Merge tag 'for-v4.6-rc/omap-fixes-a' of git://git.kernel.org/pub/scm/linux/kernel/git/pjw/omap-pending into omap-for-v4.6/fixes

ARM: OMAP2+: first hwmod fix for v4.6-rc

Fix a longstanding bug in the hwmod code that could cause
hardware SYSCONFIG register values to not match the kernel's
idea of what they should be, and that could result in lower
performance during IP block idle entry.

Basic build, boot, and PM test logs are available here:

http://www.pwsan.com/omap/testlogs/omap-hwmod-fixes-a-for-v4.6-rc/20160326231727/

90 files changed, 5291 insertions, 139 deletions

diff --git a/tools/testing/nvdimm/test/iomap.c b/tools/testing/nvdimm/test/iomap.c
index 7ec7df9e7fc7..0c1a7e65bb81 100644
--- a/tools/testing/nvdimm/test/iomap.c
+++ b/tools/testing/nvdimm/test/iomap.c
@@ -113,7 +113,7 @@ void *__wrap_devm_memremap_pages(struct device *dev, struct resource *res,
 }
 EXPORT_SYMBOL(__wrap_devm_memremap_pages);
 
-pfn_t __wrap_phys_to_pfn_t(dma_addr_t addr, unsigned long flags)
+pfn_t __wrap_phys_to_pfn_t(phys_addr_t addr, unsigned long flags)
 {
 	struct nfit_test_resource *nfit_res = get_nfit_res(addr);
 
diff --git a/tools/testing/nvdimm/test/nfit.c b/tools/testing/nvdimm/test/nfit.c
index 90bd2ea41032..3187322eeed7 100644
--- a/tools/testing/nvdimm/test/nfit.c
+++ b/tools/testing/nvdimm/test/nfit.c
@@ -151,6 +151,11 @@ struct nfit_test {
 	int (*alloc)(struct nfit_test *t);
 	void (*setup)(struct nfit_test *t);
 	int setup_hotplug;
+	struct ars_state {
+		struct nd_cmd_ars_status *ars_status;
+		unsigned long deadline;
+		spinlock_t lock;
+	} ars_state;
 };
 
 static struct nfit_test *to_nfit_test(struct device *dev)
@@ -217,51 +222,125 @@ static int nfit_test_cmd_set_config_data(struct nd_cmd_set_config_hdr *nd_cmd,
 	return rc;
 }
 
+#define NFIT_TEST_ARS_RECORDS 4
+#define NFIT_TEST_CLEAR_ERR_UNIT 256
+
 static int nfit_test_cmd_ars_cap(struct nd_cmd_ars_cap *nd_cmd,
 		unsigned int buf_len)
 {
 	if (buf_len < sizeof(*nd_cmd))
 		return -EINVAL;
 
-	nd_cmd->max_ars_out = 256;
+	nd_cmd->max_ars_out = sizeof(struct nd_cmd_ars_status)
+		+ NFIT_TEST_ARS_RECORDS * sizeof(struct nd_ars_record);
 	nd_cmd->status = (ND_ARS_PERSISTENT | ND_ARS_VOLATILE) << 16;
+	nd_cmd->clear_err_unit = NFIT_TEST_CLEAR_ERR_UNIT;
 
 	return 0;
 }
 
-static int nfit_test_cmd_ars_start(struct nd_cmd_ars_start *nd_cmd,
-		unsigned int buf_len)
+/*
+ * Initialize the ars_state to return an ars_result 1 second in the future with
+ * a 4K error range in the middle of the requested address range.
+ */
+static void post_ars_status(struct ars_state *ars_state, u64 addr, u64 len)
 {
-	if (buf_len < sizeof(*nd_cmd))
+	struct nd_cmd_ars_status *ars_status;
+	struct nd_ars_record *ars_record;
+
+	ars_state->deadline = jiffies + 1*HZ;
+	ars_status = ars_state->ars_status;
+	ars_status->status = 0;
+	ars_status->out_length = sizeof(struct nd_cmd_ars_status)
+		+ sizeof(struct nd_ars_record);
+	ars_status->address = addr;
+	ars_status->length = len;
+	ars_status->type = ND_ARS_PERSISTENT;
+	ars_status->num_records = 1;
+	ars_record = &ars_status->records[0];
+	ars_record->handle = 0;
+	ars_record->err_address = addr + len / 2;
+	ars_record->length = SZ_4K;
+}
+
+static int nfit_test_cmd_ars_start(struct ars_state *ars_state,
+		struct nd_cmd_ars_start *ars_start, unsigned int buf_len,
+		int *cmd_rc)
+{
+	if (buf_len < sizeof(*ars_start))
 		return -EINVAL;
 
-	nd_cmd->status = 0;
+	spin_lock(&ars_state->lock);
+	if (time_before(jiffies, ars_state->deadline)) {
+		ars_start->status = NFIT_ARS_START_BUSY;
+		*cmd_rc = -EBUSY;
+	} else {
+		ars_start->status = 0;
+		ars_start->scrub_time = 1;
+		post_ars_status(ars_state, ars_start->address,
+				ars_start->length);
+		*cmd_rc = 0;
+	}
+	spin_unlock(&ars_state->lock);
 
 	return 0;
 }
 
-static int nfit_test_cmd_ars_status(struct nd_cmd_ars_status *nd_cmd,
-		unsigned int buf_len)
+static int nfit_test_cmd_ars_status(struct ars_state *ars_state,
+		struct nd_cmd_ars_status *ars_status, unsigned int buf_len,
+		int *cmd_rc)
 {
-	if (buf_len < sizeof(*nd_cmd))
+	if (buf_len < ars_state->ars_status->out_length)
 		return -EINVAL;
 
-	nd_cmd->out_length = 256;
-	nd_cmd->num_records = 0;
-	nd_cmd->address = 0;
-	nd_cmd->length = -1ULL;
-	nd_cmd->status = 0;
+	spin_lock(&ars_state->lock);
+	if (time_before(jiffies, ars_state->deadline)) {
+		memset(ars_status, 0, buf_len);
+		ars_status->status = NFIT_ARS_STATUS_BUSY;
+		ars_status->out_length = sizeof(*ars_status);
+		*cmd_rc = -EBUSY;
+	} else {
+		memcpy(ars_status, ars_state->ars_status,
+				ars_state->ars_status->out_length);
+		*cmd_rc = 0;
+	}
+	spin_unlock(&ars_state->lock);
+	return 0;
+}
+
+static int nfit_test_cmd_clear_error(struct nd_cmd_clear_error *clear_err,
+		unsigned int buf_len, int *cmd_rc)
+{
+	const u64 mask = NFIT_TEST_CLEAR_ERR_UNIT - 1;
+	if (buf_len < sizeof(*clear_err))
+		return -EINVAL;
+
+	if ((clear_err->address & mask) || (clear_err->length & mask))
+		return -EINVAL;
 
+	/*
+	 * Report 'all clear' success for all commands even though a new
+	 * scrub will find errors again.  This is enough to have the
+	 * error removed from the 'badblocks' tracking in the pmem
+	 * driver.
+	 */
+	clear_err->status = 0;
+	clear_err->cleared = clear_err->length;
+	*cmd_rc = 0;
 	return 0;
 }
 
 static int nfit_test_ctl(struct nvdimm_bus_descriptor *nd_desc,
 		struct nvdimm *nvdimm, unsigned int cmd, void *buf,
-		unsigned int buf_len)
+		unsigned int buf_len, int *cmd_rc)
 {
 	struct acpi_nfit_desc *acpi_desc = to_acpi_desc(nd_desc);
 	struct nfit_test *t = container_of(acpi_desc, typeof(*t), acpi_desc);
-	int i, rc = 0;
+	int i, rc = 0, __cmd_rc;
+
+	if (!cmd_rc)
+		cmd_rc = &__cmd_rc;
+	*cmd_rc = 0;
 
 	if (nvdimm) {
 		struct nfit_mem *nfit_mem = nvdimm_provider_data(nvdimm);
@@ -293,6 +372,8 @@ static int nfit_test_ctl(struct nvdimm_bus_descriptor *nd_desc,
 			return -ENOTTY;
 		}
 	} else {
+		struct ars_state *ars_state = &t->ars_state;
+
 		if (!nd_desc || !test_bit(cmd, &nd_desc->dsm_mask))
 			return -ENOTTY;
 
@@ -301,10 +382,15 @@ static int nfit_test_ctl(struct nvdimm_bus_descriptor *nd_desc,
 			rc = nfit_test_cmd_ars_cap(buf, buf_len);
 			break;
 		case ND_CMD_ARS_START:
-			rc = nfit_test_cmd_ars_start(buf, buf_len);
+			rc = nfit_test_cmd_ars_start(ars_state, buf, buf_len,
+					cmd_rc);
 			break;
 		case ND_CMD_ARS_STATUS:
-			rc = nfit_test_cmd_ars_status(buf, buf_len);
+			rc = nfit_test_cmd_ars_status(ars_state, buf, buf_len,
+					cmd_rc);
+			break;
+		case ND_CMD_CLEAR_ERROR:
+			rc = nfit_test_cmd_clear_error(buf, buf_len, cmd_rc);
 			break;
 		default:
 			return -ENOTTY;
@@ -420,11 +506,25 @@ static struct nfit_test_resource *nfit_test_lookup(resource_size_t addr)
 	return NULL;
 }
 
+static int ars_state_init(struct device *dev, struct ars_state *ars_state)
+{
+	ars_state->ars_status = devm_kzalloc(dev,
+			sizeof(struct nd_cmd_ars_status)
+			+ sizeof(struct nd_ars_record) * NFIT_TEST_ARS_RECORDS,
+			GFP_KERNEL);
+	if (!ars_state->ars_status)
+		return -ENOMEM;
+	spin_lock_init(&ars_state->lock);
+	return 0;
+}
+
 static int nfit_test0_alloc(struct nfit_test *t)
 {
 	size_t nfit_size = sizeof(struct acpi_nfit_system_address) * NUM_SPA
 			+ sizeof(struct acpi_nfit_memory_map) * NUM_MEM
 			+ sizeof(struct acpi_nfit_control_region) * NUM_DCR
+			+ offsetof(struct acpi_nfit_control_region,
+					window_size) * NUM_DCR
 			+ sizeof(struct acpi_nfit_data_region) * NUM_BDW
 			+ sizeof(struct acpi_nfit_flush_address) * NUM_DCR;
 	int i;
@@ -467,14 +567,14 @@ static int nfit_test0_alloc(struct nfit_test *t)
 			return -ENOMEM;
 	}
 
-	return 0;
+	return ars_state_init(&t->pdev.dev, &t->ars_state);
 }
 
 static int nfit_test1_alloc(struct nfit_test *t)
 {
 	size_t nfit_size = sizeof(struct acpi_nfit_system_address)
 		+ sizeof(struct acpi_nfit_memory_map)
-		+ sizeof(struct acpi_nfit_control_region);
+		+ offsetof(struct acpi_nfit_control_region, window_size);
 
 	t->nfit_buf = test_alloc(t, nfit_size, &t->nfit_dma);
 	if (!t->nfit_buf)
@@ -485,12 +585,11 @@ static int nfit_test1_alloc(struct nfit_test *t)
 	if (!t->spa_set[0])
 		return -ENOMEM;
 
-	return 0;
+	return ars_state_init(&t->pdev.dev, &t->ars_state);
 }
 
 static void nfit_test0_setup(struct nfit_test *t)
 {
-	struct nvdimm_bus_descriptor *nd_desc;
 	struct acpi_nfit_desc *acpi_desc;
 	struct acpi_nfit_memory_map *memdev;
 	void *nfit_buf = t->nfit_buf;
@@ -607,7 +706,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	memdev->physical_id = 0;
 	memdev->region_id = 0;
 	memdev->range_index = 0+1;
-	memdev->region_index = 0+1;
+	memdev->region_index = 4+1;
 	memdev->region_size = SPA0_SIZE/2;
 	memdev->region_offset = t->spa_set_dma[0];
 	memdev->address = 0;
@@ -622,7 +721,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	memdev->physical_id = 1;
 	memdev->region_id = 0;
 	memdev->range_index = 0+1;
-	memdev->region_index = 1+1;
+	memdev->region_index = 5+1;
 	memdev->region_size = SPA0_SIZE/2;
 	memdev->region_offset = t->spa_set_dma[0] + SPA0_SIZE/2;
 	memdev->address = 0;
@@ -637,7 +736,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	memdev->physical_id = 0;
 	memdev->region_id = 1;
 	memdev->range_index = 1+1;
-	memdev->region_index = 0+1;
+	memdev->region_index = 4+1;
 	memdev->region_size = SPA1_SIZE/4;
 	memdev->region_offset = t->spa_set_dma[1];
 	memdev->address = SPA0_SIZE/2;
@@ -652,7 +751,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	memdev->physical_id = 1;
 	memdev->region_id = 1;
 	memdev->range_index = 1+1;
-	memdev->region_index = 1+1;
+	memdev->region_index = 5+1;
 	memdev->region_size = SPA1_SIZE/4;
 	memdev->region_offset = t->spa_set_dma[1] + SPA1_SIZE/4;
 	memdev->address = SPA0_SIZE/2;
@@ -667,7 +766,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	memdev->physical_id = 2;
 	memdev->region_id = 0;
 	memdev->range_index = 1+1;
-	memdev->region_index = 2+1;
+	memdev->region_index = 6+1;
 	memdev->region_size = SPA1_SIZE/4;
 	memdev->region_offset = t->spa_set_dma[1] + 2*SPA1_SIZE/4;
 	memdev->address = SPA0_SIZE/2;
@@ -682,7 +781,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	memdev->physical_id = 3;
 	memdev->region_id = 0;
 	memdev->range_index = 1+1;
-	memdev->region_index = 3+1;
+	memdev->region_index = 7+1;
 	memdev->region_size = SPA1_SIZE/4;
 	memdev->region_offset = t->spa_set_dma[1] + 3*SPA1_SIZE/4;
 	memdev->address = SPA0_SIZE/2;
@@ -810,7 +909,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	memdev->interleave_ways = 1;
 
 	offset = offset + sizeof(struct acpi_nfit_memory_map) * 14;
-	/* dcr-descriptor0 */
+	/* dcr-descriptor0: blk */
 	dcr = nfit_buf + offset;
 	dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
 	dcr->header.length = sizeof(struct acpi_nfit_control_region);
@@ -819,6 +918,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	dcr->device_id = 0;
 	dcr->revision_id = 1;
 	dcr->serial_number = ~handle[0];
+	dcr->code = NFIT_FIC_BLK;
 	dcr->windows = 1;
 	dcr->window_size = DCR_SIZE;
 	dcr->command_offset = 0;
@@ -826,7 +926,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	dcr->status_offset = 8;
 	dcr->status_size = 4;
 
-	/* dcr-descriptor1 */
+	/* dcr-descriptor1: blk */
 	dcr = nfit_buf + offset + sizeof(struct acpi_nfit_control_region);
 	dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
 	dcr->header.length = sizeof(struct acpi_nfit_control_region);
@@ -835,6 +935,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	dcr->device_id = 0;
 	dcr->revision_id = 1;
 	dcr->serial_number = ~handle[1];
+	dcr->code = NFIT_FIC_BLK;
 	dcr->windows = 1;
 	dcr->window_size = DCR_SIZE;
 	dcr->command_offset = 0;
@@ -842,7 +943,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	dcr->status_offset = 8;
 	dcr->status_size = 4;
 
-	/* dcr-descriptor2 */
+	/* dcr-descriptor2: blk */
 	dcr = nfit_buf + offset + sizeof(struct acpi_nfit_control_region) * 2;
 	dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
 	dcr->header.length = sizeof(struct acpi_nfit_control_region);
@@ -851,6 +952,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	dcr->device_id = 0;
 	dcr->revision_id = 1;
 	dcr->serial_number = ~handle[2];
+	dcr->code = NFIT_FIC_BLK;
 	dcr->windows = 1;
 	dcr->window_size = DCR_SIZE;
 	dcr->command_offset = 0;
@@ -858,7 +960,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	dcr->status_offset = 8;
 	dcr->status_size = 4;
 
-	/* dcr-descriptor3 */
+	/* dcr-descriptor3: blk */
 	dcr = nfit_buf + offset + sizeof(struct acpi_nfit_control_region) * 3;
 	dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
 	dcr->header.length = sizeof(struct acpi_nfit_control_region);
@@ -867,6 +969,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	dcr->device_id = 0;
 	dcr->revision_id = 1;
 	dcr->serial_number = ~handle[3];
+	dcr->code = NFIT_FIC_BLK;
 	dcr->windows = 1;
 	dcr->window_size = DCR_SIZE;
 	dcr->command_offset = 0;
@@ -875,6 +978,63 @@ static void nfit_test0_setup(struct nfit_test *t)
 	dcr->status_size = 4;
 
 	offset = offset + sizeof(struct acpi_nfit_control_region) * 4;
+	/* dcr-descriptor0: pmem */
+	dcr = nfit_buf + offset;
+	dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
+	dcr->header.length = offsetof(struct acpi_nfit_control_region,
+			window_size);
+	dcr->region_index = 4+1;
+	dcr->vendor_id = 0xabcd;
+	dcr->device_id = 0;
+	dcr->revision_id = 1;
+	dcr->serial_number = ~handle[0];
+	dcr->code = NFIT_FIC_BYTEN;
+	dcr->windows = 0;
+
+	/* dcr-descriptor1: pmem */
+	dcr = nfit_buf + offset + offsetof(struct acpi_nfit_control_region,
+			window_size);
+	dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
+	dcr->header.length = offsetof(struct acpi_nfit_control_region,
+			window_size);
+	dcr->region_index = 5+1;
+	dcr->vendor_id = 0xabcd;
+	dcr->device_id = 0;
+	dcr->revision_id = 1;
+	dcr->serial_number = ~handle[1];
+	dcr->code = NFIT_FIC_BYTEN;
+	dcr->windows = 0;
+
+	/* dcr-descriptor2: pmem */
+	dcr = nfit_buf + offset + offsetof(struct acpi_nfit_control_region,
+			window_size) * 2;
+	dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
+	dcr->header.length = offsetof(struct acpi_nfit_control_region,
+			window_size);
+	dcr->region_index = 6+1;
+	dcr->vendor_id = 0xabcd;
+	dcr->device_id = 0;
+	dcr->revision_id = 1;
+	dcr->serial_number = ~handle[2];
+	dcr->code = NFIT_FIC_BYTEN;
+	dcr->windows = 0;
+
+	/* dcr-descriptor3: pmem */
+	dcr = nfit_buf + offset + offsetof(struct acpi_nfit_control_region,
+			window_size) * 3;
+	dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
+	dcr->header.length = offsetof(struct acpi_nfit_control_region,
+			window_size);
+	dcr->region_index = 7+1;
+	dcr->vendor_id = 0xabcd;
+	dcr->device_id = 0;
+	dcr->revision_id = 1;
+	dcr->serial_number = ~handle[3];
+	dcr->code = NFIT_FIC_BYTEN;
+	dcr->windows = 0;
+
+	offset = offset + offsetof(struct acpi_nfit_control_region,
+			window_size) * 4;
 	/* bdw0 (spa/dcr0, dimm0) */
 	bdw = nfit_buf + offset;
 	bdw->header.type = ACPI_NFIT_TYPE_DATA_REGION;
@@ -954,15 +1114,16 @@ static void nfit_test0_setup(struct nfit_test *t)
 
 	if (t->setup_hotplug) {
 		offset = offset + sizeof(struct acpi_nfit_flush_address) * 4;
-		/* dcr-descriptor4 */
+		/* dcr-descriptor4: blk */
 		dcr = nfit_buf + offset;
 		dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
 		dcr->header.length = sizeof(struct acpi_nfit_control_region);
-		dcr->region_index = 4+1;
+		dcr->region_index = 8+1;
 		dcr->vendor_id = 0xabcd;
 		dcr->device_id = 0;
 		dcr->revision_id = 1;
 		dcr->serial_number = ~handle[4];
+		dcr->code = NFIT_FIC_BLK;
 		dcr->windows = 1;
 		dcr->window_size = DCR_SIZE;
 		dcr->command_offset = 0;
@@ -971,11 +1132,26 @@ static void nfit_test0_setup(struct nfit_test *t)
 		dcr->status_size = 4;
 
 		offset = offset + sizeof(struct acpi_nfit_control_region);
+		/* dcr-descriptor4: pmem */
+		dcr = nfit_buf + offset;
+		dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
+		dcr->header.length = offsetof(struct acpi_nfit_control_region,
+				window_size);
+		dcr->region_index = 9+1;
+		dcr->vendor_id = 0xabcd;
+		dcr->device_id = 0;
+		dcr->revision_id = 1;
+		dcr->serial_number = ~handle[4];
+		dcr->code = NFIT_FIC_BYTEN;
+		dcr->windows = 0;
+
+		offset = offset + offsetof(struct acpi_nfit_control_region,
+				window_size);
 		/* bdw4 (spa/dcr4, dimm4) */
 		bdw = nfit_buf + offset;
 		bdw->header.type = ACPI_NFIT_TYPE_DATA_REGION;
 		bdw->header.length = sizeof(struct acpi_nfit_data_region);
-		bdw->region_index = 4+1;
+		bdw->region_index = 8+1;
 		bdw->windows = 1;
 		bdw->offset = 0;
 		bdw->size = BDW_SIZE;
@@ -1023,7 +1199,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 		memdev->physical_id = 4;
 		memdev->region_id = 0;
 		memdev->range_index = 10+1;
-		memdev->region_index = 4+1;
+		memdev->region_index = 8+1;
 		memdev->region_size = 0;
 		memdev->region_offset = 0;
 		memdev->address = 0;
@@ -1039,14 +1215,14 @@ static void nfit_test0_setup(struct nfit_test *t)
 		memdev->physical_id = 4;
 		memdev->region_id = 0;
 		memdev->range_index = 11+1;
-		memdev->region_index = 4+1;
+		memdev->region_index = 9+1;
 		memdev->region_size = SPA0_SIZE;
 		memdev->region_offset = t->spa_set_dma[2];
 		memdev->address = 0;
 		memdev->interleave_index = 0;
 		memdev->interleave_ways = 1;
 
-		/* mem-region16 (spa/dcr4, dimm4) */
+		/* mem-region16 (spa/bdw4, dimm4) */
 		memdev = nfit_buf + offset +
 				sizeof(struct acpi_nfit_memory_map) * 2;
 		memdev->header.type = ACPI_NFIT_TYPE_MEMORY_MAP;
@@ -1055,7 +1231,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 		memdev->physical_id = 4;
 		memdev->region_id = 0;
 		memdev->range_index = 12+1;
-		memdev->region_index = 4+1;
+		memdev->region_index = 8+1;
 		memdev->region_size = 0;
 		memdev->region_offset = 0;
 		memdev->address = 0;
@@ -1072,6 +1248,8 @@ static void nfit_test0_setup(struct nfit_test *t)
 		flush->hint_address[0] = t->flush_dma[4];
 	}
 
+	post_ars_status(&t->ars_state, t->spa_set_dma[0], SPA0_SIZE);
+
 	acpi_desc = &t->acpi_desc;
 	set_bit(ND_CMD_GET_CONFIG_SIZE, &acpi_desc->dimm_dsm_force_en);
 	set_bit(ND_CMD_GET_CONFIG_DATA, &acpi_desc->dimm_dsm_force_en);
@@ -1079,8 +1257,7 @@ static void nfit_test0_setup(struct nfit_test *t)
 	set_bit(ND_CMD_ARS_CAP, &acpi_desc->bus_dsm_force_en);
 	set_bit(ND_CMD_ARS_START, &acpi_desc->bus_dsm_force_en);
 	set_bit(ND_CMD_ARS_STATUS, &acpi_desc->bus_dsm_force_en);
-	nd_desc = &acpi_desc->nd_desc;
-	nd_desc->ndctl = nfit_test_ctl;
+	set_bit(ND_CMD_CLEAR_ERROR, &acpi_desc->bus_dsm_force_en);
 }
 
 static void nfit_test1_setup(struct nfit_test *t)
@@ -1090,7 +1267,6 @@ static void nfit_test1_setup(struct nfit_test *t)
 	struct acpi_nfit_memory_map *memdev;
 	struct acpi_nfit_control_region *dcr;
 	struct acpi_nfit_system_address *spa;
-	struct nvdimm_bus_descriptor *nd_desc;
 	struct acpi_nfit_desc *acpi_desc;
 
 	offset = 0;
@@ -1126,26 +1302,23 @@ static void nfit_test1_setup(struct nfit_test *t)
 	/* dcr-descriptor0 */
 	dcr = nfit_buf + offset;
 	dcr->header.type = ACPI_NFIT_TYPE_CONTROL_REGION;
-	dcr->header.length = sizeof(struct acpi_nfit_control_region);
+	dcr->header.length = offsetof(struct acpi_nfit_control_region,
+			window_size);
 	dcr->region_index = 0+1;
 	dcr->vendor_id = 0xabcd;
 	dcr->device_id = 0;
 	dcr->revision_id = 1;
 	dcr->serial_number = ~0;
-	dcr->code = 0x201;
+	dcr->code = NFIT_FIC_BYTE;
 	dcr->windows = 0;
-	dcr->window_size = 0;
-	dcr->command_offset = 0;
-	dcr->command_size = 0;
-	dcr->status_offset = 0;
-	dcr->status_size = 0;
+
+	post_ars_status(&t->ars_state, t->spa_set_dma[0], SPA2_SIZE);
 
 	acpi_desc = &t->acpi_desc;
 	set_bit(ND_CMD_ARS_CAP, &acpi_desc->bus_dsm_force_en);
 	set_bit(ND_CMD_ARS_START, &acpi_desc->bus_dsm_force_en);
 	set_bit(ND_CMD_ARS_STATUS, &acpi_desc->bus_dsm_force_en);
-	nd_desc = &acpi_desc->nd_desc;
-	nd_desc->ndctl = nfit_test_ctl;
+	set_bit(ND_CMD_CLEAR_ERROR, &acpi_desc->bus_dsm_force_en);
 }
 
 static int nfit_test_blk_do_io(struct nd_blk_region *ndbr, resource_size_t dpa,
@@ -1228,26 +1401,16 @@ static int nfit_test_probe(struct platform_device *pdev)
 
 	nfit_test->setup(nfit_test);
 	acpi_desc = &nfit_test->acpi_desc;
-	acpi_desc->dev = &pdev->dev;
+	acpi_nfit_desc_init(acpi_desc, &pdev->dev);
 	acpi_desc->nfit = nfit_test->nfit_buf;
 	acpi_desc->blk_do_io = nfit_test_blk_do_io;
 	nd_desc = &acpi_desc->nd_desc;
-	nd_desc->attr_groups = acpi_nfit_attribute_groups;
+	nd_desc->provider_name = NULL;
+	nd_desc->ndctl = nfit_test_ctl;
 	acpi_desc->nvdimm_bus = nvdimm_bus_register(&pdev->dev, nd_desc);
 	if (!acpi_desc->nvdimm_bus)
 		return -ENXIO;
 
-	INIT_LIST_HEAD(&acpi_desc->spa_maps);
-	INIT_LIST_HEAD(&acpi_desc->spas);
-	INIT_LIST_HEAD(&acpi_desc->dcrs);
-	INIT_LIST_HEAD(&acpi_desc->bdws);
-	INIT_LIST_HEAD(&acpi_desc->idts);
-	INIT_LIST_HEAD(&acpi_desc->flushes);
-	INIT_LIST_HEAD(&acpi_desc->memdevs);
-	INIT_LIST_HEAD(&acpi_desc->dimms);
-	mutex_init(&acpi_desc->spa_map_mutex);
-	mutex_init(&acpi_desc->init_mutex);
-
 	rc = acpi_nfit_init(acpi_desc, nfit_test->nfit_size);
 	if (rc) {
 		nvdimm_bus_unregister(acpi_desc->nvdimm_bus);
diff --git a/tools/testing/radix-tree/.gitignore b/tools/testing/radix-tree/.gitignore
new file mode 100644
index 000000000000..11d888ca6a92
--- /dev/null
+++ b/tools/testing/radix-tree/.gitignore
@@ -0,0 +1,2 @@
+main
+radix-tree.c
diff --git a/tools/testing/radix-tree/Makefile b/tools/testing/radix-tree/Makefile
new file mode 100644
index 000000000000..604212db9d4b
--- /dev/null
+++ b/tools/testing/radix-tree/Makefile
@@ -0,0 +1,19 @@
+
+CFLAGS += -I. -g -Wall -D_LGPL_SOURCE
+LDFLAGS += -lpthread -lurcu
+TARGETS = main
+OFILES = main.o radix-tree.o linux.o test.o tag_check.o find_next_bit.o \
+	 regression1.o regression2.o regression3.o
+
+targets: $(TARGETS)
+
+main:	$(OFILES)
+	$(CC) $(CFLAGS) $(LDFLAGS) $(OFILES) -o main
+
+clean:
+	$(RM) -f $(TARGETS) *.o radix-tree.c
+
+$(OFILES): *.h */*.h
+
+radix-tree.c: ../../../lib/radix-tree.c
+	sed -e 's/^static //' -e 's/__always_inline //' -e 's/inline //' < $< > $@
diff --git a/tools/testing/radix-tree/find_next_bit.c b/tools/testing/radix-tree/find_next_bit.c
new file mode 100644
index 000000000000..d1c2178bb2d4
--- /dev/null
+++ b/tools/testing/radix-tree/find_next_bit.c
@@ -0,0 +1,57 @@
+/* find_next_bit.c: fallback find next bit implementation
+ *
+ * Copyright (C) 2004 Red Hat, Inc. All Rights Reserved.
+ * Written by David Howells ([email protected])
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/types.h>
+#include <linux/bitops.h>
+
+#define BITOP_WORD(nr)		((nr) / BITS_PER_LONG)
+
+/*
+ * Find the next set bit in a memory region.
+ */
+unsigned long find_next_bit(const unsigned long *addr, unsigned long size,
+			    unsigned long offset)
+{
+	const unsigned long *p = addr + BITOP_WORD(offset);
+	unsigned long result = offset & ~(BITS_PER_LONG-1);
+	unsigned long tmp;
+
+	if (offset >= size)
+		return size;
+	size -= result;
+	offset %= BITS_PER_LONG;
+	if (offset) {
+		tmp = *(p++);
+		tmp &= (~0UL << offset);
+		if (size < BITS_PER_LONG)
+			goto found_first;
+		if (tmp)
+			goto found_middle;
+		size -= BITS_PER_LONG;
+		result += BITS_PER_LONG;
+	}
+	while (size & ~(BITS_PER_LONG-1)) {
+		if ((tmp = *(p++)))
+			goto found_middle;
+		result += BITS_PER_LONG;
+		size -= BITS_PER_LONG;
+	}
+	if (!size)
+		return result;
+	tmp = *p;
+
+found_first:
+	tmp &= (~0UL >> (BITS_PER_LONG - size));
+	if (tmp == 0UL)		/* Are any bits set? */
+		return result + size;	/* Nope. */
+found_middle:
+	return result + __ffs(tmp);
+}
diff --git a/tools/testing/radix-tree/linux.c b/tools/testing/radix-tree/linux.c
new file mode 100644
index 000000000000..154823737b20
--- /dev/null
+++ b/tools/testing/radix-tree/linux.c
@@ -0,0 +1,60 @@
+#include <stdlib.h>
+#include <string.h>
+#include <malloc.h>
+#include <unistd.h>
+#include <assert.h>
+
+#include <linux/mempool.h>
+#include <linux/slab.h>
+#include <urcu/uatomic.h>
+
+int nr_allocated;
+
+void *mempool_alloc(mempool_t *pool, int gfp_mask)
+{
+	return pool->alloc(gfp_mask, pool->data);
+}
+
+void mempool_free(void *element, mempool_t *pool)
+{
+	pool->free(element, pool->data);
+}
+
+mempool_t *mempool_create(int min_nr, mempool_alloc_t *alloc_fn,
+			mempool_free_t *free_fn, void *pool_data)
+{
+	mempool_t *ret = malloc(sizeof(*ret));
+
+	ret->alloc = alloc_fn;
+	ret->free = free_fn;
+	ret->data = pool_data;
+	return ret;
+}
+
+void *kmem_cache_alloc(struct kmem_cache *cachep, int flags)
+{
+	void *ret = malloc(cachep->size);
+	if (cachep->ctor)
+		cachep->ctor(ret);
+	uatomic_inc(&nr_allocated);
+	return ret;
+}
+
+void kmem_cache_free(struct kmem_cache *cachep, void *objp)
+{
+	assert(objp);
+	uatomic_dec(&nr_allocated);
+	memset(objp, 0, cachep->size);
+	free(objp);
+}
+
+struct kmem_cache *
+kmem_cache_create(const char *name, size_t size, size_t offset,
+	unsigned long flags, void (*ctor)(void *))
+{
+	struct kmem_cache *ret = malloc(sizeof(*ret));
+
+	ret->size = size;
+	ret->ctor = ctor;
+	return ret;
+}
diff --git a/tools/testing/radix-tree/linux/bitops.h b/tools/testing/radix-tree/linux/bitops.h
new file mode 100644
index 000000000000..71d58427ab60
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops.h
@@ -0,0 +1,150 @@
+#ifndef _ASM_GENERIC_BITOPS_NON_ATOMIC_H_
+#define _ASM_GENERIC_BITOPS_NON_ATOMIC_H_
+
+#include <linux/types.h>
+
+#define BITOP_MASK(nr)		(1UL << ((nr) % BITS_PER_LONG))
+#define BITOP_WORD(nr)		((nr) / BITS_PER_LONG)
+
+/**
+ * __set_bit - Set a bit in memory
+ * @nr: the bit to set
+ * @addr: the address to start counting from
+ *
+ * Unlike set_bit(), this function is non-atomic and may be reordered.
+ * If it's called on the same region of memory simultaneously, the effect
+ * may be that only one operation succeeds.
+ */
+static inline void __set_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+
+	*p  |= mask;
+}
+
+static inline void __clear_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+
+	*p &= ~mask;
+}
+
+/**
+ * __change_bit - Toggle a bit in memory
+ * @nr: the bit to change
+ * @addr: the address to start counting from
+ *
+ * Unlike change_bit(), this function is non-atomic and may be reordered.
+ * If it's called on the same region of memory simultaneously, the effect
+ * may be that only one operation succeeds.
+ */
+static inline void __change_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+
+	*p ^= mask;
+}
+
+/**
+ * __test_and_set_bit - Set a bit and return its old value
+ * @nr: Bit to set
+ * @addr: Address to count from
+ *
+ * This operation is non-atomic and can be reordered.
+ * If two examples of this operation race, one can appear to succeed
+ * but actually fail.  You must protect multiple accesses with a lock.
+ */
+static inline int __test_and_set_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+	unsigned long old = *p;
+
+	*p = old | mask;
+	return (old & mask) != 0;
+}
+
+/**
+ * __test_and_clear_bit - Clear a bit and return its old value
+ * @nr: Bit to clear
+ * @addr: Address to count from
+ *
+ * This operation is non-atomic and can be reordered.
+ * If two examples of this operation race, one can appear to succeed
+ * but actually fail.  You must protect multiple accesses with a lock.
+ */
+static inline int __test_and_clear_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+	unsigned long old = *p;
+
+	*p = old & ~mask;
+	return (old & mask) != 0;
+}
+
+/* WARNING: non atomic and it can be reordered! */
+static inline int __test_and_change_bit(int nr,
+					    volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+	unsigned long old = *p;
+
+	*p = old ^ mask;
+	return (old & mask) != 0;
+}
+
+/**
+ * test_bit - Determine whether a bit is set
+ * @nr: bit number to test
+ * @addr: Address to start counting from
+ */
+static inline int test_bit(int nr, const volatile unsigned long *addr)
+{
+	return 1UL & (addr[BITOP_WORD(nr)] >> (nr & (BITS_PER_LONG-1)));
+}
+
+/**
+ * __ffs - find first bit in word.
+ * @word: The word to search
+ *
+ * Undefined if no bit exists, so code should check against 0 first.
+ */
+static inline unsigned long __ffs(unsigned long word)
+{
+	int num = 0;
+
+	if ((word & 0xffffffff) == 0) {
+		num += 32;
+		word >>= 32;
+	}
+	if ((word & 0xffff) == 0) {
+		num += 16;
+		word >>= 16;
+	}
+	if ((word & 0xff) == 0) {
+		num += 8;
+		word >>= 8;
+	}
+	if ((word & 0xf) == 0) {
+		num += 4;
+		word >>= 4;
+	}
+	if ((word & 0x3) == 0) {
+		num += 2;
+		word >>= 2;
+	}
+	if ((word & 0x1) == 0)
+		num += 1;
+	return num;
+}
+
+unsigned long find_next_bit(const unsigned long *addr,
+			    unsigned long size,
+			    unsigned long offset);
+
+#endif /* _ASM_GENERIC_BITOPS_NON_ATOMIC_H_ */
diff --git a/tools/testing/radix-tree/linux/bitops/__ffs.h b/tools/testing/radix-tree/linux/bitops/__ffs.h
new file mode 100644
index 000000000000..9a3274aecf83
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops/__ffs.h
@@ -0,0 +1,43 @@
+#ifndef _ASM_GENERIC_BITOPS___FFS_H_
+#define _ASM_GENERIC_BITOPS___FFS_H_
+
+#include <asm/types.h>
+
+/**
+ * __ffs - find first bit in word.
+ * @word: The word to search
+ *
+ * Undefined if no bit exists, so code should check against 0 first.
+ */
+static inline unsigned long __ffs(unsigned long word)
+{
+	int num = 0;
+
+#if BITS_PER_LONG == 64
+	if ((word & 0xffffffff) == 0) {
+		num += 32;
+		word >>= 32;
+	}
+#endif
+	if ((word & 0xffff) == 0) {
+		num += 16;
+		word >>= 16;
+	}
+	if ((word & 0xff) == 0) {
+		num += 8;
+		word >>= 8;
+	}
+	if ((word & 0xf) == 0) {
+		num += 4;
+		word >>= 4;
+	}
+	if ((word & 0x3) == 0) {
+		num += 2;
+		word >>= 2;
+	}
+	if ((word & 0x1) == 0)
+		num += 1;
+	return num;
+}
+
+#endif /* _ASM_GENERIC_BITOPS___FFS_H_ */
diff --git a/tools/testing/radix-tree/linux/bitops/ffs.h b/tools/testing/radix-tree/linux/bitops/ffs.h
new file mode 100644
index 000000000000..fbbb43af7dc0
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops/ffs.h
@@ -0,0 +1,41 @@
+#ifndef _ASM_GENERIC_BITOPS_FFS_H_
+#define _ASM_GENERIC_BITOPS_FFS_H_
+
+/**
+ * ffs - find first bit set
+ * @x: the word to search
+ *
+ * This is defined the same way as
+ * the libc and compiler builtin ffs routines, therefore
+ * differs in spirit from the above ffz (man ffs).
+ */
+static inline int ffs(int x)
+{
+	int r = 1;
+
+	if (!x)
+		return 0;
+	if (!(x & 0xffff)) {
+		x >>= 16;
+		r += 16;
+	}
+	if (!(x & 0xff)) {
+		x >>= 8;
+		r += 8;
+	}
+	if (!(x & 0xf)) {
+		x >>= 4;
+		r += 4;
+	}
+	if (!(x & 3)) {
+		x >>= 2;
+		r += 2;
+	}
+	if (!(x & 1)) {
+		x >>= 1;
+		r += 1;
+	}
+	return r;
+}
+
+#endif /* _ASM_GENERIC_BITOPS_FFS_H_ */
diff --git a/tools/testing/radix-tree/linux/bitops/ffz.h b/tools/testing/radix-tree/linux/bitops/ffz.h
new file mode 100644
index 000000000000..6744bd4cdf46
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops/ffz.h
@@ -0,0 +1,12 @@
+#ifndef _ASM_GENERIC_BITOPS_FFZ_H_
+#define _ASM_GENERIC_BITOPS_FFZ_H_
+
+/*
+ * ffz - find first zero in word.
+ * @word: The word to search
+ *
+ * Undefined if no zero exists, so code should check against ~0UL first.
+ */
+#define ffz(x)  __ffs(~(x))
+
+#endif /* _ASM_GENERIC_BITOPS_FFZ_H_ */
diff --git a/tools/testing/radix-tree/linux/bitops/find.h b/tools/testing/radix-tree/linux/bitops/find.h
new file mode 100644
index 000000000000..72a51e5a12ef
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops/find.h
@@ -0,0 +1,13 @@
+#ifndef _ASM_GENERIC_BITOPS_FIND_H_
+#define _ASM_GENERIC_BITOPS_FIND_H_
+
+extern unsigned long find_next_bit(const unsigned long *addr, unsigned long
+		size, unsigned long offset);
+
+extern unsigned long find_next_zero_bit(const unsigned long *addr, unsigned
+		long size, unsigned long offset);
+
+#define find_first_bit(addr, size) find_next_bit((addr), (size), 0)
+#define find_first_zero_bit(addr, size) find_next_zero_bit((addr), (size), 0)
+
+#endif /*_ASM_GENERIC_BITOPS_FIND_H_ */
diff --git a/tools/testing/radix-tree/linux/bitops/fls.h b/tools/testing/radix-tree/linux/bitops/fls.h
new file mode 100644
index 000000000000..850859bc5069
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops/fls.h
@@ -0,0 +1,41 @@
+#ifndef _ASM_GENERIC_BITOPS_FLS_H_
+#define _ASM_GENERIC_BITOPS_FLS_H_
+
+/**
+ * fls - find last (most-significant) bit set
+ * @x: the word to search
+ *
+ * This is defined the same way as ffs.
+ * Note fls(0) = 0, fls(1) = 1, fls(0x80000000) = 32.
+ */
+
+static inline int fls(int x)
+{
+	int r = 32;
+
+	if (!x)
+		return 0;
+	if (!(x & 0xffff0000u)) {
+		x <<= 16;
+		r -= 16;
+	}
+	if (!(x & 0xff000000u)) {
+		x <<= 8;
+		r -= 8;
+	}
+	if (!(x & 0xf0000000u)) {
+		x <<= 4;
+		r -= 4;
+	}
+	if (!(x & 0xc0000000u)) {
+		x <<= 2;
+		r -= 2;
+	}
+	if (!(x & 0x80000000u)) {
+		x <<= 1;
+		r -= 1;
+	}
+	return r;
+}
+
+#endif /* _ASM_GENERIC_BITOPS_FLS_H_ */
diff --git a/tools/testing/radix-tree/linux/bitops/fls64.h b/tools/testing/radix-tree/linux/bitops/fls64.h
new file mode 100644
index 000000000000..1b6b17ce2428
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops/fls64.h
@@ -0,0 +1,14 @@
+#ifndef _ASM_GENERIC_BITOPS_FLS64_H_
+#define _ASM_GENERIC_BITOPS_FLS64_H_
+
+#include <asm/types.h>
+
+static inline int fls64(__u64 x)
+{
+	__u32 h = x >> 32;
+	if (h)
+		return fls(h) + 32;
+	return fls(x);
+}
+
+#endif /* _ASM_GENERIC_BITOPS_FLS64_H_ */
diff --git a/tools/testing/radix-tree/linux/bitops/hweight.h b/tools/testing/radix-tree/linux/bitops/hweight.h
new file mode 100644
index 000000000000..fbbc383771da
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops/hweight.h
@@ -0,0 +1,11 @@
+#ifndef _ASM_GENERIC_BITOPS_HWEIGHT_H_
+#define _ASM_GENERIC_BITOPS_HWEIGHT_H_
+
+#include <asm/types.h>
+
+extern unsigned int hweight32(unsigned int w);
+extern unsigned int hweight16(unsigned int w);
+extern unsigned int hweight8(unsigned int w);
+extern unsigned long hweight64(__u64 w);
+
+#endif /* _ASM_GENERIC_BITOPS_HWEIGHT_H_ */
diff --git a/tools/testing/radix-tree/linux/bitops/le.h b/tools/testing/radix-tree/linux/bitops/le.h
new file mode 100644
index 000000000000..b9c7e5d2d2ad
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops/le.h
@@ -0,0 +1,53 @@
+#ifndef _ASM_GENERIC_BITOPS_LE_H_
+#define _ASM_GENERIC_BITOPS_LE_H_
+
+#include <asm/types.h>
+#include <asm/byteorder.h>
+
+#define BITOP_WORD(nr)		((nr) / BITS_PER_LONG)
+#define BITOP_LE_SWIZZLE	((BITS_PER_LONG-1) & ~0x7)
+
+#if defined(__LITTLE_ENDIAN)
+
+#define generic_test_le_bit(nr, addr) test_bit(nr, addr)
+#define generic___set_le_bit(nr, addr) __set_bit(nr, addr)
+#define generic___clear_le_bit(nr, addr) __clear_bit(nr, addr)
+
+#define generic_test_and_set_le_bit(nr, addr) test_and_set_bit(nr, addr)
+#define generic_test_and_clear_le_bit(nr, addr) test_and_clear_bit(nr, addr)
+
+#define generic___test_and_set_le_bit(nr, addr) __test_and_set_bit(nr, addr)
+#define generic___test_and_clear_le_bit(nr, addr) __test_and_clear_bit(nr, addr)
+
+#define generic_find_next_zero_le_bit(addr, size, offset) find_next_zero_bit(addr, size, offset)
+
+#elif defined(__BIG_ENDIAN)
+
+#define generic_test_le_bit(nr, addr) \
+	test_bit((nr) ^ BITOP_LE_SWIZZLE, (addr))
+#define generic___set_le_bit(nr, addr) \
+	__set_bit((nr) ^ BITOP_LE_SWIZZLE, (addr))
+#define generic___clear_le_bit(nr, addr) \
+	__clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr))
+
+#define generic_test_and_set_le_bit(nr, addr) \
+	test_and_set_bit((nr) ^ BITOP_LE_SWIZZLE, (addr))
+#define generic_test_and_clear_le_bit(nr, addr) \
+	test_and_clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr))
+
+#define generic___test_and_set_le_bit(nr, addr) \
+	__test_and_set_bit((nr) ^ BITOP_LE_SWIZZLE, (addr))
+#define generic___test_and_clear_le_bit(nr, addr) \
+	__test_and_clear_bit((nr) ^ BITOP_LE_SWIZZLE, (addr))
+
+extern unsigned long generic_find_next_zero_le_bit(const unsigned long *addr,
+		unsigned long size, unsigned long offset);
+
+#else
+#error "Please fix <asm/byteorder.h>"
+#endif
+
+#define generic_find_first_zero_le_bit(addr, size) \
+        generic_find_next_zero_le_bit((addr), (size), 0)
+
+#endif /* _ASM_GENERIC_BITOPS_LE_H_ */
diff --git a/tools/testing/radix-tree/linux/bitops/non-atomic.h b/tools/testing/radix-tree/linux/bitops/non-atomic.h
new file mode 100644
index 000000000000..46a825cf2ae1
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bitops/non-atomic.h
@@ -0,0 +1,111 @@
+#ifndef _ASM_GENERIC_BITOPS_NON_ATOMIC_H_
+#define _ASM_GENERIC_BITOPS_NON_ATOMIC_H_
+
+#include <asm/types.h>
+
+#define BITOP_MASK(nr)		(1UL << ((nr) % BITS_PER_LONG))
+#define BITOP_WORD(nr)		((nr) / BITS_PER_LONG)
+
+/**
+ * __set_bit - Set a bit in memory
+ * @nr: the bit to set
+ * @addr: the address to start counting from
+ *
+ * Unlike set_bit(), this function is non-atomic and may be reordered.
+ * If it's called on the same region of memory simultaneously, the effect
+ * may be that only one operation succeeds.
+ */
+static inline void __set_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+
+	*p  |= mask;
+}
+
+static inline void __clear_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+
+	*p &= ~mask;
+}
+
+/**
+ * __change_bit - Toggle a bit in memory
+ * @nr: the bit to change
+ * @addr: the address to start counting from
+ *
+ * Unlike change_bit(), this function is non-atomic and may be reordered.
+ * If it's called on the same region of memory simultaneously, the effect
+ * may be that only one operation succeeds.
+ */
+static inline void __change_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+
+	*p ^= mask;
+}
+
+/**
+ * __test_and_set_bit - Set a bit and return its old value
+ * @nr: Bit to set
+ * @addr: Address to count from
+ *
+ * This operation is non-atomic and can be reordered.
+ * If two examples of this operation race, one can appear to succeed
+ * but actually fail.  You must protect multiple accesses with a lock.
+ */
+static inline int __test_and_set_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+	unsigned long old = *p;
+
+	*p = old | mask;
+	return (old & mask) != 0;
+}
+
+/**
+ * __test_and_clear_bit - Clear a bit and return its old value
+ * @nr: Bit to clear
+ * @addr: Address to count from
+ *
+ * This operation is non-atomic and can be reordered.
+ * If two examples of this operation race, one can appear to succeed
+ * but actually fail.  You must protect multiple accesses with a lock.
+ */
+static inline int __test_and_clear_bit(int nr, volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+	unsigned long old = *p;
+
+	*p = old & ~mask;
+	return (old & mask) != 0;
+}
+
+/* WARNING: non atomic and it can be reordered! */
+static inline int __test_and_change_bit(int nr,
+					    volatile unsigned long *addr)
+{
+	unsigned long mask = BITOP_MASK(nr);
+	unsigned long *p = ((unsigned long *)addr) + BITOP_WORD(nr);
+	unsigned long old = *p;
+
+	*p = old ^ mask;
+	return (old & mask) != 0;
+}
+
+/**
+ * test_bit - Determine whether a bit is set
+ * @nr: bit number to test
+ * @addr: Address to start counting from
+ */
+static inline int test_bit(int nr, const volatile unsigned long *addr)
+{
+	return 1UL & (addr[BITOP_WORD(nr)] >> (nr & (BITS_PER_LONG-1)));
+}
+
+#endif /* _ASM_GENERIC_BITOPS_NON_ATOMIC_H_ */
diff --git a/tools/testing/radix-tree/linux/bug.h b/tools/testing/radix-tree/linux/bug.h
new file mode 100644
index 000000000000..ccbe444977df
--- /dev/null
+++ b/tools/testing/radix-tree/linux/bug.h
@@ -0,0 +1 @@
+#define WARN_ON_ONCE(x)		assert(x)
diff --git a/tools/testing/radix-tree/linux/cpu.h b/tools/testing/radix-tree/linux/cpu.h
new file mode 100644
index 000000000000..60a40459f269
--- /dev/null
+++ b/tools/testing/radix-tree/linux/cpu.h
@@ -0,0 +1,34 @@
+
+#define hotcpu_notifier(a, b)
+
+#define CPU_ONLINE		0x0002 /* CPU (unsigned)v is up */
+#define CPU_UP_PREPARE		0x0003 /* CPU (unsigned)v coming up */
+#define CPU_UP_CANCELED		0x0004 /* CPU (unsigned)v NOT coming up */
+#define CPU_DOWN_PREPARE	0x0005 /* CPU (unsigned)v going down */
+#define CPU_DOWN_FAILED		0x0006 /* CPU (unsigned)v NOT going down */
+#define CPU_DEAD		0x0007 /* CPU (unsigned)v dead */
+#define CPU_DYING		0x0008 /* CPU (unsigned)v not running any task,
+					* not handling interrupts, soon dead.
+					* Called on the dying cpu, interrupts
+					* are already disabled. Must not
+					* sleep, must not fail */
+#define CPU_POST_DEAD		0x0009 /* CPU (unsigned)v dead, cpu_hotplug
+					* lock is dropped */
+#define CPU_STARTING		0x000A /* CPU (unsigned)v soon running.
+					* Called on the new cpu, just before
+					* enabling interrupts. Must not sleep,
+					* must not fail */
+#define CPU_DYING_IDLE		0x000B /* CPU (unsigned)v dying, reached
+					* idle loop. */
+#define CPU_BROKEN		0x000C /* CPU (unsigned)v did not die properly,
+					* perhaps due to preemption. */
+#define CPU_TASKS_FROZEN	0x0010
+
+#define CPU_ONLINE_FROZEN	(CPU_ONLINE | CPU_TASKS_FROZEN)
+#define CPU_UP_PREPARE_FROZEN	(CPU_UP_PREPARE | CPU_TASKS_FROZEN)
+#define CPU_UP_CANCELED_FROZEN  (CPU_UP_CANCELED | CPU_TASKS_FROZEN)
+#define CPU_DOWN_PREPARE_FROZEN (CPU_DOWN_PREPARE | CPU_TASKS_FROZEN)
+#define CPU_DOWN_FAILED_FROZEN  (CPU_DOWN_FAILED | CPU_TASKS_FROZEN)
+#define CPU_DEAD_FROZEN		(CPU_DEAD | CPU_TASKS_FROZEN)
+#define CPU_DYING_FROZEN	(CPU_DYING | CPU_TASKS_FROZEN)
+#define CPU_STARTING_FROZEN	(CPU_STARTING | CPU_TASKS_FROZEN)
diff --git a/tools/testing/radix-tree/linux/export.h b/tools/testing/radix-tree/linux/export.h
new file mode 100644
index 000000000000..b6afd131998d
--- /dev/null
+++ b/tools/testing/radix-tree/linux/export.h
@@ -0,0 +1,2 @@
+
+#define EXPORT_SYMBOL(sym)
diff --git a/tools/testing/radix-tree/linux/gfp.h b/tools/testing/radix-tree/linux/gfp.h
new file mode 100644
index 000000000000..0e37f7a760eb
--- /dev/null
+++ b/tools/testing/radix-tree/linux/gfp.h
@@ -0,0 +1,10 @@
+#ifndef _GFP_H
+#define _GFP_H
+
+#define __GFP_BITS_SHIFT 22
+#define __GFP_BITS_MASK ((gfp_t)((1 << __GFP_BITS_SHIFT) - 1))
+#define __GFP_WAIT 1
+#define __GFP_ACCOUNT 0
+#define __GFP_NOWARN 0
+
+#endif
diff --git a/tools/testing/radix-tree/linux/kernel.h b/tools/testing/radix-tree/linux/kernel.h
new file mode 100644
index 000000000000..ae013b0160ac
--- /dev/null
+++ b/tools/testing/radix-tree/linux/kernel.h
@@ -0,0 +1,35 @@
+#ifndef _KERNEL_H
+#define _KERNEL_H
+
+#include <assert.h>
+#include <string.h>
+#include <stdio.h>
+#include <stddef.h>
+#include <limits.h>
+
+#ifndef NULL
+#define NULL	0
+#endif
+
+#define BUG_ON(expr)	assert(!(expr))
+#define __init
+#define __must_check
+#define panic(expr)
+#define printk printf
+#define __force
+#define likely(c) (c)
+#define unlikely(c) (c)
+#define DIV_ROUND_UP(n,d) (((n) + (d) - 1) / (d))
+
+#define ARRAY_SIZE(arr) (sizeof(arr) / sizeof((arr)[0]))
+
+#define container_of(ptr, type, member) ({                      \
+	const typeof( ((type *)0)->member ) *__mptr = (ptr);    \
+	(type *)( (char *)__mptr - offsetof(type, member) );})
+#define min(a, b) ((a) < (b) ? (a) : (b))
+
+static inline int in_interrupt(void)
+{
+	return 0;
+}
+#endif /* _KERNEL_H */
diff --git a/tools/testing/radix-tree/linux/kmemleak.h b/tools/testing/radix-tree/linux/kmemleak.h
new file mode 100644
index 000000000000..155f112786c4
--- /dev/null
+++ b/tools/testing/radix-tree/linux/kmemleak.h
@@ -0,0 +1 @@
+static inline void kmemleak_update_trace(const void *ptr) { }
diff --git a/tools/testing/radix-tree/linux/mempool.h b/tools/testing/radix-tree/linux/mempool.h
new file mode 100644
index 000000000000..6a2dc55b41d6
--- /dev/null
+++ b/tools/testing/radix-tree/linux/mempool.h
@@ -0,0 +1,16 @@
+
+#include <linux/slab.h>
+
+typedef void *(mempool_alloc_t)(int gfp_mask, void *pool_data);
+typedef void (mempool_free_t)(void *element, void *pool_data);
+
+typedef struct {
+	mempool_alloc_t *alloc;
+	mempool_free_t *free;
+	void *data;
+} mempool_t;
+
+void *mempool_alloc(mempool_t *pool, int gfp_mask);
+void mempool_free(void *element, mempool_t *pool);
+mempool_t *mempool_create(int min_nr, mempool_alloc_t *alloc_fn,
+			mempool_free_t *free_fn, void *pool_data);
diff --git a/tools/testing/radix-tree/linux/notifier.h b/tools/testing/radix-tree/linux/notifier.h
new file mode 100644
index 000000000000..70e4797d5a46
--- /dev/null
+++ b/tools/testing/radix-tree/linux/notifier.h
@@ -0,0 +1,8 @@
+#ifndef _NOTIFIER_H
+#define _NOTIFIER_H
+
+struct notifier_block;
+
+#define NOTIFY_OK              0x0001          /* Suits me */
+
+#endif
diff --git a/tools/testing/radix-tree/linux/percpu.h b/tools/testing/radix-tree/linux/percpu.h
new file mode 100644
index 000000000000..5837f1d56f17
--- /dev/null
+++ b/tools/testing/radix-tree/linux/percpu.h
@@ -0,0 +1,7 @@
+
+#define DEFINE_PER_CPU(type, val) type val
+
+#define __get_cpu_var(var)	var
+#define this_cpu_ptr(var)	var
+#define per_cpu_ptr(ptr, cpu)   ({ (void)(cpu); (ptr); })
+#define per_cpu(var, cpu)	(*per_cpu_ptr(&(var), cpu))
diff --git a/tools/testing/radix-tree/linux/preempt.h b/tools/testing/radix-tree/linux/preempt.h
new file mode 100644
index 000000000000..6210672e3baa
--- /dev/null
+++ b/tools/testing/radix-tree/linux/preempt.h
@@ -0,0 +1,4 @@
+/* */
+
+#define preempt_disable() do { } while (0)
+#define preempt_enable() do { } while (0)
diff --git a/tools/testing/radix-tree/linux/radix-tree.h b/tools/testing/radix-tree/linux/radix-tree.h
new file mode 100644
index 000000000000..ce694ddd4aea
--- /dev/null
+++ b/tools/testing/radix-tree/linux/radix-tree.h
@@ -0,0 +1 @@
+#include "../../../../include/linux/radix-tree.h"
diff --git a/tools/testing/radix-tree/linux/rcupdate.h b/tools/testing/radix-tree/linux/rcupdate.h
new file mode 100644
index 000000000000..f7129ea2a899
--- /dev/null
+++ b/tools/testing/radix-tree/linux/rcupdate.h
@@ -0,0 +1,9 @@
+#ifndef _RCUPDATE_H
+#define _RCUPDATE_H
+
+#include <urcu.h>
+
+#define rcu_dereference_raw(p) rcu_dereference(p)
+#define rcu_dereference_protected(p, cond) rcu_dereference(p)
+
+#endif
diff --git a/tools/testing/radix-tree/linux/slab.h b/tools/testing/radix-tree/linux/slab.h
new file mode 100644
index 000000000000..57282506c21d
--- /dev/null
+++ b/tools/testing/radix-tree/linux/slab.h
@@ -0,0 +1,28 @@
+#ifndef SLAB_H
+#define SLAB_H
+
+#include <linux/types.h>
+
+#define GFP_KERNEL 1
+#define SLAB_HWCACHE_ALIGN 1
+#define SLAB_PANIC 2
+#define SLAB_RECLAIM_ACCOUNT    0x00020000UL            /* Objects are reclaimable */
+
+static inline int gfpflags_allow_blocking(gfp_t mask)
+{
+	return 1;
+}
+
+struct kmem_cache {
+	int size;
+	void (*ctor)(void *);
+};
+
+void *kmem_cache_alloc(struct kmem_cache *cachep, int flags);
+void kmem_cache_free(struct kmem_cache *cachep, void *objp);
+
+struct kmem_cache *
+kmem_cache_create(const char *name, size_t size, size_t offset,
+	unsigned long flags, void (*ctor)(void *));
+
+#endif		/* SLAB_H */
diff --git a/tools/testing/radix-tree/linux/types.h b/tools/testing/radix-tree/linux/types.h
new file mode 100644
index 000000000000..72a9d85f6c76
--- /dev/null
+++ b/tools/testing/radix-tree/linux/types.h
@@ -0,0 +1,28 @@
+#ifndef _TYPES_H
+#define _TYPES_H
+
+#define __rcu
+#define __read_mostly
+
+#define BITS_PER_LONG (sizeof(long) * 8)
+
+struct list_head {
+	struct list_head *next, *prev;
+};
+
+static inline void INIT_LIST_HEAD(struct list_head *list)
+{
+	list->next = list;
+	list->prev = list;
+}
+
+typedef struct {
+	unsigned int x;
+} spinlock_t;
+
+#define uninitialized_var(x) x = x
+
+typedef unsigned gfp_t;
+#include <linux/gfp.h>
+
+#endif
diff --git a/tools/testing/radix-tree/main.c b/tools/testing/radix-tree/main.c
new file mode 100644
index 000000000000..0e83cad27a9f
--- /dev/null
+++ b/tools/testing/radix-tree/main.c
@@ -0,0 +1,272 @@
+#include <stdio.h>
+#include <stdlib.h>
+#include <unistd.h>
+#include <time.h>
+#include <assert.h>
+
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+
+#include "test.h"
+#include "regression.h"
+
+void __gang_check(unsigned long middle, long down, long up, int chunk, int hop)
+{
+	long idx;
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	middle = 1 << 30;
+
+	for (idx = -down; idx < up; idx++)
+		item_insert(&tree, middle + idx);
+
+	item_check_absent(&tree, middle - down - 1);
+	for (idx = -down; idx < up; idx++)
+		item_check_present(&tree, middle + idx);
+	item_check_absent(&tree, middle + up);
+
+	item_gang_check_present(&tree, middle - down,
+			up + down, chunk, hop);
+	item_full_scan(&tree, middle - down, down + up, chunk);
+	item_kill_tree(&tree);
+}
+
+void gang_check(void)
+{
+	__gang_check(1 << 30, 128, 128, 35, 2);
+	__gang_check(1 << 31, 128, 128, 32, 32);
+	__gang_check(1 << 31, 128, 128, 32, 100);
+	__gang_check(1 << 31, 128, 128, 17, 7);
+	__gang_check(0xffff0000, 0, 65536, 17, 7);
+	__gang_check(0xfffffffe, 1, 1, 17, 7);
+}
+
+void __big_gang_check(void)
+{
+	unsigned long start;
+	int wrapped = 0;
+
+	start = 0;
+	do {
+		unsigned long old_start;
+
+//		printf("0x%08lx\n", start);
+		__gang_check(start, rand() % 113 + 1, rand() % 71,
+				rand() % 157, rand() % 91 + 1);
+		old_start = start;
+		start += rand() % 1000000;
+		start %= 1ULL << 33;
+		if (start < old_start)
+			wrapped = 1;
+	} while (!wrapped);
+}
+
+void big_gang_check(void)
+{
+	int i;
+
+	for (i = 0; i < 1000; i++) {
+		__big_gang_check();
+		srand(time(0));
+		printf("%d ", i);
+		fflush(stdout);
+	}
+}
+
+void add_and_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert(&tree, 44);
+	item_check_present(&tree, 44);
+	item_check_absent(&tree, 43);
+	item_kill_tree(&tree);
+}
+
+void dynamic_height_check(void)
+{
+	int i;
+	RADIX_TREE(tree, GFP_KERNEL);
+	tree_verify_min_height(&tree, 0);
+
+	item_insert(&tree, 42);
+	tree_verify_min_height(&tree, 42);
+
+	item_insert(&tree, 1000000);
+	tree_verify_min_height(&tree, 1000000);
+
+	assert(item_delete(&tree, 1000000));
+	tree_verify_min_height(&tree, 42);
+
+	assert(item_delete(&tree, 42));
+	tree_verify_min_height(&tree, 0);
+
+	for (i = 0; i < 1000; i++) {
+		item_insert(&tree, i);
+		tree_verify_min_height(&tree, i);
+	}
+
+	i--;
+	for (;;) {
+		assert(item_delete(&tree, i));
+		if (i == 0) {
+			tree_verify_min_height(&tree, 0);
+			break;
+		}
+		i--;
+		tree_verify_min_height(&tree, i);
+	}
+
+	item_kill_tree(&tree);
+}
+
+void check_copied_tags(struct radix_tree_root *tree, unsigned long start, unsigned long end, unsigned long *idx, int count, int fromtag, int totag)
+{
+	int i;
+
+	for (i = 0; i < count; i++) {
+/*		if (i % 1000 == 0)
+			putchar('.'); */
+		if (idx[i] < start || idx[i] > end) {
+			if (item_tag_get(tree, idx[i], totag)) {
+				printf("%lu-%lu: %lu, tags %d-%d\n", start, end, idx[i], item_tag_get(tree, idx[i], fromtag), item_tag_get(tree, idx[i], totag));
+			}
+			assert(!item_tag_get(tree, idx[i], totag));
+			continue;
+		}
+		if (item_tag_get(tree, idx[i], fromtag) ^
+			item_tag_get(tree, idx[i], totag)) {
+			printf("%lu-%lu: %lu, tags %d-%d\n", start, end, idx[i], item_tag_get(tree, idx[i], fromtag), item_tag_get(tree, idx[i], totag));
+		}
+		assert(!(item_tag_get(tree, idx[i], fromtag) ^
+			 item_tag_get(tree, idx[i], totag)));
+	}
+}
+
+#define ITEMS 50000
+
+void copy_tag_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	unsigned long idx[ITEMS];
+	unsigned long start, end, count = 0, tagged, cur, tmp;
+	int i;
+
+//	printf("generating radix tree indices...\n");
+	start = rand();
+	end = rand();
+	if (start > end && (rand() % 10)) {
+		cur = start;
+		start = end;
+		end = cur;
+	}
+	/* Specifically create items around the start and the end of the range
+	 * with high probability to check for off by one errors */
+	cur = rand();
+	if (cur & 1) {
+		item_insert(&tree, start);
+		if (cur & 2) {
+			if (start <= end)
+				count++;
+			item_tag_set(&tree, start, 0);
+		}
+	}
+	if (cur & 4) {
+		item_insert(&tree, start-1);
+		if (cur & 8)
+			item_tag_set(&tree, start-1, 0);
+	}
+	if (cur & 16) {
+		item_insert(&tree, end);
+		if (cur & 32) {
+			if (start <= end)
+				count++;
+			item_tag_set(&tree, end, 0);
+		}
+	}
+	if (cur & 64) {
+		item_insert(&tree, end+1);
+		if (cur & 128)
+			item_tag_set(&tree, end+1, 0);
+	}
+
+	for (i = 0; i < ITEMS; i++) {
+		do {
+			idx[i] = rand();
+		} while (item_lookup(&tree, idx[i]));
+
+		item_insert(&tree, idx[i]);
+		if (rand() & 1) {
+			item_tag_set(&tree, idx[i], 0);
+			if (idx[i] >= start && idx[i] <= end)
+				count++;
+		}
+/*		if (i % 1000 == 0)
+			putchar('.'); */
+	}
+
+//	printf("\ncopying tags...\n");
+	cur = start;
+	tagged = radix_tree_range_tag_if_tagged(&tree, &cur, end, ITEMS, 0, 1);
+
+//	printf("checking copied tags\n");
+	assert(tagged == count);
+	check_copied_tags(&tree, start, end, idx, ITEMS, 0, 1);
+
+	/* Copy tags in several rounds */
+//	printf("\ncopying tags...\n");
+	cur = start;
+	do {
+		tmp = rand() % (count/10+2);
+		tagged = radix_tree_range_tag_if_tagged(&tree, &cur, end, tmp, 0, 2);
+	} while (tmp == tagged);
+
+//	printf("%lu %lu %lu\n", tagged, tmp, count);
+//	printf("checking copied tags\n");
+	check_copied_tags(&tree, start, end, idx, ITEMS, 0, 2);
+	assert(tagged < tmp);
+	verify_tag_consistency(&tree, 0);
+	verify_tag_consistency(&tree, 1);
+	verify_tag_consistency(&tree, 2);
+//	printf("\n");
+	item_kill_tree(&tree);
+}
+
+static void single_thread_tests(void)
+{
+	int i;
+
+	tag_check();
+	printf("after tag_check: %d allocated\n", nr_allocated);
+	gang_check();
+	printf("after gang_check: %d allocated\n", nr_allocated);
+	add_and_check();
+	printf("after add_and_check: %d allocated\n", nr_allocated);
+	dynamic_height_check();
+	printf("after dynamic_height_check: %d allocated\n", nr_allocated);
+	big_gang_check();
+	printf("after big_gang_check: %d allocated\n", nr_allocated);
+	for (i = 0; i < 2000; i++) {
+		copy_tag_check();
+		printf("%d ", i);
+		fflush(stdout);
+	}
+	printf("after copy_tag_check: %d allocated\n", nr_allocated);
+}
+
+int main(void)
+{
+	rcu_register_thread();
+	radix_tree_init();
+
+	regression1_test();
+	regression2_test();
+	regression3_test();
+	single_thread_tests();
+
+	sleep(1);
+	printf("after sleep(1): %d allocated\n", nr_allocated);
+	rcu_unregister_thread();
+
+	exit(0);
+}
diff --git a/tools/testing/radix-tree/rcupdate.c b/tools/testing/radix-tree/rcupdate.c
new file mode 100644
index 000000000000..31a2d14225d6
--- /dev/null
+++ b/tools/testing/radix-tree/rcupdate.c
@@ -0,0 +1,86 @@
+#include <linux/rcupdate.h>
+#include <pthread.h>
+#include <stdio.h>
+#include <assert.h>
+
+static pthread_mutex_t rculock = PTHREAD_MUTEX_INITIALIZER;
+static struct rcu_head *rcuhead_global = NULL;
+static __thread int nr_rcuhead = 0;
+static __thread struct rcu_head *rcuhead = NULL;
+static __thread struct rcu_head *rcutail = NULL;
+
+static pthread_cond_t rcu_worker_cond = PTHREAD_COND_INITIALIZER;
+
+/* switch to urcu implementation when it is merged. */
+void call_rcu(struct rcu_head *head, void (*func)(struct rcu_head *head))
+{
+	head->func = func;
+	head->next = rcuhead;
+	rcuhead = head;
+	if (!rcutail)
+		rcutail = head;
+	nr_rcuhead++;
+	if (nr_rcuhead >= 1000) {
+		int signal = 0;
+
+		pthread_mutex_lock(&rculock);
+		if (!rcuhead_global)
+			signal = 1;
+		rcutail->next = rcuhead_global;
+		rcuhead_global = head;
+		pthread_mutex_unlock(&rculock);
+
+		nr_rcuhead = 0;
+		rcuhead = NULL;
+		rcutail = NULL;
+
+		if (signal) {
+			pthread_cond_signal(&rcu_worker_cond);
+		}
+	}
+}
+
+static void *rcu_worker(void *arg)
+{
+	struct rcu_head *r;
+
+	rcupdate_thread_init();
+
+	while (1) {
+		pthread_mutex_lock(&rculock);
+		while (!rcuhead_global) {
+			pthread_cond_wait(&rcu_worker_cond, &rculock);
+		}
+		r = rcuhead_global;
+		rcuhead_global = NULL;
+
+		pthread_mutex_unlock(&rculock);
+
+		synchronize_rcu();
+
+		while (r) {
+			struct rcu_head *tmp = r->next;
+			r->func(r);
+			r = tmp;
+		}
+	}
+
+	rcupdate_thread_exit();
+
+	return NULL;
+}
+
+static pthread_t worker_thread;
+void rcupdate_init(void)
+{
+	pthread_create(&worker_thread, NULL, rcu_worker, NULL);
+}
+
+void rcupdate_thread_init(void)
+{
+	rcu_register_thread();
+}
+void rcupdate_thread_exit(void)
+{
+	rcu_unregister_thread();
+}
diff --git a/tools/testing/radix-tree/regression.h b/tools/testing/radix-tree/regression.h
new file mode 100644
index 000000000000..e018c4816688
--- /dev/null
+++ b/tools/testing/radix-tree/regression.h
@@ -0,0 +1,8 @@
+#ifndef __REGRESSION_H__
+#define __REGRESSION_H__
+
+void regression1_test(void);
+void regression2_test(void);
+void regression3_test(void);
+
+#endif
diff --git a/tools/testing/radix-tree/regression1.c b/tools/testing/radix-tree/regression1.c
new file mode 100644
index 000000000000..2d03a63bb79c
--- /dev/null
+++ b/tools/testing/radix-tree/regression1.c
@@ -0,0 +1,220 @@
+/*
+ * Regression1
+ * Description:
+ * Salman Qazi describes the following radix-tree bug:
+ *
+ * In the following case, we get can get a deadlock:
+ *
+ * 0.  The radix tree contains two items, one has the index 0.
+ * 1.  The reader (in this case find_get_pages) takes the rcu_read_lock.
+ * 2.  The reader acquires slot(s) for item(s) including the index 0 item.
+ * 3.  The non-zero index item is deleted, and as a consequence the other item
+ *     is moved to the root of the tree. The place where it used to be is queued
+ *     for deletion after the readers finish.
+ * 3b. The zero item is deleted, removing it from the direct slot, it remains in
+ *     the rcu-delayed indirect node.
+ * 4.  The reader looks at the index 0 slot, and finds that the page has 0 ref
+ *     count
+ * 5.  The reader looks at it again, hoping that the item will either be freed
+ *     or the ref count will increase. This never happens, as the slot it is
+ *     looking at will never be updated. Also, this slot can never be reclaimed
+ *     because the reader is holding rcu_read_lock and is in an infinite loop.
+ *
+ * The fix is to re-use the same "indirect" pointer case that requires a slot
+ * lookup retry into a general "retry the lookup" bit.
+ *
+ * Running:
+ * This test should run to completion in a few seconds. The above bug would
+ * cause it to hang indefinitely.
+ *
+ * Upstream commit:
+ * Not yet
+ */
+#include <linux/kernel.h>
+#include <linux/gfp.h>
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+#include <linux/rcupdate.h>
+#include <stdlib.h>
+#include <pthread.h>
+#include <stdio.h>
+#include <assert.h>
+
+#include "regression.h"
+
+static RADIX_TREE(mt_tree, GFP_KERNEL);
+static pthread_mutex_t mt_lock;
+
+struct page {
+	pthread_mutex_t lock;
+	struct rcu_head rcu;
+	int count;
+	unsigned long index;
+};
+
+static struct page *page_alloc(void)
+{
+	struct page *p;
+	p = malloc(sizeof(struct page));
+	p->count = 1;
+	p->index = 1;
+	pthread_mutex_init(&p->lock, NULL);
+
+	return p;
+}
+
+static void page_rcu_free(struct rcu_head *rcu)
+{
+	struct page *p = container_of(rcu, struct page, rcu);
+	assert(!p->count);
+	pthread_mutex_destroy(&p->lock);
+	free(p);
+}
+
+static void page_free(struct page *p)
+{
+	call_rcu(&p->rcu, page_rcu_free);
+}
+
+static unsigned find_get_pages(unsigned long start,
+			    unsigned int nr_pages, struct page **pages)
+{
+	unsigned int i;
+	unsigned int ret;
+	unsigned int nr_found;
+
+	rcu_read_lock();
+restart:
+	nr_found = radix_tree_gang_lookup_slot(&mt_tree,
+				(void ***)pages, NULL, start, nr_pages);
+	ret = 0;
+	for (i = 0; i < nr_found; i++) {
+		struct page *page;
+repeat:
+		page = radix_tree_deref_slot((void **)pages[i]);
+		if (unlikely(!page))
+			continue;
+
+		if (radix_tree_exception(page)) {
+			if (radix_tree_deref_retry(page)) {
+				/*
+				 * Transient condition which can only trigger
+				 * when entry at index 0 moves out of or back
+				 * to root: none yet gotten, safe to restart.
+				 */
+				assert((start | i) == 0);
+				goto restart;
+			}
+			/*
+			 * No exceptional entries are inserted in this test.
+			 */
+			assert(0);
+		}
+
+		pthread_mutex_lock(&page->lock);
+		if (!page->count) {
+			pthread_mutex_unlock(&page->lock);
+			goto repeat;
+		}
+		/* don't actually update page refcount */
+		pthread_mutex_unlock(&page->lock);
+
+		/* Has the page moved? */
+		if (unlikely(page != *((void **)pages[i]))) {
+			goto repeat;
+		}
+
+		pages[ret] = page;
+		ret++;
+	}
+	rcu_read_unlock();
+	return ret;
+}
+
+static pthread_barrier_t worker_barrier;
+
+static void *regression1_fn(void *arg)
+{
+	rcu_register_thread();
+
+	if (pthread_barrier_wait(&worker_barrier) ==
+			PTHREAD_BARRIER_SERIAL_THREAD) {
+		int j;
+
+		for (j = 0; j < 1000000; j++) {
+			struct page *p;
+
+			p = page_alloc();
+			pthread_mutex_lock(&mt_lock);
+			radix_tree_insert(&mt_tree, 0, p);
+			pthread_mutex_unlock(&mt_lock);
+
+			p = page_alloc();
+			pthread_mutex_lock(&mt_lock);
+			radix_tree_insert(&mt_tree, 1, p);
+			pthread_mutex_unlock(&mt_lock);
+
+			pthread_mutex_lock(&mt_lock);
+			p = radix_tree_delete(&mt_tree, 1);
+			pthread_mutex_lock(&p->lock);
+			p->count--;
+			pthread_mutex_unlock(&p->lock);
+			pthread_mutex_unlock(&mt_lock);
+			page_free(p);
+
+			pthread_mutex_lock(&mt_lock);
+			p = radix_tree_delete(&mt_tree, 0);
+			pthread_mutex_lock(&p->lock);
+			p->count--;
+			pthread_mutex_unlock(&p->lock);
+			pthread_mutex_unlock(&mt_lock);
+			page_free(p);
+		}
+	} else {
+		int j;
+
+		for (j = 0; j < 100000000; j++) {
+			struct page *pages[10];
+
+			find_get_pages(0, 10, pages);
+		}
+	}
+
+	rcu_unregister_thread();
+
+	return NULL;
+}
+
+static pthread_t *threads;
+void regression1_test(void)
+{
+	int nr_threads;
+	int i;
+	long arg;
+
+	/* Regression #1 */
+	printf("running regression test 1, should finish in under a minute\n");
+	nr_threads = 2;
+	pthread_barrier_init(&worker_barrier, NULL, nr_threads);
+
+	threads = malloc(nr_threads * sizeof(pthread_t *));
+
+	for (i = 0; i < nr_threads; i++) {
+		arg = i;
+		if (pthread_create(&threads[i], NULL, regression1_fn, (void *)arg)) {
+			perror("pthread_create");
+			exit(1);
+		}
+	}
+
+	for (i = 0; i < nr_threads; i++) {
+		if (pthread_join(threads[i], NULL)) {
+			perror("pthread_join");
+			exit(1);
+		}
+	}
+
+	free(threads);
+
+	printf("regression test 1, done\n");
+}
diff --git a/tools/testing/radix-tree/regression2.c b/tools/testing/radix-tree/regression2.c
new file mode 100644
index 000000000000..5d2fa28cdca3
--- /dev/null
+++ b/tools/testing/radix-tree/regression2.c
@@ -0,0 +1,126 @@
+/*
+ * Regression2
+ * Description:
+ * Toshiyuki Okajima describes the following radix-tree bug:
+ *
+ * In the following case, we can get a hangup on
+ *   radix_radix_tree_gang_lookup_tag_slot.
+ *
+ * 0.  The radix tree contains RADIX_TREE_MAP_SIZE items. And the tag of
+ *     a certain item has PAGECACHE_TAG_DIRTY.
+ * 1.  radix_tree_range_tag_if_tagged(, start, end, , PAGECACHE_TAG_DIRTY,
+ *     PAGECACHE_TAG_TOWRITE) is called to add PAGECACHE_TAG_TOWRITE tag
+ *     for the tag which has PAGECACHE_TAG_DIRTY. However, there is no tag with
+ *     PAGECACHE_TAG_DIRTY within the range from start to end. As the result,
+ *     There is no tag with PAGECACHE_TAG_TOWRITE but the root tag has
+ *     PAGECACHE_TAG_TOWRITE.
+ * 2.  An item is added into the radix tree and then the level of it is
+ *     extended into 2 from 1. At that time, the new radix tree node succeeds
+ *     the tag status of the root tag. Therefore the tag of the new radix tree
+ *     node has PAGECACHE_TAG_TOWRITE but there is not slot with
+ *     PAGECACHE_TAG_TOWRITE tag in the child node of the new radix tree node.
+ * 3.  The tag of a certain item is cleared with PAGECACHE_TAG_DIRTY.
+ * 4.  All items within the index range from 0 to RADIX_TREE_MAP_SIZE - 1 are
+ *     released. (Only the item which index is RADIX_TREE_MAP_SIZE exist in the
+ *     radix tree.) As the result, the slot of the radix tree node is NULL but
+ *     the tag which corresponds to the slot has PAGECACHE_TAG_TOWRITE.
+ * 5.  radix_tree_gang_lookup_tag_slot(PAGECACHE_TAG_TOWRITE) calls
+ *     __lookup_tag. __lookup_tag returns with 0. And __lookup_tag doesn't
+ *     change the index that is the input and output parameter. Because the 1st
+ *     slot of the radix tree node is NULL, but the tag which corresponds to
+ *     the slot has PAGECACHE_TAG_TOWRITE.
+ *     Therefore radix_tree_gang_lookup_tag_slot tries to get some items by
+ *     calling __lookup_tag, but it cannot get any items forever.
+ *
+ * The fix is to change that radix_tree_tag_if_tagged doesn't tag the root tag
+ * if it doesn't set any tags within the specified range.
+ *
+ * Running:
+ * This test should run to completion immediately. The above bug would cause it
+ * to hang indefinitely.
+ *
+ * Upstream commit:
+ * Not yet
+ */
+#include <linux/kernel.h>
+#include <linux/gfp.h>
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+#include <stdlib.h>
+#include <stdio.h>
+
+#include "regression.h"
+
+#ifdef __KERNEL__
+#define RADIX_TREE_MAP_SHIFT    (CONFIG_BASE_SMALL ? 4 : 6)
+#else
+#define RADIX_TREE_MAP_SHIFT    3       /* For more stressful testing */
+#endif
+
+#define RADIX_TREE_MAP_SIZE     (1UL << RADIX_TREE_MAP_SHIFT)
+#define PAGECACHE_TAG_DIRTY     0
+#define PAGECACHE_TAG_WRITEBACK 1
+#define PAGECACHE_TAG_TOWRITE   2
+
+static RADIX_TREE(mt_tree, GFP_KERNEL);
+unsigned long page_count = 0;
+
+struct page {
+	unsigned long index;
+};
+
+static struct page *page_alloc(void)
+{
+	struct page *p;
+	p = malloc(sizeof(struct page));
+	p->index = page_count++;
+
+	return p;
+}
+
+void regression2_test(void)
+{
+	int i;
+	struct page *p;
+	int max_slots = RADIX_TREE_MAP_SIZE;
+	unsigned long int start, end;
+	struct page *pages[1];
+
+	printf("running regression test 2 (should take milliseconds)\n");
+	/* 0. */
+	for (i = 0; i <= max_slots - 1; i++) {
+		p = page_alloc();
+		radix_tree_insert(&mt_tree, i, p);
+	}
+	radix_tree_tag_set(&mt_tree, max_slots - 1, PAGECACHE_TAG_DIRTY);
+
+	/* 1. */
+	start = 0;
+	end = max_slots - 2;
+	radix_tree_range_tag_if_tagged(&mt_tree, &start, end, 1,
+				PAGECACHE_TAG_DIRTY, PAGECACHE_TAG_TOWRITE);
+
+	/* 2. */
+	p = page_alloc();
+	radix_tree_insert(&mt_tree, max_slots, p);
+
+	/* 3. */
+	radix_tree_tag_clear(&mt_tree, max_slots - 1, PAGECACHE_TAG_DIRTY);
+
+	/* 4. */
+	for (i = max_slots - 1; i >= 0; i--)
+		radix_tree_delete(&mt_tree, i);
+
+	/* 5. */
+	// NOTE: start should not be 0 because radix_tree_gang_lookup_tag_slot
+	//       can return.
+	start = 1;
+	end = max_slots - 2;
+	radix_tree_gang_lookup_tag_slot(&mt_tree, (void ***)pages, start, end,
+		PAGECACHE_TAG_TOWRITE);
+
+	/* We remove all the remained nodes */
+	radix_tree_delete(&mt_tree, max_slots);
+
+	printf("regression test 2, done\n");
+}
diff --git a/tools/testing/radix-tree/regression3.c b/tools/testing/radix-tree/regression3.c
new file mode 100644
index 000000000000..1f06ed73d0a8
--- /dev/null
+++ b/tools/testing/radix-tree/regression3.c
@@ -0,0 +1,117 @@
+/*
+ * Regression3
+ * Description:
+ * Helper radix_tree_iter_retry resets next_index to the current index.
+ * In following radix_tree_next_slot current chunk size becomes zero.
+ * This isn't checked and it tries to dereference null pointer in slot.
+ *
+ * Helper radix_tree_iter_next reset slot to NULL and next_index to index + 1,
+ * for tagger iteraction it also must reset cached tags in iterator to abort
+ * next radix_tree_next_slot and go to slow-path into radix_tree_next_chunk.
+ *
+ * Running:
+ * This test should run to completion immediately. The above bug would
+ * cause it to segfault.
+ *
+ * Upstream commit:
+ * Not yet
+ */
+#include <linux/kernel.h>
+#include <linux/gfp.h>
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+#include <stdlib.h>
+#include <stdio.h>
+
+#include "regression.h"
+
+void regression3_test(void)
+{
+	RADIX_TREE(root, GFP_KERNEL);
+	void *ptr0 = (void *)4ul;
+	void *ptr = (void *)8ul;
+	struct radix_tree_iter iter;
+	void **slot;
+	bool first;
+
+	printf("running regression test 3 (should take milliseconds)\n");
+
+	radix_tree_insert(&root, 0, ptr0);
+	radix_tree_tag_set(&root, 0, 0);
+
+	first = true;
+	radix_tree_for_each_tagged(slot, &root, &iter, 0, 0) {
+		printf("tagged %ld %p\n", iter.index, *slot);
+		if (first) {
+			radix_tree_insert(&root, 1, ptr);
+			radix_tree_tag_set(&root, 1, 0);
+			first = false;
+		}
+		if (radix_tree_deref_retry(*slot)) {
+			printf("retry at %ld\n", iter.index);
+			slot = radix_tree_iter_retry(&iter);
+			continue;
+		}
+	}
+	radix_tree_delete(&root, 1);
+
+	first = true;
+	radix_tree_for_each_slot(slot, &root, &iter, 0) {
+		printf("slot %ld %p\n", iter.index, *slot);
+		if (first) {
+			radix_tree_insert(&root, 1, ptr);
+			first = false;
+		}
+		if (radix_tree_deref_retry(*slot)) {
+			printk("retry at %ld\n", iter.index);
+			slot = radix_tree_iter_retry(&iter);
+			continue;
+		}
+	}
+	radix_tree_delete(&root, 1);
+
+	first = true;
+	radix_tree_for_each_contig(slot, &root, &iter, 0) {
+		printk("contig %ld %p\n", iter.index, *slot);
+		if (first) {
+			radix_tree_insert(&root, 1, ptr);
+			first = false;
+		}
+		if (radix_tree_deref_retry(*slot)) {
+			printk("retry at %ld\n", iter.index);
+			slot = radix_tree_iter_retry(&iter);
+			continue;
+		}
+	}
+
+	radix_tree_for_each_slot(slot, &root, &iter, 0) {
+		printf("slot %ld %p\n", iter.index, *slot);
+		if (!iter.index) {
+			printf("next at %ld\n", iter.index);
+			slot = radix_tree_iter_next(&iter);
+		}
+	}
+
+	radix_tree_for_each_contig(slot, &root, &iter, 0) {
+		printf("contig %ld %p\n", iter.index, *slot);
+		if (!iter.index) {
+			printf("next at %ld\n", iter.index);
+			slot = radix_tree_iter_next(&iter);
+		}
+	}
+
+	radix_tree_tag_set(&root, 0, 0);
+	radix_tree_tag_set(&root, 1, 0);
+	radix_tree_for_each_tagged(slot, &root, &iter, 0, 0) {
+		printf("tagged %ld %p\n", iter.index, *slot);
+		if (!iter.index) {
+			printf("next at %ld\n", iter.index);
+			slot = radix_tree_iter_next(&iter);
+		}
+	}
+
+	radix_tree_delete(&root, 0);
+	radix_tree_delete(&root, 1);
+
+	printf("regression test 3 passed\n");
+}
diff --git a/tools/testing/radix-tree/tag_check.c b/tools/testing/radix-tree/tag_check.c
new file mode 100644
index 000000000000..83136be552a0
--- /dev/null
+++ b/tools/testing/radix-tree/tag_check.c
@@ -0,0 +1,332 @@
+#include <stdlib.h>
+#include <assert.h>
+#include <stdio.h>
+#include <string.h>
+
+#include <linux/slab.h>
+#include <linux/radix-tree.h>
+
+#include "test.h"
+
+
+static void
+__simple_checks(struct radix_tree_root *tree, unsigned long index, int tag)
+{
+	int ret;
+
+	item_check_absent(tree, index);
+	assert(item_tag_get(tree, index, tag) == 0);
+
+	item_insert(tree, index);
+	assert(item_tag_get(tree, index, tag) == 0);
+	item_tag_set(tree, index, tag);
+	ret = item_tag_get(tree, index, tag);
+	assert(ret != 0);
+	ret = item_delete(tree, index);
+	assert(ret != 0);
+	item_insert(tree, index);
+	ret = item_tag_get(tree, index, tag);
+	assert(ret == 0);
+	ret = item_delete(tree, index);
+	assert(ret != 0);
+	ret = item_delete(tree, index);
+	assert(ret == 0);
+}
+
+void simple_checks(void)
+{
+	unsigned long index;
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	for (index = 0; index < 10000; index++) {
+		__simple_checks(&tree, index, 0);
+		__simple_checks(&tree, index, 1);
+	}
+	verify_tag_consistency(&tree, 0);
+	verify_tag_consistency(&tree, 1);
+	printf("before item_kill_tree: %d allocated\n", nr_allocated);
+	item_kill_tree(&tree);
+	printf("after item_kill_tree: %d allocated\n", nr_allocated);
+}
+
+/*
+ * Check that tags propagate correctly when extending a tree.
+ */
+static void extend_checks(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert(&tree, 43);
+	assert(item_tag_get(&tree, 43, 0) == 0);
+	item_tag_set(&tree, 43, 0);
+	assert(item_tag_get(&tree, 43, 0) == 1);
+	item_insert(&tree, 1000000);
+	assert(item_tag_get(&tree, 43, 0) == 1);
+
+	item_insert(&tree, 0);
+	item_tag_set(&tree, 0, 0);
+	item_delete(&tree, 1000000);
+	assert(item_tag_get(&tree, 43, 0) != 0);
+	item_delete(&tree, 43);
+	assert(item_tag_get(&tree, 43, 0) == 0);	/* crash */
+	assert(item_tag_get(&tree, 0, 0) == 1);
+
+	verify_tag_consistency(&tree, 0);
+
+	item_kill_tree(&tree);
+}
+
+/*
+ * Check that tags propagate correctly when contracting a tree.
+ */
+static void contract_checks(void)
+{
+	struct item *item;
+	int tmp;
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	tmp = 1<<RADIX_TREE_MAP_SHIFT;
+	item_insert(&tree, tmp);
+	item_insert(&tree, tmp+1);
+	item_tag_set(&tree, tmp, 0);
+	item_tag_set(&tree, tmp, 1);
+	item_tag_set(&tree, tmp+1, 0);
+	item_delete(&tree, tmp+1);
+	item_tag_clear(&tree, tmp, 1);
+
+	assert(radix_tree_gang_lookup_tag(&tree, (void **)&item, 0, 1, 0) == 1);
+	assert(radix_tree_gang_lookup_tag(&tree, (void **)&item, 0, 1, 1) == 0);
+
+	assert(item_tag_get(&tree, tmp, 0) == 1);
+	assert(item_tag_get(&tree, tmp, 1) == 0);
+
+	verify_tag_consistency(&tree, 0);
+	item_kill_tree(&tree);
+}
+
+/*
+ * Stupid tag thrasher
+ *
+ * Create a large linear array corresponding to the tree.   Each element in
+ * the array is coherent with each node in the tree
+ */
+
+enum {
+	NODE_ABSENT = 0,
+	NODE_PRESENT = 1,
+	NODE_TAGGED = 2,
+};
+
+#define THRASH_SIZE		1000 * 1000
+#define N 127
+#define BATCH	33
+
+static void gang_check(struct radix_tree_root *tree,
+			char *thrash_state, int tag)
+{
+	struct item *items[BATCH];
+	int nr_found;
+	unsigned long index = 0;
+	unsigned long last_index = 0;
+
+	while ((nr_found = radix_tree_gang_lookup_tag(tree, (void **)items,
+					index, BATCH, tag))) {
+		int i;
+
+		for (i = 0; i < nr_found; i++) {
+			struct item *item = items[i];
+
+			while (last_index < item->index) {
+				assert(thrash_state[last_index] != NODE_TAGGED);
+				last_index++;
+			}
+			assert(thrash_state[last_index] == NODE_TAGGED);
+			last_index++;
+		}
+		index = items[nr_found - 1]->index + 1;
+	}
+}
+
+static void do_thrash(struct radix_tree_root *tree, char *thrash_state, int tag)
+{
+	int insert_chunk;
+	int delete_chunk;
+	int tag_chunk;
+	int untag_chunk;
+	int total_tagged = 0;
+	int total_present = 0;
+
+	for (insert_chunk = 1; insert_chunk < THRASH_SIZE; insert_chunk *= N)
+	for (delete_chunk = 1; delete_chunk < THRASH_SIZE; delete_chunk *= N)
+	for (tag_chunk = 1; tag_chunk < THRASH_SIZE; tag_chunk *= N)
+	for (untag_chunk = 1; untag_chunk < THRASH_SIZE; untag_chunk *= N) {
+		int i;
+		unsigned long index;
+		int nr_inserted = 0;
+		int nr_deleted = 0;
+		int nr_tagged = 0;
+		int nr_untagged = 0;
+		int actual_total_tagged;
+		int actual_total_present;
+
+		for (i = 0; i < insert_chunk; i++) {
+			index = rand() % THRASH_SIZE;
+			if (thrash_state[index] != NODE_ABSENT)
+				continue;
+			item_check_absent(tree, index);
+			item_insert(tree, index);
+			assert(thrash_state[index] != NODE_PRESENT);
+			thrash_state[index] = NODE_PRESENT;
+			nr_inserted++;
+			total_present++;
+		}
+
+		for (i = 0; i < delete_chunk; i++) {
+			index = rand() % THRASH_SIZE;
+			if (thrash_state[index] == NODE_ABSENT)
+				continue;
+			item_check_present(tree, index);
+			if (item_tag_get(tree, index, tag)) {
+				assert(thrash_state[index] == NODE_TAGGED);
+				total_tagged--;
+			} else {
+				assert(thrash_state[index] == NODE_PRESENT);
+			}
+			item_delete(tree, index);
+			assert(thrash_state[index] != NODE_ABSENT);
+			thrash_state[index] = NODE_ABSENT;
+			nr_deleted++;
+			total_present--;
+		}
+
+		for (i = 0; i < tag_chunk; i++) {
+			index = rand() % THRASH_SIZE;
+			if (thrash_state[index] != NODE_PRESENT) {
+				if (item_lookup(tree, index))
+					assert(item_tag_get(tree, index, tag));
+				continue;
+			}
+			item_tag_set(tree, index, tag);
+			item_tag_set(tree, index, tag);
+			assert(thrash_state[index] != NODE_TAGGED);
+			thrash_state[index] = NODE_TAGGED;
+			nr_tagged++;
+			total_tagged++;
+		}
+
+		for (i = 0; i < untag_chunk; i++) {
+			index = rand() % THRASH_SIZE;
+			if (thrash_state[index] != NODE_TAGGED)
+				continue;
+			item_check_present(tree, index);
+			assert(item_tag_get(tree, index, tag));
+			item_tag_clear(tree, index, tag);
+			item_tag_clear(tree, index, tag);
+			assert(thrash_state[index] != NODE_PRESENT);
+			thrash_state[index] = NODE_PRESENT;
+			nr_untagged++;
+			total_tagged--;
+		}
+
+		actual_total_tagged = 0;
+		actual_total_present = 0;
+		for (index = 0; index < THRASH_SIZE; index++) {
+			switch (thrash_state[index]) {
+			case NODE_ABSENT:
+				item_check_absent(tree, index);
+				break;
+			case NODE_PRESENT:
+				item_check_present(tree, index);
+				assert(!item_tag_get(tree, index, tag));
+				actual_total_present++;
+				break;
+			case NODE_TAGGED:
+				item_check_present(tree, index);
+				assert(item_tag_get(tree, index, tag));
+				actual_total_present++;
+				actual_total_tagged++;
+				break;
+			}
+		}
+
+		gang_check(tree, thrash_state, tag);
+
+		printf("%d(%d) %d(%d) %d(%d) %d(%d) / "
+				"%d(%d) present, %d(%d) tagged\n",
+			insert_chunk, nr_inserted,
+			delete_chunk, nr_deleted,
+			tag_chunk, nr_tagged,
+			untag_chunk, nr_untagged,
+			total_present, actual_total_present,
+			total_tagged, actual_total_tagged);
+	}
+}
+
+static void thrash_tags(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+	char *thrash_state;
+
+	thrash_state = malloc(THRASH_SIZE);
+	memset(thrash_state, 0, THRASH_SIZE);
+
+	do_thrash(&tree, thrash_state, 0);
+
+	verify_tag_consistency(&tree, 0);
+	item_kill_tree(&tree);
+	free(thrash_state);
+}
+
+static void leak_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	item_insert(&tree, 1000000);
+	item_delete(&tree, 1000000);
+	item_kill_tree(&tree);
+}
+
+static void __leak_check(void)
+{
+	RADIX_TREE(tree, GFP_KERNEL);
+
+	printf("%d: nr_allocated=%d\n", __LINE__, nr_allocated);
+	item_insert(&tree, 1000000);
+	printf("%d: nr_allocated=%d\n", __LINE__, nr_allocated);
+	item_delete(&tree, 1000000);
+	printf("%d: nr_allocated=%d\n", __LINE__, nr_allocated);
+	item_kill_tree(&tree);
+	printf("%d: nr_allocated=%d\n", __LINE__, nr_allocated);
+}
+
+static void single_check(void)
+{
+	struct item *items[BATCH];
+	RADIX_TREE(tree, GFP_KERNEL);
+	int ret;
+
+	item_insert(&tree, 0);
+	item_tag_set(&tree, 0, 0);
+	ret = radix_tree_gang_lookup_tag(&tree, (void **)items, 0, BATCH, 0);
+	assert(ret == 1);
+	ret = radix_tree_gang_lookup_tag(&tree, (void **)items, 1, BATCH, 0);
+	assert(ret == 0);
+	verify_tag_consistency(&tree, 0);
+	verify_tag_consistency(&tree, 1);
+	item_kill_tree(&tree);
+}
+
+void tag_check(void)
+{
+	single_check();
+	extend_checks();
+	contract_checks();
+	printf("after extend_checks: %d allocated\n", nr_allocated);
+	__leak_check();
+	leak_check();
+	printf("after leak_check: %d allocated\n", nr_allocated);
+	simple_checks();
+	printf("after simple_checks: %d allocated\n", nr_allocated);
+	thrash_tags();
+	printf("after thrash_tags: %d allocated\n", nr_allocated);
+}
diff --git a/tools/testing/radix-tree/test.c b/tools/testing/radix-tree/test.c
new file mode 100644
index 000000000000..2bebf34cdc27
--- /dev/null
+++ b/tools/testing/radix-tree/test.c
@@ -0,0 +1,219 @@
+#include <stdlib.h>
+#include <assert.h>
+#include <stdio.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/bitops.h>
+
+#include "test.h"
+
+struct item *
+item_tag_set(struct radix_tree_root *root, unsigned long index, int tag)
+{
+	return radix_tree_tag_set(root, index, tag);
+}
+
+struct item *
+item_tag_clear(struct radix_tree_root *root, unsigned long index, int tag)
+{
+	return radix_tree_tag_clear(root, index, tag);
+}
+
+int item_tag_get(struct radix_tree_root *root, unsigned long index, int tag)
+{
+	return radix_tree_tag_get(root, index, tag);
+}
+
+int __item_insert(struct radix_tree_root *root, struct item *item)
+{
+	return radix_tree_insert(root, item->index, item);
+}
+
+int item_insert(struct radix_tree_root *root, unsigned long index)
+{
+	return __item_insert(root, item_create(index));
+}
+
+int item_delete(struct radix_tree_root *root, unsigned long index)
+{
+	struct item *item = radix_tree_delete(root, index);
+
+	if (item) {
+		assert(item->index == index);
+		free(item);
+		return 1;
+	}
+	return 0;
+}
+
+struct item *item_create(unsigned long index)
+{
+	struct item *ret = malloc(sizeof(*ret));
+
+	ret->index = index;
+	return ret;
+}
+
+void item_check_present(struct radix_tree_root *root, unsigned long index)
+{
+	struct item *item;
+
+	item = radix_tree_lookup(root, index);
+	assert(item != 0);
+	assert(item->index == index);
+}
+
+struct item *item_lookup(struct radix_tree_root *root, unsigned long index)
+{
+	return radix_tree_lookup(root, index);
+}
+
+void item_check_absent(struct radix_tree_root *root, unsigned long index)
+{
+	struct item *item;
+
+	item = radix_tree_lookup(root, index);
+	assert(item == 0);
+}
+
+/*
+ * Scan only the passed (start, start+nr] for present items
+ */
+void item_gang_check_present(struct radix_tree_root *root,
+			unsigned long start, unsigned long nr,
+			int chunk, int hop)
+{
+	struct item *items[chunk];
+	unsigned long into;
+
+	for (into = 0; into < nr; ) {
+		int nfound;
+		int nr_to_find = chunk;
+		int i;
+
+		if (nr_to_find > (nr - into))
+			nr_to_find = nr - into;
+
+		nfound = radix_tree_gang_lookup(root, (void **)items,
+						start + into, nr_to_find);
+		assert(nfound == nr_to_find);
+		for (i = 0; i < nfound; i++)
+			assert(items[i]->index == start + into + i);
+		into += hop;
+	}
+}
+
+/*
+ * Scan the entire tree, only expecting present items (start, start+nr]
+ */
+void item_full_scan(struct radix_tree_root *root, unsigned long start,
+			unsigned long nr, int chunk)
+{
+	struct item *items[chunk];
+	unsigned long into = 0;
+	unsigned long this_index = start;
+	int nfound;
+	int i;
+
+//	printf("%s(0x%08lx, 0x%08lx, %d)\n", __FUNCTION__, start, nr, chunk);
+
+	while ((nfound = radix_tree_gang_lookup(root, (void **)items, into,
+					chunk))) {
+//		printf("At 0x%08lx, nfound=%d\n", into, nfound);
+		for (i = 0; i < nfound; i++) {
+			assert(items[i]->index == this_index);
+			this_index++;
+		}
+//		printf("Found 0x%08lx->0x%08lx\n",
+//			items[0]->index, items[nfound-1]->index);
+		into = this_index;
+	}
+	if (chunk)
+		assert(this_index == start + nr);
+	nfound = radix_tree_gang_lookup(root, (void **)items,
+					this_index, chunk);
+	assert(nfound == 0);
+}
+
+static int verify_node(struct radix_tree_node *slot, unsigned int tag,
+			unsigned int height, int tagged)
+{
+	int anyset = 0;
+	int i;
+	int j;
+
+	slot = indirect_to_ptr(slot);
+
+	/* Verify consistency at this level */
+	for (i = 0; i < RADIX_TREE_TAG_LONGS; i++) {
+		if (slot->tags[tag][i]) {
+			anyset = 1;
+			break;
+		}
+	}
+	if (tagged != anyset) {
+		printf("tag: %u, height %u, tagged: %d, anyset: %d\n", tag, height, tagged, anyset);
+		for (j = 0; j < RADIX_TREE_MAX_TAGS; j++) {
+			printf("tag %d: ", j);
+			for (i = 0; i < RADIX_TREE_TAG_LONGS; i++)
+				printf("%016lx ", slot->tags[j][i]);
+			printf("\n");
+		}
+		return 1;
+	}
+	assert(tagged == anyset);
+
+	/* Go for next level */
+	if (height > 1) {
+		for (i = 0; i < RADIX_TREE_MAP_SIZE; i++)
+			if (slot->slots[i])
+				if (verify_node(slot->slots[i], tag, height - 1,
+					    !!test_bit(i, slot->tags[tag]))) {
+					printf("Failure at off %d\n", i);
+					for (j = 0; j < RADIX_TREE_MAX_TAGS; j++) {
+						printf("tag %d: ", j);
+						for (i = 0; i < RADIX_TREE_TAG_LONGS; i++)
+							printf("%016lx ", slot->tags[j][i]);
+						printf("\n");
+					}
+					return 1;
+				}
+	}
+	return 0;
+}
+
+void verify_tag_consistency(struct radix_tree_root *root, unsigned int tag)
+{
+	if (!root->height)
+		return;
+	verify_node(root->rnode, tag, root->height, !!root_tag_get(root, tag));
+}
+
+void item_kill_tree(struct radix_tree_root *root)
+{
+	struct item *items[32];
+	int nfound;
+
+	while ((nfound = radix_tree_gang_lookup(root, (void **)items, 0, 32))) {
+		int i;
+
+		for (i = 0; i < nfound; i++) {
+			void *ret;
+
+			ret = radix_tree_delete(root, items[i]->index);
+			assert(ret == items[i]);
+			free(items[i]);
+		}
+	}
+	assert(radix_tree_gang_lookup(root, (void **)items, 0, 32) == 0);
+	assert(root->rnode == NULL);
+}
+
+void tree_verify_min_height(struct radix_tree_root *root, int maxindex)
+{
+	assert(radix_tree_maxindex(root->height) >= maxindex);
+	if (root->height > 1)
+		assert(radix_tree_maxindex(root->height-1) < maxindex);
+	else if (root->height == 1)
+		assert(radix_tree_maxindex(root->height-1) <= maxindex);
+}
diff --git a/tools/testing/radix-tree/test.h b/tools/testing/radix-tree/test.h
new file mode 100644
index 000000000000..4e1d95faaa94
--- /dev/null
+++ b/tools/testing/radix-tree/test.h
@@ -0,0 +1,40 @@
+#include <linux/gfp.h>
+#include <linux/types.h>
+#include <linux/radix-tree.h>
+#include <linux/rcupdate.h>
+
+struct item {
+	unsigned long index;
+};
+
+struct item *item_create(unsigned long index);
+int __item_insert(struct radix_tree_root *root, struct item *item);
+int item_insert(struct radix_tree_root *root, unsigned long index);
+int item_delete(struct radix_tree_root *root, unsigned long index);
+struct item *item_lookup(struct radix_tree_root *root, unsigned long index);
+
+void item_check_present(struct radix_tree_root *root, unsigned long index);
+void item_check_absent(struct radix_tree_root *root, unsigned long index);
+void item_gang_check_present(struct radix_tree_root *root,
+			unsigned long start, unsigned long nr,
+			int chunk, int hop);
+void item_full_scan(struct radix_tree_root *root, unsigned long start,
+			unsigned long nr, int chunk);
+void item_kill_tree(struct radix_tree_root *root);
+
+void tag_check(void);
+
+struct item *
+item_tag_set(struct radix_tree_root *root, unsigned long index, int tag);
+struct item *
+item_tag_clear(struct radix_tree_root *root, unsigned long index, int tag);
+int item_tag_get(struct radix_tree_root *root, unsigned long index, int tag);
+void tree_verify_min_height(struct radix_tree_root *root, int maxindex);
+void verify_tag_consistency(struct radix_tree_root *root, unsigned int tag);
+
+extern int nr_allocated;
+
+/* Normally private parts of lib/radix-tree.c */
+void *indirect_to_ptr(void *ptr);
+int root_tag_get(struct radix_tree_root *root, unsigned int tag);
+unsigned long radix_tree_maxindex(unsigned int height);
diff --git a/tools/testing/selftests/breakpoints/.gitignore b/tools/testing/selftests/breakpoints/.gitignore
index 9b3193d06608..a23bb4a6f06c 100644
--- a/tools/testing/selftests/breakpoints/.gitignore
+++ b/tools/testing/selftests/breakpoints/.gitignore
@@ -1 +1,2 @@
 breakpoint_test
+step_after_suspend_test
diff --git a/tools/testing/selftests/breakpoints/Makefile b/tools/testing/selftests/breakpoints/Makefile
index c0d957015f52..74e533fd4bc5 100644
--- a/tools/testing/selftests/breakpoints/Makefile
+++ b/tools/testing/selftests/breakpoints/Makefile
@@ -6,9 +6,11 @@ ifeq ($(ARCH),x86)
 TEST_PROGS := breakpoint_test
 endif
 
+TEST_PROGS += step_after_suspend_test
+
 all: $(TEST_PROGS)
 
 include ../lib.mk
 
 clean:
-	rm -fr breakpoint_test
+	rm -fr breakpoint_test step_after_suspend_test
diff --git a/tools/testing/selftests/breakpoints/step_after_suspend_test.c b/tools/testing/selftests/breakpoints/step_after_suspend_test.c
new file mode 100644
index 000000000000..60b8a95dac26
--- /dev/null
+++ b/tools/testing/selftests/breakpoints/step_after_suspend_test.c
@@ -0,0 +1,218 @@
+/*
+ * Copyright (C) 2016 Google, Inc.
+ *
+ * This software is licensed under the terms of the GNU General Public
+ * License version 2, as published by the Free Software Foundation, and
+ * may be copied, distributed, and modified under those terms.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ */
+
+#define _GNU_SOURCE
+
+#include <errno.h>
+#include <fcntl.h>
+#include <sched.h>
+#include <signal.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <string.h>
+#include <unistd.h>
+#include <sys/ptrace.h>
+#include <sys/stat.h>
+#include <sys/timerfd.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+
+#include "../kselftest.h"
+
+void child(int cpu)
+{
+	cpu_set_t set;
+
+	CPU_ZERO(&set);
+	CPU_SET(cpu, &set);
+	if (sched_setaffinity(0, sizeof(set), &set) != 0) {
+		perror("sched_setaffinity() failed");
+		_exit(1);
+	}
+
+	if (ptrace(PTRACE_TRACEME, 0, NULL, NULL) != 0) {
+		perror("ptrace(PTRACE_TRACEME) failed");
+		_exit(1);
+	}
+
+	if (raise(SIGSTOP) != 0) {
+		perror("raise(SIGSTOP) failed");
+		_exit(1);
+	}
+
+	_exit(0);
+}
+
+bool run_test(int cpu)
+{
+	int status;
+	pid_t pid = fork();
+	pid_t wpid;
+
+	if (pid < 0) {
+		perror("fork() failed");
+		return false;
+	}
+	if (pid == 0)
+		child(cpu);
+
+	wpid = waitpid(pid, &status, __WALL);
+	if (wpid != pid) {
+		perror("waitpid() failed");
+		return false;
+	}
+	if (!WIFSTOPPED(status)) {
+		printf("child did not stop\n");
+		return false;
+	}
+	if (WSTOPSIG(status) != SIGSTOP) {
+		printf("child did not stop with SIGSTOP\n");
+		return false;
+	}
+
+	if (ptrace(PTRACE_SINGLESTEP, pid, NULL, NULL) < 0) {
+		if (errno == EIO) {
+			printf("ptrace(PTRACE_SINGLESTEP) not supported on this architecture\n");
+			ksft_exit_skip();
+		}
+		perror("ptrace(PTRACE_SINGLESTEP) failed");
+		return false;
+	}
+
+	wpid = waitpid(pid, &status, __WALL);
+	if (wpid != pid) {
+		perror("waitpid() failed");
+		return false;
+	}
+	if (WIFEXITED(status)) {
+		printf("child did not single-step\n");
+		return false;
+	}
+	if (!WIFSTOPPED(status)) {
+		printf("child did not stop\n");
+		return false;
+	}
+	if (WSTOPSIG(status) != SIGTRAP) {
+		printf("child did not stop with SIGTRAP\n");
+		return false;
+	}
+
+	if (ptrace(PTRACE_CONT, pid, NULL, NULL) < 0) {
+		perror("ptrace(PTRACE_CONT) failed");
+		return false;
+	}
+
+	wpid = waitpid(pid, &status, __WALL);
+	if (wpid != pid) {
+		perror("waitpid() failed");
+		return false;
+	}
+	if (!WIFEXITED(status)) {
+		printf("child did not exit after PTRACE_CONT\n");
+		return false;
+	}
+
+	return true;
+}
+
+void suspend(void)
+{
+	int power_state_fd;
+	struct sigevent event = {};
+	int timerfd;
+	int err;
+	struct itimerspec spec = {};
+
+	power_state_fd = open("/sys/power/state", O_RDWR);
+	if (power_state_fd < 0) {
+		perror("open(\"/sys/power/state\") failed (is this test running as root?)");
+		ksft_exit_fail();
+	}
+
+	timerfd = timerfd_create(CLOCK_BOOTTIME_ALARM, 0);
+	if (timerfd < 0) {
+		perror("timerfd_create() failed");
+		ksft_exit_fail();
+	}
+
+	spec.it_value.tv_sec = 5;
+	err = timerfd_settime(timerfd, 0, &spec, NULL);
+	if (err < 0) {
+		perror("timerfd_settime() failed");
+		ksft_exit_fail();
+	}
+
+	if (write(power_state_fd, "mem", strlen("mem")) != strlen("mem")) {
+		perror("entering suspend failed");
+		ksft_exit_fail();
+	}
+
+	close(timerfd);
+	close(power_state_fd);
+}
+
+int main(int argc, char **argv)
+{
+	int opt;
+	bool do_suspend = true;
+	bool succeeded = true;
+	cpu_set_t available_cpus;
+	int err;
+	int cpu;
+
+	while ((opt = getopt(argc, argv, "n")) != -1) {
+		switch (opt) {
+		case 'n':
+			do_suspend = false;
+			break;
+		default:
+			printf("Usage: %s [-n]\n", argv[0]);
+			printf("        -n: do not trigger a suspend/resume cycle before the test\n");
+			return -1;
+		}
+	}
+
+	if (do_suspend)
+		suspend();
+
+	err = sched_getaffinity(0, sizeof(available_cpus), &available_cpus);
+	if (err < 0) {
+		perror("sched_getaffinity() failed");
+		ksft_exit_fail();
+	}
+
+	for (cpu = 0; cpu < CPU_SETSIZE; cpu++) {
+		bool test_success;
+
+		if (!CPU_ISSET(cpu, &available_cpus))
+			continue;
+
+		test_success = run_test(cpu);
+		printf("CPU %d: ", cpu);
+		if (test_success) {
+			printf("[OK]\n");
+			ksft_inc_pass_cnt();
+		} else {
+			printf("[FAILED]\n");
+			ksft_inc_fail_cnt();
+			succeeded = false;
+		}
+	}
+
+	ksft_print_cnts();
+	if (succeeded)
+		ksft_exit_pass();
+	else
+		ksft_exit_fail();
+}
diff --git a/tools/testing/selftests/cpu-hotplug/config b/tools/testing/selftests/cpu-hotplug/config
new file mode 100644
index 000000000000..e6ab090cfbf3
--- /dev/null
+++ b/tools/testing/selftests/cpu-hotplug/config
@@ -0,0 +1,2 @@
+CONFIG_NOTIFIER_ERROR_INJECTION=y
+CONFIG_CPU_NOTIFIER_ERROR_INJECT=m
diff --git a/tools/testing/selftests/efivarfs/efivarfs.sh b/tools/testing/selftests/efivarfs/efivarfs.sh
index 77edcdcc016b..057278448515 100755
--- a/tools/testing/selftests/efivarfs/efivarfs.sh
+++ b/tools/testing/selftests/efivarfs/efivarfs.sh
@@ -88,7 +88,11 @@ test_delete()
 		exit 1
 	fi
 
-	rm $file
+	rm $file 2>/dev/null
+	if [ $? -ne 0 ]; then
+		chattr -i $file
+		rm $file
+	fi
 
 	if [ -e $file ]; then
 		echo "$file couldn't be deleted" >&2
@@ -111,6 +115,7 @@ test_zero_size_delete()
 		exit 1
 	fi
 
+	chattr -i $file
 	printf "$attrs" > $file
 
 	if [ -e $file ]; then
@@ -141,7 +146,11 @@ test_valid_filenames()
 			echo "$file could not be created" >&2
 			ret=1
 		else
-			rm $file
+			rm $file 2>/dev/null
+			if [ $? -ne 0 ]; then
+				chattr -i $file
+				rm $file
+			fi
 		fi
 	done
 
@@ -174,7 +183,11 @@ test_invalid_filenames()
 
 		if [ -e $file ]; then
 			echo "Creating $file should have failed" >&2
-			rm $file
+			rm $file 2>/dev/null
+			if [ $? -ne 0 ]; then
+				chattr -i $file
+				rm $file
+			fi
 			ret=1
 		fi
 	done
diff --git a/tools/testing/selftests/efivarfs/open-unlink.c b/tools/testing/selftests/efivarfs/open-unlink.c
index 8c0764407b3c..4af74f733036 100644
--- a/tools/testing/selftests/efivarfs/open-unlink.c
+++ b/tools/testing/selftests/efivarfs/open-unlink.c
@@ -1,10 +1,68 @@
+#include <errno.h>
 #include <stdio.h>
 #include <stdint.h>
 #include <stdlib.h>
 #include <unistd.h>
+#include <sys/ioctl.h>
 #include <sys/types.h>
 #include <sys/stat.h>
 #include <fcntl.h>
+#include <linux/fs.h>
+
+static int set_immutable(const char *path, int immutable)
+{
+	unsigned int flags;
+	int fd;
+	int rc;
+	int error;
+
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	rc = ioctl(fd, FS_IOC_GETFLAGS, &flags);
+	if (rc < 0) {
+		error = errno;
+		close(fd);
+		errno = error;
+		return rc;
+	}
+
+	if (immutable)
+		flags |= FS_IMMUTABLE_FL;
+	else
+		flags &= ~FS_IMMUTABLE_FL;
+
+	rc = ioctl(fd, FS_IOC_SETFLAGS, &flags);
+	error = errno;
+	close(fd);
+	errno = error;
+	return rc;
+}
+
+static int get_immutable(const char *path)
+{
+	unsigned int flags;
+	int fd;
+	int rc;
+	int error;
+
+	fd = open(path, O_RDONLY);
+	if (fd < 0)
+		return fd;
+
+	rc = ioctl(fd, FS_IOC_GETFLAGS, &flags);
+	if (rc < 0) {
+		error = errno;
+		close(fd);
+		errno = error;
+		return rc;
+	}
+	close(fd);
+	if (flags & FS_IMMUTABLE_FL)
+		return 1;
+	return 0;
+}
 
 int main(int argc, char **argv)
 {
@@ -27,7 +85,7 @@ int main(int argc, char **argv)
 	buf[4] = 0;
 
 	/* create a test variable */
-	fd = open(path, O_WRONLY | O_CREAT);
+	fd = open(path, O_WRONLY | O_CREAT, 0600);
 	if (fd < 0) {
 		perror("open(O_WRONLY)");
 		return EXIT_FAILURE;
@@ -41,6 +99,18 @@ int main(int argc, char **argv)
 
 	close(fd);
 
+	rc = get_immutable(path);
+	if (rc < 0) {
+		perror("ioctl(FS_IOC_GETFLAGS)");
+		return EXIT_FAILURE;
+	} else if (rc) {
+		rc = set_immutable(path, 0);
+		if (rc < 0) {
+			perror("ioctl(FS_IOC_SETFLAGS)");
+			return EXIT_FAILURE;
+		}
+	}
+
 	fd = open(path, O_RDONLY);
 	if (fd < 0) {
 		perror("open");
diff --git a/tools/testing/selftests/firmware/config b/tools/testing/selftests/firmware/config
new file mode 100644
index 000000000000..c8137f70e291
--- /dev/null
+++ b/tools/testing/selftests/firmware/config
@@ -0,0 +1 @@
+CONFIG_TEST_FIRMWARE=y
diff --git a/tools/testing/selftests/ftrace/config b/tools/testing/selftests/ftrace/config
new file mode 100644
index 000000000000..ef8214661612
--- /dev/null
+++ b/tools/testing/selftests/ftrace/config
@@ -0,0 +1 @@
+CONFIG_FTRACE=y
diff --git a/tools/testing/selftests/ftrace/test.d/instances/instance.tc b/tools/testing/selftests/ftrace/test.d/instances/instance.tc
index 773e276ff90b..1e1abe0ad354 100644
--- a/tools/testing/selftests/ftrace/test.d/instances/instance.tc
+++ b/tools/testing/selftests/ftrace/test.d/instances/instance.tc
@@ -39,28 +39,23 @@ instance_slam() {
 }
 
 instance_slam &
-x=`jobs -l`
-p1=`echo $x | cut -d' ' -f2`
+p1=$!
 echo $p1
 
 instance_slam &
-x=`jobs -l | tail -1`
-p2=`echo $x | cut -d' ' -f2`
+p2=$!
 echo $p2
 
 instance_slam &
-x=`jobs -l | tail -1`
-p3=`echo $x | cut -d' ' -f2`
+p3=$!
 echo $p3
 
 instance_slam &
-x=`jobs -l | tail -1`
-p4=`echo $x | cut -d' ' -f2`
+p4=$!
 echo $p4
 
 instance_slam &
-x=`jobs -l | tail -1`
-p5=`echo $x | cut -d' ' -f2`
+p5=$!
 echo $p5
 
 ls -lR >/dev/null
diff --git a/tools/testing/selftests/ipc/.gitignore b/tools/testing/selftests/ipc/.gitignore
new file mode 100644
index 000000000000..84b66a3c1f74
--- /dev/null
+++ b/tools/testing/selftests/ipc/.gitignore
@@ -0,0 +1 @@
+msgque_test
diff --git a/tools/testing/selftests/ipc/config b/tools/testing/selftests/ipc/config
new file mode 100644
index 000000000000..0702447109f5
--- /dev/null
+++ b/tools/testing/selftests/ipc/config
@@ -0,0 +1,2 @@
+CONFIG_EXPERT=y
+CONFIG_CHECKPOINT_RESTORE=y
diff --git a/tools/testing/selftests/lib/Makefile b/tools/testing/selftests/lib/Makefile
index 47147b968514..08360060ab14 100644
--- a/tools/testing/selftests/lib/Makefile
+++ b/tools/testing/selftests/lib/Makefile
@@ -3,6 +3,6 @@
 # No binaries, but make sure arg-less "make" doesn't trigger "run_tests"
 all:
 
-TEST_PROGS := printf.sh
+TEST_PROGS := printf.sh bitmap.sh
 
 include ../lib.mk
diff --git a/tools/testing/selftests/lib/bitmap.sh b/tools/testing/selftests/lib/bitmap.sh
new file mode 100755
index 000000000000..2da187b6ddad
--- /dev/null
+++ b/tools/testing/selftests/lib/bitmap.sh
@@ -0,0 +1,10 @@
+#!/bin/sh
+# Runs bitmap infrastructure tests using test_bitmap kernel module
+
+if /sbin/modprobe -q test_bitmap; then
+	/sbin/modprobe -q -r test_bitmap
+	echo "bitmap: ok"
+else
+	echo "bitmap: [FAIL]"
+	exit 1
+fi
diff --git a/tools/testing/selftests/media_tests/.gitignore b/tools/testing/selftests/media_tests/.gitignore
new file mode 100644
index 000000000000..1c0711708b98
--- /dev/null
+++ b/tools/testing/selftests/media_tests/.gitignore
@@ -0,0 +1 @@
+media_device_test
diff --git a/tools/testing/selftests/media_tests/Makefile b/tools/testing/selftests/media_tests/Makefile
new file mode 100644
index 000000000000..7071bcc1d066
--- /dev/null
+++ b/tools/testing/selftests/media_tests/Makefile
@@ -0,0 +1,7 @@
+TEST_PROGS := media_device_test
+all: $(TEST_PROGS)
+
+include ../lib.mk
+
+clean:
+	rm -fr media_device_test
diff --git a/tools/testing/selftests/media_tests/media_device_test.c b/tools/testing/selftests/media_tests/media_device_test.c
new file mode 100644
index 000000000000..cbf53a032ab5
--- /dev/null
+++ b/tools/testing/selftests/media_tests/media_device_test.c
@@ -0,0 +1,95 @@
+/*
+ * media_devkref_test.c - Media Controller Device Kref API Test
+ *
+ * Copyright (c) 2016 Shuah Khan <[email protected]>
+ * Copyright (c) 2016 Samsung Electronics Co., Ltd.
+ *
+ * This file is released under the GPLv2.
+ */
+
+/*
+ * This file adds a test for Media Controller API.
+ * This test should be run as root and should not be
+ * included in the Kselftest run. This test should be
+ * run when hardware and driver that makes use Media
+ * Controller API are present in the system.
+ *
+ * This test opens user specified Media Device and calls
+ * MEDIA_IOC_DEVICE_INFO ioctl in a loop once every 10
+ * seconds.
+ *
+ * Usage:
+ *	sudo ./media_device_test -d /dev/mediaX
+ *
+ *	While test is running, remove the device and
+ *	ensure there are no use after free errors and
+ *	other Oops in the dmesg. Enable KaSan kernel
+ *	config option for use-after-free error detection.
+*/
+
+#include <stdio.h>
+#include <unistd.h>
+#include <stdlib.h>
+#include <errno.h>
+#include <string.h>
+#include <fcntl.h>
+#include <sys/ioctl.h>
+#include <sys/stat.h>
+#include <linux/media.h>
+
+int main(int argc, char **argv)
+{
+	int opt;
+	char media_device[256];
+	int count = 0;
+	struct media_device_info mdi;
+	int ret;
+	int fd;
+
+	if (argc < 2) {
+		printf("Usage: %s [-d </dev/mediaX>]\n", argv[0]);
+		exit(-1);
+	}
+
+	/* Process arguments */
+	while ((opt = getopt(argc, argv, "d:")) != -1) {
+		switch (opt) {
+		case 'd':
+			strncpy(media_device, optarg, sizeof(media_device) - 1);
+			media_device[sizeof(media_device)-1] = '\0';
+			break;
+		default:
+			printf("Usage: %s [-d </dev/mediaX>]\n", argv[0]);
+			exit(-1);
+		}
+	}
+
+	if (getuid() != 0) {
+		printf("Please run the test as root - Exiting.\n");
+		exit(-1);
+	}
+
+	/* Open Media device and keep it open */
+	fd = open(media_device, O_RDWR);
+	if (fd == -1) {
+		printf("Media Device open errno %s\n", strerror(errno));
+		exit(-1);
+	}
+
+	printf("\nNote:\n"
+	       "While test is running, remove the device and\n"
+	       "ensure there are no use after free errors and\n"
+	       "other Oops in the dmesg. Enable KaSan kernel\n"
+	       "config option for use-after-free error detection.\n\n");
+
+	while (count < 100) {
+		ret = ioctl(fd, MEDIA_IOC_DEVICE_INFO, &mdi);
+		if (ret < 0)
+			printf("Media Device Info errno %s\n", strerror(errno));
+		else
+			printf("Media device model %s driver %s\n",
+				mdi.model, mdi.driver);
+		sleep(10);
+		count++;
+	}
+}
diff --git a/tools/testing/selftests/memory-hotplug/config b/tools/testing/selftests/memory-hotplug/config
new file mode 100644
index 000000000000..2fde30191a47
--- /dev/null
+++ b/tools/testing/selftests/memory-hotplug/config
@@ -0,0 +1,4 @@
+CONFIG_MEMORY_HOTPLUG=y
+CONFIG_MEMORY_HOTPLUG_SPARSE=y
+CONFIG_NOTIFIER_ERROR_INJECTION=y
+CONFIG_MEMORY_NOTIFIER_ERROR_INJECT=m
diff --git a/tools/testing/selftests/mount/config b/tools/testing/selftests/mount/config
new file mode 100644
index 000000000000..b5d881e48548
--- /dev/null
+++ b/tools/testing/selftests/mount/config
@@ -0,0 +1,2 @@
+CONFIG_USER_NS=y
+CONFIG_DEVPTS_MULTIPLE_INSTANCES=y
diff --git a/tools/testing/selftests/net/.gitignore b/tools/testing/selftests/net/.gitignore
index 6fb23366b258..69bb3fc38fb2 100644
--- a/tools/testing/selftests/net/.gitignore
+++ b/tools/testing/selftests/net/.gitignore
@@ -2,3 +2,4 @@ socket
 psock_fanout
 psock_tpacket
 reuseport_bpf
+reuseport_bpf_cpu
diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index 41449b5ad0a9..c658792d47b4 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -4,7 +4,7 @@ CFLAGS = -Wall -O2 -g
 
 CFLAGS += -I../../../../usr/include/
 
-NET_PROGS = socket psock_fanout psock_tpacket reuseport_bpf
+NET_PROGS = socket psock_fanout psock_tpacket reuseport_bpf reuseport_bpf_cpu
 
 all: $(NET_PROGS)
 %: %.c
diff --git a/tools/testing/selftests/net/config b/tools/testing/selftests/net/config
new file mode 100644
index 000000000000..e57b4ac40e72
--- /dev/null
+++ b/tools/testing/selftests/net/config
@@ -0,0 +1,3 @@
+CONFIG_USER_NS=y
+CONFIG_BPF_SYSCALL=y
+CONFIG_TEST_BPF=m
diff --git a/tools/testing/selftests/net/reuseport_bpf.c b/tools/testing/selftests/net/reuseport_bpf.c
index bec1b5dd2530..96ba386b1b7b 100644
--- a/tools/testing/selftests/net/reuseport_bpf.c
+++ b/tools/testing/selftests/net/reuseport_bpf.c
@@ -9,10 +9,12 @@
 
 #include <errno.h>
 #include <error.h>
+#include <fcntl.h>
 #include <linux/bpf.h>
 #include <linux/filter.h>
 #include <linux/unistd.h>
 #include <netinet/in.h>
+#include <netinet/tcp.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <string.h>
@@ -169,9 +171,15 @@ static void build_recv_group(const struct test_params p, int fd[], uint16_t mod,
 		if (bind(fd[i], addr, sockaddr_size()))
 			error(1, errno, "failed to bind recv socket %d", i);
 
-		if (p.protocol == SOCK_STREAM)
+		if (p.protocol == SOCK_STREAM) {
+			opt = 4;
+			if (setsockopt(fd[i], SOL_TCP, TCP_FASTOPEN, &opt,
+				       sizeof(opt)))
+				error(1, errno,
+				      "failed to set TCP_FASTOPEN on %d", i);
 			if (listen(fd[i], p.recv_socks * 10))
 				error(1, errno, "failed to listen on socket");
+		}
 	}
 	free(addr);
 }
@@ -189,10 +197,8 @@ static void send_from(struct test_params p, uint16_t sport, char *buf,
 
 	if (bind(fd, saddr, sockaddr_size()))
 		error(1, errno, "failed to bind send socket");
-	if (connect(fd, daddr, sockaddr_size()))
-		error(1, errno, "failed to connect");
 
-	if (send(fd, buf, len, 0) < 0)
+	if (sendto(fd, buf, len, MSG_FASTOPEN, daddr, sockaddr_size()) < 0)
 		error(1, errno, "failed to send message");
 
 	close(fd);
@@ -260,7 +266,7 @@ static void test_recv_order(const struct test_params p, int fd[], int mod)
 	}
 }
 
-static void test_reuseport_ebpf(const struct test_params p)
+static void test_reuseport_ebpf(struct test_params p)
 {
 	int i, fd[p.recv_socks];
 
@@ -268,6 +274,7 @@ static void test_reuseport_ebpf(const struct test_params p)
 	build_recv_group(p, fd, p.recv_socks, attach_ebpf);
 	test_recv_order(p, fd, p.recv_socks);
 
+	p.send_port_min += p.recv_socks * 2;
 	fprintf(stderr, "Reprograming, testing mod %zd...\n", p.recv_socks / 2);
 	attach_ebpf(fd[0], p.recv_socks / 2);
 	test_recv_order(p, fd, p.recv_socks / 2);
@@ -276,7 +283,7 @@ static void test_reuseport_ebpf(const struct test_params p)
 		close(fd[i]);
 }
 
-static void test_reuseport_cbpf(const struct test_params p)
+static void test_reuseport_cbpf(struct test_params p)
 {
 	int i, fd[p.recv_socks];
 
@@ -284,6 +291,7 @@ static void test_reuseport_cbpf(const struct test_params p)
 	build_recv_group(p, fd, p.recv_socks, attach_cbpf);
 	test_recv_order(p, fd, p.recv_socks);
 
+	p.send_port_min += p.recv_socks * 2;
 	fprintf(stderr, "Reprograming, testing mod %zd...\n", p.recv_socks / 2);
 	attach_cbpf(fd[0], p.recv_socks / 2);
 	test_recv_order(p, fd, p.recv_socks / 2);
@@ -377,7 +385,7 @@ static void test_filter_no_reuseport(const struct test_params p)
 
 static void test_filter_without_bind(void)
 {
-	int fd1, fd2;
+	int fd1, fd2, opt = 1;
 
 	fprintf(stderr, "Testing filter add without bind...\n");
 	fd1 = socket(AF_INET, SOCK_DGRAM, 0);
@@ -386,6 +394,10 @@ static void test_filter_without_bind(void)
 	fd2 = socket(AF_INET, SOCK_DGRAM, 0);
 	if (fd2 < 0)
 		error(1, errno, "failed to create socket 2");
+	if (setsockopt(fd1, SOL_SOCKET, SO_REUSEPORT, &opt, sizeof(opt)))
+		error(1, errno, "failed to set SO_REUSEPORT on socket 1");
+	if (setsockopt(fd2, SOL_SOCKET, SO_REUSEPORT, &opt, sizeof(opt)))
+		error(1, errno, "failed to set SO_REUSEPORT on socket 2");
 
 	attach_ebpf(fd1, 10);
 	attach_cbpf(fd2, 10);
@@ -394,6 +406,32 @@ static void test_filter_without_bind(void)
 	close(fd2);
 }
 
+void enable_fastopen(void)
+{
+	int fd = open("/proc/sys/net/ipv4/tcp_fastopen", 0);
+	int rw_mask = 3;  /* bit 1: client side; bit-2 server side */
+	int val, size;
+	char buf[16];
+
+	if (fd < 0)
+		error(1, errno, "Unable to open tcp_fastopen sysctl");
+	if (read(fd, buf, sizeof(buf)) <= 0)
+		error(1, errno, "Unable to read tcp_fastopen sysctl");
+	val = atoi(buf);
+	close(fd);
+
+	if ((val & rw_mask) != rw_mask) {
+		fd = open("/proc/sys/net/ipv4/tcp_fastopen", O_RDWR);
+		if (fd < 0)
+			error(1, errno,
+			      "Unable to open tcp_fastopen sysctl for writing");
+		val |= rw_mask;
+		size = snprintf(buf, 16, "%d", val);
+		if (write(fd, buf, size) <= 0)
+			error(1, errno, "Unable to write tcp_fastopen sysctl");
+		close(fd);
+	}
+}
 
 int main(void)
 {
@@ -506,6 +544,71 @@ int main(void)
 		.recv_port = 8007,
 		.send_port_min = 9100});
 
+	/* TCP fastopen is required for the TCP tests */
+	enable_fastopen();
+	fprintf(stderr, "---- IPv4 TCP ----\n");
+	test_reuseport_ebpf((struct test_params) {
+		.recv_family = AF_INET,
+		.send_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8008,
+		.send_port_min = 9120});
+	test_reuseport_cbpf((struct test_params) {
+		.recv_family = AF_INET,
+		.send_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8009,
+		.send_port_min = 9160});
+	test_extra_filter((struct test_params) {
+		.recv_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_port = 8010});
+	test_filter_no_reuseport((struct test_params) {
+		.recv_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_port = 8011});
+
+	fprintf(stderr, "---- IPv6 TCP ----\n");
+	test_reuseport_ebpf((struct test_params) {
+		.recv_family = AF_INET6,
+		.send_family = AF_INET6,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8012,
+		.send_port_min = 9200});
+	test_reuseport_cbpf((struct test_params) {
+		.recv_family = AF_INET6,
+		.send_family = AF_INET6,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8013,
+		.send_port_min = 9240});
+	test_extra_filter((struct test_params) {
+		.recv_family = AF_INET6,
+		.protocol = SOCK_STREAM,
+		.recv_port = 8014});
+	test_filter_no_reuseport((struct test_params) {
+		.recv_family = AF_INET6,
+		.protocol = SOCK_STREAM,
+		.recv_port = 8015});
+
+	fprintf(stderr, "---- IPv6 TCP w/ mapped IPv4 ----\n");
+	test_reuseport_ebpf((struct test_params) {
+		.recv_family = AF_INET6,
+		.send_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8016,
+		.send_port_min = 9320});
+	test_reuseport_cbpf((struct test_params) {
+		.recv_family = AF_INET6,
+		.send_family = AF_INET,
+		.protocol = SOCK_STREAM,
+		.recv_socks = 10,
+		.recv_port = 8017,
+		.send_port_min = 9360});
 
 	test_filter_without_bind();
 
diff --git a/tools/testing/selftests/net/reuseport_bpf_cpu.c b/tools/testing/selftests/net/reuseport_bpf_cpu.c
new file mode 100644
index 000000000000..b23d6f54de7b
--- /dev/null
+++ b/tools/testing/selftests/net/reuseport_bpf_cpu.c
@@ -0,0 +1,258 @@
+/*
+ * Test functionality of BPF filters with SO_REUSEPORT.  This program creates
+ * an SO_REUSEPORT receiver group containing one socket per CPU core. It then
+ * creates a BPF program that will select a socket from this group based
+ * on the core id that receives the packet.  The sending code artificially
+ * moves itself to run on different core ids and sends one message from
+ * each core.  Since these packets are delivered over loopback, they should
+ * arrive on the same core that sent them.  The receiving code then ensures
+ * that the packet was received on the socket for the corresponding core id.
+ * This entire process is done for several different core id permutations
+ * and for each IPv4/IPv6 and TCP/UDP combination.
+ */
+
+#define _GNU_SOURCE
+
+#include <arpa/inet.h>
+#include <errno.h>
+#include <error.h>
+#include <linux/filter.h>
+#include <linux/in.h>
+#include <linux/unistd.h>
+#include <sched.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <sys/epoll.h>
+#include <sys/types.h>
+#include <sys/socket.h>
+#include <unistd.h>
+
+static const int PORT = 8888;
+
+static void build_rcv_group(int *rcv_fd, size_t len, int family, int proto)
+{
+	struct sockaddr_storage addr;
+	struct sockaddr_in  *addr4;
+	struct sockaddr_in6 *addr6;
+	size_t i;
+	int opt;
+
+	switch (family) {
+	case AF_INET:
+		addr4 = (struct sockaddr_in *)&addr;
+		addr4->sin_family = AF_INET;
+		addr4->sin_addr.s_addr = htonl(INADDR_ANY);
+		addr4->sin_port = htons(PORT);
+		break;
+	case AF_INET6:
+		addr6 = (struct sockaddr_in6 *)&addr;
+		addr6->sin6_family = AF_INET6;
+		addr6->sin6_addr = in6addr_any;
+		addr6->sin6_port = htons(PORT);
+		break;
+	default:
+		error(1, 0, "Unsupported family %d", family);
+	}
+
+	for (i = 0; i < len; ++i) {
+		rcv_fd[i] = socket(family, proto, 0);
+		if (rcv_fd[i] < 0)
+			error(1, errno, "failed to create receive socket");
+
+		opt = 1;
+		if (setsockopt(rcv_fd[i], SOL_SOCKET, SO_REUSEPORT, &opt,
+			       sizeof(opt)))
+			error(1, errno, "failed to set SO_REUSEPORT");
+
+		if (bind(rcv_fd[i], (struct sockaddr *)&addr, sizeof(addr)))
+			error(1, errno, "failed to bind receive socket");
+
+		if (proto == SOCK_STREAM && listen(rcv_fd[i], len * 10))
+			error(1, errno, "failed to listen on receive port");
+	}
+}
+
+static void attach_bpf(int fd)
+{
+	struct sock_filter code[] = {
+		/* A = raw_smp_processor_id() */
+		{ BPF_LD  | BPF_W | BPF_ABS, 0, 0, SKF_AD_OFF + SKF_AD_CPU },
+		/* return A */
+		{ BPF_RET | BPF_A, 0, 0, 0 },
+	};
+	struct sock_fprog p = {
+		.len = 2,
+		.filter = code,
+	};
+
+	if (setsockopt(fd, SOL_SOCKET, SO_ATTACH_REUSEPORT_CBPF, &p, sizeof(p)))
+		error(1, errno, "failed to set SO_ATTACH_REUSEPORT_CBPF");
+}
+
+static void send_from_cpu(int cpu_id, int family, int proto)
+{
+	struct sockaddr_storage saddr, daddr;
+	struct sockaddr_in  *saddr4, *daddr4;
+	struct sockaddr_in6 *saddr6, *daddr6;
+	cpu_set_t cpu_set;
+	int fd;
+
+	switch (family) {
+	case AF_INET:
+		saddr4 = (struct sockaddr_in *)&saddr;
+		saddr4->sin_family = AF_INET;
+		saddr4->sin_addr.s_addr = htonl(INADDR_ANY);
+		saddr4->sin_port = 0;
+
+		daddr4 = (struct sockaddr_in *)&daddr;
+		daddr4->sin_family = AF_INET;
+		daddr4->sin_addr.s_addr = htonl(INADDR_LOOPBACK);
+		daddr4->sin_port = htons(PORT);
+		break;
+	case AF_INET6:
+		saddr6 = (struct sockaddr_in6 *)&saddr;
+		saddr6->sin6_family = AF_INET6;
+		saddr6->sin6_addr = in6addr_any;
+		saddr6->sin6_port = 0;
+
+		daddr6 = (struct sockaddr_in6 *)&daddr;
+		daddr6->sin6_family = AF_INET6;
+		daddr6->sin6_addr = in6addr_loopback;
+		daddr6->sin6_port = htons(PORT);
+		break;
+	default:
+		error(1, 0, "Unsupported family %d", family);
+	}
+
+	memset(&cpu_set, 0, sizeof(cpu_set));
+	CPU_SET(cpu_id, &cpu_set);
+	if (sched_setaffinity(0, sizeof(cpu_set), &cpu_set) < 0)
+		error(1, errno, "failed to pin to cpu");
+
+	fd = socket(family, proto, 0);
+	if (fd < 0)
+		error(1, errno, "failed to create send socket");
+
+	if (bind(fd, (struct sockaddr *)&saddr, sizeof(saddr)))
+		error(1, errno, "failed to bind send socket");
+
+	if (connect(fd, (struct sockaddr *)&daddr, sizeof(daddr)))
+		error(1, errno, "failed to connect send socket");
+
+	if (send(fd, "a", 1, 0) < 0)
+		error(1, errno, "failed to send message");
+
+	close(fd);
+}
+
+static
+void receive_on_cpu(int *rcv_fd, int len, int epfd, int cpu_id, int proto)
+{
+	struct epoll_event ev;
+	int i, fd;
+	char buf[8];
+
+	i = epoll_wait(epfd, &ev, 1, -1);
+	if (i < 0)
+		error(1, errno, "epoll_wait failed");
+
+	if (proto == SOCK_STREAM) {
+		fd = accept(ev.data.fd, NULL, NULL);
+		if (fd < 0)
+			error(1, errno, "failed to accept");
+		i = recv(fd, buf, sizeof(buf), 0);
+		close(fd);
+	} else {
+		i = recv(ev.data.fd, buf, sizeof(buf), 0);
+	}
+
+	if (i < 0)
+		error(1, errno, "failed to recv");
+
+	for (i = 0; i < len; ++i)
+		if (ev.data.fd == rcv_fd[i])
+			break;
+	if (i == len)
+		error(1, 0, "failed to find socket");
+	fprintf(stderr, "send cpu %d, receive socket %d\n", cpu_id, i);
+	if (cpu_id != i)
+		error(1, 0, "cpu id/receive socket mismatch");
+}
+
+static void test(int *rcv_fd, int len, int family, int proto)
+{
+	struct epoll_event ev;
+	int epfd, cpu;
+
+	build_rcv_group(rcv_fd, len, family, proto);
+	attach_bpf(rcv_fd[0]);
+
+	epfd = epoll_create(1);
+	if (epfd < 0)
+		error(1, errno, "failed to create epoll");
+	for (cpu = 0; cpu < len; ++cpu) {
+		ev.events = EPOLLIN;
+		ev.data.fd = rcv_fd[cpu];
+		if (epoll_ctl(epfd, EPOLL_CTL_ADD, rcv_fd[cpu], &ev))
+			error(1, errno, "failed to register sock epoll");
+	}
+
+	/* Forward iterate */
+	for (cpu = 0; cpu < len; ++cpu) {
+		send_from_cpu(cpu, family, proto);
+		receive_on_cpu(rcv_fd, len, epfd, cpu, proto);
+	}
+
+	/* Reverse iterate */
+	for (cpu = len - 1; cpu >= 0; --cpu) {
+		send_from_cpu(cpu, family, proto);
+		receive_on_cpu(rcv_fd, len, epfd, cpu, proto);
+	}
+
+	/* Even cores */
+	for (cpu = 0; cpu < len; cpu += 2) {
+		send_from_cpu(cpu, family, proto);
+		receive_on_cpu(rcv_fd, len, epfd, cpu, proto);
+	}
+
+	/* Odd cores */
+	for (cpu = 1; cpu < len; cpu += 2) {
+		send_from_cpu(cpu, family, proto);
+		receive_on_cpu(rcv_fd, len, epfd, cpu, proto);
+	}
+
+	close(epfd);
+	for (cpu = 0; cpu < len; ++cpu)
+		close(rcv_fd[cpu]);
+}
+
+int main(void)
+{
+	int *rcv_fd, cpus;
+
+	cpus = sysconf(_SC_NPROCESSORS_ONLN);
+	if (cpus <= 0)
+		error(1, errno, "failed counting cpus");
+
+	rcv_fd = calloc(cpus, sizeof(int));
+	if (!rcv_fd)
+		error(1, 0, "failed to allocate array");
+
+	fprintf(stderr, "---- IPv4 UDP ----\n");
+	test(rcv_fd, cpus, AF_INET, SOCK_DGRAM);
+
+	fprintf(stderr, "---- IPv6 UDP ----\n");
+	test(rcv_fd, cpus, AF_INET6, SOCK_DGRAM);
+
+	fprintf(stderr, "---- IPv4 TCP ----\n");
+	test(rcv_fd, cpus, AF_INET, SOCK_STREAM);
+
+	fprintf(stderr, "---- IPv6 TCP ----\n");
+	test(rcv_fd, cpus, AF_INET6, SOCK_STREAM);
+
+	free(rcv_fd);
+
+	fprintf(stderr, "SUCCESS\n");
+	return 0;
+}
diff --git a/tools/testing/selftests/powerpc/Makefile b/tools/testing/selftests/powerpc/Makefile
index 0c2706bda330..b08f77cbe31b 100644
--- a/tools/testing/selftests/powerpc/Makefile
+++ b/tools/testing/selftests/powerpc/Makefile
@@ -8,7 +8,7 @@ ifeq ($(ARCH),powerpc)
 
 GIT_VERSION = $(shell git describe --always --long --dirty || echo "unknown")
 
-CFLAGS := -Wall -O2 -flto -Wall -Werror -DGIT_VERSION='"$(GIT_VERSION)"' -I$(CURDIR) $(CFLAGS)
+CFLAGS := -Wall -O2 -Wall -Werror -DGIT_VERSION='"$(GIT_VERSION)"' -I$(CURDIR) $(CFLAGS)
 
 export CFLAGS
 
@@ -22,7 +22,8 @@ SUB_DIRS = benchmarks 		\
 	   switch_endian	\
 	   syscalls		\
 	   tm			\
-	   vphn
+	   vphn         \
+	   math
 
 endif
 
diff --git a/tools/testing/selftests/powerpc/basic_asm.h b/tools/testing/selftests/powerpc/basic_asm.h
new file mode 100644
index 000000000000..3349a0704d1a
--- /dev/null
+++ b/tools/testing/selftests/powerpc/basic_asm.h
@@ -0,0 +1,70 @@
+#ifndef _SELFTESTS_POWERPC_BASIC_ASM_H
+#define _SELFTESTS_POWERPC_BASIC_ASM_H
+
+#include <ppc-asm.h>
+#include <asm/unistd.h>
+
+#define LOAD_REG_IMMEDIATE(reg,expr) \
+	lis	reg,(expr)@highest;	\
+	ori	reg,reg,(expr)@higher;	\
+	rldicr	reg,reg,32,31;	\
+	oris	reg,reg,(expr)@high;	\
+	ori	reg,reg,(expr)@l;
+
+/*
+ * Note: These macros assume that variables being stored on the stack are
+ * doublewords, while this is usually the case it may not always be the
+ * case for each use case.
+ */
+#if defined(_CALL_ELF) && _CALL_ELF == 2
+#define STACK_FRAME_MIN_SIZE 32
+#define STACK_FRAME_TOC_POS  24
+#define __STACK_FRAME_PARAM(_param)  (32 + ((_param)*8))
+#define __STACK_FRAME_LOCAL(_num_params,_var_num)  ((STACK_FRAME_PARAM(_num_params)) + ((_var_num)*8))
+#else
+#define STACK_FRAME_MIN_SIZE 112
+#define STACK_FRAME_TOC_POS  40
+#define __STACK_FRAME_PARAM(i)  (48 + ((i)*8))
+
+/*
+ * Caveat: if a function passed more than 8 doublewords, the caller will have
+ * made more space... which would render the 112 incorrect.
+ */
+#define __STACK_FRAME_LOCAL(_num_params,_var_num)  (112 + ((_var_num)*8))
+#endif
+
+/* Parameter x saved to the stack */
+#define STACK_FRAME_PARAM(var)    __STACK_FRAME_PARAM(var)
+
+/* Local variable x saved to the stack after x parameters */
+#define STACK_FRAME_LOCAL(num_params,var)    __STACK_FRAME_LOCAL(num_params,var)
+#define STACK_FRAME_LR_POS   16
+#define STACK_FRAME_CR_POS   8
+
+/*
+ * It is very important to note here that _extra is the extra amount of
+ * stack space needed. This space can be accessed using STACK_FRAME_PARAM()
+ * or STACK_FRAME_LOCAL() macros.
+ *
+ * r1 and r2 are not defined in ppc-asm.h (instead they are defined as sp
+ * and toc). Kernel programmers tend to prefer rX even for r1 and r2, hence
+ * %1 and %r2. r0 is defined in ppc-asm.h and therefore %r0 gets
+ * preprocessed incorrectly, hence r0.
+ */
+#define PUSH_BASIC_STACK(_extra) \
+	mflr	r0; \
+	std	r0,STACK_FRAME_LR_POS(%r1); \
+	stdu	%r1,-(_extra + STACK_FRAME_MIN_SIZE)(%r1); \
+	mfcr	r0; \
+	stw	r0,STACK_FRAME_CR_POS(%r1); \
+	std	%r2,STACK_FRAME_TOC_POS(%r1);
+
+#define POP_BASIC_STACK(_extra) \
+	ld	%r2,STACK_FRAME_TOC_POS(%r1); \
+	lwz	r0,STACK_FRAME_CR_POS(%r1); \
+	mtcr	r0; \
+	addi	%r1,%r1,(_extra + STACK_FRAME_MIN_SIZE); \
+	ld	r0,STACK_FRAME_LR_POS(%r1); \
+	mtlr	r0;
+
+#endif /* _SELFTESTS_POWERPC_BASIC_ASM_H */
diff --git a/tools/testing/selftests/powerpc/math/.gitignore b/tools/testing/selftests/powerpc/math/.gitignore
new file mode 100644
index 000000000000..4fe13a439fd7
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/.gitignore
@@ -0,0 +1,6 @@
+fpu_syscall
+vmx_syscall
+fpu_preempt
+vmx_preempt
+fpu_signal
+vmx_signal
diff --git a/tools/testing/selftests/powerpc/math/Makefile b/tools/testing/selftests/powerpc/math/Makefile
new file mode 100644
index 000000000000..5b88875d5955
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/Makefile
@@ -0,0 +1,19 @@
+TEST_PROGS := fpu_syscall fpu_preempt fpu_signal vmx_syscall vmx_preempt vmx_signal
+
+all: $(TEST_PROGS)
+
+$(TEST_PROGS): ../harness.c
+$(TEST_PROGS): CFLAGS += -O2 -g -pthread -m64 -maltivec
+
+fpu_syscall: fpu_asm.S
+fpu_preempt: fpu_asm.S
+fpu_signal:  fpu_asm.S
+
+vmx_syscall: vmx_asm.S
+vmx_preempt: vmx_asm.S
+vmx_signal: vmx_asm.S
+
+include ../../lib.mk
+
+clean:
+	rm -f $(TEST_PROGS) *.o
diff --git a/tools/testing/selftests/powerpc/math/fpu_asm.S b/tools/testing/selftests/powerpc/math/fpu_asm.S
new file mode 100644
index 000000000000..f3711d80e709
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/fpu_asm.S
@@ -0,0 +1,198 @@
+/*
+ * Copyright 2015, Cyril Bur, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include "../basic_asm.h"
+
+#define PUSH_FPU(pos) \
+	stfd	f14,pos(sp); \
+	stfd	f15,pos+8(sp); \
+	stfd	f16,pos+16(sp); \
+	stfd	f17,pos+24(sp); \
+	stfd	f18,pos+32(sp); \
+	stfd	f19,pos+40(sp); \
+	stfd	f20,pos+48(sp); \
+	stfd	f21,pos+56(sp); \
+	stfd	f22,pos+64(sp); \
+	stfd	f23,pos+72(sp); \
+	stfd	f24,pos+80(sp); \
+	stfd	f25,pos+88(sp); \
+	stfd	f26,pos+96(sp); \
+	stfd	f27,pos+104(sp); \
+	stfd	f28,pos+112(sp); \
+	stfd	f29,pos+120(sp); \
+	stfd	f30,pos+128(sp); \
+	stfd	f31,pos+136(sp);
+
+#define POP_FPU(pos) \
+	lfd	f14,pos(sp); \
+	lfd	f15,pos+8(sp); \
+	lfd	f16,pos+16(sp); \
+	lfd	f17,pos+24(sp); \
+	lfd	f18,pos+32(sp); \
+	lfd	f19,pos+40(sp); \
+	lfd	f20,pos+48(sp); \
+	lfd	f21,pos+56(sp); \
+	lfd	f22,pos+64(sp); \
+	lfd	f23,pos+72(sp); \
+	lfd	f24,pos+80(sp); \
+	lfd	f25,pos+88(sp); \
+	lfd	f26,pos+96(sp); \
+	lfd	f27,pos+104(sp); \
+	lfd	f28,pos+112(sp); \
+	lfd	f29,pos+120(sp); \
+	lfd	f30,pos+128(sp); \
+	lfd	f31,pos+136(sp);
+
+# Careful calling this, it will 'clobber' fpu (by design)
+# Don't call this from C
+FUNC_START(load_fpu)
+	lfd	f14,0(r3)
+	lfd	f15,8(r3)
+	lfd	f16,16(r3)
+	lfd	f17,24(r3)
+	lfd	f18,32(r3)
+	lfd	f19,40(r3)
+	lfd	f20,48(r3)
+	lfd	f21,56(r3)
+	lfd	f22,64(r3)
+	lfd	f23,72(r3)
+	lfd	f24,80(r3)
+	lfd	f25,88(r3)
+	lfd	f26,96(r3)
+	lfd	f27,104(r3)
+	lfd	f28,112(r3)
+	lfd	f29,120(r3)
+	lfd	f30,128(r3)
+	lfd	f31,136(r3)
+	blr
+FUNC_END(load_fpu)
+
+FUNC_START(check_fpu)
+	mr r4,r3
+	li	r3,1 # assume a bad result
+	lfd	f0,0(r4)
+	fcmpu	cr1,f0,f14
+	bne	cr1,1f
+	lfd	f0,8(r4)
+	fcmpu	cr1,f0,f15
+	bne	cr1,1f
+	lfd	f0,16(r4)
+	fcmpu	cr1,f0,f16
+	bne	cr1,1f
+	lfd	f0,24(r4)
+	fcmpu	cr1,f0,f17
+	bne	cr1,1f
+	lfd	f0,32(r4)
+	fcmpu	cr1,f0,f18
+	bne	cr1,1f
+	lfd	f0,40(r4)
+	fcmpu	cr1,f0,f19
+	bne	cr1,1f
+	lfd	f0,48(r4)
+	fcmpu	cr1,f0,f20
+	bne	cr1,1f
+	lfd	f0,56(r4)
+	fcmpu	cr1,f0,f21
+	bne	cr1,1f
+	lfd	f0,64(r4)
+	fcmpu	cr1,f0,f22
+	bne	cr1,1f
+	lfd	f0,72(r4)
+	fcmpu	cr1,f0,f23
+	bne	cr1,1f
+	lfd	f0,80(r4)
+	fcmpu	cr1,f0,f24
+	bne	cr1,1f
+	lfd	f0,88(r4)
+	fcmpu	cr1,f0,f25
+	bne	cr1,1f
+	lfd	f0,96(r4)
+	fcmpu	cr1,f0,f26
+	bne	cr1,1f
+	lfd	f0,104(r4)
+	fcmpu	cr1,f0,f27
+	bne	cr1,1f
+	lfd	f0,112(r4)
+	fcmpu	cr1,f0,f28
+	bne	cr1,1f
+	lfd	f0,120(r4)
+	fcmpu	cr1,f0,f29
+	bne	cr1,1f
+	lfd	f0,128(r4)
+	fcmpu	cr1,f0,f30
+	bne	cr1,1f
+	lfd	f0,136(r4)
+	fcmpu	cr1,f0,f31
+	bne	cr1,1f
+	li	r3,0 # Success!!!
+1:	blr
+
+FUNC_START(test_fpu)
+	# r3 holds pointer to where to put the result of fork
+	# r4 holds pointer to the pid
+	# f14-f31 are non volatiles
+	PUSH_BASIC_STACK(256)
+	std	r3,STACK_FRAME_PARAM(0)(sp) # Address of darray
+	std r4,STACK_FRAME_PARAM(1)(sp) # Address of pid
+	PUSH_FPU(STACK_FRAME_LOCAL(2,0))
+
+	bl load_fpu
+	nop
+	li	r0,__NR_fork
+	sc
+
+	# pass the result of the fork to the caller
+	ld	r9,STACK_FRAME_PARAM(1)(sp)
+	std	r3,0(r9)
+
+	ld r3,STACK_FRAME_PARAM(0)(sp)
+	bl check_fpu
+	nop
+
+	POP_FPU(STACK_FRAME_LOCAL(2,0))
+	POP_BASIC_STACK(256)
+	blr
+FUNC_END(test_fpu)
+
+# int preempt_fpu(double *darray, int *threads_running, int *running)
+# On starting will (atomically) decrement not_ready as a signal that the FPU
+# has been loaded with darray. Will proceed to check the validity of the FPU
+# registers while running is not zero.
+FUNC_START(preempt_fpu)
+	PUSH_BASIC_STACK(256)
+	std r3,STACK_FRAME_PARAM(0)(sp) # double *darray
+	std r4,STACK_FRAME_PARAM(1)(sp) # int *threads_starting
+	std r5,STACK_FRAME_PARAM(2)(sp) # int *running
+	PUSH_FPU(STACK_FRAME_LOCAL(3,0))
+
+	bl load_fpu
+	nop
+
+	sync
+	# Atomic DEC
+	ld r3,STACK_FRAME_PARAM(1)(sp)
+1:	lwarx r4,0,r3
+	addi r4,r4,-1
+	stwcx. r4,0,r3
+	bne- 1b
+
+2:	ld r3,STACK_FRAME_PARAM(0)(sp)
+	bl check_fpu
+	nop
+	cmpdi r3,0
+	bne 3f
+	ld r4,STACK_FRAME_PARAM(2)(sp)
+	ld r5,0(r4)
+	cmpwi r5,0
+	bne 2b
+
+3:	POP_FPU(STACK_FRAME_LOCAL(3,0))
+	POP_BASIC_STACK(256)
+	blr
+FUNC_END(preempt_fpu)
diff --git a/tools/testing/selftests/powerpc/math/fpu_preempt.c b/tools/testing/selftests/powerpc/math/fpu_preempt.c
new file mode 100644
index 000000000000..0f85b79d883d
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/fpu_preempt.c
@@ -0,0 +1,113 @@
+/*
+ * Copyright 2015, Cyril Bur, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This test attempts to see if the FPU registers change across preemption.
+ * Two things should be noted here a) The check_fpu function in asm only checks
+ * the non volatile registers as it is reused from the syscall test b) There is
+ * no way to be sure preemption happened so this test just uses many threads
+ * and a long wait. As such, a successful test doesn't mean much but a failure
+ * is bad.
+ */
+
+#include <stdio.h>
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <sys/time.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <stdlib.h>
+#include <pthread.h>
+
+#include "utils.h"
+
+/* Time to wait for workers to get preempted (seconds) */
+#define PREEMPT_TIME 20
+/*
+ * Factor by which to multiply number of online CPUs for total number of
+ * worker threads
+ */
+#define THREAD_FACTOR 8
+
+
+__thread double darray[] = {0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0,
+		     1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0,
+		     2.1};
+
+int threads_starting;
+int running;
+
+extern void preempt_fpu(double *darray, int *threads_starting, int *running);
+
+void *preempt_fpu_c(void *p)
+{
+	int i;
+	srand(pthread_self());
+	for (i = 0; i < 21; i++)
+		darray[i] = rand();
+
+	/* Test failed if it ever returns */
+	preempt_fpu(darray, &threads_starting, &running);
+
+	return p;
+}
+
+int test_preempt_fpu(void)
+{
+	int i, rc, threads;
+	pthread_t *tids;
+
+	threads = sysconf(_SC_NPROCESSORS_ONLN) * THREAD_FACTOR;
+	tids = malloc((threads) * sizeof(pthread_t));
+	FAIL_IF(!tids);
+
+	running = true;
+	threads_starting = threads;
+	for (i = 0; i < threads; i++) {
+		rc = pthread_create(&tids[i], NULL, preempt_fpu_c, NULL);
+		FAIL_IF(rc);
+	}
+
+	setbuf(stdout, NULL);
+	/* Not really necessary but nice to wait for every thread to start */
+	printf("\tWaiting for all workers to start...");
+	while(threads_starting)
+		asm volatile("": : :"memory");
+	printf("done\n");
+
+	printf("\tWaiting for %d seconds to let some workers get preempted...", PREEMPT_TIME);
+	sleep(PREEMPT_TIME);
+	printf("done\n");
+
+	printf("\tStopping workers...");
+	/*
+	 * Working are checking this value every loop. In preempt_fpu 'cmpwi r5,0; bne 2b'.
+	 * r5 will have loaded the value of running.
+	 */
+	running = 0;
+	for (i = 0; i < threads; i++) {
+		void *rc_p;
+		pthread_join(tids[i], &rc_p);
+
+		/*
+		 * Harness will say the fail was here, look at why preempt_fpu
+		 * returned
+		 */
+		if ((long) rc_p)
+			printf("oops\n");
+		FAIL_IF((long) rc_p);
+	}
+	printf("done\n");
+
+	free(tids);
+	return 0;
+}
+
+int main(int argc, char *argv[])
+{
+	return test_harness(test_preempt_fpu, "fpu_preempt");
+}
diff --git a/tools/testing/selftests/powerpc/math/fpu_signal.c b/tools/testing/selftests/powerpc/math/fpu_signal.c
new file mode 100644
index 000000000000..888aa51b4204
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/fpu_signal.c
@@ -0,0 +1,135 @@
+/*
+ * Copyright 2015, Cyril Bur, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This test attempts to see if the FPU registers are correctly reported in a
+ * signal context. Each worker just spins checking its FPU registers, at some
+ * point a signal will interrupt it and C code will check the signal context
+ * ensuring it is also the same.
+ */
+
+#include <stdio.h>
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <sys/time.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <stdlib.h>
+#include <pthread.h>
+
+#include "utils.h"
+
+/* Number of times each thread should receive the signal */
+#define ITERATIONS 10
+/*
+ * Factor by which to multiply number of online CPUs for total number of
+ * worker threads
+ */
+#define THREAD_FACTOR 8
+
+__thread double darray[] = {0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0,
+		     1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0,
+		     2.1};
+
+bool bad_context;
+int threads_starting;
+int running;
+
+extern long preempt_fpu(double *darray, int *threads_starting, int *running);
+
+void signal_fpu_sig(int sig, siginfo_t *info, void *context)
+{
+	int i;
+	ucontext_t *uc = context;
+	mcontext_t *mc = &uc->uc_mcontext;
+
+	/* Only the non volatiles were loaded up */
+	for (i = 14; i < 32; i++) {
+		if (mc->fp_regs[i] != darray[i - 14]) {
+			bad_context = true;
+			break;
+		}
+	}
+}
+
+void *signal_fpu_c(void *p)
+{
+	int i;
+	long rc;
+	struct sigaction act;
+	act.sa_sigaction = signal_fpu_sig;
+	act.sa_flags = SA_SIGINFO;
+	rc = sigaction(SIGUSR1, &act, NULL);
+	if (rc)
+		return p;
+
+	srand(pthread_self());
+	for (i = 0; i < 21; i++)
+		darray[i] = rand();
+
+	rc = preempt_fpu(darray, &threads_starting, &running);
+
+	return (void *) rc;
+}
+
+int test_signal_fpu(void)
+{
+	int i, j, rc, threads;
+	void *rc_p;
+	pthread_t *tids;
+
+	threads = sysconf(_SC_NPROCESSORS_ONLN) * THREAD_FACTOR;
+	tids = malloc(threads * sizeof(pthread_t));
+	FAIL_IF(!tids);
+
+	running = true;
+	threads_starting = threads;
+	for (i = 0; i < threads; i++) {
+		rc = pthread_create(&tids[i], NULL, signal_fpu_c, NULL);
+		FAIL_IF(rc);
+	}
+
+	setbuf(stdout, NULL);
+	printf("\tWaiting for all workers to start...");
+	while (threads_starting)
+		asm volatile("": : :"memory");
+	printf("done\n");
+
+	printf("\tSending signals to all threads %d times...", ITERATIONS);
+	for (i = 0; i < ITERATIONS; i++) {
+		for (j = 0; j < threads; j++) {
+			pthread_kill(tids[j], SIGUSR1);
+		}
+		sleep(1);
+	}
+	printf("done\n");
+
+	printf("\tStopping workers...");
+	running = 0;
+	for (i = 0; i < threads; i++) {
+		pthread_join(tids[i], &rc_p);
+
+		/*
+		 * Harness will say the fail was here, look at why signal_fpu
+		 * returned
+		 */
+		if ((long) rc_p || bad_context)
+			printf("oops\n");
+		if (bad_context)
+			fprintf(stderr, "\t!! bad_context is true\n");
+		FAIL_IF((long) rc_p || bad_context);
+	}
+	printf("done\n");
+
+	free(tids);
+	return 0;
+}
+
+int main(int argc, char *argv[])
+{
+	return test_harness(test_signal_fpu, "fpu_signal");
+}
diff --git a/tools/testing/selftests/powerpc/math/fpu_syscall.c b/tools/testing/selftests/powerpc/math/fpu_syscall.c
new file mode 100644
index 000000000000..949e6721256d
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/fpu_syscall.c
@@ -0,0 +1,90 @@
+/*
+ * Copyright 2015, Cyril Bur, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This test attempts to see if the FPU registers change across a syscall (fork).
+ */
+
+#include <stdio.h>
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <sys/time.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <stdlib.h>
+
+#include "utils.h"
+
+extern int test_fpu(double *darray, pid_t *pid);
+
+double darray[] = {0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8, 0.9, 1.0,
+		     1.1, 1.2, 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0,
+		     2.1};
+
+int syscall_fpu(void)
+{
+	pid_t fork_pid;
+	int i;
+	int ret;
+	int child_ret;
+	for (i = 0; i < 1000; i++) {
+		/* test_fpu will fork() */
+		ret = test_fpu(darray, &fork_pid);
+		if (fork_pid == -1)
+			return -1;
+		if (fork_pid == 0)
+			exit(ret);
+		waitpid(fork_pid, &child_ret, 0);
+		if (ret || child_ret)
+			return 1;
+	}
+
+	return 0;
+}
+
+int test_syscall_fpu(void)
+{
+	/*
+	 * Setup an environment with much context switching
+	 */
+	pid_t pid2;
+	pid_t pid = fork();
+	int ret;
+	int child_ret;
+	FAIL_IF(pid == -1);
+
+	pid2 = fork();
+	/* Can't FAIL_IF(pid2 == -1); because already forked once */
+	if (pid2 == -1) {
+		/*
+		 * Couldn't fork, ensure test is a fail
+		 */
+		child_ret = ret = 1;
+	} else {
+		ret = syscall_fpu();
+		if (pid2)
+			waitpid(pid2, &child_ret, 0);
+		else
+			exit(ret);
+	}
+
+	ret |= child_ret;
+
+	if (pid)
+		waitpid(pid, &child_ret, 0);
+	else
+		exit(ret);
+
+	FAIL_IF(ret || child_ret);
+	return 0;
+}
+
+int main(int argc, char *argv[])
+{
+	return test_harness(test_syscall_fpu, "syscall_fpu");
+
+}
diff --git a/tools/testing/selftests/powerpc/math/vmx_asm.S b/tools/testing/selftests/powerpc/math/vmx_asm.S
new file mode 100644
index 000000000000..1b8c248b3ac1
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/vmx_asm.S
@@ -0,0 +1,235 @@
+/*
+ * Copyright 2015, Cyril Bur, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include "../basic_asm.h"
+
+# POS MUST BE 16 ALIGNED!
+#define PUSH_VMX(pos,reg) \
+	li	reg,pos; \
+	stvx	v20,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v21,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v22,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v23,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v24,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v25,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v26,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v27,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v28,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v29,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v30,reg,sp; \
+	addi	reg,reg,16; \
+	stvx	v31,reg,sp;
+
+# POS MUST BE 16 ALIGNED!
+#define POP_VMX(pos,reg) \
+	li	reg,pos; \
+	lvx	v20,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v21,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v22,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v23,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v24,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v25,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v26,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v27,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v28,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v29,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v30,reg,sp; \
+	addi	reg,reg,16; \
+	lvx	v31,reg,sp;
+
+# Carefull this will 'clobber' vmx (by design)
+# Don't call this from C
+FUNC_START(load_vmx)
+	li	r5,0
+	lvx	v20,r5,r3
+	addi	r5,r5,16
+	lvx	v21,r5,r3
+	addi	r5,r5,16
+	lvx	v22,r5,r3
+	addi	r5,r5,16
+	lvx	v23,r5,r3
+	addi	r5,r5,16
+	lvx	v24,r5,r3
+	addi	r5,r5,16
+	lvx	v25,r5,r3
+	addi	r5,r5,16
+	lvx	v26,r5,r3
+	addi	r5,r5,16
+	lvx	v27,r5,r3
+	addi	r5,r5,16
+	lvx	v28,r5,r3
+	addi	r5,r5,16
+	lvx	v29,r5,r3
+	addi	r5,r5,16
+	lvx	v30,r5,r3
+	addi	r5,r5,16
+	lvx	v31,r5,r3
+	blr
+FUNC_END(load_vmx)
+
+# Should be safe from C, only touches r4, r5 and v0,v1,v2
+FUNC_START(check_vmx)
+	PUSH_BASIC_STACK(32)
+	mr r4,r3
+	li	r3,1 # assume a bad result
+	li	r5,0
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v20
+	vmr	v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v21
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v22
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v23
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v24
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v25
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v26
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v27
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v28
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v29
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v30
+	vand	v2,v2,v1
+
+	addi	r5,r5,16
+	lvx	v0,r5,r4
+	vcmpequd.	v1,v0,v31
+	vand	v2,v2,v1
+
+	li	r5,STACK_FRAME_LOCAL(0,0)
+	stvx	v2,r5,sp
+	ldx	r0,r5,sp
+	cmpdi	r0,0xffffffffffffffff
+	bne	1f
+	li	r3,0
+1:	POP_BASIC_STACK(32)
+	blr
+FUNC_END(check_vmx)
+
+# Safe from C
+FUNC_START(test_vmx)
+	# r3 holds pointer to where to put the result of fork
+	# r4 holds pointer to the pid
+	# v20-v31 are non-volatile
+	PUSH_BASIC_STACK(512)
+	std	r3,STACK_FRAME_PARAM(0)(sp) # Address of varray
+	std r4,STACK_FRAME_PARAM(1)(sp) # address of pid
+	PUSH_VMX(STACK_FRAME_LOCAL(2,0),r4)
+
+	bl load_vmx
+	nop
+
+	li	r0,__NR_fork
+	sc
+	# Pass the result of fork back to the caller
+	ld	r9,STACK_FRAME_PARAM(1)(sp)
+	std	r3,0(r9)
+
+	ld r3,STACK_FRAME_PARAM(0)(sp)
+	bl check_vmx
+	nop
+
+	POP_VMX(STACK_FRAME_LOCAL(2,0),r4)
+	POP_BASIC_STACK(512)
+	blr
+FUNC_END(test_vmx)
+
+# int preempt_vmx(vector int *varray, int *threads_starting, int *running)
+# On starting will (atomically) decrement threads_starting as a signal that
+# the VMX have been loaded with varray. Will proceed to check the validity of
+# the VMX registers while running is not zero.
+FUNC_START(preempt_vmx)
+	PUSH_BASIC_STACK(512)
+	std r3,STACK_FRAME_PARAM(0)(sp) # vector int *varray
+	std r4,STACK_FRAME_PARAM(1)(sp) # int *threads_starting
+	std r5,STACK_FRAME_PARAM(2)(sp) # int *running
+	# VMX need to write to 16 byte aligned addresses, skip STACK_FRAME_LOCAL(3,0)
+	PUSH_VMX(STACK_FRAME_LOCAL(4,0),r4)
+
+	bl load_vmx
+	nop
+
+	sync
+	# Atomic DEC
+	ld r3,STACK_FRAME_PARAM(1)(sp)
+1:	lwarx r4,0,r3
+	addi r4,r4,-1
+	stwcx. r4,0,r3
+	bne- 1b
+
+2:	ld r3,STACK_FRAME_PARAM(0)(sp)
+	bl check_vmx
+	nop
+	cmpdi r3,0
+	bne 3f
+	ld r4,STACK_FRAME_PARAM(2)(sp)
+	ld r5,0(r4)
+	cmpwi r5,0
+	bne 2b
+
+3:	POP_VMX(STACK_FRAME_LOCAL(4,0),r4)
+	POP_BASIC_STACK(512)
+	blr
+FUNC_END(preempt_vmx)
diff --git a/tools/testing/selftests/powerpc/math/vmx_preempt.c b/tools/testing/selftests/powerpc/math/vmx_preempt.c
new file mode 100644
index 000000000000..9ef376c55b13
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/vmx_preempt.c
@@ -0,0 +1,112 @@
+/*
+ * Copyright 2015, Cyril Bur, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This test attempts to see if the VMX registers change across preemption.
+ * Two things should be noted here a) The check_vmx function in asm only checks
+ * the non volatile registers as it is reused from the syscall test b) There is
+ * no way to be sure preemption happened so this test just uses many threads
+ * and a long wait. As such, a successful test doesn't mean much but a failure
+ * is bad.
+ */
+
+#include <stdio.h>
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <sys/time.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <stdlib.h>
+#include <pthread.h>
+
+#include "utils.h"
+
+/* Time to wait for workers to get preempted (seconds) */
+#define PREEMPT_TIME 20
+/*
+ * Factor by which to multiply number of online CPUs for total number of
+ * worker threads
+ */
+#define THREAD_FACTOR 8
+
+__thread vector int varray[] = {{1, 2, 3, 4}, {5, 6, 7, 8}, {9, 10,11,12},
+	{13,14,15,16},{17,18,19,20},{21,22,23,24},
+	{25,26,27,28},{29,30,31,32},{33,34,35,36},
+	{37,38,39,40},{41,42,43,44},{45,46,47,48}};
+
+int threads_starting;
+int running;
+
+extern void preempt_vmx(vector int *varray, int *threads_starting, int *running);
+
+void *preempt_vmx_c(void *p)
+{
+	int i, j;
+	srand(pthread_self());
+	for (i = 0; i < 12; i++)
+		for (j = 0; j < 4; j++)
+			varray[i][j] = rand();
+
+	/* Test fails if it ever returns */
+	preempt_vmx(varray, &threads_starting, &running);
+	return p;
+}
+
+int test_preempt_vmx(void)
+{
+	int i, rc, threads;
+	pthread_t *tids;
+
+	threads = sysconf(_SC_NPROCESSORS_ONLN) * THREAD_FACTOR;
+	tids = malloc(threads * sizeof(pthread_t));
+	FAIL_IF(!tids);
+
+	running = true;
+	threads_starting = threads;
+	for (i = 0; i < threads; i++) {
+		rc = pthread_create(&tids[i], NULL, preempt_vmx_c, NULL);
+		FAIL_IF(rc);
+	}
+
+	setbuf(stdout, NULL);
+	/* Not really nessesary but nice to wait for every thread to start */
+	printf("\tWaiting for all workers to start...");
+	while(threads_starting)
+		asm volatile("": : :"memory");
+	printf("done\n");
+
+	printf("\tWaiting for %d seconds to let some workers get preempted...", PREEMPT_TIME);
+	sleep(PREEMPT_TIME);
+	printf("done\n");
+
+	printf("\tStopping workers...");
+	/*
+	 * Working are checking this value every loop. In preempt_vmx 'cmpwi r5,0; bne 2b'.
+	 * r5 will have loaded the value of running.
+	 */
+	running = 0;
+	for (i = 0; i < threads; i++) {
+		void *rc_p;
+		pthread_join(tids[i], &rc_p);
+
+		/*
+		 * Harness will say the fail was here, look at why preempt_vmx
+		 * returned
+		 */
+		if ((long) rc_p)
+			printf("oops\n");
+		FAIL_IF((long) rc_p);
+	}
+	printf("done\n");
+
+	return 0;
+}
+
+int main(int argc, char *argv[])
+{
+	return test_harness(test_preempt_vmx, "vmx_preempt");
+}
diff --git a/tools/testing/selftests/powerpc/math/vmx_signal.c b/tools/testing/selftests/powerpc/math/vmx_signal.c
new file mode 100644
index 000000000000..671d7533a557
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/vmx_signal.c
@@ -0,0 +1,156 @@
+/*
+ * Copyright 2015, Cyril Bur, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This test attempts to see if the VMX registers are correctly reported in a
+ * signal context. Each worker just spins checking its VMX registers, at some
+ * point a signal will interrupt it and C code will check the signal context
+ * ensuring it is also the same.
+ */
+
+#include <stdio.h>
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <sys/time.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <stdlib.h>
+#include <string.h>
+#include <pthread.h>
+#include <altivec.h>
+
+#include "utils.h"
+
+/* Number of times each thread should receive the signal */
+#define ITERATIONS 10
+/*
+ * Factor by which to multiply number of online CPUs for total number of
+ * worker threads
+ */
+#define THREAD_FACTOR 8
+
+__thread vector int varray[] = {{1, 2, 3, 4}, {5, 6, 7, 8}, {9, 10,11,12},
+	{13,14,15,16},{17,18,19,20},{21,22,23,24},
+	{25,26,27,28},{29,30,31,32},{33,34,35,36},
+	{37,38,39,40},{41,42,43,44},{45,46,47,48}};
+
+bool bad_context;
+int running;
+int threads_starting;
+
+extern int preempt_vmx(vector int *varray, int *threads_starting, int *sentinal);
+
+void signal_vmx_sig(int sig, siginfo_t *info, void *context)
+{
+	int i;
+	ucontext_t *uc = context;
+	mcontext_t *mc = &uc->uc_mcontext;
+
+	/* Only the non volatiles were loaded up */
+	for (i = 20; i < 32; i++) {
+		if (memcmp(mc->v_regs->vrregs[i], &varray[i - 20], 16)) {
+			int j;
+			/*
+			 * Shouldn't printf() in a signal handler, however, this is a
+			 * test and we've detected failure. Understanding what failed
+			 * is paramount. All that happens after this is tests exit with
+			 * failure.
+			 */
+			printf("VMX mismatch at reg %d!\n", i);
+			printf("Reg | Actual                  | Expected\n");
+			for (j = 20; j < 32; j++) {
+				printf("%d  | 0x%04x%04x%04x%04x      | 0x%04x%04x%04x%04x\n", j, mc->v_regs->vrregs[j][0],
+					   mc->v_regs->vrregs[j][1], mc->v_regs->vrregs[j][2], mc->v_regs->vrregs[j][3],
+					   varray[j - 20][0], varray[j - 20][1], varray[j - 20][2], varray[j - 20][3]);
+			}
+			bad_context = true;
+			break;
+		}
+	}
+}
+
+void *signal_vmx_c(void *p)
+{
+	int i, j;
+	long rc;
+	struct sigaction act;
+	act.sa_sigaction = signal_vmx_sig;
+	act.sa_flags = SA_SIGINFO;
+	rc = sigaction(SIGUSR1, &act, NULL);
+	if (rc)
+		return p;
+
+	srand(pthread_self());
+	for (i = 0; i < 12; i++)
+		for (j = 0; j < 4; j++)
+			varray[i][j] = rand();
+
+	rc = preempt_vmx(varray, &threads_starting, &running);
+
+	return (void *) rc;
+}
+
+int test_signal_vmx(void)
+{
+	int i, j, rc, threads;
+	void *rc_p;
+	pthread_t *tids;
+
+	threads = sysconf(_SC_NPROCESSORS_ONLN) * THREAD_FACTOR;
+	tids = malloc(threads * sizeof(pthread_t));
+	FAIL_IF(!tids);
+
+	running = true;
+	threads_starting = threads;
+	for (i = 0; i < threads; i++) {
+		rc = pthread_create(&tids[i], NULL, signal_vmx_c, NULL);
+		FAIL_IF(rc);
+	}
+
+	setbuf(stdout, NULL);
+	printf("\tWaiting for %d workers to start... %d", threads, threads_starting);
+	while (threads_starting) {
+		asm volatile("": : :"memory");
+		usleep(1000);
+		printf(", %d", threads_starting);
+	}
+	printf(" ...done\n");
+
+	printf("\tSending signals to all threads %d times...", ITERATIONS);
+	for (i = 0; i < ITERATIONS; i++) {
+		for (j = 0; j < threads; j++) {
+			pthread_kill(tids[j], SIGUSR1);
+		}
+		sleep(1);
+	}
+	printf("done\n");
+
+	printf("\tKilling workers...");
+	running = 0;
+	for (i = 0; i < threads; i++) {
+		pthread_join(tids[i], &rc_p);
+
+		/*
+		 * Harness will say the fail was here, look at why signal_vmx
+		 * returned
+		 */
+		if ((long) rc_p || bad_context)
+			printf("oops\n");
+		if (bad_context)
+			fprintf(stderr, "\t!! bad_context is true\n");
+		FAIL_IF((long) rc_p || bad_context);
+	}
+	printf("done\n");
+
+	free(tids);
+	return 0;
+}
+
+int main(int argc, char *argv[])
+{
+	return test_harness(test_signal_vmx, "vmx_signal");
+}
diff --git a/tools/testing/selftests/powerpc/math/vmx_syscall.c b/tools/testing/selftests/powerpc/math/vmx_syscall.c
new file mode 100644
index 000000000000..a017918ee1ca
--- /dev/null
+++ b/tools/testing/selftests/powerpc/math/vmx_syscall.c
@@ -0,0 +1,91 @@
+/*
+ * Copyright 2015, Cyril Bur, IBM Corp.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * This test attempts to see if the VMX registers change across a syscall (fork).
+ */
+
+#include <altivec.h>
+#include <stdio.h>
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <sys/time.h>
+#include <stdlib.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include "utils.h"
+
+vector int varray[] = {{1, 2, 3, 4}, {5, 6, 7, 8}, {9, 10,11,12},
+	{13,14,15,16},{17,18,19,20},{21,22,23,24},
+	{25,26,27,28},{29,30,31,32},{33,34,35,36},
+	{37,38,39,40},{41,42,43,44},{45,46,47,48}};
+
+extern int test_vmx(vector int *varray, pid_t *pid);
+
+int vmx_syscall(void)
+{
+	pid_t fork_pid;
+	int i;
+	int ret;
+	int child_ret;
+	for (i = 0; i < 1000; i++) {
+		/* test_vmx will fork() */
+		ret = test_vmx(varray, &fork_pid);
+		if (fork_pid == -1)
+			return -1;
+		if (fork_pid == 0)
+			exit(ret);
+		waitpid(fork_pid, &child_ret, 0);
+		if (ret || child_ret)
+			return 1;
+	}
+
+	return 0;
+}
+
+int test_vmx_syscall(void)
+{
+	/*
+	 * Setup an environment with much context switching
+	 */
+	pid_t pid2;
+	pid_t pid = fork();
+	int ret;
+	int child_ret;
+	FAIL_IF(pid == -1);
+
+	pid2 = fork();
+	ret = vmx_syscall();
+	/* Can't FAIL_IF(pid2 == -1); because we've already forked */
+	if (pid2 == -1) {
+		/*
+		 * Couldn't fork, ensure child_ret is set and is a fail
+		 */
+		ret = child_ret = 1;
+	} else {
+		if (pid2)
+			waitpid(pid2, &child_ret, 0);
+		else
+			exit(ret);
+	}
+
+	ret |= child_ret;
+
+	if (pid)
+		waitpid(pid, &child_ret, 0);
+	else
+		exit(ret);
+
+	FAIL_IF(ret || child_ret);
+	return 0;
+}
+
+int main(int argc, char *argv[])
+{
+	return test_harness(test_vmx_syscall, "vmx_syscall");
+
+}
diff --git a/tools/testing/selftests/powerpc/tm/tm-signal-msr-resv.c b/tools/testing/selftests/powerpc/tm/tm-signal-msr-resv.c
index d86653f282b1..8c54d18b3e9a 100644
--- a/tools/testing/selftests/powerpc/tm/tm-signal-msr-resv.c
+++ b/tools/testing/selftests/powerpc/tm/tm-signal-msr-resv.c
@@ -40,7 +40,7 @@ void signal_usr1(int signum, siginfo_t *info, void *uc)
 #ifdef __powerpc64__
 	ucp->uc_mcontext.gp_regs[PT_MSR] |= (7ULL << 32);
 #else
-	ucp->uc_mcontext.regs->gpr[PT_MSR] |= (7ULL);
+	ucp->uc_mcontext.uc_regs->gregs[PT_MSR] |= (7ULL);
 #endif
 	/* Should segv on return becuase of invalid context */
 	segv_expected = 1;
diff --git a/tools/testing/selftests/pstore/config b/tools/testing/selftests/pstore/config
new file mode 100644
index 000000000000..6a8e5a9bfc10
--- /dev/null
+++ b/tools/testing/selftests/pstore/config
@@ -0,0 +1,4 @@
+CONFIG_MISC_FILESYSTEMS=y
+CONFIG_PSTORE=y
+CONFIG_PSTORE_PMSG=y
+CONFIG_PSTORE_CONSOLE=y
diff --git a/tools/testing/selftests/rcutorture/bin/parse-console.sh b/tools/testing/selftests/rcutorture/bin/parse-console.sh
index 844787a0d7be..5eb49b7f864c 100755
--- a/tools/testing/selftests/rcutorture/bin/parse-console.sh
+++ b/tools/testing/selftests/rcutorture/bin/parse-console.sh
@@ -33,7 +33,7 @@ if grep -Pq '\x00' < $file
 then
 	print_warning Console output contains nul bytes, old qemu still running?
 fi
-egrep 'Badness|WARNING:|Warn|BUG|===========|Call Trace:|Oops:|detected stalls on CPUs/tasks:|Stall ended before state dump start' < $file | grep -v 'ODEBUG: ' | grep -v 'Warning: unable to open an initial console' > $1.diags
+egrep 'Badness|WARNING:|Warn|BUG|===========|Call Trace:|Oops:|detected stalls on CPUs/tasks:|self-detected stall on CPU|Stall ended before state dump start|\?\?\? Writer stall state' < $file | grep -v 'ODEBUG: ' | grep -v 'Warning: unable to open an initial console' > $1.diags
 if test -s $1.diags
 then
 	print_warning Assertion failure in $file $title
@@ -64,10 +64,12 @@ then
 	then
 		summary="$summary  lockdep: $n_badness"
 	fi
-	n_stalls=`egrep -c 'detected stalls on CPUs/tasks:|Stall ended before state dump start' $1`
+	n_stalls=`egrep -c 'detected stalls on CPUs/tasks:|self-detected stall on CPU|Stall ended before state dump start|\?\?\? Writer stall state' $1`
 	if test "$n_stalls" -ne 0
 	then
 		summary="$summary  Stalls: $n_stalls"
 	fi
 	print_warning Summary: $summary
+else
+	rm $1.diags
 fi
diff --git a/tools/testing/selftests/seccomp/config b/tools/testing/selftests/seccomp/config
new file mode 100644
index 000000000000..db1e11b08c8a
--- /dev/null
+++ b/tools/testing/selftests/seccomp/config
@@ -0,0 +1,2 @@
+CONFIG_SECCOMP=y
+CONFIG_SECCOMP_FILTER=y
diff --git a/tools/testing/selftests/static_keys/config b/tools/testing/selftests/static_keys/config
new file mode 100644
index 000000000000..d538fb774b96
--- /dev/null
+++ b/tools/testing/selftests/static_keys/config
@@ -0,0 +1 @@
+CONFIG_TEST_STATIC_KEYS=m
diff --git a/tools/testing/selftests/timers/alarmtimer-suspend.c b/tools/testing/selftests/timers/alarmtimer-suspend.c
index 72cacf5383dd..2b361b830395 100644
--- a/tools/testing/selftests/timers/alarmtimer-suspend.c
+++ b/tools/testing/selftests/timers/alarmtimer-suspend.c
@@ -153,7 +153,7 @@ int main(void)
 
 		alarmcount = 0;
 		if (timer_create(alarm_clock_id, &se, &tm1) == -1) {
-			printf("timer_create failled, %s unspported?\n",
+			printf("timer_create failed, %s unsupported?\n",
 					clockstring(alarm_clock_id));
 			break;
 		}
diff --git a/tools/testing/selftests/timers/valid-adjtimex.c b/tools/testing/selftests/timers/valid-adjtimex.c
index e86d937cc22c..60fe3c569bd9 100644
--- a/tools/testing/selftests/timers/valid-adjtimex.c
+++ b/tools/testing/selftests/timers/valid-adjtimex.c
@@ -45,7 +45,17 @@ static inline int ksft_exit_fail(void)
 }
 #endif
 
-#define NSEC_PER_SEC 1000000000L
+#define NSEC_PER_SEC 1000000000LL
+#define USEC_PER_SEC 1000000LL
+
+#define ADJ_SETOFFSET 0x0100
+
+#include <sys/syscall.h>
+static int clock_adjtime(clockid_t id, struct timex *tx)
+{
+	return syscall(__NR_clock_adjtime, id, tx);
+}
+
 
 /* clear NTP time_status & time_state */
 int clear_time_state(void)
@@ -193,10 +203,137 @@ out:
 }
 
 
+int set_offset(long long offset, int use_nano)
+{
+	struct timex tmx = {};
+	int ret;
+
+	tmx.modes = ADJ_SETOFFSET;
+	if (use_nano) {
+		tmx.modes |= ADJ_NANO;
+
+		tmx.time.tv_sec = offset / NSEC_PER_SEC;
+		tmx.time.tv_usec = offset % NSEC_PER_SEC;
+
+		if (offset < 0 && tmx.time.tv_usec) {
+			tmx.time.tv_sec -= 1;
+			tmx.time.tv_usec += NSEC_PER_SEC;
+		}
+	} else {
+		tmx.time.tv_sec = offset / USEC_PER_SEC;
+		tmx.time.tv_usec = offset % USEC_PER_SEC;
+
+		if (offset < 0 && tmx.time.tv_usec) {
+			tmx.time.tv_sec -= 1;
+			tmx.time.tv_usec += USEC_PER_SEC;
+		}
+	}
+
+	ret = clock_adjtime(CLOCK_REALTIME, &tmx);
+	if (ret < 0) {
+		printf("(sec: %ld  usec: %ld) ", tmx.time.tv_sec, tmx.time.tv_usec);
+		printf("[FAIL]\n");
+		return -1;
+	}
+	return 0;
+}
+
+int set_bad_offset(long sec, long usec, int use_nano)
+{
+	struct timex tmx = {};
+	int ret;
+
+	tmx.modes = ADJ_SETOFFSET;
+	if (use_nano)
+		tmx.modes |= ADJ_NANO;
+
+	tmx.time.tv_sec = sec;
+	tmx.time.tv_usec = usec;
+	ret = clock_adjtime(CLOCK_REALTIME, &tmx);
+	if (ret >= 0) {
+		printf("Invalid (sec: %ld  usec: %ld) did not fail! ", tmx.time.tv_sec, tmx.time.tv_usec);
+		printf("[FAIL]\n");
+		return -1;
+	}
+	return 0;
+}
+
+int validate_set_offset(void)
+{
+	printf("Testing ADJ_SETOFFSET... ");
+
+	/* Test valid values */
+	if (set_offset(NSEC_PER_SEC - 1, 1))
+		return -1;
+
+	if (set_offset(-NSEC_PER_SEC + 1, 1))
+		return -1;
+
+	if (set_offset(-NSEC_PER_SEC - 1, 1))
+		return -1;
+
+	if (set_offset(5 * NSEC_PER_SEC, 1))
+		return -1;
+
+	if (set_offset(-5 * NSEC_PER_SEC, 1))
+		return -1;
+
+	if (set_offset(5 * NSEC_PER_SEC + NSEC_PER_SEC / 2, 1))
+		return -1;
+
+	if (set_offset(-5 * NSEC_PER_SEC - NSEC_PER_SEC / 2, 1))
+		return -1;
+
+	if (set_offset(USEC_PER_SEC - 1, 0))
+		return -1;
+
+	if (set_offset(-USEC_PER_SEC + 1, 0))
+		return -1;
+
+	if (set_offset(-USEC_PER_SEC - 1, 0))
+		return -1;
+
+	if (set_offset(5 * USEC_PER_SEC, 0))
+		return -1;
+
+	if (set_offset(-5 * USEC_PER_SEC, 0))
+		return -1;
+
+	if (set_offset(5 * USEC_PER_SEC + USEC_PER_SEC / 2, 0))
+		return -1;
+
+	if (set_offset(-5 * USEC_PER_SEC - USEC_PER_SEC / 2, 0))
+		return -1;
+
+	/* Test invalid values */
+	if (set_bad_offset(0, -1, 1))
+		return -1;
+	if (set_bad_offset(0, -1, 0))
+		return -1;
+	if (set_bad_offset(0, 2 * NSEC_PER_SEC, 1))
+		return -1;
+	if (set_bad_offset(0, 2 * USEC_PER_SEC, 0))
+		return -1;
+	if (set_bad_offset(0, NSEC_PER_SEC, 1))
+		return -1;
+	if (set_bad_offset(0, USEC_PER_SEC, 0))
+		return -1;
+	if (set_bad_offset(0, -NSEC_PER_SEC, 1))
+		return -1;
+	if (set_bad_offset(0, -USEC_PER_SEC, 0))
+		return -1;
+
+	printf("[OK]\n");
+	return 0;
+}
+
 int main(int argc, char **argv)
 {
 	if (validate_freq())
 		return ksft_exit_fail();
 
+	if (validate_set_offset())
+		return ksft_exit_fail();
+
 	return ksft_exit_pass();
 }
diff --git a/tools/testing/selftests/user/config b/tools/testing/selftests/user/config
new file mode 100644
index 000000000000..784ed8416324
--- /dev/null
+++ b/tools/testing/selftests/user/config
@@ -0,0 +1 @@
+CONFIG_TEST_USER_COPY=m
diff --git a/tools/testing/selftests/vm/config b/tools/testing/selftests/vm/config
new file mode 100644
index 000000000000..698c7ed28a26
--- /dev/null
+++ b/tools/testing/selftests/vm/config
@@ -0,0 +1 @@
+CONFIG_USERFAULTFD=y
diff --git a/tools/testing/selftests/x86/Makefile b/tools/testing/selftests/x86/Makefile
index d0c473f65850..b47ebd170690 100644
--- a/tools/testing/selftests/x86/Makefile
+++ b/tools/testing/selftests/x86/Makefile
@@ -4,15 +4,16 @@ include ../lib.mk
 
 .PHONY: all all_32 all_64 warn_32bit_failure clean
 
-TARGETS_C_BOTHBITS := single_step_syscall sysret_ss_attrs syscall_nt ptrace_syscall
-TARGETS_C_32BIT_ONLY := entry_from_vm86 syscall_arg_fault sigreturn test_syscall_vdso unwind_vdso \
+TARGETS_C_BOTHBITS := single_step_syscall sysret_ss_attrs syscall_nt ptrace_syscall \
+			check_initial_reg_state sigreturn ldt_gdt iopl
+TARGETS_C_32BIT_ONLY := entry_from_vm86 syscall_arg_fault test_syscall_vdso unwind_vdso \
 			test_FCMOV test_FCOMI test_FISTTP \
-			ldt_gdt \
 			vdso_restorer
 
 TARGETS_C_32BIT_ALL := $(TARGETS_C_BOTHBITS) $(TARGETS_C_32BIT_ONLY)
+TARGETS_C_64BIT_ALL := $(TARGETS_C_BOTHBITS) $(TARGETS_C_64BIT_ONLY)
 BINARIES_32 := $(TARGETS_C_32BIT_ALL:%=%_32)
-BINARIES_64 := $(TARGETS_C_BOTHBITS:%=%_64)
+BINARIES_64 := $(TARGETS_C_64BIT_ALL:%=%_64)
 
 CFLAGS := -O2 -g -std=gnu99 -pthread -Wall
 
@@ -40,7 +41,7 @@ clean:
 $(TARGETS_C_32BIT_ALL:%=%_32): %_32: %.c
 	$(CC) -m32 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ -lrt -ldl -lm
 
-$(TARGETS_C_BOTHBITS:%=%_64): %_64: %.c
+$(TARGETS_C_64BIT_ALL:%=%_64): %_64: %.c
 	$(CC) -m64 -o $@ $(CFLAGS) $(EXTRA_CFLAGS) $^ -lrt -ldl
 
 # x86_64 users should be encouraged to install 32-bit libraries
@@ -65,3 +66,9 @@ endif
 sysret_ss_attrs_64: thunks.S
 ptrace_syscall_32: raw_syscall_helper_32.S
 test_syscall_vdso_32: thunks_32.S
+
+# check_initial_reg_state is special: it needs a custom entry, and it
+# needs to be static so that its interpreter doesn't destroy its initial
+# state.
+check_initial_reg_state_32: CFLAGS += -Wl,-ereal_start -static
+check_initial_reg_state_64: CFLAGS += -Wl,-ereal_start -static
diff --git a/tools/testing/selftests/x86/check_initial_reg_state.c b/tools/testing/selftests/x86/check_initial_reg_state.c
new file mode 100644
index 000000000000..6aaed9b85baf
--- /dev/null
+++ b/tools/testing/selftests/x86/check_initial_reg_state.c
@@ -0,0 +1,109 @@
+/*
+ * check_initial_reg_state.c - check that execve sets the correct state
+ * Copyright (c) 2014-2016 Andrew Lutomirski
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but
+ * WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the GNU
+ * General Public License for more details.
+ */
+
+#define _GNU_SOURCE
+
+#include <stdio.h>
+
+unsigned long ax, bx, cx, dx, si, di, bp, sp, flags;
+unsigned long r8, r9, r10, r11, r12, r13, r14, r15;
+
+asm (
+	".pushsection .text\n\t"
+	".type real_start, @function\n\t"
+	".global real_start\n\t"
+	"real_start:\n\t"
+#ifdef __x86_64__
+	"mov %rax, ax\n\t"
+	"mov %rbx, bx\n\t"
+	"mov %rcx, cx\n\t"
+	"mov %rdx, dx\n\t"
+	"mov %rsi, si\n\t"
+	"mov %rdi, di\n\t"
+	"mov %rbp, bp\n\t"
+	"mov %rsp, sp\n\t"
+	"mov %r8, r8\n\t"
+	"mov %r9, r9\n\t"
+	"mov %r10, r10\n\t"
+	"mov %r11, r11\n\t"
+	"mov %r12, r12\n\t"
+	"mov %r13, r13\n\t"
+	"mov %r14, r14\n\t"
+	"mov %r15, r15\n\t"
+	"pushfq\n\t"
+	"popq flags\n\t"
+#else
+	"mov %eax, ax\n\t"
+	"mov %ebx, bx\n\t"
+	"mov %ecx, cx\n\t"
+	"mov %edx, dx\n\t"
+	"mov %esi, si\n\t"
+	"mov %edi, di\n\t"
+	"mov %ebp, bp\n\t"
+	"mov %esp, sp\n\t"
+	"pushfl\n\t"
+	"popl flags\n\t"
+#endif
+	"jmp _start\n\t"
+	".size real_start, . - real_start\n\t"
+	".popsection");
+
+int main()
+{
+	int nerrs = 0;
+
+	if (sp == 0) {
+		printf("[FAIL]\tTest was built incorrectly\n");
+		return 1;
+	}
+
+	if (ax || bx || cx || dx || si || di || bp
+#ifdef __x86_64__
+	    || r8 || r9 || r10 || r11 || r12 || r13 || r14 || r15
+#endif
+		) {
+		printf("[FAIL]\tAll GPRs except SP should be 0\n");
+#define SHOW(x) printf("\t" #x " = 0x%lx\n", x);
+		SHOW(ax);
+		SHOW(bx);
+		SHOW(cx);
+		SHOW(dx);
+		SHOW(si);
+		SHOW(di);
+		SHOW(bp);
+		SHOW(sp);
+#ifdef __x86_64__
+		SHOW(r8);
+		SHOW(r9);
+		SHOW(r10);
+		SHOW(r11);
+		SHOW(r12);
+		SHOW(r13);
+		SHOW(r14);
+		SHOW(r15);
+#endif
+		nerrs++;
+	} else {
+		printf("[OK]\tAll GPRs except SP are 0\n");
+	}
+
+	if (flags != 0x202) {
+		printf("[FAIL]\tFLAGS is 0x%lx, but it should be 0x202\n", flags);
+		nerrs++;
+	} else {
+		printf("[OK]\tFLAGS is 0x202\n");
+	}
+
+	return nerrs ? 1 : 0;
+}
diff --git a/tools/testing/selftests/x86/iopl.c b/tools/testing/selftests/x86/iopl.c
new file mode 100644
index 000000000000..c496ca97bc18
--- /dev/null
+++ b/tools/testing/selftests/x86/iopl.c
@@ -0,0 +1,135 @@
+/*
+ * iopl.c - Test case for a Linux on Xen 64-bit bug
+ * Copyright (c) 2015 Andrew Lutomirski
+ */
+
+#define _GNU_SOURCE
+#include <err.h>
+#include <stdio.h>
+#include <stdint.h>
+#include <signal.h>
+#include <setjmp.h>
+#include <stdlib.h>
+#include <string.h>
+#include <errno.h>
+#include <unistd.h>
+#include <sys/types.h>
+#include <sys/wait.h>
+#include <stdbool.h>
+#include <sched.h>
+#include <sys/io.h>
+
+static int nerrs = 0;
+
+static void sethandler(int sig, void (*handler)(int, siginfo_t *, void *),
+		       int flags)
+{
+	struct sigaction sa;
+	memset(&sa, 0, sizeof(sa));
+	sa.sa_sigaction = handler;
+	sa.sa_flags = SA_SIGINFO | flags;
+	sigemptyset(&sa.sa_mask);
+	if (sigaction(sig, &sa, 0))
+		err(1, "sigaction");
+
+}
+
+static jmp_buf jmpbuf;
+
+static void sigsegv(int sig, siginfo_t *si, void *ctx_void)
+{
+	siglongjmp(jmpbuf, 1);
+}
+
+int main(void)
+{
+	cpu_set_t cpuset;
+	CPU_ZERO(&cpuset);
+	CPU_SET(0, &cpuset);
+	if (sched_setaffinity(0, sizeof(cpuset), &cpuset) != 0)
+		err(1, "sched_setaffinity to CPU 0");
+
+	/* Probe for iopl support.  Note that iopl(0) works even as nonroot. */
+	if (iopl(3) != 0) {
+		printf("[OK]\tiopl(3) failed (%d) -- try running as root\n",
+		       errno);
+		return 0;
+	}
+
+	/* Restore our original state prior to starting the test. */
+	if (iopl(0) != 0)
+		err(1, "iopl(0)");
+
+	pid_t child = fork();
+	if (child == -1)
+		err(1, "fork");
+
+	if (child == 0) {
+		printf("\tchild: set IOPL to 3\n");
+		if (iopl(3) != 0)
+			err(1, "iopl");
+
+		printf("[RUN]\tchild: write to 0x80\n");
+		asm volatile ("outb %%al, $0x80" : : "a" (0));
+
+		return 0;
+	} else {
+		int status;
+		if (waitpid(child, &status, 0) != child ||
+		    !WIFEXITED(status)) {
+			printf("[FAIL]\tChild died\n");
+			nerrs++;
+		} else if (WEXITSTATUS(status) != 0) {
+			printf("[FAIL]\tChild failed\n");
+			nerrs++;
+		} else {
+			printf("[OK]\tChild succeeded\n");
+		}
+	}
+
+	printf("[RUN]\tparent: write to 0x80 (should fail)\n");
+
+	sethandler(SIGSEGV, sigsegv, 0);
+	if (sigsetjmp(jmpbuf, 1) != 0) {
+		printf("[OK]\twrite was denied\n");
+	} else {
+		asm volatile ("outb %%al, $0x80" : : "a" (0));
+		printf("[FAIL]\twrite was allowed\n");
+		nerrs++;
+	}
+
+	/* Test the capability checks. */
+	printf("\tiopl(3)\n");
+	if (iopl(3) != 0)
+		err(1, "iopl(3)");
+
+	printf("\tDrop privileges\n");
+	if (setresuid(1, 1, 1) != 0) {
+		printf("[WARN]\tDropping privileges failed\n");
+		goto done;
+	}
+
+	printf("[RUN]\tiopl(3) unprivileged but with IOPL==3\n");
+	if (iopl(3) != 0) {
+		printf("[FAIL]\tiopl(3) should work if iopl is already 3 even if unprivileged\n");
+		nerrs++;
+	}
+
+	printf("[RUN]\tiopl(0) unprivileged\n");
+	if (iopl(0) != 0) {
+		printf("[FAIL]\tiopl(0) should work if iopl is already 3 even if unprivileged\n");
+		nerrs++;
+	}
+
+	printf("[RUN]\tiopl(3) unprivileged\n");
+	if (iopl(3) == 0) {
+		printf("[FAIL]\tiopl(3) should fail if when unprivileged if iopl==0\n");
+		nerrs++;
+	} else {
+		printf("[OK]\tFailed as expected\n");
+	}
+
+done:
+	return nerrs ? 1 : 0;
+}
+
diff --git a/tools/testing/selftests/x86/ptrace_syscall.c b/tools/testing/selftests/x86/ptrace_syscall.c
index 5105b49cd8aa..421456784bc6 100644
--- a/tools/testing/selftests/x86/ptrace_syscall.c
+++ b/tools/testing/selftests/x86/ptrace_syscall.c
@@ -103,6 +103,17 @@ static void sethandler(int sig, void (*handler)(int, siginfo_t *, void *),
 		err(1, "sigaction");
 }
 
+static void setsigign(int sig, int flags)
+{
+	struct sigaction sa;
+	memset(&sa, 0, sizeof(sa));
+	sa.sa_sigaction = (void *)SIG_IGN;
+	sa.sa_flags = flags;
+	sigemptyset(&sa.sa_mask);
+	if (sigaction(sig, &sa, 0))
+		err(1, "sigaction");
+}
+
 static void clearhandler(int sig)
 {
 	struct sigaction sa;
@@ -187,7 +198,7 @@ static void test_ptrace_syscall_restart(void)
 
 	printf("[RUN]\tSYSEMU\n");
 	if (ptrace(PTRACE_SYSEMU, chld, 0, 0) != 0)
-		err(1, "PTRACE_SYSCALL");
+		err(1, "PTRACE_SYSEMU");
 	wait_trap(chld);
 
 	if (ptrace(PTRACE_GETREGS, chld, 0, &regs) != 0)
@@ -218,7 +229,7 @@ static void test_ptrace_syscall_restart(void)
 		err(1, "PTRACE_SETREGS");
 
 	if (ptrace(PTRACE_SYSEMU, chld, 0, 0) != 0)
-		err(1, "PTRACE_SYSCALL");
+		err(1, "PTRACE_SYSEMU");
 	wait_trap(chld);
 
 	if (ptrace(PTRACE_GETREGS, chld, 0, &regs) != 0)
@@ -250,7 +261,7 @@ static void test_ptrace_syscall_restart(void)
 		err(1, "PTRACE_SETREGS");
 
 	if (ptrace(PTRACE_SYSEMU, chld, 0, 0) != 0)
-		err(1, "PTRACE_SYSCALL");
+		err(1, "PTRACE_SYSEMU");
 	wait_trap(chld);
 
 	if (ptrace(PTRACE_GETREGS, chld, 0, &regs) != 0)
@@ -277,6 +288,119 @@ static void test_ptrace_syscall_restart(void)
 	}
 }
 
+static void test_restart_under_ptrace(void)
+{
+	printf("[RUN]\tkernel syscall restart under ptrace\n");
+	pid_t chld = fork();
+	if (chld < 0)
+		err(1, "fork");
+
+	if (chld == 0) {
+		if (ptrace(PTRACE_TRACEME, 0, 0, 0) != 0)
+			err(1, "PTRACE_TRACEME");
+
+		printf("\tChild will take a nap until signaled\n");
+		setsigign(SIGUSR1, SA_RESTART);
+		raise(SIGSTOP);
+
+		syscall(SYS_pause, 0, 0, 0, 0, 0, 0);
+		_exit(0);
+	}
+
+	int status;
+
+	/* Wait for SIGSTOP. */
+	if (waitpid(chld, &status, 0) != chld || !WIFSTOPPED(status))
+		err(1, "waitpid");
+
+	struct user_regs_struct regs;
+
+	printf("[RUN]\tSYSCALL\n");
+	if (ptrace(PTRACE_SYSCALL, chld, 0, 0) != 0)
+		err(1, "PTRACE_SYSCALL");
+	wait_trap(chld);
+
+	/* We should be stopped at pause(2) entry. */
+
+	if (ptrace(PTRACE_GETREGS, chld, 0, &regs) != 0)
+		err(1, "PTRACE_GETREGS");
+
+	if (regs.user_syscall_nr != SYS_pause ||
+	    regs.user_arg0 != 0 || regs.user_arg1 != 0 ||
+	    regs.user_arg2 != 0 || regs.user_arg3 != 0 ||
+	    regs.user_arg4 != 0 || regs.user_arg5 != 0) {
+		printf("[FAIL]\tInitial args are wrong (nr=%lu, args=%lu %lu %lu %lu %lu %lu)\n", (unsigned long)regs.user_syscall_nr, (unsigned long)regs.user_arg0, (unsigned long)regs.user_arg1, (unsigned long)regs.user_arg2, (unsigned long)regs.user_arg3, (unsigned long)regs.user_arg4, (unsigned long)regs.user_arg5);
+		nerrs++;
+	} else {
+		printf("[OK]\tInitial nr and args are correct\n");
+	}
+
+	/* Interrupt it. */
+	kill(chld, SIGUSR1);
+
+	/* Advance.  We should be stopped at exit. */
+	printf("[RUN]\tSYSCALL\n");
+	if (ptrace(PTRACE_SYSCALL, chld, 0, 0) != 0)
+		err(1, "PTRACE_SYSCALL");
+	wait_trap(chld);
+
+	if (ptrace(PTRACE_GETREGS, chld, 0, &regs) != 0)
+		err(1, "PTRACE_GETREGS");
+
+	if (regs.user_syscall_nr != SYS_pause ||
+	    regs.user_arg0 != 0 || regs.user_arg1 != 0 ||
+	    regs.user_arg2 != 0 || regs.user_arg3 != 0 ||
+	    regs.user_arg4 != 0 || regs.user_arg5 != 0) {
+		printf("[FAIL]\tArgs after SIGUSR1 are wrong (nr=%lu, args=%lu %lu %lu %lu %lu %lu)\n", (unsigned long)regs.user_syscall_nr, (unsigned long)regs.user_arg0, (unsigned long)regs.user_arg1, (unsigned long)regs.user_arg2, (unsigned long)regs.user_arg3, (unsigned long)regs.user_arg4, (unsigned long)regs.user_arg5);
+		nerrs++;
+	} else {
+		printf("[OK]\tArgs after SIGUSR1 are correct (ax = %ld)\n",
+		       (long)regs.user_ax);
+	}
+
+	/* Poke the regs back in.  This must not break anything. */
+	if (ptrace(PTRACE_SETREGS, chld, 0, &regs) != 0)
+		err(1, "PTRACE_SETREGS");
+
+	/* Catch the (ignored) SIGUSR1. */
+	if (ptrace(PTRACE_CONT, chld, 0, 0) != 0)
+		err(1, "PTRACE_CONT");
+	if (waitpid(chld, &status, 0) != chld)
+		err(1, "waitpid");
+	if (!WIFSTOPPED(status)) {
+		printf("[FAIL]\tChild was stopped for SIGUSR1 (status = 0x%x)\n", status);
+		nerrs++;
+	} else {
+		printf("[OK]\tChild got SIGUSR1\n");
+	}
+
+	/* The next event should be pause(2) again. */
+	printf("[RUN]\tStep again\n");
+	if (ptrace(PTRACE_SYSCALL, chld, 0, 0) != 0)
+		err(1, "PTRACE_SYSCALL");
+	wait_trap(chld);
+
+	/* We should be stopped at pause(2) entry. */
+
+	if (ptrace(PTRACE_GETREGS, chld, 0, &regs) != 0)
+		err(1, "PTRACE_GETREGS");
+
+	if (regs.user_syscall_nr != SYS_pause ||
+	    regs.user_arg0 != 0 || regs.user_arg1 != 0 ||
+	    regs.user_arg2 != 0 || regs.user_arg3 != 0 ||
+	    regs.user_arg4 != 0 || regs.user_arg5 != 0) {
+		printf("[FAIL]\tpause did not restart (nr=%lu, args=%lu %lu %lu %lu %lu %lu)\n", (unsigned long)regs.user_syscall_nr, (unsigned long)regs.user_arg0, (unsigned long)regs.user_arg1, (unsigned long)regs.user_arg2, (unsigned long)regs.user_arg3, (unsigned long)regs.user_arg4, (unsigned long)regs.user_arg5);
+		nerrs++;
+	} else {
+		printf("[OK]\tpause(2) restarted correctly\n");
+	}
+
+	/* Kill it. */
+	kill(chld, SIGKILL);
+	if (waitpid(chld, &status, 0) != chld)
+		err(1, "waitpid");
+}
+
 int main()
 {
 	printf("[RUN]\tCheck int80 return regs\n");
@@ -290,5 +414,7 @@ int main()
 
 	test_ptrace_syscall_restart();
 
+	test_restart_under_ptrace();
+
 	return 0;
 }
diff --git a/tools/testing/selftests/x86/sigreturn.c b/tools/testing/selftests/x86/sigreturn.c
index b5aa1bab7416..8a577e7070c6 100644
--- a/tools/testing/selftests/x86/sigreturn.c
+++ b/tools/testing/selftests/x86/sigreturn.c
@@ -54,6 +54,37 @@
 #include <sys/ptrace.h>
 #include <sys/user.h>
 
+/* Pull in AR_xyz defines. */
+typedef unsigned int u32;
+typedef unsigned short u16;
+#include "../../../../arch/x86/include/asm/desc_defs.h"
+
+/*
+ * Copied from asm/ucontext.h, as asm/ucontext.h conflicts badly with the glibc
+ * headers.
+ */
+#ifdef __x86_64__
+/*
+ * UC_SIGCONTEXT_SS will be set when delivering 64-bit or x32 signals on
+ * kernels that save SS in the sigcontext.  All kernels that set
+ * UC_SIGCONTEXT_SS will correctly restore at least the low 32 bits of esp
+ * regardless of SS (i.e. they implement espfix).
+ *
+ * Kernels that set UC_SIGCONTEXT_SS will also set UC_STRICT_RESTORE_SS
+ * when delivering a signal that came from 64-bit code.
+ *
+ * Sigreturn restores SS as follows:
+ *
+ * if (saved SS is valid || UC_STRICT_RESTORE_SS is set ||
+ *     saved CS is not 64-bit)
+ *         new SS = saved SS  (will fail IRET and signal if invalid)
+ * else
+ *         new SS = a flat 32-bit data segment
+ */
+#define UC_SIGCONTEXT_SS       0x2
+#define UC_STRICT_RESTORE_SS   0x4
+#endif
+
 /*
  * In principle, this test can run on Linux emulation layers (e.g.
  * Illumos "LX branded zones").  Solaris-based kernels reserve LDT
@@ -267,6 +298,9 @@ static gregset_t initial_regs, requested_regs, resulting_regs;
 /* Instructions for the SIGUSR1 handler. */
 static volatile unsigned short sig_cs, sig_ss;
 static volatile sig_atomic_t sig_trapped, sig_err, sig_trapno;
+#ifdef __x86_64__
+static volatile sig_atomic_t sig_corrupt_final_ss;
+#endif
 
 /* Abstractions for some 32-bit vs 64-bit differences. */
 #ifdef __x86_64__
@@ -305,9 +339,105 @@ static greg_t *csptr(ucontext_t *ctx)
 }
 #endif
 
+/*
+ * Checks a given selector for its code bitness or returns -1 if it's not
+ * a usable code segment selector.
+ */
+int cs_bitness(unsigned short cs)
+{
+	uint32_t valid = 0, ar;
+	asm ("lar %[cs], %[ar]\n\t"
+	     "jnz 1f\n\t"
+	     "mov $1, %[valid]\n\t"
+	     "1:"
+	     : [ar] "=r" (ar), [valid] "+rm" (valid)
+	     : [cs] "r" (cs));
+
+	if (!valid)
+		return -1;
+
+	bool db = (ar & (1 << 22));
+	bool l = (ar & (1 << 21));
+
+	if (!(ar & (1<<11)))
+	    return -1;	/* Not code. */
+
+	if (l && !db)
+		return 64;
+	else if (!l && db)
+		return 32;
+	else if (!l && !db)
+		return 16;
+	else
+		return -1;	/* Unknown bitness. */
+}
+
+/*
+ * Checks a given selector for its code bitness or returns -1 if it's not
+ * a usable code segment selector.
+ */
+bool is_valid_ss(unsigned short cs)
+{
+	uint32_t valid = 0, ar;
+	asm ("lar %[cs], %[ar]\n\t"
+	     "jnz 1f\n\t"
+	     "mov $1, %[valid]\n\t"
+	     "1:"
+	     : [ar] "=r" (ar), [valid] "+rm" (valid)
+	     : [cs] "r" (cs));
+
+	if (!valid)
+		return false;
+
+	if ((ar & AR_TYPE_MASK) != AR_TYPE_RWDATA &&
+	    (ar & AR_TYPE_MASK) != AR_TYPE_RWDATA_EXPDOWN)
+		return false;
+
+	return (ar & AR_P);
+}
+
 /* Number of errors in the current test case. */
 static volatile sig_atomic_t nerrs;
 
+static void validate_signal_ss(int sig, ucontext_t *ctx)
+{
+#ifdef __x86_64__
+	bool was_64bit = (cs_bitness(*csptr(ctx)) == 64);
+
+	if (!(ctx->uc_flags & UC_SIGCONTEXT_SS)) {
+		printf("[FAIL]\tUC_SIGCONTEXT_SS was not set\n");
+		nerrs++;
+
+		/*
+		 * This happens on Linux 4.1.  The rest will fail, too, so
+		 * return now to reduce the noise.
+		 */
+		return;
+	}
+
+	/* UC_STRICT_RESTORE_SS is set iff we came from 64-bit mode. */
+	if (!!(ctx->uc_flags & UC_STRICT_RESTORE_SS) != was_64bit) {
+		printf("[FAIL]\tUC_STRICT_RESTORE_SS was wrong in signal %d\n",
+		       sig);
+		nerrs++;
+	}
+
+	if (is_valid_ss(*ssptr(ctx))) {
+		/*
+		 * DOSEMU was written before 64-bit sigcontext had SS, and
+		 * it tries to figure out the signal source SS by looking at
+		 * the physical register.  Make sure that keeps working.
+		 */
+		unsigned short hw_ss;
+		asm ("mov %%ss, %0" : "=rm" (hw_ss));
+		if (hw_ss != *ssptr(ctx)) {
+			printf("[FAIL]\tHW SS didn't match saved SS\n");
+			nerrs++;
+		}
+	}
+#endif
+}
+
 /*
  * SIGUSR1 handler.  Sets CS and SS as requested and points IP to the
  * int3 trampoline.  Sets SP to a large known value so that we can see
@@ -317,6 +447,8 @@ static void sigusr1(int sig, siginfo_t *info, void *ctx_void)
 {
 	ucontext_t *ctx = (ucontext_t*)ctx_void;
 
+	validate_signal_ss(sig, ctx);
+
 	memcpy(&initial_regs, &ctx->uc_mcontext.gregs, sizeof(gregset_t));
 
 	*csptr(ctx) = sig_cs;
@@ -334,13 +466,16 @@ static void sigusr1(int sig, siginfo_t *info, void *ctx_void)
 }
 
 /*
- * Called after a successful sigreturn.  Restores our state so that
- * the original raise(SIGUSR1) returns.
+ * Called after a successful sigreturn (via int3) or from a failed
+ * sigreturn (directly by kernel).  Restores our state so that the
+ * original raise(SIGUSR1) returns.
  */
 static void sigtrap(int sig, siginfo_t *info, void *ctx_void)
 {
 	ucontext_t *ctx = (ucontext_t*)ctx_void;
 
+	validate_signal_ss(sig, ctx);
+
 	sig_err = ctx->uc_mcontext.gregs[REG_ERR];
 	sig_trapno = ctx->uc_mcontext.gregs[REG_TRAPNO];
 
@@ -358,41 +493,62 @@ static void sigtrap(int sig, siginfo_t *info, void *ctx_void)
 	memcpy(&resulting_regs, &ctx->uc_mcontext.gregs, sizeof(gregset_t));
 	memcpy(&ctx->uc_mcontext.gregs, &initial_regs, sizeof(gregset_t));
 
+#ifdef __x86_64__
+	if (sig_corrupt_final_ss) {
+		if (ctx->uc_flags & UC_STRICT_RESTORE_SS) {
+			printf("[FAIL]\tUC_STRICT_RESTORE_SS was set inappropriately\n");
+			nerrs++;
+		} else {
+			/*
+			 * DOSEMU transitions from 32-bit to 64-bit mode by
+			 * adjusting sigcontext, and it requires that this work
+			 * even if the saved SS is bogus.
+			 */
+			printf("\tCorrupting SS on return to 64-bit mode\n");
+			*ssptr(ctx) = 0;
+		}
+	}
+#endif
+
 	sig_trapped = sig;
 }
 
-/*
- * Checks a given selector for its code bitness or returns -1 if it's not
- * a usable code segment selector.
- */
-int cs_bitness(unsigned short cs)
+#ifdef __x86_64__
+/* Tests recovery if !UC_STRICT_RESTORE_SS */
+static void sigusr2(int sig, siginfo_t *info, void *ctx_void)
 {
-	uint32_t valid = 0, ar;
-	asm ("lar %[cs], %[ar]\n\t"
-	     "jnz 1f\n\t"
-	     "mov $1, %[valid]\n\t"
-	     "1:"
-	     : [ar] "=r" (ar), [valid] "+rm" (valid)
-	     : [cs] "r" (cs));
+	ucontext_t *ctx = (ucontext_t*)ctx_void;
 
-	if (!valid)
-		return -1;
+	if (!(ctx->uc_flags & UC_STRICT_RESTORE_SS)) {
+		printf("[FAIL]\traise(2) didn't set UC_STRICT_RESTORE_SS\n");
+		nerrs++;
+		return;  /* We can't do the rest. */
+	}
 
-	bool db = (ar & (1 << 22));
-	bool l = (ar & (1 << 21));
+	ctx->uc_flags &= ~UC_STRICT_RESTORE_SS;
+	*ssptr(ctx) = 0;
 
-	if (!(ar & (1<<11)))
-	    return -1;	/* Not code. */
+	/* Return.  The kernel should recover without sending another signal. */
+}
 
-	if (l && !db)
-		return 64;
-	else if (!l && db)
-		return 32;
-	else if (!l && !db)
-		return 16;
-	else
-		return -1;	/* Unknown bitness. */
+static int test_nonstrict_ss(void)
+{
+	clearhandler(SIGUSR1);
+	clearhandler(SIGTRAP);
+	clearhandler(SIGSEGV);
+	clearhandler(SIGILL);
+	sethandler(SIGUSR2, sigusr2, 0);
+
+	nerrs = 0;
+
+	printf("[RUN]\tClear UC_STRICT_RESTORE_SS and corrupt SS\n");
+	raise(SIGUSR2);
+	if (!nerrs)
+		printf("[OK]\tIt worked\n");
+
+	return nerrs;
 }
+#endif
 
 /* Finds a usable code segment of the requested bitness. */
 int find_cs(int bitness)
@@ -576,6 +732,12 @@ static int test_bad_iret(int cs_bits, unsigned short ss, int force_cs)
 		       errdesc, strsignal(sig_trapped));
 		return 0;
 	} else {
+		/*
+		 * This also implicitly tests UC_STRICT_RESTORE_SS:
+		 * We check that these signals set UC_STRICT_RESTORE_SS and,
+		 * if UC_STRICT_RESTORE_SS doesn't cause strict behavior,
+		 * then we won't get SIGSEGV.
+		 */
 		printf("[FAIL]\tDid not get SIGSEGV\n");
 		return 1;
 	}
@@ -632,6 +794,14 @@ int main()
 						    GDT3(gdt_data16_idx));
 	}
 
+#ifdef __x86_64__
+	/* Nasty ABI case: check SS corruption handling. */
+	sig_corrupt_final_ss = 1;
+	total_nerrs += test_valid_sigreturn(32, false, -1);
+	total_nerrs += test_valid_sigreturn(32, true, -1);
+	sig_corrupt_final_ss = 0;
+#endif
+
 	/*
 	 * We're done testing valid sigreturn cases.  Now we test states
 	 * for which sigreturn itself will succeed but the subsequent
@@ -680,5 +850,9 @@ int main()
 	if (gdt_npdata32_idx)
 		test_bad_iret(32, GDT3(gdt_npdata32_idx), -1);
 
+#ifdef __x86_64__
+	total_nerrs += test_nonstrict_ss();
+#endif
+
 	return total_nerrs ? 1 : 0;
 }
diff --git a/tools/testing/selftests/x86/syscall_nt.c b/tools/testing/selftests/x86/syscall_nt.c
index 60c06af4646a..43fcab367fb0 100644
--- a/tools/testing/selftests/x86/syscall_nt.c
+++ b/tools/testing/selftests/x86/syscall_nt.c
@@ -17,6 +17,9 @@
 
 #include <stdio.h>
 #include <unistd.h>
+#include <string.h>
+#include <signal.h>
+#include <err.h>
 #include <sys/syscall.h>
 #include <asm/processor-flags.h>
 
@@ -26,6 +29,8 @@
 # define WIDTH "l"
 #endif
 
+static unsigned int nerrs;
+
 static unsigned long get_eflags(void)
 {
 	unsigned long eflags;
@@ -39,16 +44,52 @@ static void set_eflags(unsigned long eflags)
 		      : : "rm" (eflags) : "flags");
 }
 
-int main()
+static void sethandler(int sig, void (*handler)(int, siginfo_t *, void *),
+		       int flags)
 {
-	printf("[RUN]\tSet NT and issue a syscall\n");
-	set_eflags(get_eflags() | X86_EFLAGS_NT);
+	struct sigaction sa;
+	memset(&sa, 0, sizeof(sa));
+	sa.sa_sigaction = handler;
+	sa.sa_flags = SA_SIGINFO | flags;
+	sigemptyset(&sa.sa_mask);
+	if (sigaction(sig, &sa, 0))
+		err(1, "sigaction");
+}
+
+static void sigtrap(int sig, siginfo_t *si, void *ctx_void)
+{
+}
+
+static void do_it(unsigned long extraflags)
+{
+	unsigned long flags;
+
+	set_eflags(get_eflags() | extraflags);
 	syscall(SYS_getpid);
-	if (get_eflags() & X86_EFLAGS_NT) {
-		printf("[OK]\tThe syscall worked and NT is still set\n");
-		return 0;
+	flags = get_eflags();
+	if ((flags & extraflags) == extraflags) {
+		printf("[OK]\tThe syscall worked and flags are still set\n");
 	} else {
-		printf("[FAIL]\tThe syscall worked but NT was cleared\n");
-		return 1;
+		printf("[FAIL]\tThe syscall worked but flags were cleared (flags = 0x%lx but expected 0x%lx set)\n",
+		       flags, extraflags);
+		nerrs++;
 	}
 }
+
+int main(void)
+{
+	printf("[RUN]\tSet NT and issue a syscall\n");
+	do_it(X86_EFLAGS_NT);
+
+	/*
+	 * Now try it again with TF set -- TF forces returns via IRET in all
+	 * cases except non-ptregs-using 64-bit full fast path syscalls.
+	 */
+
+	sethandler(SIGTRAP, sigtrap, 0);
+
+	printf("[RUN]\tSet NT|TF and issue a syscall\n");
+	do_it(X86_EFLAGS_NT | X86_EFLAGS_TF);
+
+	return nerrs == 0 ? 0 : 1;
+}
diff --git a/tools/testing/selftests/zram/config b/tools/testing/selftests/zram/config
new file mode 100644
index 000000000000..e0cc47e2c7e2
--- /dev/null
+++ b/tools/testing/selftests/zram/config
@@ -0,0 +1,2 @@
+CONFIG_ZSMALLOC=y
+CONFIG_ZRAM=m