@@ -1,6 +1,6 @@

 Summary:	Linux API header files

 Name:		linux-api-headers

-Version:	4.9.26

+Version:	4.9.27

 Release:	1%{?dist}

 License:	GPLv2

 URL:		http://www.kernel.org/

@@ -8,7 +8,7 @@ Group:		System Environment/Kernel

 Vendor:		VMware, Inc.

 Distribution: Photon

 Source0:        http://www.kernel.org/pub/linux/kernel/v4.x/linux-%{version}.tar.xz

-%define sha1 linux=b244ab8ee3d7a0385c7bc1b1dc1d55f0920df997

+%define sha1 linux=d248ff518f822d4324a482a39f3159b1ee86cc84

 BuildArch:	noarch

 %description

 The Linux API Headers expose the kernel's API for use by Glibc.

@@ -25,6 +25,8 @@ find /%{buildroot}%{_includedir} \( -name .install -o -name ..install.cmd \) -de

 %defattr(-,root,root)

 %{_includedir}/*

 %changelog

+*   Wed May 10 2017 Alexey Makhalov <amakhalov@vmware.com> 4.9.27-1

+-   Update to linux-4.9.27

 *   Sun May 7 2017 Alexey Makhalov <amakhalov@vmware.com> 4.9.26-1

 -   Update to linux-4.9.26

 *   Tue Apr 25 2017 Alexey Makhalov <amakhalov@vmware.com> 4.9.24-1

@@ -1,5 +1,17 @@

+From 7d80ae3e4f1dd73fb61010e5a84135d3e92ba47e Mon Sep 17 00:00:00 2001

+From: Alexey Makhalov <amakhalov@vmware.com>

+Date: Tue, 9 May 2017 12:26:58 -0700

+Subject: [PATCH] Patch from Clear Linux

+

+---

+ drivers/video/console/fbcon.c | 2 +-

+ include/linux/blkdev.h        | 2 +-

+ init/do_mounts.c              | 4 +++-

+ mm/vmstat.c                   | 2 +-

+ 4 files changed, 6 insertions(+), 4 deletions(-)

+

 diff --git a/drivers/video/console/fbcon.c b/drivers/video/console/fbcon.c

-index 92f3949..4ff61d7 100644

+index 4db10d7990c9..1f84959cc87e 100644

 --- a/drivers/video/console/fbcon.c

 +++ b/drivers/video/console/fbcon.c

 @@ -146,7 +146,7 @@ static const struct consw fb_con;

@@ -12,10 +24,10 @@ index 92f3949..4ff61d7 100644

  #define divides(a, b)	((!(a) || (b)%(a)) ? 0 : 1)

 diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h

-index c70e358..72f56a8 100644

+index f6a816129856..f4fb458a91dd 100644

 --- a/include/linux/blkdev.h

 +++ b/include/linux/blkdev.h

-@@ -38,7 +38,7 @@ struct blk_flush_queue;

+@@ -39,7 +39,7 @@ struct blk_flush_queue;

  struct pr_ops;

  #define BLKDEV_MIN_RQ	4

@@ -25,7 +37,7 @@ index c70e358..72f56a8 100644

  /*

   * Maximum number of blkcg policies allowed to be registered concurrently.

 diff --git a/init/do_mounts.c b/init/do_mounts.c

-index dea5de9..da84094 100644

+index dea5de95c2dd..da840946d90b 100644

 --- a/init/do_mounts.c

 +++ b/init/do_mounts.c

 @@ -28,6 +28,7 @@

@@ -47,10 +59,10 @@ index dea5de9..da84094 100644

  	md_run_setup();

 diff --git a/mm/vmstat.c b/mm/vmstat.c

-index c54fd29..f0baa1a 100644

+index 604f26a4f696..bfbbcd76df57 100644

 --- a/mm/vmstat.c

 +++ b/mm/vmstat.c

-@@ -1381,7 +1381,7 @@ static const struct file_operations proc_vmstat_file_operations = {

+@@ -1549,7 +1549,7 @@ static const struct file_operations proc_vmstat_file_operations = {

  #ifdef CONFIG_SMP

  static struct workqueue_struct *vmstat_wq;

  static DEFINE_PER_CPU(struct delayed_work, vmstat_work);

@@ -60,5 +72,5 @@ index c54fd29..f0baa1a 100644

  #ifdef CONFIG_PROC_FS

  static void refresh_vm_stats(struct work_struct *work)

 -- 

-1.9.1

+2.11.0

@@ -1,20 +1,55 @@

-From a2d9f1f747247f31f769930a8198b12a0667d21f Mon Sep 17 00:00:00 2001

+From c6e73db5fafa151b81a969a74c3138488a4310fa Mon Sep 17 00:00:00 2001

 From: Alexey Makhalov <amakhalov@vmware.com>

 Date: Tue, 29 Sep 2015 15:55:49 -0700

 Subject: [PATCH] pci/probe.c: Hardcodded pci probe.

+Added pci=scan_all cmdline parameter to verify hardcodded pci values

 ---

- drivers/pci/probe.c | 312 +++++++++++++++++++++++++++++++++++++++++++++++++++-

- 1 file changed, 310 insertions(+), 2 deletions(-)

+ drivers/pci/pci.c   |   2 +

+ drivers/pci/pci.h   |   2 +

+ drivers/pci/probe.c | 377 +++++++++++++++++++++++++++++++++++++++++++++++++++-

+ 3 files changed, 377 insertions(+), 4 deletions(-)

+diff --git a/drivers/pci/pci.c b/drivers/pci/pci.c

+index 6922964e3dff..e4db16e7c1d2 100644

+--- a/drivers/pci/pci.c

+@@ -5303,6 +5303,8 @@ static int __init pci_setup(char *str)

+ 				pcie_bus_config = PCIE_BUS_PEER2PEER;

+ 			} else if (!strncmp(str, "pcie_scan_all", 13)) {

+ 				pci_add_flags(PCI_SCAN_ALL_PCIE_DEVS);

++			} else if (!strncmp(str, "scan_all", 8)) {

++				pci_scan_all();

+ 			} else {

+ 				printk(KERN_ERR "PCI: Unknown option `%s'\n",

+ 						str);

+diff --git a/drivers/pci/pci.h b/drivers/pci/pci.h

+index a5d37f6a9fb5..aaa456c9bf95 100644

+--- a/drivers/pci/pci.h

+@@ -163,6 +163,8 @@ void pci_no_msi(void);

+ static inline void pci_no_msi(void) { }

+ #endif

+ 

++void pci_scan_all(void);

++

+ static inline void pci_msi_set_enable(struct pci_dev *dev, int enable)

+ {

+ 	u16 control;

 diff --git a/drivers/pci/probe.c b/drivers/pci/probe.c

-index edb1984..522bdaa 100644

+index d266d800f246..11544ddd6e5c 100644

 --- a/drivers/pci/probe.c

 +++ b/drivers/pci/probe.c

-@@ -162,6 +162,281 @@ static inline unsigned long decode_bar(struct pci_dev *dev, u32 bar)

+@@ -163,6 +163,346 @@ static inline unsigned long decode_bar(struct pci_dev *dev, u32 bar)

  #define PCI_COMMAND_DECODE_ENABLE	(PCI_COMMAND_MEMORY | PCI_COMMAND_IO)

++static int pci_scan_and_verify = 0;

++

++void pci_scan_all(void)

++{

++	pci_scan_and_verify = 1;

++}

 +

 +/* shortcut version of __pci_read_base where we know the sizes already */

 +int __pci_read_base_shortcut(struct pci_dev *dev, enum pci_bar_type type,

@@ -58,7 +93,8 @@ index edb1984..522bdaa 100644

 +			mask64 = (u32)PCI_BASE_ADDRESS_MEM_MASK;

 +		}

 +	} else {

-+		res->flags |= (l & IORESOURCE_ROM_ENABLE);

++		if (l & PCI_ROM_ADDRESS_ENABLE)

++			res->flags |= IORESOURCE_ROM_ENABLE;

 +		l64 = l & PCI_ROM_ADDRESS_MASK;

 +		sz64 = sz & PCI_ROM_ADDRESS_MASK;

 +		mask64 = (u32)PCI_ROM_ADDRESS_MASK;

@@ -142,6 +178,35 @@ index edb1984..522bdaa 100644

 +	return (res->flags & IORESOURCE_MEM_64) ? 1 : 0;

 +}

 +

++static int guess_bar_count(int class)

++{

++        if (class == 0x088000)

++            return 2;

++        if (class == 0x068000)

++            return 0;

++        if (class == 0x060100)

++            return 0;

++        if (class == 0x060000)

++            return 0;

++        if (class == 0x030000)

++            return 3;

++        if (class == 0x020000 || class == 0x010000)

++            return 4;

++        if (class == 0x00ff00)

++            return 1;

++        return 6;

++}

++

++static int has_rom(int class, int rom)

++{

++        if (class == 0x088000 ||

++	    class == 0x060100 ||

++	    class == 0x060000 ||

++	    class == 0x00ff00)

++          return 0;

++        return rom;

++}

++

 +static int is_known_device(struct pci_dev *dev, int pos, int *sz)

 +{

 +	switch (dev->vendor) {

@@ -192,8 +257,9 @@ index edb1984..522bdaa 100644

 +					*sz = 0xfffffff1;

 +					return 1;

 +				case 0x14:

-+					*sz = 0xfc000000;

-+					return 1;

++					/* variable mem size */

++					*sz = 0xffffffff;

++					return 0;

 +				case 0x18:

 +					*sz = 0xff800000;

 +					return 1;

@@ -215,7 +281,7 @@ index edb1984..522bdaa 100644

 +					*sz = 0xffffffc1;

 +					return 1;

 +				case 0x14:

-+					*sz = 0xffffe000;

++					*sz = 0xffffe004;

 +					return 1;

 +				case 0x1c:

 +				case 0x20:

@@ -286,19 +352,55 @@ index edb1984..522bdaa 100644

 +		}

 +		break;

 +

++		/* SCSI storage controller */

++		case 0x1000:

++		switch (dev->device) {

++			/* LSI Logic */

++			case 0x0030:

++			switch (pos) {

++				case 0x10:

++					*sz = 0xff01;

++					return 1;

++				case 0x14:

++					*sz = 0xfffe0004;

++					return 1;

++				case 0x1c:

++					*sz = 0xfffe0004;

++					return 1;

++				case 0x18:

++				case 0x20:

++				case 0x24:

++					*sz = 0; /* Not implemented */

++					return 1;

++				case 0x30:

++					*sz = 0xffffc000;

++					return 1;

++			}

++			break;

++		}

++		break;

 +	}

++	*sz = 0xffffffff;

 +        return 0;

 +}

 +

  /**

   * pci_read_base - read a PCI BAR

   * @dev: the PCI device

-@@ -181,6 +456,13 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,

+@@ -175,13 +515,20 @@ static inline unsigned long decode_bar(struct pci_dev *dev, u32 bar)

+ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,

+ 		    struct resource *res, unsigned int pos)

+ {

+-	u32 l, sz, mask;

++	u32 l, sz, known_sz, mask;

+ 	u64 l64, sz64, mask64;

+ 	u16 orig_cmd;

+ 	struct pci_bus_region region, inverted_region;

  	mask = type ? PCI_ROM_ADDRESS_MASK : ~0;

-+	if (is_known_device(dev, pos, &sz))

-+		return __pci_read_base_shortcut(dev, type, res, pos, sz, 0);

++	if (is_known_device(dev, pos, &known_sz) && !pci_scan_and_verify)

++		return __pci_read_base_shortcut(dev, type, res, pos, known_sz, 0);

 +

 +	res->name = pci_name(dev);

 +

@@ -307,7 +409,7 @@ index edb1984..522bdaa 100644

  	/* No printks while decoding is disabled! */

  	if (!dev->mmio_always_on) {

  		pci_read_config_word(dev, PCI_COMMAND, &orig_cmd);

-@@ -190,7 +472,6 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,

+@@ -191,7 +538,6 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,

  		}

  	}

@@ -315,56 +417,55 @@ index edb1984..522bdaa 100644

  	pci_read_config_dword(dev, pos, &l);

  	pci_write_config_dword(dev, pos, l | mask);

-@@ -1145,6 +1426,28 @@ int pci_cfg_space_size(struct pci_dev *dev)

+@@ -207,6 +553,11 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,

+ 	if (sz == 0xffffffff)

+ 		sz = 0;

- #define LEGACY_IO_RESOURCE	(IORESOURCE_IO | IORESOURCE_PCI_FIXED)

- 

-+static int guess_bar_count(int class)

-+{

-+        if (class == 0x068000)

-+            return 0;

-+        if (class == 0x020000)

-+            return 3;

-+        if (class == 0x010000)

-+            return 2;

-+        if (class == 0x00ff00)

-+            return 1;

-+        return 6;

-+}

++	if (pci_scan_and_verify && known_sz != 0xffffffff && known_sz != sz)

++		dev_err(&dev->dev,

++			"wrong known bar size: %x != %x\n",

++			known_sz, sz);

 +

-+static int has_rom(int class, int rom)

-+{

-+        if (class == 0x020000)

-+          return 0;

-+        if (class == 0x010000 || class == 0x00ff00)

-+          return 0;

-+        return rom;

-+}

-+

- static void pci_msi_setup_pci_dev(struct pci_dev *dev)

- {

  	/*

-@@ -1178,6 +1481,7 @@ int pci_setup_device(struct pci_dev *dev)

- 	int pos = 0;

- 	struct pci_bus_region region;

- 	struct resource *res;

-+	int maxbar;

+ 	 * I don't know how l can have all bits set.  Copied from old code.

+ 	 * Maybe it fixes a bug on some ancient platform.

+@@ -311,8 +662,19 @@ int __pci_read_base(struct pci_dev *dev, enum pci_bar_type type,

+ fail:

+ 	res->flags = 0;

+ out:

+-	if (res->flags)

++	if (res->flags) {

+ 		dev_printk(KERN_DEBUG, &dev->dev, "reg 0x%x: %pR\n", pos, res);

++		if (pci_scan_and_verify) {

++			int bar = (pos - PCI_BASE_ADDRESS_0) >> 2;

++			int maxbar = guess_bar_count(dev->class);

++			if (bar < 6 && bar >= maxbar)

++				dev_err(&dev->dev,

++					"wrong guess_bar_count value %d\n",

++					maxbar);

++			if (pos == PCI_ROM_ADDRESS && !has_rom(dev->class, 1))

++				dev_err(&dev->dev, "wrong has_rom value\n");

++		}

++	}

- 	if (pci_read_config_byte(dev, PCI_HEADER_TYPE, &hdr_type))

- 		return -EIO;

-@@ -1224,7 +1528,11 @@ int pci_setup_device(struct pci_dev *dev)

+ 	return (res->flags & IORESOURCE_MEM_64) ? 1 : 0;

+ }

+@@ -1265,7 +1627,14 @@ int pci_setup_device(struct pci_dev *dev)

  		if (class == PCI_CLASS_BRIDGE_PCI)

  			goto bad;

  		pci_read_irq(dev);

 -		pci_read_bases(dev, 6, PCI_ROM_ADDRESS);

 +

-+		maxbar = guess_bar_count(dev->class);

++		if (pci_scan_and_verify)

++			pci_read_bases(dev, 6, PCI_ROM_ADDRESS);

++		else if (class != PCI_CLASS_STORAGE_IDE)

++			pci_read_bases(dev,

++			guess_bar_count(dev->class),

++			has_rom(dev->class, PCI_ROM_ADDRESS));

 +

-+		if (class != PCI_CLASS_STORAGE_IDE)

-+			pci_read_bases(dev, maxbar, has_rom(dev->class, PCI_ROM_ADDRESS));

  		pci_read_config_word(dev, PCI_SUBSYSTEM_VENDOR_ID, &dev->subsystem_vendor);

  		pci_read_config_word(dev, PCI_SUBSYSTEM_ID, &dev->subsystem_device);

 -- 

-1.9.1

+2.11.0

@@ -1,5 +1,14 @@

+From 725ac7c54935bbff858092e2db302afe88062162 Mon Sep 17 00:00:00 2001

+From: Alexey Makhalov <amakhalov@vmware.com>

+Date: Tue, 9 May 2017 12:39:18 -0700

+Subject: [PATCH] x86/vmware: pv-ops clocksource

+

+---

+ arch/x86/kernel/cpu/vmware.c | 21 +++++++++++++++++++++

+ 1 file changed, 21 insertions(+)

+

 diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c

-index c998fba..8129f99 100644

+index 70b8c4614e22..987ac571d16c 100644

 --- a/arch/x86/kernel/cpu/vmware.c

 +++ b/arch/x86/kernel/cpu/vmware.c

 @@ -155,6 +155,26 @@ static void __init vmware_cyc2ns_setup(void)

@@ -37,3 +46,6 @@ index c998fba..8129f99 100644

  }

  #else

  #define vmware_paravirt_ops_setup() do {} while (0)

+-- 

+2.11.0

+

@@ -1,5 +1,19 @@

+From b2ce2307e8547e420fcce9fbad46119814f7f810 Mon Sep 17 00:00:00 2001

+From: Alexey Makhalov <amakhalov@vmware.com>

+Date: Tue, 9 May 2017 12:39:57 -0700

+Subject: [PATCH] x86/vmware: pv-ops boot_clock

+

+---

+ arch/x86/include/asm/paravirt.h       |  5 +++++

+ arch/x86/include/asm/paravirt_types.h |  5 +++++

+ arch/x86/kernel/cpu/vmware.c          | 24 +++++++++++++++++++++++-

+ arch/x86/kernel/head_64.S             |  8 ++++++++

+ arch/x86/kernel/paravirt.c            |  7 +++++++

+ arch/x86/kernel/setup.c               |  9 +++++++++

+ 6 files changed, 57 insertions(+), 1 deletion(-)

+

 diff --git a/arch/x86/include/asm/paravirt.h b/arch/x86/include/asm/paravirt.h

-index ce93281..3d4ba2a 100644

+index ce932812f142..3d4ba2a999cb 100644

 --- a/arch/x86/include/asm/paravirt.h

 +++ b/arch/x86/include/asm/paravirt.h

 @@ -193,6 +193,11 @@ static inline u64 paravirt_steal_clock(int cpu)

@@ -15,7 +29,7 @@ index ce93281..3d4ba2a 100644

  {

  	return PVOP_CALL1(u64, pv_cpu_ops.read_pmc, counter);

 diff --git a/arch/x86/include/asm/paravirt_types.h b/arch/x86/include/asm/paravirt_types.h

-index 0f400c0..5f09979 100644

+index 0f400c0e4979..5f09979f612d 100644

 --- a/arch/x86/include/asm/paravirt_types.h

 +++ b/arch/x86/include/asm/paravirt_types.h

 @@ -51,6 +51,10 @@ struct mm_struct;

@@ -38,7 +52,7 @@ index 0f400c0..5f09979 100644

  struct pv_cpu_ops {

 diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c

-index 8129f99..387ac46 100644

+index 987ac571d16c..0e1fc6e17efc 100644

 --- a/arch/x86/kernel/cpu/vmware.c

 +++ b/arch/x86/kernel/cpu/vmware.c

 @@ -124,6 +124,7 @@ static struct kmsg_dumper kmsg_dumper = {

@@ -94,7 +108,7 @@ index 8129f99..387ac46 100644

  #else

  #define vmware_paravirt_ops_setup() do {} while (0)

 diff --git a/arch/x86/kernel/head_64.S b/arch/x86/kernel/head_64.S

-index b4421cc..f7b8017 100644

+index b4421cc191b0..f7b8017fa90a 100644

 --- a/arch/x86/kernel/head_64.S

 +++ b/arch/x86/kernel/head_64.S

 @@ -184,6 +184,14 @@ ENTRY(secondary_startup_64)

@@ -113,7 +127,7 @@ index b4421cc..f7b8017 100644

  	call verify_cpu

 diff --git a/arch/x86/kernel/paravirt.c b/arch/x86/kernel/paravirt.c

-index bbf3d59..10bbdb3 100644

+index bbf3d5933eaa..10bbdb3e6651 100644

 --- a/arch/x86/kernel/paravirt.c

 +++ b/arch/x86/kernel/paravirt.c

 @@ -203,6 +203,12 @@ static u64 native_steal_clock(int cpu)

@@ -138,7 +152,7 @@ index bbf3d59..10bbdb3 100644

  __visible struct pv_irq_ops pv_irq_ops = {

 diff --git a/arch/x86/kernel/setup.c b/arch/x86/kernel/setup.c

-index 9c337b0..67a42e5 100644

+index 9c337b0e8ba7..67a42e58b118 100644

 --- a/arch/x86/kernel/setup.c

 +++ b/arch/x86/kernel/setup.c

 @@ -1302,3 +1302,12 @@ void arch_show_smap(struct seq_file *m, struct vm_area_struct *vma)

@@ -154,3 +168,6 @@ index 9c337b0..67a42e5 100644

 +	paravirt_read_boot_clock64(ts);

 +}

 +#endif

+-- 

+2.11.0

+

@@ -1,8 +1,20 @@

+From ffb4b625e8b2f0ce0f31433564973eb2c80a7ef5 Mon Sep 17 00:00:00 2001

+From: Alexey Makhalov <amakhalov@vmware.com>

+Date: Tue, 9 May 2017 12:41:17 -0700

+Subject: [PATCH] vmware only

+

+---

+ arch/x86/kernel/cpu/common.c | 2 ++

+ arch/x86/kernel/cpu/rdrand.c | 2 ++

+ arch/x86/kernel/cpu/vmware.c | 5 +++++

+ arch/x86/kernel/smpboot.c    | 2 +-

+ 4 files changed, 10 insertions(+), 1 deletion(-)

+

 diff --git a/arch/x86/kernel/cpu/common.c b/arch/x86/kernel/cpu/common.c

-index cc9e980..be408d4 100644

+index 4eece91ada37..5b8e45f48bf3 100644

 --- a/arch/x86/kernel/cpu/common.c

 +++ b/arch/x86/kernel/cpu/common.c

-@@ -1095,7 +1095,9 @@ static void identify_cpu(struct cpuinfo_x86 *c)

+@@ -1089,7 +1089,9 @@ static void identify_cpu(struct cpuinfo_x86 *c)

  #endif

  	init_hypervisor(c);

@@ -13,7 +25,7 @@ index cc9e980..be408d4 100644

  	setup_pku(c);

 diff --git a/arch/x86/kernel/cpu/rdrand.c b/arch/x86/kernel/cpu/rdrand.c

-index cfa97ff..e31cfe6 100644

+index cfa97ff67bda..e31cfe683da2 100644

 --- a/arch/x86/kernel/cpu/rdrand.c

 +++ b/arch/x86/kernel/cpu/rdrand.c

 @@ -32,6 +32,7 @@ static int __init x86_rdrand_setup(char *s)

@@ -30,7 +42,7 @@ index cfa97ff..e31cfe6 100644

  #endif

 +#endif

 diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c

-index 387ac46..b9e3cad 100644

+index 0e1fc6e17efc..59e2cc6a70d2 100644

 --- a/arch/x86/kernel/cpu/vmware.c

 +++ b/arch/x86/kernel/cpu/vmware.c

 @@ -30,6 +30,7 @@

@@ -53,10 +65,10 @@ index 387ac46..b9e3cad 100644

  /*

 diff --git a/arch/x86/kernel/smpboot.c b/arch/x86/kernel/smpboot.c

-index 42f5eb7..693ba78 100644

+index 36171bcd91f8..f7f3d0b4b882 100644

 --- a/arch/x86/kernel/smpboot.c

 +++ b/arch/x86/kernel/smpboot.c

-@@ -690,7 +690,7 @@ wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip)

+@@ -683,7 +683,7 @@ wakeup_secondary_cpu_via_nmi(int apicid, unsigned long start_eip)

  	/*

  	 * Give the other CPU some time to accept the IPI.

  	 */

@@ -65,3 +77,6 @@ index 42f5eb7..693ba78 100644

  	if (APIC_INTEGRATED(boot_cpu_apic_version)) {

  		maxlvt = lapic_get_maxlvt();

  		if (maxlvt > 3)			/* Due to the Pentium erratum 3AP.  */

+-- 

+2.11.0

+

deleted file mode 100644

@@ -1,174 +0,0 @@

-From 63b0e9edceec10fa41ec33393a1515a5ff444277 Mon Sep 17 00:00:00 2001

-From: Mike Galbraith <umgwanakikbuti@gmail.com>

-Date: Tue, 14 Jul 2015 17:39:50 +0200

-Subject: [PATCH] sched/fair: Beef up wake_wide()

-

-Josef Bacik reported that Facebook sees better performance with their

-1:N load (1 dispatch/node, N workers/node) when carrying an old patch

-to try very hard to wake to an idle CPU.  While looking at wake_wide(),

-I noticed that it doesn't pay attention to the wakeup of a many partner

-waker, returning 1 only when waking one of its many partners.

-

-Correct that, letting explicit domain flags override the heuristic.

-

-While at it, adjust task_struct bits, we don't need a 64-bit counter.

-

-Tested-by: Josef Bacik <jbacik@fb.com>

-Signed-off-by: Mike Galbraith <umgwanakikbuti@gmail.com>

-[ Tidy things up. ]

-Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>

-Cc: Linus Torvalds <torvalds@linux-foundation.org>

-Cc: Mike Galbraith <efault@gmx.de>

-Cc: Peter Zijlstra <peterz@infradead.org>

-Cc: Thomas Gleixner <tglx@linutronix.de>

-Cc: kernel-team<Kernel-team@fb.com>

-Cc: morten.rasmussen@arm.com

-Cc: riel@redhat.com

-Link: http://lkml.kernel.org/r/1436888390.7983.49.camel@gmail.com

-Signed-off-by: Ingo Molnar <mingo@kernel.org>

- include/linux/sched.h |  4 +--

- kernel/sched/fair.c   | 67 ++++++++++++++++++++++++++-------------------------

- 2 files changed, 36 insertions(+), 35 deletions(-)

-

-diff --git b/include/linux/sched.h a/include/linux/sched.h

-index 65a8a86..7412070 100644

-+++ a/include/linux/sched.h

-@@ -1387,9 +1387,9 @@ struct task_struct {

- #ifdef CONFIG_SMP

- 	struct llist_node wake_entry;

- 	int on_cpu;

--	unsigned int wakee_flips;

--	unsigned long wakee_flip_decay_ts;

- 	struct task_struct *last_wakee;

-+	unsigned long wakee_flips;

-+	unsigned long wakee_flip_decay_ts;

- 

- 	int wake_cpu;

- #endif

-diff --git b/kernel/sched/fair.c a/kernel/sched/fair.c

-index ea23f9f..8b384b8d 100644

-+++ a/kernel/sched/fair.c

-@@ -4659,29 +4659,26 @@ static long effective_load(struct task_group *tg, int cpu, long wl, long wg)

- 

- #endif

- 

--/*

-- * Detect M:N waker/wakee relationships via a switching-frequency heuristic.

-- * A waker of many should wake a different task than the one last awakened

-- * at a frequency roughly N times higher than one of its wakees.  In order

-- * to determine whether we should let the load spread vs consolodating to

-- * shared cache, we look for a minimum 'flip' frequency of llc_size in one

-- * partner, and a factor of lls_size higher frequency in the other.  With

-- * both conditions met, we can be relatively sure that the relationship is

-- * non-monogamous, with partner count exceeding socket size.  Waker/wakee

-- * being client/server, worker/dispatcher, interrupt source or whatever is

-- * irrelevant, spread criteria is apparent partner count exceeds socket size.

-- */

- static int wake_wide(struct task_struct *p)

- {

--	unsigned int master = current->wakee_flips;

--	unsigned int slave = p->wakee_flips;

- 	int factor = this_cpu_read(sd_llc_size);

- 

--	if (master < slave)

--		swap(master, slave);

--	if (slave < factor || master < slave * factor)

--		return 0;

--	return 1;

-+	/*

-+	 * Yeah, it's the switching-frequency, could means many wakee or

-+	 * rapidly switch, use factor here will just help to automatically

-+	 * adjust the loose-degree, so bigger node will lead to more pull.

-+	 */

-+	if (p->wakee_flips > factor) {

-+		/*

-+		 * wakee is somewhat hot, it needs certain amount of cpu

-+		 * resource, so if waker is far more hot, prefer to leave

-+		 * it alone.

-+		 */

-+		if (current->wakee_flips > (factor * p->wakee_flips))

-+			return 1;

-+	}

-+

-+	return 0;

- }

- 

- static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)

-@@ -4693,6 +4690,13 @@ static int wake_affine(struct sched_domain *sd, struct task_struct *p, int sync)

- 	unsigned long weight;

- 	int balanced;

- 

-+	/*

-+	 * If we wake multiple tasks be careful to not bounce

-+	 * ourselves around too much.

-+	 */

-+	if (wake_wide(p))

-+		return 0;

-+

- 	idx	  = sd->wake_idx;

- 	this_cpu  = smp_processor_id();

- 	prev_cpu  = task_cpu(p);

-@@ -4953,17 +4957,17 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f

- {

- 	struct sched_domain *tmp, *affine_sd = NULL, *sd = NULL;

- 	int cpu = smp_processor_id();

--	int new_cpu = prev_cpu;

-+	int new_cpu = cpu;

- 	int want_affine = 0;

- 	int sync = wake_flags & WF_SYNC;

- 

- 	if (sd_flag & SD_BALANCE_WAKE)

--		want_affine = !wake_wide(p) && cpumask_test_cpu(cpu, tsk_cpus_allowed(p));

-+		want_affine = cpumask_test_cpu(cpu, tsk_cpus_allowed(p));

- 

- 	rcu_read_lock();

- 	for_each_domain(cpu, tmp) {

- 		if (!(tmp->flags & SD_LOAD_BALANCE))

--			break;

-+			continue;

- 

- 		/*

- 		 * If both cpu and prev_cpu are part of this domain,

-@@ -4977,21 +4981,17 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f

- 

- 		if (tmp->flags & sd_flag)

- 			sd = tmp;

--		else if (!want_affine)

--			break;

- 	}

- 

--	if (affine_sd) {

--		sd = NULL; /* Prefer wake_affine over balance flags */

--		if (cpu != prev_cpu && wake_affine(affine_sd, p, sync))

--			new_cpu = cpu;

--	}

-+	if (affine_sd && cpu != prev_cpu && wake_affine(affine_sd, p, sync))

-+		prev_cpu = cpu;

- 

--	if (!sd) {

--		if (sd_flag & SD_BALANCE_WAKE) /* XXX always ? */

--			new_cpu = select_idle_sibling(p, new_cpu);

-+	if (sd_flag & SD_BALANCE_WAKE) {

-+		new_cpu = select_idle_sibling(p, prev_cpu);

-+		goto unlock;

-+	}

- 

--	} else while (sd) {

-+	while (sd) {

- 		struct sched_group *group;

- 		int weight;

- 

-@@ -5025,6 +5025,7 @@ select_task_rq_fair(struct task_struct *p, int prev_cpu, int sd_flag, int wake_f

- 		}

- 		/* while loop will break here if sd == NULL */

- 	}

-+unlock:

- 	rcu_read_unlock();

- 

- 	return new_cpu;

-1.9.1

-

@@ -1,7 +1,17 @@

-diff -Naur a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c

-+++ b/net/sunrpc/xprtsock.c	2016-09-01 15:06:02.939461738 -0700

-@@ -2333,6 +2333,7 @@

+From 440ba4f4203ab271cec9d02d799dd310caac7efd Mon Sep 17 00:00:00 2001

+From: Alexey Makhalov <amakhalov@vmware.com>

+Date: Tue, 9 May 2017 12:36:15 -0700

+Subject: [PATCH] SUNRPC: Do not reuse srcport for TIME_WAIT socket

+

+---

+ net/sunrpc/xprtsock.c | 1 +

+ 1 file changed, 1 insertion(+)

+

+diff --git a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c

+index e01c825bc683..eff5a482f35a 100644

+--- a/net/sunrpc/xprtsock.c

+@@ -2367,6 +2367,7 @@ static void xs_tcp_setup_socket(struct work_struct *work)

  		/* We're probably in TIME_WAIT. Get rid of existing socket,

  		 * and retry

  		 */

@@ -9,3 +19,6 @@ diff -Naur a/net/sunrpc/xprtsock.c b/net/sunrpc/xprtsock.c

  		xs_tcp_force_close(xprt);

  		break;

  	case 0:

+-- 

+2.11.0

+

@@ -1,6 +1,17 @@

-+++ b/net/ipv4/tcp.c	2016-09-01 14:52:17.239459557 -0700

-@@ -3104,7 +3104,7 @@

+From d64863f3ff0a643b0c79c9d0a2b451b20ddd2912 Mon Sep 17 00:00:00 2001

+From: Alexey Makhalov <amakhalov@vmware.com>

+Date: Tue, 9 May 2017 12:32:51 -0700

+Subject: [PATCH] Double tcp_mem limits

+

+---

+ net/ipv4/tcp.c | 2 +-

+ 1 file changed, 1 insertion(+), 1 deletion(-)

+

+diff --git a/net/ipv4/tcp.c b/net/ipv4/tcp.c

+index eb142ca71fc5..70068387dfab 100644

+--- a/net/ipv4/tcp.c

+@@ -3272,7 +3272,7 @@ __setup("thash_entries=", set_thash_entries);

  static void __init tcp_init_mem(void)

  {

@@ -9,3 +20,6 @@

  	limit = max(limit, 128UL);

  	sysctl_tcp_mem[0] = limit / 4 * 3;		/* 4.68 % */

+-- 

+2.11.0

+

deleted file mode 100644

@@ -1,53 +0,0 @@

-diff -Naur a/include/linux/sched/sysctl.h b/include/linux/sched/sysctl.h

-+++ b/include/linux/sched/sysctl.h	2016-09-01 14:52:39.335459615 -0700

-@@ -39,6 +39,7 @@

- extern unsigned int sysctl_sched_min_granularity;

- extern unsigned int sysctl_sched_wakeup_granularity;

- extern unsigned int sysctl_sched_child_runs_first;

-+extern unsigned int sysctl_sched_weighted_cpuload_uses_rla;

- 

- enum sched_tunable_scaling {

- 	SCHED_TUNABLESCALING_NONE,

-diff -Naur a/kernel/sched/fair.c b/kernel/sched/fair.c

-+++ b/kernel/sched/fair.c	2016-09-01 14:52:39.339459615 -0700

-@@ -35,6 +35,10 @@

- 

- #include "sched.h"

- 

-+#ifdef CONFIG_SMP

-+unsigned int sysctl_sched_weighted_cpuload_uses_rla = 1;

-+#endif

-+

- /*

-  * Targeted preemption latency for CPU-bound tasks:

-  * (default: 6ms * (1 + ilog(ncpus)), units: nanoseconds)

-@@ -4360,7 +4364,9 @@

- /* Used instead of source_load when we know the type == 0 */

- static unsigned long weighted_cpuload(const int cpu)

- {

--	return cfs_rq_runnable_load_avg(&cpu_rq(cpu)->cfs);

-+	if (sysctl_sched_weighted_cpuload_uses_rla)

-+		return cfs_rq_runnable_load_avg(&cpu_rq(cpu)->cfs);

-+	return cpu_rq(cpu)->load.weight;

- }

- 

- #ifdef CONFIG_NO_HZ_COMMON

-diff -Naur a/kernel/sysctl.c b/kernel/sysctl.c

-+++ b/kernel/sysctl.c	2016-09-01 14:52:39.339459615 -0700

-@@ -350,6 +350,13 @@

- 		.mode		= 0644,

- 		.proc_handler	= proc_dointvec,

- 	},

-+	{

-+		.procname	= "sched_weighted_cpuload_uses_rla",

-+		.data		= &sysctl_sched_weighted_cpuload_uses_rla,

-+		.maxlen		= sizeof(unsigned int),

-+		.mode		= 0644,

-+		.proc_handler	= proc_dointvec,

-+	},

- #ifdef CONFIG_SCHEDSTATS

- 	{

- 		.procname	= "sched_schedstats",

deleted file mode 100644

@@ -1,179 +0,0 @@

-diff --git a/arch/x86/kernel/cpu/vmware.c b/arch/x86/kernel/cpu/vmware.c

-index 628a059..a964013 100644

-+++ b/arch/x86/kernel/cpu/vmware.c

-@@ -26,16 +26,20 @@

- #include <asm/div64.h>

- #include <asm/x86_init.h>

- #include <asm/hypervisor.h>

-+#include <linux/kmsg_dump.h>

- 

--#define CPUID_VMWARE_INFO_LEAF	0x40000000

--#define VMWARE_HYPERVISOR_MAGIC	0x564D5868

--#define VMWARE_HYPERVISOR_PORT	0x5658

-+#define CPUID_VMWARE_INFO_LEAF		0x40000000

-+#define VMWARE_HYPERVISOR_MAGIC		0x564D5868

-+#define VMWARE_HYPERVISOR_PORT		0x5658

-+#define VMWARE_HYPERVISOR_HB_PORT	0x5659

- 

- #define VMWARE_PORT_CMD_GETVERSION	10

- #define VMWARE_PORT_CMD_GETHZ		45

- #define VMWARE_PORT_CMD_GETVCPU_INFO	68

- #define VMWARE_PORT_CMD_LEGACY_X2APIC	3

- #define VMWARE_PORT_CMD_VCPU_RESERVED	31

-+#define VMWARE_PORT_CMD_MESSAGE		30

-+#define VMWARE_HB_PORT_CMD_MESSAGE	0

- 

- #define VMWARE_PORT(cmd, eax, ebx, ecx, edx)				\

- 	__asm__("inl (%%dx)" :						\

-@@ -75,6 +79,13 @@ static unsigned long vmware_get_tsc_khz(void)

- 	return tsc_hz;

- }

- 

-+static void kmsg_dumper_vmware_log(struct kmsg_dumper *dumper,

-+				enum kmsg_dump_reason reason);

-+

-+static struct kmsg_dumper kmsg_dumper = {

-+	.dump = kmsg_dumper_vmware_log

-+};

-+

- static void __init vmware_platform_setup(void)

- {

- 	uint32_t eax, ebx, ecx, edx;

-@@ -86,6 +97,8 @@ static void __init vmware_platform_setup(void)

- 		x86_platform.calibrate_tsc = vmware_get_tsc_khz;

- 	else

- 		pr_warn("Failed to get TSC freq from the hypervisor\n");

-+

-+	kmsg_dump_register(&kmsg_dumper);

- }

- 

- /*

-@@ -145,3 +158,127 @@ const __refconst struct hypervisor_x86 x86_hyper_vmware = {

- 	.x2apic_available	= vmware_legacy_x2apic_available,

- };

- EXPORT_SYMBOL(x86_hyper_vmware);

-+

-+#define MESSAGE_STATUS_SUCCESS  (0x01 << 16)

-+#define MESSAGE_STATUS_CPT      (0x10 << 16)

-+#define MESSAGE_STATUS_HB       (0x80 << 16)

-+

-+#define RPCI_PROTOCOL_NUM       0x49435052 /* 'RPCI' */

-+#define GUESTMSG_FLAG_COOKIE    0x80000000

-+

-+#define MESSAGE_TYPE_OPEN	(0 << 16)