperv_mmio; iter; iter = iter->sibling) {
+		if ((iter->start >= max) || (iter->end <= min))
+			continue;
+
+		range_min = iter->start;
+		range_max = iter->end;
+
+		/* If this range overlaps the frame buffer, split it into
+		   two tries. */
+		for (i = 0; i < 2; i++) {
+			local_min = range_min;
+			local_max = range_max;
+			if (fb_overlap_ok || (range_min >= fb_end) ||
+			    (range_max <= screen_info.lfb_base)) {
+				i++;
+			} else {
+				if ((range_min <= screen_info.lfb_base) &&
+				    (range_max >= screen_info.lfb_base)) {
+					/*
+					 * The frame buffer is in this window,
+					 * so trim this into the part that
+					 * preceeds the frame buffer.
+					 */
+					local_max = screen_info.lfb_base - 1;
+					range_min = fb_end;
+				} else {
+					range_min = fb_end;
+					continue;
+				}
+			}
+
+			start = (local_min + align - 1) & ~(align - 1);
+			for (; start + size - 1 <= local_max; start += align) {
+				*new = request_mem_region_exclusive(start, size,
+								    dev_n);
+				if (*new)
+					return 0;
+			}
+		}
+	}
+
+	return -ENXIO;
+}
+EXPORT_SYMBOL_GPL(vmbus_allocate_mmio);
+
 static int vmbus_acpi_add(struct acpi_device *device)
 {
 	acpi_status result;
diff --git a/drivers/video/fbdev/hyperv_fb.c b/drivers/video/fbdev/hyperv_fb.c
index b54ee1c..e2451bd 100644
--- a/drivers/video/fbdev/hyperv_fb.c
+++ b/drivers/video/fbdev/hyperv_fb.c
@@ -213,7 +213,7 @@ struct synthvid_msg {
 
 struct hvfb_par {
 	struct fb_info *info;
-	struct resource mem;
+	struct resource *mem;
 	bool fb_ready; /* fb device is ready */
 	struct completion wait;
 	u32 synthvid_version;
@@ -677,26 +677,18 @@ static void hvfb_get_option(struct fb_info *info)
 
 
 /* Get framebuffer memory from Hyper-V video pci space */
-static int hvfb_getmem(struct fb_info *info)
+static int hvfb_getmem(struct hv_device *hdev, struct fb_info *info)
 {
 	struct hvfb_par *par = info->par;
 	struct pci_dev *pdev  = NULL;
 	void __iomem *fb_virt;
 	int gen2vm = efi_enabled(EFI_BOOT);
+	resource_size_t pot_start, pot_end;
 	int ret;
 
-	par->mem.name = KBUILD_MODNAME;
-	par->mem.flags = IORESOURCE_MEM | IORESOURCE_BUSY;
 	if (gen2vm) {
-		ret = allocate_resource(hyperv_mmio, &par->mem,
-					screen_fb_size,
-					0, -1,
-					screen_fb_size,
-					NULL, NULL);
-		if (ret != 0) {
-			pr_err("Unable to allocate framebuffer memory\n");
-			return -ENODEV;
-		}
+		pot_start = 0;
+		pot_end = -1;
 	} else {
 		pdev = pci_get_device(PCI_VENDOR_ID_MICROSOFT,
 			      PCI_DEVICE_ID_HYPERV_VIDEO, NULL);
@@ -709,16 +701,18 @@ static int hvfb_getmem(struct fb_info *info)
 		    pci_resource_len(pdev, 0) < screen_fb_size)
 			goto err1;
 
-		par->mem.end = pci_resource_end(pdev, 0);
-		par->mem.start = par->mem.end - screen_fb_size + 1;
-		ret = request_resource(&pdev->resource[0], &par->mem);
-		if (ret != 0) {
-			pr_err("Unable to request framebuffer memory\n");
-			goto err1;
-		}
+		pot_end = pci_resource_end(pdev, 0);
+		pot_start = pot_end - screen_fb_size + 1;
+	}
+
+	ret = vmbus_allocate_mmio(&par->mem, hdev, pot_start, pot_end,
+				  screen_fb_size, 0x100000, true);
+	if (ret != 0) {
+		pr_err("Unable to allocate framebuffer memory\n");
+		goto err1;
 	}
 
-	fb_virt = ioremap(par->mem.start, screen_fb_size);
+	fb_virt = ioremap(par->mem->start, screen_fb_size);
 	if (!fb_virt)
 		goto err2;
 
@@ -736,7 +730,7 @@ static int hvfb_getmem(struct fb_info *info)
 		info->apertures->ranges[0].size = pci_resource_len(pdev, 0);
 	}
 
-	info->fix.smem_start = par->mem.start;
+	info->fix.smem_start = par->mem->start;
 	info->fix.smem_len = screen_fb_size;
 	info->screen_base = fb_virt;
 	info->screen_size = screen_fb_size;
@@ -749,7 +743,8 @@ static int hvfb_getmem(struct fb_info *info)
 err3:
 	iounmap(fb_virt);
 err2:
-	release_resource(&par->mem);
+	release_mem_region(par->mem->start, screen_fb_size);
+	par->mem = NULL;
 err1:
 	if (!gen2vm)
 		pci_dev_put(pdev);
@@ -763,7 +758,8 @@ static void hvfb_putmem(struct fb_info *info)
 	struct hvfb_par *par = info->par;
 
 	iounmap(info->screen_base);
-	release_resource(&par->mem);
+	release_mem_region(par->mem->start, screen_fb_size);
+	par->mem = NULL;
 }
 
 
@@ -794,7 +790,7 @@ static int hvfb_probe(struct hv_device *hdev,
 		goto error1;
 	}
 
-	ret = hvfb_getmem(info);
+	ret = hvfb_getmem(hdev, info);
 	if (ret) {
 		pr_err("No memory for framebuffer\n");
 		goto error2;
diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h
index 217e14b..54733d5 100644
--- a/include/linux/hyperv.h
+++ b/include/linux/hyperv.h
@@ -977,6 +977,11 @@ int __must_check __vmbus_driver_register(struct hv_driver *hv_driver,
 					 const char *mod_name);
 void vmbus_driver_unregister(struct hv_driver *hv_driver);
 
+int vmbus_allocate_mmio(struct resource **new, struct hv_device *device_obj,
+			resource_size_t min, resource_size_t max,
+			resource_size_t size, resource_size_t align,
+			bool fb_overlap_ok);
+
 /**
  * VMBUS_DEVICE - macro used to describe a specific hyperv vmbus device
  *
@@ -1233,8 +1238,6 @@ extern bool vmbus_prep_negotiate_resp(struct icmsg_hdr *,
 
 void hv_process_channel_removal(struct vmbus_channel *channel, u32 relid);
 
-extern struct resource *hyperv_mmio;
-
 /*
  * Negotiated version with the Host.
  */
-- 
1.7.4.1

[PATCH 13/28] hv: util: checking the wrong variable

[K. Y. Srinivasan]

From: Dan Carpenter <dan.carpenter@oracle.com>

We don't catch this allocation failure because there is a typo and we
check the wrong variable.

Fixes: 14b50f80c32d ('Drivers: hv: util: introduce hv_utils_transport abstraction')

Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/hv_utils_transport.c |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/hv/hv_utils_transport.c b/drivers/hv/hv_utils_transport.c
index ea7ba5e..6a9d80a 100644
--- a/drivers/hv/hv_utils_transport.c
+++ b/drivers/hv/hv_utils_transport.c
@@ -186,7 +186,7 @@ int hvutil_transport_send(struct hvutil_transport *hvt, void *msg, int len)
 		return -EINVAL;
 	} else if (hvt->mode == HVUTIL_TRANSPORT_NETLINK) {
 		cn_msg = kzalloc(sizeof(*cn_msg) + len, GFP_ATOMIC);
-		if (!msg)
+		if (!cn_msg)
 			return -ENOMEM;
 		cn_msg->id.idx = hvt->cn_id.idx;
 		cn_msg->id.val = hvt->cn_id.val;
-- 
1.7.4.1

[PATCH 14/28] Drivers: hv: vmbus: fix typo in hv_port_info struct

[K. Y. Srinivasan]

From: Nik Nyby <nikolas@gnu.org>

This fixes a typo: base_flag_bumber to base_flag_number

Signed-off-by: Nik Nyby <nikolas@gnu.org>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/hyperv_vmbus.h |    2 +-
 1 files changed, 1 insertions(+), 1 deletions(-)

diff --git a/drivers/hv/hyperv_vmbus.h b/drivers/hv/hyperv_vmbus.h
index cddc0c9..6383707 100644
--- a/drivers/hv/hyperv_vmbus.h
+++ b/drivers/hv/hyperv_vmbus.h
@@ -141,7 +141,7 @@ struct hv_port_info {
 		struct {
 			u32 target_sint;
 			u32 target_vp;
-			u16 base_flag_bumber;
+			u16 base_flag_number;
 			u16 flag_count;
 			u32 rsvdz;
 		} event_port_info;
-- 
1.7.4.1

[PATCH 15/28] Drivers: hv: vmbus: don't send CHANNELMSG_UNLOAD on pre-Win2012R2 hosts

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

Pre-Win2012R2 hosts don't properly handle CHANNELMSG_UNLOAD and
wait_for_completion() hangs. Avoid sending such request on old hosts.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/channel_mgmt.c |    4 ++++
 1 files changed, 4 insertions(+), 0 deletions(-)

diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
index 4506a66..00ba3f3 100644
--- a/drivers/hv/channel_mgmt.c
+++ b/drivers/hv/channel_mgmt.c
@@ -469,6 +469,10 @@ void vmbus_initiate_unload(void)
 {
 	struct vmbus_channel_message_header hdr;
 
+	/* Pre-Win2012R2 hosts don't support reconnect */
+	if (vmbus_proto_version < VERSION_WIN8_1)
+		return;
+
 	init_completion(&vmbus_connection.unload_event);
 	memset(&hdr, 0, sizeof(struct vmbus_channel_message_header));
 	hdr.msgtype = CHANNELMSG_UNLOAD;
-- 
1.7.4.1

[PATCH 16/28] mshyperv: fix recognition of Hyper-V guest crash MSR's

[K. Y. Srinivasan]

From: Denis V. Lunev <den@openvz.org>

Hypervisor Top Level Functional Specification v3.1/4.0 notes that cpuid
(0x40000003) EDX's 10th bit should be used to check that Hyper-V guest
crash MSR's functionality available.

This patch should fix this recognition. Currently the code checks EAX
register instead of EDX.

Signed-off-by: Andrey Smetanin <asmetanin@virtuozzo.com>
Signed-off-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 arch/x86/include/asm/mshyperv.h |    1 +
 arch/x86/kernel/cpu/mshyperv.c  |    1 +
 drivers/hv/vmbus_drv.c          |    4 ++--
 3 files changed, 4 insertions(+), 2 deletions(-)

diff --git a/arch/x86/include/asm/mshyperv.h b/arch/x86/include/asm/mshyperv.h
index d02f9c9..aaf59b7 100644
--- a/arch/x86/include/asm/mshyperv.h
+++ b/arch/x86/include/asm/mshyperv.h
@@ -7,6 +7,7 @@
 
 struct ms_hyperv_info {
 	u32 features;
+	u32 misc_features;
 	u32 hints;
 };
 
diff --git a/arch/x86/kernel/cpu/mshyperv.c b/arch/x86/kernel/cpu/mshyperv.c
index 5ea5bbc..f794bfa 100644
--- a/arch/x86/kernel/cpu/mshyperv.c
+++ b/arch/x86/kernel/cpu/mshyperv.c
@@ -158,6 +158,7 @@ static void __init ms_hyperv_init_platform(void)
 	 * Extract the features and hints
 	 */
 	ms_hyperv.features = cpuid_eax(HYPERV_CPUID_FEATURES);
+	ms_hyperv.misc_features = cpuid_edx(HYPERV_CPUID_FEATURES);
 	ms_hyperv.hints    = cpuid_eax(HYPERV_CPUID_ENLIGHTMENT_INFO);
 
 	printk(KERN_INFO "HyperV: features 0x%x, hints 0x%x\n",
diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index 83759c1..8c3eaee 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -868,7 +868,7 @@ static int vmbus_bus_init(int irq)
 	/*
 	 * Only register if the crash MSRs are available
 	 */
-	if (ms_hyperv.features & HV_FEATURE_GUEST_CRASH_MSR_AVAILABLE) {
+	if (ms_hyperv.misc_features & HV_FEATURE_GUEST_CRASH_MSR_AVAILABLE) {
 		register_die_notifier(&hyperv_die_block);
 		atomic_notifier_chain_register(&panic_notifier_list,
 					       &hyperv_panic_block);
@@ -1313,7 +1313,7 @@ static void __exit vmbus_exit(void)
 	hv_remove_vmbus_irq();
 	tasklet_kill(&msg_dpc);
 	vmbus_free_channels();
-	if (ms_hyperv.features & HV_FEATURE_GUEST_CRASH_MSR_AVAILABLE) {
+	if (ms_hyperv.misc_features & HV_FEATURE_GUEST_CRASH_MSR_AVAILABLE) {
 		unregister_die_notifier(&hyperv_die_block);
 		atomic_notifier_chain_unregister(&panic_notifier_list,
 						 &hyperv_panic_block);
-- 
1.7.4.1

[PATCH 17/28] Drivers: hv: vmbus: Consider ND NIC in binding channels to CPUs

[K. Y. Srinivasan]

We cycle through all the "high performance" channels to distribute
load across the available CPUs. Process the NetworkDirect as a
high performance device.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/channel_mgmt.c |    1 +
 1 files changed, 1 insertions(+), 0 deletions(-)

diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
index 00ba3f3..30613df 100644
--- a/drivers/hv/channel_mgmt.c
+++ b/drivers/hv/channel_mgmt.c
@@ -347,6 +347,7 @@ enum {
 	IDE = 0,
 	SCSI,
 	NIC,
+	ND_NIC,
 	MAX_PERF_CHN,
 };
 
-- 
1.7.4.1

[PATCH 18/28] Drivers: hv: vmbus: Improve the CPU affiliation for channels

[K. Y. Srinivasan]

The current code tracks the assigned CPUs within a NUMA node in the context of
the primary channel. So, if we have a VM with a single NUMA node with 8 VCPUs, we may
end up unevenly distributing the channel load. Fix the issue by tracking affiliations
globally.

Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/channel_mgmt.c |   11 ++++++-----
 drivers/hv/hv.c           |    9 +++++++++
 drivers/hv/hyperv_vmbus.h |    5 +++++
 include/linux/hyperv.h    |    1 -
 4 files changed, 20 insertions(+), 6 deletions(-)

diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
index 30613df..39c5afc 100644
--- a/drivers/hv/channel_mgmt.c
+++ b/drivers/hv/channel_mgmt.c
@@ -392,6 +392,7 @@ static void init_vp_index(struct vmbus_channel *channel, const uuid_le *type_gui
 	struct vmbus_channel *primary = channel->primary_channel;
 	int next_node;
 	struct cpumask available_mask;
+	struct cpumask *alloced_mask;
 
 	for (i = IDE; i < MAX_PERF_CHN; i++) {
 		if (!memcmp(type_guid->b, hp_devs[i].guid,
@@ -409,7 +410,6 @@ static void init_vp_index(struct vmbus_channel *channel, const uuid_le *type_gui
 		 * channel, bind it to cpu 0.
 		 */
 		channel->numa_node = 0;
-		cpumask_set_cpu(0, &channel->alloced_cpus_in_node);
 		channel->target_cpu = 0;
 		channel->target_vp = hv_context.vp_index[0];
 		return;
@@ -434,21 +434,22 @@ static void init_vp_index(struct vmbus_channel *channel, const uuid_le *type_gui
 		channel->numa_node = next_node;
 		primary = channel;
 	}
+	alloced_mask = &hv_context.hv_numa_map[primary->numa_node];
 
-	if (cpumask_weight(&primary->alloced_cpus_in_node) ==
+	if (cpumask_weight(alloced_mask) ==
 	    cpumask_weight(cpumask_of_node(primary->numa_node))) {
 		/*
 		 * We have cycled through all the CPUs in the node;
 		 * reset the alloced map.
 		 */
-		cpumask_clear(&primary->alloced_cpus_in_node);
+		cpumask_clear(alloced_mask);
 	}
 
-	cpumask_xor(&available_mask, &primary->alloced_cpus_in_node,
+	cpumask_xor(&available_mask, alloced_mask,
 		    cpumask_of_node(primary->numa_node));
 
 	cur_cpu = cpumask_next(-1, &available_mask);
-	cpumask_set_cpu(cur_cpu, &primary->alloced_cpus_in_node);
+	cpumask_set_cpu(cur_cpu, alloced_mask);
 
 	channel->target_cpu = cur_cpu;
 	channel->target_vp = hv_context.vp_index[cur_cpu];
diff --git a/drivers/hv/hv.c b/drivers/hv/hv.c
index 41d8072..fd93cfd 100644
--- a/drivers/hv/hv.c
+++ b/drivers/hv/hv.c
@@ -332,6 +332,13 @@ int hv_synic_alloc(void)
 	size_t ced_size = sizeof(struct clock_event_device);
 	int cpu;
 
+	hv_context.hv_numa_map = kzalloc(sizeof(struct cpumask) * nr_node_ids,
+					 GFP_ATOMIC);
+	if (hv_context.hv_numa_map == NULL) {
+		pr_err("Unable to allocate NUMA map\n");
+		goto err;
+	}
+
 	for_each_online_cpu(cpu) {
 		hv_context.event_dpc[cpu] = kmalloc(size, GFP_ATOMIC);
 		if (hv_context.event_dpc[cpu] == NULL) {
@@ -345,6 +352,7 @@ int hv_synic_alloc(void)
 			pr_err("Unable to allocate clock event device\n");
 			goto err;
 		}
+
 		hv_init_clockevent_device(hv_context.clk_evt[cpu], cpu);
 
 		hv_context.synic_message_page[cpu] =
@@ -393,6 +401,7 @@ void hv_synic_free(void)
 {
 	int cpu;
 
+	kfree(hv_context.hv_numa_map);
 	for_each_online_cpu(cpu)
 		hv_synic_free_cpu(cpu);
 }
diff --git a/drivers/hv/hyperv_vmbus.h b/drivers/hv/hyperv_vmbus.h
index 6383707..6f25825 100644
--- a/drivers/hv/hyperv_vmbus.h
+++ b/drivers/hv/hyperv_vmbus.h
@@ -551,6 +551,11 @@ struct hv_context {
 	 * Support PV clockevent device.
 	 */
 	struct clock_event_device *clk_evt[NR_CPUS];
+	/*
+	 * To manage allocations in a NUMA node.
+	 * Array indexed by numa node ID.
+	 */
+	struct cpumask *hv_numa_map;
 };
 
 extern struct hv_context hv_context;
diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h
index 54733d5..5a3df5a 100644
--- a/include/linux/hyperv.h
+++ b/include/linux/hyperv.h
@@ -699,7 +699,6 @@ struct vmbus_channel {
 	/*
 	 * State to manage the CPU affiliation of channels.
 	 */
-	struct cpumask alloced_cpus_in_node;
 	int numa_node;
 	/*
 	 * Support for sub-channels. For high performance devices,
-- 
1.7.4.1

[PATCH 19/28] Drivers: hv: vmbus: Further improve CPU affiliation logic

[K. Y. Srinivasan]

From: Dexuan Cui <decui@microsoft.com>

Keep track of CPU affiliations of sub-channels within the scope of the primary
channel. This will allow us to better distribute the load amongst available
CPUs.

Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/channel_mgmt.c |   20 ++++++++++++++++++--
 include/linux/hyperv.h    |    1 +
 2 files changed, 19 insertions(+), 2 deletions(-)

diff --git a/drivers/hv/channel_mgmt.c b/drivers/hv/channel_mgmt.c
index 39c5afc..2f9aead 100644
--- a/drivers/hv/channel_mgmt.c
+++ b/drivers/hv/channel_mgmt.c
@@ -448,8 +448,24 @@ static void init_vp_index(struct vmbus_channel *channel, const uuid_le *type_gui
 	cpumask_xor(&available_mask, alloced_mask,
 		    cpumask_of_node(primary->numa_node));
 
-	cur_cpu = cpumask_next(-1, &available_mask);
-	cpumask_set_cpu(cur_cpu, alloced_mask);
+	cur_cpu = -1;
+	while (true) {
+		cur_cpu = cpumask_next(cur_cpu, &available_mask);
+		if (cur_cpu >= nr_cpu_ids) {
+			cur_cpu = -1;
+			cpumask_copy(&available_mask,
+				     cpumask_of_node(primary->numa_node));
+			continue;
+		}
+
+		if (!cpumask_test_cpu(cur_cpu,
+				&primary->alloced_cpus_in_node)) {
+			cpumask_set_cpu(cur_cpu,
+					&primary->alloced_cpus_in_node);
+			cpumask_set_cpu(cur_cpu, alloced_mask);
+			break;
+		}
+	}
 
 	channel->target_cpu = cur_cpu;
 	channel->target_vp = hv_context.vp_index[cur_cpu];
diff --git a/include/linux/hyperv.h b/include/linux/hyperv.h
index 5a3df5a..54733d5 100644
--- a/include/linux/hyperv.h
+++ b/include/linux/hyperv.h
@@ -699,6 +699,7 @@ struct vmbus_channel {
 	/*
 	 * State to manage the CPU affiliation of channels.
 	 */
+	struct cpumask alloced_cpus_in_node;
 	int numa_node;
 	/*
 	 * Support for sub-channels. For high performance devices,
-- 
1.7.4.1

[PATCH 20/28] Drivers: hv_vmbus: Fix signal to host condition

[K. Y. Srinivasan]

From: Christopher Oo <t-chriso@microsoft.com>

Fixes a bug where previously hv_ringbuffer_read would pass in the old
number of bytes available to read instead of the expected old read index
when calculating when to signal to the host that the ringbuffer is empty.
Since the previous write size is already saved, also changes the
hv_need_to_signal_on_read to use the previously read value rather than
recalculating it.

Signed-off-by: Christopher Oo <t-chriso@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/ring_buffer.c |   14 +++-----------
 1 files changed, 3 insertions(+), 11 deletions(-)

diff --git a/drivers/hv/ring_buffer.c b/drivers/hv/ring_buffer.c
index 6361d12..70a1a9a 100644
--- a/drivers/hv/ring_buffer.c
+++ b/drivers/hv/ring_buffer.c
@@ -103,10 +103,9 @@ static bool hv_need_to_signal(u32 old_write, struct hv_ring_buffer_info *rbi)
  *    there is room for the producer to send the pending packet.
  */
 
-static bool hv_need_to_signal_on_read(u32 old_rd,
-					 struct hv_ring_buffer_info *rbi)
+static bool hv_need_to_signal_on_read(u32 prev_write_sz,
+				      struct hv_ring_buffer_info *rbi)
 {
-	u32 prev_write_sz;
 	u32 cur_write_sz;
 	u32 r_size;
 	u32 write_loc = rbi->ring_buffer->write_index;
@@ -123,10 +122,6 @@ static bool hv_need_to_signal_on_read(u32 old_rd,
 	cur_write_sz = write_loc >= read_loc ? r_size - (write_loc - read_loc) :
 			read_loc - write_loc;
 
-	prev_write_sz = write_loc >= old_rd ? r_size - (write_loc - old_rd) :
-			old_rd - write_loc;
-
-
 	if ((prev_write_sz < pending_sz) && (cur_write_sz >= pending_sz))
 		return true;
 
@@ -517,7 +512,6 @@ int hv_ringbuffer_read(struct hv_ring_buffer_info *inring_info, void *buffer,
 	u32 next_read_location = 0;
 	u64 prev_indices = 0;
 	unsigned long flags;
-	u32 old_read;
 
 	if (buflen <= 0)
 		return -EINVAL;
@@ -528,8 +522,6 @@ int hv_ringbuffer_read(struct hv_ring_buffer_info *inring_info, void *buffer,
 				&bytes_avail_toread,
 				&bytes_avail_towrite);
 
-	old_read = bytes_avail_toread;
-
 	/* Make sure there is something to read */
 	if (bytes_avail_toread < buflen) {
 		spin_unlock_irqrestore(&inring_info->ring_lock, flags);
@@ -560,7 +552,7 @@ int hv_ringbuffer_read(struct hv_ring_buffer_info *inring_info, void *buffer,
 
 	spin_unlock_irqrestore(&inring_info->ring_lock, flags);
 
-	*signal = hv_need_to_signal_on_read(old_read, inring_info);
+	*signal = hv_need_to_signal_on_read(bytes_avail_towrite, inring_info);
 
 	return 0;
 }
-- 
1.7.4.1

[PATCH 21/28] drivers/hv: Migrate to new 'set-state' interface

[K. Y. Srinivasan]

From: Viresh Kumar <viresh.kumar@linaro.org>

Migrate hv driver to the new 'set-state' interface provided by
clockevents core, the earlier 'set-mode' interface is marked obsolete
now.

This also enables us to implement callbacks for new states of clockevent
devices, for example: ONESHOT_STOPPED.

Cc: "K. Y. Srinivasan" <kys@microsoft.com>
Cc: Haiyang Zhang <haiyangz@microsoft.com>
Cc: devel@linuxdriverproject.org
Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/hv.c |   45 +++++++++++++++++++--------------------------
 1 files changed, 19 insertions(+), 26 deletions(-)

diff --git a/drivers/hv/hv.c b/drivers/hv/hv.c
index fd93cfd..c641faf 100644
--- a/drivers/hv/hv.c
+++ b/drivers/hv/hv.c
@@ -274,7 +274,7 @@ static int hv_ce_set_next_event(unsigned long delta,
 {
 	cycle_t current_tick;
 
-	WARN_ON(evt->mode != CLOCK_EVT_MODE_ONESHOT);
+	WARN_ON(!clockevent_state_oneshot(evt));
 
 	rdmsrl(HV_X64_MSR_TIME_REF_COUNT, current_tick);
 	current_tick += delta;
@@ -282,31 +282,24 @@ static int hv_ce_set_next_event(unsigned long delta,
 	return 0;
 }
 
-static void hv_ce_setmode(enum clock_event_mode mode,
-			  struct clock_event_device *evt)
+static int hv_ce_shutdown(struct clock_event_device *evt)
+{
+	wrmsrl(HV_X64_MSR_STIMER0_COUNT, 0);
+	wrmsrl(HV_X64_MSR_STIMER0_CONFIG, 0);
+
+	return 0;
+}
+
+static int hv_ce_set_oneshot(struct clock_event_device *evt)
 {
 	union hv_timer_config timer_cfg;
 
-	switch (mode) {
-	case CLOCK_EVT_MODE_PERIODIC:
-		/* unsupported */
-		break;
-
-	case CLOCK_EVT_MODE_ONESHOT:
-		timer_cfg.enable = 1;
-		timer_cfg.auto_enable = 1;
-		timer_cfg.sintx = VMBUS_MESSAGE_SINT;
-		wrmsrl(HV_X64_MSR_STIMER0_CONFIG, timer_cfg.as_uint64);
-		break;
-
-	case CLOCK_EVT_MODE_UNUSED:
-	case CLOCK_EVT_MODE_SHUTDOWN:
-		wrmsrl(HV_X64_MSR_STIMER0_COUNT, 0);
-		wrmsrl(HV_X64_MSR_STIMER0_CONFIG, 0);
-		break;
-	case CLOCK_EVT_MODE_RESUME:
-		break;
-	}
+	timer_cfg.enable = 1;
+	timer_cfg.auto_enable = 1;
+	timer_cfg.sintx = VMBUS_MESSAGE_SINT;
+	wrmsrl(HV_X64_MSR_STIMER0_CONFIG, timer_cfg.as_uint64);
+
+	return 0;
 }
 
 static void hv_init_clockevent_device(struct clock_event_device *dev, int cpu)
@@ -321,7 +314,8 @@ static void hv_init_clockevent_device(struct clock_event_device *dev, int cpu)
 	 * references to the hv_vmbus module making it impossible to unload.
 	 */
 
-	dev->set_mode = hv_ce_setmode;
+	dev->set_state_shutdown = hv_ce_shutdown;
+	dev->set_state_oneshot = hv_ce_set_oneshot;
 	dev->set_next_event = hv_ce_set_next_event;
 }
 
@@ -515,8 +509,7 @@ void hv_synic_cleanup(void *arg)
 
 	/* Turn off clockevent device */
 	if (ms_hyperv.features & HV_X64_MSR_SYNTIMER_AVAILABLE)
-		hv_ce_setmode(CLOCK_EVT_MODE_SHUTDOWN,
-			      hv_context.clk_evt[cpu]);
+		hv_ce_shutdown(hv_context.clk_evt[cpu]);
 
 	rdmsrl(HV_X64_MSR_SINT0 + VMBUS_MESSAGE_SINT, shared_sint.as_uint64);
 
-- 
1.7.4.1

[PATCH 22/28] Drivers: hv: vmbus: Implement a clocksource based on the TSC page

[K. Y. Srinivasan]

The current Hyper-V clock source is based on the per-partition reference counter
and this counter is being accessed via s synthetic MSR - HV_X64_MSR_TIME_REF_COUNT.
Hyper-V has a more efficient way of computing the per-partition reference
counter value that does not involve reading a synthetic MSR. We implement
a time source based on this mechanism.

Tested-by: Vivek Yadav <vyadav@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 arch/x86/include/uapi/asm/hyperv.h |    2 +
 drivers/hv/hv.c                    |   83 ++++++++++++++++++++++++++++++++++++
 drivers/hv/hyperv_vmbus.h          |    9 ++++
 3 files changed, 94 insertions(+), 0 deletions(-)

diff --git a/arch/x86/include/uapi/asm/hyperv.h b/arch/x86/include/uapi/asm/hyperv.h
index 8fba544..89f15e3 100644
--- a/arch/x86/include/uapi/asm/hyperv.h
+++ b/arch/x86/include/uapi/asm/hyperv.h
@@ -27,6 +27,8 @@
 #define HV_X64_MSR_VP_RUNTIME_AVAILABLE		(1 << 0)
 /* Partition Reference Counter (HV_X64_MSR_TIME_REF_COUNT) available*/
 #define HV_X64_MSR_TIME_REF_COUNT_AVAILABLE	(1 << 1)
+/* Partition reference TSC MSR is available */
+#define HV_X64_MSR_REFERENCE_TSC_AVAILABLE              (1 << 9)
 
 /* A partition's reference time stamp counter (TSC) page */
 #define HV_X64_MSR_REFERENCE_TSC		0x40000021
diff --git a/drivers/hv/hv.c b/drivers/hv/hv.c
index c641faf..6341be8 100644
--- a/drivers/hv/hv.c
+++ b/drivers/hv/hv.c
@@ -133,6 +133,56 @@ static u64 do_hypercall(u64 control, void *input, void *output)
 #endif /* !x86_64 */
 }
 
+#ifdef CONFIG_X86_64
+static cycle_t read_hv_clock_tsc(struct clocksource *arg)
+{
+	cycle_t current_tick;
+	struct ms_hyperv_tsc_page *tsc_pg = hv_context.tsc_page;
+
+	if (tsc_pg->tsc_sequence != -1) {
+		/*
+		 * Use the tsc page to compute the value.
+		 */
+
+		while (1) {
+			cycle_t tmp;
+			u32 sequence = tsc_pg->tsc_sequence;
+			u64 cur_tsc;
+			u64 scale = tsc_pg->tsc_scale;
+			s64 offset = tsc_pg->tsc_offset;
+
+			rdtscll(cur_tsc);
+			/* current_tick = ((cur_tsc *scale) >> 64) + offset */
+			asm("mulq %3"
+				: "=d" (current_tick), "=a" (tmp)
+				: "a" (cur_tsc), "r" (scale));
+
+			current_tick += offset;
+			if (tsc_pg->tsc_sequence == sequence)
+				return current_tick;
+
+			if (tsc_pg->tsc_sequence != -1)
+				continue;
+			/*
+			 * Fallback using MSR method.
+			 */
+			break;
+		}
+	}
+	rdmsrl(HV_X64_MSR_TIME_REF_COUNT, current_tick);
+	return current_tick;
+}
+
+static struct clocksource hyperv_cs_tsc = {
+		.name           = "hyperv_clocksource_tsc_page",
+		.rating         = 425,
+		.read           = read_hv_clock_tsc,
+		.mask           = CLOCKSOURCE_MASK(64),
+		.flags          = CLOCK_SOURCE_IS_CONTINUOUS,
+};
+#endif
+
+
 /*
  * hv_init - Main initialization routine.
  *
@@ -142,7 +192,9 @@ int hv_init(void)
 {
 	int max_leaf;
 	union hv_x64_msr_hypercall_contents hypercall_msr;
+	union hv_x64_msr_hypercall_contents tsc_msr;
 	void *virtaddr = NULL;
+	void *va_tsc = NULL;
 
 	memset(hv_context.synic_event_page, 0, sizeof(void *) * NR_CPUS);
 	memset(hv_context.synic_message_page, 0,
@@ -186,6 +238,22 @@ int hv_init(void)
 
 	hv_context.hypercall_page = virtaddr;
 
+#ifdef CONFIG_X86_64
+	if (ms_hyperv.features & HV_X64_MSR_REFERENCE_TSC_AVAILABLE) {
+		va_tsc = __vmalloc(PAGE_SIZE, GFP_KERNEL, PAGE_KERNEL);
+		if (!va_tsc)
+			goto cleanup;
+		hv_context.tsc_page = va_tsc;
+
+		rdmsrl(HV_X64_MSR_REFERENCE_TSC, tsc_msr.as_uint64);
+
+		tsc_msr.enable = 1;
+		tsc_msr.guest_physical_address = vmalloc_to_pfn(va_tsc);
+
+		wrmsrl(HV_X64_MSR_REFERENCE_TSC, tsc_msr.as_uint64);
+		clocksource_register_hz(&hyperv_cs_tsc, NSEC_PER_SEC/100);
+	}
+#endif
 	return 0;
 
 cleanup:
@@ -219,6 +287,21 @@ void hv_cleanup(void)
 		vfree(hv_context.hypercall_page);
 		hv_context.hypercall_page = NULL;
 	}
+
+#ifdef CONFIG_X86_64
+	/*
+	 * Cleanup the TSC page based CS.
+	 */
+	if (ms_hyperv.features & HV_X64_MSR_REFERENCE_TSC_AVAILABLE) {
+		clocksource_change_rating(&hyperv_cs_tsc, 10);
+		clocksource_unregister(&hyperv_cs_tsc);
+
+		hypercall_msr.as_uint64 = 0;
+		wrmsrl(HV_X64_MSR_REFERENCE_TSC, hypercall_msr.as_uint64);
+		vfree(hv_context.tsc_page);
+		hv_context.tsc_page = NULL;
+	}
+#endif
 }
 
 /*
diff --git a/drivers/hv/hyperv_vmbus.h b/drivers/hv/hyperv_vmbus.h
index 6f25825..3d70e36 100644
--- a/drivers/hv/hyperv_vmbus.h
+++ b/drivers/hv/hyperv_vmbus.h
@@ -517,6 +517,7 @@ struct hv_context {
 	u64 guestid;
 
 	void *hypercall_page;
+	void *tsc_page;
 
 	bool synic_initialized;
 
@@ -560,6 +561,14 @@ struct hv_context {
 
 extern struct hv_context hv_context;
 
+struct ms_hyperv_tsc_page {
+	volatile u32 tsc_sequence;
+	u32 reserved1;
+	volatile u64 tsc_scale;
+	volatile s64 tsc_offset;
+	u64 reserved2[509];
+};
+
 struct hv_ring_buffer_debug_info {
 	u32 current_interrupt_mask;
 	u32 current_read_index;
-- 
1.7.4.1

[PATCH 23/28] Drivers: hv: vmbus: add a sysfs attr to show the binding of channel/VP

[K. Y. Srinivasan]

From: Dexuan Cui <decui@microsoft.com>

This is useful to analyze performance issue.

Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/vmbus_drv.c |   38 ++++++++++++++++++++++++++++++++++++++
 1 files changed, 38 insertions(+), 0 deletions(-)

diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index 8c3eaee..45d6d7a 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -441,6 +441,43 @@ static ssize_t in_write_bytes_avail_show(struct device *dev,
 }
 static DEVICE_ATTR_RO(in_write_bytes_avail);
 
+static ssize_t channel_vp_mapping_show(struct device *dev,
+				       struct device_attribute *dev_attr,
+				       char *buf)
+{
+	struct hv_device *hv_dev = device_to_hv_device(dev);
+	struct vmbus_channel *channel = hv_dev->channel, *cur_sc;
+	unsigned long flags;
+	int buf_size = PAGE_SIZE, n_written, tot_written;
+	struct list_head *cur;
+
+	if (!channel)
+		return -ENODEV;
+
+	tot_written = snprintf(buf, buf_size, "%u:%u\n",
+		channel->offermsg.child_relid, channel->target_cpu);
+
+	spin_lock_irqsave(&channel->lock, flags);
+
+	list_for_each(cur, &channel->sc_list) {
+		if (tot_written >= buf_size - 1)
+			break;
+
+		cur_sc = list_entry(cur, struct vmbus_channel, sc_list);
+		n_written = scnprintf(buf + tot_written,
+				     buf_size - tot_written,
+				     "%u:%u\n",
+				     cur_sc->offermsg.child_relid,
+				     cur_sc->target_cpu);
+		tot_written += n_written;
+	}
+
+	spin_unlock_irqrestore(&channel->lock, flags);
+
+	return tot_written;
+}
+static DEVICE_ATTR_RO(channel_vp_mapping);
+
 /* Set up per device attributes in /sys/bus/vmbus/devices/<bus device> */
 static struct attribute *vmbus_attrs[] = {
 	&dev_attr_id.attr,
@@ -465,6 +502,7 @@ static struct attribute *vmbus_attrs[] = {
 	&dev_attr_in_write_index.attr,
 	&dev_attr_in_read_bytes_avail.attr,
 	&dev_attr_in_write_bytes_avail.attr,
+	&dev_attr_channel_vp_mapping.attr,
 	NULL,
 };
 ATTRIBUTE_GROUPS(vmbus);
-- 
1.7.4.1

[PATCH 24/28] tools: hv: add a python script lsvmbus to list VMBus devices

[K. Y. Srinivasan]

From: Dexuan Cui <decui@microsoft.com>

By default lsvmbus lists all the devices in the VMBus.
With -v or -vv, more information is printed, including the VMBus
Rel_ID, class ID, device ID and which channel is bound to which
virtual processor, etc.

Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 tools/hv/lsvmbus |  101 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
 1 files changed, 101 insertions(+), 0 deletions(-)
 create mode 100644 tools/hv/lsvmbus

diff --git a/tools/hv/lsvmbus b/tools/hv/lsvmbus
new file mode 100644
index 0000000..162a378
--- /dev/null
+++ b/tools/hv/lsvmbus
@@ -0,0 +1,101 @@
+#!/usr/bin/env python
+
+import os
+from optparse import OptionParser
+
+parser = OptionParser()
+parser.add_option("-v", "--verbose", dest="verbose",
+		   help="print verbose messages. Try -vv, -vvv for \
+			more verbose messages", action="count")
+
+(options, args) = parser.parse_args()
+
+verbose = 0
+if options.verbose is not None:
+	verbose = options.verbose
+
+vmbus_sys_path = '/sys/bus/vmbus/devices'
+if not os.path.isdir(vmbus_sys_path):
+	print "%s doesn't exist: exiting..." % vmbus_sys_path
+	exit(-1)
+
+vmbus_dev_dict = {
+	'{0e0b6031-5213-4934-818b-38d90ced39db}' : '[Operating system shutdown]',
+	'{9527e630-d0ae-497b-adce-e80ab0175caf}' : '[Time Synchronization]',
+	'{57164f39-9115-4e78-ab55-382f3bd5422d}' : '[Heartbeat]',
+	'{a9a0f4e7-5a45-4d96-b827-8a841e8c03e6}' : '[Data Exchange]',
+	'{35fa2e29-ea23-4236-96ae-3a6ebacba440}' : '[Backup (volume checkpoint)]',
+	'{34d14be3-dee4-41c8-9ae7-6b174977c192}' : '[Guest services]',
+	'{525074dc-8985-46e2-8057-a307dc18a502}' : '[Dynamic Memory]',
+	'{cfa8b69e-5b4a-4cc0-b98b-8ba1a1f3f95a}' : 'Synthetic mouse',
+	'{f912ad6d-2b17-48ea-bd65-f927a61c7684}' : 'Synthetic keyboard',
+	'{da0a7802-e377-4aac-8e77-0558eb1073f8}' : 'Synthetic framebuffer adapter',
+	'{f8615163-df3e-46c5-913f-f2d2f965ed0e}' : 'Synthetic network adapter',
+	'{32412632-86cb-44a2-9b5c-50d1417354f5}' : 'Synthetic IDE Controller',
+	'{ba6163d9-04a1-4d29-b605-72e2ffb1dc7f}' : 'Synthetic SCSI Controller',
+	'{2f9bcc4a-0069-4af3-b76b-6fd0be528cda}' : 'Synthetic fiber channel adapter',
+	'{8c2eaf3d-32a7-4b09-ab99-bd1f1c86b501}' : 'Synthetic RDMA adapter',
+	'{276aacf4-ac15-426c-98dd-7521ad3f01fe}' : '[Reserved system device]',
+	'{f8e65716-3cb3-4a06-9a60-1889c5cccab5}' : '[Reserved system device]',
+	'{3375baf4-9e15-4b30-b765-67acb10d607b}' : '[Reserved system device]',
+}
+
+def get_vmbus_dev_attr(dev_name, attr):
+	try:
+		f = open('%s/%s/%s' % (vmbus_sys_path, dev_name, attr), 'r')
+		lines = f.readlines()
+		f.close()
+	except IOError:
+		lines = []
+
+	return lines
+
+class VMBus_Dev:
+	pass
+
+
+vmbus_dev_list = []
+
+for f in os.listdir(vmbus_sys_path):
+	vmbus_id = get_vmbus_dev_attr(f, 'id')[0].strip()
+	class_id = get_vmbus_dev_attr(f, 'class_id')[0].strip()
+	device_id = get_vmbus_dev_attr(f, 'device_id')[0].strip()
+	dev_desc = vmbus_dev_dict.get(class_id, 'Unknown')
+
+	chn_vp_mapping = get_vmbus_dev_attr(f, 'channel_vp_mapping')
+	chn_vp_mapping = [c.strip() for c in chn_vp_mapping]
+	chn_vp_mapping = sorted(chn_vp_mapping,
+		key = lambda c : int(c.split(':')[0]))
+
+	chn_vp_mapping = ['\tRel_ID=%s, target_cpu=%s' %
+				(c.split(':')[0], c.split(':')[1])
+					for c in chn_vp_mapping]
+	d = VMBus_Dev()
+	d.sysfs_path = '%s/%s' % (vmbus_sys_path, f)
+	d.vmbus_id = vmbus_id
+	d.class_id = class_id
+	d.device_id = device_id
+	d.dev_desc = dev_desc
+	d.chn_vp_mapping = '\n'.join(chn_vp_mapping)
+	if d.chn_vp_mapping:
+		d.chn_vp_mapping += '\n'
+
+	vmbus_dev_list.append(d)
+
+
+vmbus_dev_list  = sorted(vmbus_dev_list, key = lambda d : int(d.vmbus_id))
+
+format0 = '%2s: %s'
+format1 = '%2s: Class_ID = %s - %s\n%s'
+format2 = '%2s: Class_ID = %s - %s\n\tDevice_ID = %s\n\tSysfs path: %s\n%s'
+
+for d in vmbus_dev_list:
+	if verbose == 0:
+		print ('VMBUS ID ' + format0) % (d.vmbus_id, d.dev_desc)
+	elif verbose == 1:
+		print ('VMBUS ID ' + format1) %	\
+			(d.vmbus_id, d.class_id, d.dev_desc, d.chn_vp_mapping)
+	else:
+		print ('VMBUS ID ' + format2) % \
+			(d.vmbus_id, d.class_id, d.dev_desc, \
+			d.device_id, d.sysfs_path, d.chn_vp_mapping)
-- 
1.7.4.1

[PATCH 25/28] Drivers: hv: vmbus: document the VMBus sysfs files

[K. Y. Srinivasan]

From: Dexuan Cui <decui@microsoft.com>

The 4 sysfs files should be stable ABIs to the user space.

Signed-off-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 Documentation/ABI/stable/sysfs-bus-vmbus |   29 +++++++++++++++++++++++++++++
 MAINTAINERS                              |    1 +
 2 files changed, 30 insertions(+), 0 deletions(-)
 create mode 100644 Documentation/ABI/stable/sysfs-bus-vmbus

diff --git a/Documentation/ABI/stable/sysfs-bus-vmbus b/Documentation/ABI/stable/sysfs-bus-vmbus
new file mode 100644
index 0000000..636e938
--- /dev/null
+++ b/Documentation/ABI/stable/sysfs-bus-vmbus
@@ -0,0 +1,29 @@
+What:		/sys/bus/vmbus/devices/vmbus_*/id
+Date:		Jul 2009
+KernelVersion:	2.6.31
+Contact:	K. Y. Srinivasan <kys@microsoft.com>
+Description:	The VMBus child_relid of the device's primary channel
+Users:		tools/hv/lsvmbus
+
+What:		/sys/bus/vmbus/devices/vmbus_*/class_id
+Date:		Jul 2009
+KernelVersion:	2.6.31
+Contact:	K. Y. Srinivasan <kys@microsoft.com>
+Description:	The VMBus interface type GUID of the device
+Users:		tools/hv/lsvmbus
+
+What:		/sys/bus/vmbus/devices/vmbus_*/device_id
+Date:		Jul 2009
+KernelVersion:	2.6.31
+Contact:	K. Y. Srinivasan <kys@microsoft.com>
+Description:	The VMBus interface instance GUID of the device
+Users:		tools/hv/lsvmbus
+
+What:		/sys/bus/vmbus/devices/vmbus_*/channel_vp_mapping
+Date:		Jul 2015
+KernelVersion:	4.2.0
+Contact:	K. Y. Srinivasan <kys@microsoft.com>
+Description:	The mapping of which primary/sub channels are bound to which
+		Virtual Processors.
+		Format: <channel's child_relid:the bound cpu's number>
+Users:		tools/hv/lsvmbus
diff --git a/MAINTAINERS b/MAINTAINERS
index fd60784..45ef1e1 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -4956,6 +4956,7 @@ F:	drivers/scsi/storvsc_drv.c
 F:	drivers/video/fbdev/hyperv_fb.c
 F:	include/linux/hyperv.h
 F:	tools/hv/
+F:	Documentation/ABI/stable/sysfs-bus-vmbus
 
 I2C OVER PARALLEL PORT
 M:	Jean Delvare <jdelvare@suse.de>
-- 
1.7.4.1

[PATCH 26/28] cpu-hotplug: convert cpu_hotplug_disabled to a counter

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

As a prerequisite to exporting cpu_hotplug_enable/cpu_hotplug_disable
functions to modules we need to convert cpu_hotplug_disabled to a counter
to properly support disable -> disable -> enable call sequences. E.g.
after Hyper-V vmbus module (which is supposed to be the first user of
exported cpu_hotplug_enable/cpu_hotplug_disable) did cpu_hotplug_disable()
hibernate path calls disable_nonboot_cpus() and if we hit an error in
_cpu_down() enable_nonboot_cpus() will be called on the failure path (thus
making cpu_hotplug_disabled = 0 and leaving cpu hotplug in 'enabled'
state). Same problem is possible if more than 1 module use
cpu_hotplug_disable/cpu_hotplug_enable on their load/unload paths. When
one of these modules is been unloaded it is logical to leave cpu hotplug
in 'disabled' state.

To support the change we need to increse cpu_hotplug_disabled counter
in disable_nonboot_cpus() unconditionally as all users of
disable_nonboot_cpus() are supposed to do enable_nonboot_cpus() in case
an error was returned.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 Documentation/power/suspend-and-cpuhotplug.txt |    6 +++---
 kernel/cpu.c                                   |   21 +++++++++++++--------
 2 files changed, 16 insertions(+), 11 deletions(-)

diff --git a/Documentation/power/suspend-and-cpuhotplug.txt b/Documentation/power/suspend-and-cpuhotplug.txt
index 2850df3..2fc9095 100644
--- a/Documentation/power/suspend-and-cpuhotplug.txt
+++ b/Documentation/power/suspend-and-cpuhotplug.txt
@@ -72,7 +72,7 @@ More details follow:
                                         |
                                         v
                            Disable regular cpu hotplug
-                        by setting cpu_hotplug_disabled=1
+                        by increasing cpu_hotplug_disabled
                                         |
                                         v
                             Release cpu_add_remove_lock
@@ -89,7 +89,7 @@ Resuming back is likewise, with the counterparts being (in the order of
 execution during resume):
 * enable_nonboot_cpus() which involves:
    |  Acquire cpu_add_remove_lock
-   |  Reset cpu_hotplug_disabled to 0, thereby enabling regular cpu hotplug
+   |  Decrease cpu_hotplug_disabled, thereby enabling regular cpu hotplug
    |  Call _cpu_up() [for all those cpus in the frozen_cpus mask, in a loop]
    |  Release cpu_add_remove_lock
    v
@@ -120,7 +120,7 @@ after the entire cycle is complete (i.e., suspend + resume).
                            Acquire cpu_add_remove_lock
                                         |
                                         v
-                          If cpu_hotplug_disabled is 1
+                          If cpu_hotplug_disabled > 0
                                 return gracefully
                                         |
                                         |
diff --git a/kernel/cpu.c b/kernel/cpu.c
index 6a37454..1225fc2 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -191,14 +191,14 @@ void cpu_hotplug_done(void)
 void cpu_hotplug_disable(void)
 {
 	cpu_maps_update_begin();
-	cpu_hotplug_disabled = 1;
+	cpu_hotplug_disabled++;
 	cpu_maps_update_done();
 }
 
 void cpu_hotplug_enable(void)
 {
 	cpu_maps_update_begin();
-	cpu_hotplug_disabled = 0;
+	WARN_ON(--cpu_hotplug_disabled < 0);
 	cpu_maps_update_done();
 }
 
@@ -617,13 +617,18 @@ int disable_nonboot_cpus(void)
 		}
 	}
 
-	if (!error) {
+	if (!error)
 		BUG_ON(num_online_cpus() > 1);
-		/* Make sure the CPUs won't be enabled by someone else */
-		cpu_hotplug_disabled = 1;
-	} else {
+	else
 		pr_err("Non-boot CPUs are not disabled\n");
-	}
+
+	/*
+	 * Make sure the CPUs won't be enabled by someone else. We need to do
+	 * this even in case of failure as all disable_nonboot_cpus() users are
+	 * supposed to do enable_nonboot_cpus() on the failure path.
+	 */
+	cpu_hotplug_disabled++;
+
 	cpu_maps_update_done();
 	return error;
 }
@@ -642,7 +647,7 @@ void __ref enable_nonboot_cpus(void)
 
 	/* Allow everyone to use the CPU hotplug again */
 	cpu_maps_update_begin();
-	cpu_hotplug_disabled = 0;
+	WARN_ON(--cpu_hotplug_disabled < 0);
 	if (cpumask_empty(frozen_cpus))
 		goto out;
 
-- 
1.7.4.1

[PATCH 27/28] cpu-hotplug: export cpu_hotplug_enable/cpu_hotplug_disable

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

Hyper-V module needs to disable cpu hotplug (offlining) as there is no
support from hypervisor side to reassign already opened event channels
to a different CPU. Currently it is been done by altering
smp_ops.cpu_disable but it is hackish.

Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 kernel/cpu.c |    3 ++-
 1 files changed, 2 insertions(+), 1 deletions(-)

diff --git a/kernel/cpu.c b/kernel/cpu.c
index 1225fc2..18f00ae 100644
--- a/kernel/cpu.c
+++ b/kernel/cpu.c
@@ -194,6 +194,7 @@ void cpu_hotplug_disable(void)
 	cpu_hotplug_disabled++;
 	cpu_maps_update_done();
 }
+EXPORT_SYMBOL_GPL(cpu_hotplug_disable);
 
 void cpu_hotplug_enable(void)
 {
@@ -201,7 +202,7 @@ void cpu_hotplug_enable(void)
 	WARN_ON(--cpu_hotplug_disabled < 0);
 	cpu_maps_update_done();
 }
-
+EXPORT_SYMBOL_GPL(cpu_hotplug_enable);
 #endif	/* CONFIG_HOTPLUG_CPU */
 
 /* Need to know about CPUs going up/down? */
-- 
1.7.4.1

[PATCH 28/28] Drivers: hv: vmbus: use cpu_hotplug_enable/disable

[K. Y. Srinivasan]

From: Vitaly Kuznetsov <vkuznets@redhat.com>

Commit e513229b4c38 ("Drivers: hv: vmbus: prevent cpu offlining on newer
hypervisors") was altering smp_ops.cpu_disable to prevent CPU offlining.
We can bo better by using cpu_hotplug_enable/disable functions instead of
such hard-coding.

Reported-by: Radim Kr.má  <rkrcmar@redhat.com>
Signed-off-by: Vitaly Kuznetsov <vkuznets@redhat.com>
Signed-off-by: K. Y. Srinivasan <kys@microsoft.com>
---
 drivers/hv/vmbus_drv.c |   38 ++++----------------------------------
 1 files changed, 4 insertions(+), 34 deletions(-)

diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
index 45d6d7a..f19b6f7 100644
--- a/drivers/hv/vmbus_drv.c
+++ b/drivers/hv/vmbus_drv.c
@@ -828,38 +828,6 @@ static void vmbus_isr(void)
 	}
 }
 
-#ifdef CONFIG_HOTPLUG_CPU
-static int hyperv_cpu_disable(void)
-{
-	return -ENOSYS;
-}
-
-static void hv_cpu_hotplug_quirk(bool vmbus_loaded)
-{
-	static void *previous_cpu_disable;
-
-	/*
-	 * Offlining a CPU when running on newer hypervisors (WS2012R2, Win8,
-	 * ...) is not supported at this moment as channel interrupts are
-	 * distributed across all of them.
-	 */
-
-	if ((vmbus_proto_version == VERSION_WS2008) ||
-	    (vmbus_proto_version == VERSION_WIN7))
-		return;
-
-	if (vmbus_loaded) {
-		previous_cpu_disable = smp_ops.cpu_disable;
-		smp_ops.cpu_disable = hyperv_cpu_disable;
-		pr_notice("CPU offlining is not supported by hypervisor\n");
-	} else if (previous_cpu_disable)
-		smp_ops.cpu_disable = previous_cpu_disable;
-}
-#else
-static void hv_cpu_hotplug_quirk(bool vmbus_loaded)
-{
-}
-#endif
 
 /*
  * vmbus_bus_init -Main vmbus driver initialization routine.
@@ -901,7 +869,8 @@ static int vmbus_bus_init(int irq)
 	if (ret)
 		goto err_alloc;
 
-	hv_cpu_hotplug_quirk(true);
+	if (vmbus_proto_version > VERSION_WIN7)
+		cpu_hotplug_disable();
 
 	/*
 	 * Only register if the crash MSRs are available
@@ -1364,7 +1333,8 @@ static void __exit vmbus_exit(void)
 	}
 	hv_synic_free();
 	acpi_bus_unregister_driver(&vmbus_acpi_driver);
-	hv_cpu_hotplug_quirk(false);
+	if (vmbus_proto_version > VERSION_WIN7)
+		cpu_hotplug_enable();
 }
 
 
-- 
1.7.4.1

Re: [PATCH 11/28] drivers:hv: Modify hv_vmbus to search for all MMIO ranges available.

[Greg KH]

On Sat, Aug 01, 2015 at 04:08:15PM -0700, K. Y. Srinivasan wrote:
> From: jakeo@microsoft.com <jakeo@microsoft.com>

That's not his name :(

I've stopped here, please fix up and resend the rest of the series.

thanks,

greg k-h

RE: [PATCH 11/28] drivers:hv: Modify hv_vmbus to search for all MMIO ranges available.

[KY Srinivasan]

> -----Original Message-----
> From: Greg KH [mailto:gregkh@linuxfoundation.org]
> Sent: Tuesday, August 4, 2015 10:29 PM
> To: KY Srinivasan <kys@microsoft.com>
> Cc: linux-kernel@vger.kernel.org; devel@linuxdriverproject.org;
> olaf@aepfle.de; apw@canonical.com; vkuznets@redhat.com;
> jasowang@redhat.com; Jake Oshins <jakeo@microsoft.com>
> Subject: Re: [PATCH 11/28] drivers:hv: Modify hv_vmbus to search for all
> MMIO ranges available.
> 
> On Sat, Aug 01, 2015 at 04:08:15PM -0700, K. Y. Srinivasan wrote:
> > From: jakeo@microsoft.com <jakeo@microsoft.com>
> 
> That's not his name :(

Sorry about that.

> 
> I've stopped here, please fix up and resend the rest of the series.
> 
 I will resend shortly.

Thanks,

K. Y