Received-SPF: pass (google.com: best guess record for domain of linux-kernel-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
Subject: [PATCH v2] libnvdimm/dimm: Add a no-BLK quirk based on NVDIMM family
From:   Dan Williams <dan.j.williams@intel.com>
To:     linux-nvdimm@lists.01.org
Cc:     Dexuan Cui <decui@microsoft.com>, Dexuan Cui <decui@microsoft.com>,
        linux-kernel@vger.kernel.org
Date:   Tue, 12 Feb 2019 20:14:10 -0800
Message-ID: <155003117983.733096.1672828968568209328.stgit@dwillia2-desk3.amr.corp.intel.com>
In-Reply-To: <154915640488.3541628.4947340559510114602.stgit@dwillia2-desk3.amr.corp.intel.com>
References: <154915640488.3541628.4947340559510114602.stgit@dwillia2-desk3.amr.corp.intel.com>
User-Agent: StGit/0.18-2-gc94f
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Precedence: bulk

As Dexuan reports the NVDIMM_FAMILY_HYPERV platform is incompatible with
the existing Linux namespace implementation because it uses
NSLABEL_FLAG_LOCAL for x1-width PMEM interleave sets. Quirk it as an
platform / DIMM that does not provide BLK-aperture access. Allow the
libnvdimm core to assume no potential for aliasing. In case other
implementations make the same mistake, provide a "noblk" module
parameter to force-enable the quirk.

Link: https://lkml.kernel.org/r/PU1P153MB0169977604493B82B662A01CBF920@PU1P153MB0169.APCP153.PROD.OUTLOOK.COM
Reported-by: Dexuan Cui <decui@microsoft.com>
Tested-by: Dexuan Cui <decui@microsoft.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
Changes since v1:
* Fix the default setting for NDD_NOBLK in __nvdimm_create().

 drivers/acpi/nfit/core.c        |    4 ++++
 drivers/nvdimm/dimm_devs.c      |    7 +++++++
 drivers/nvdimm/label.c          |    3 +++
 drivers/nvdimm/namespace_devs.c |    6 ++++++
 drivers/nvdimm/region_devs.c    |    7 +++++++
 include/linux/libnvdimm.h       |    2 ++
 6 files changed, 29 insertions(+)

diff --git a/drivers/acpi/nfit/core.c b/drivers/acpi/nfit/core.c
index 4a7e8b1fa43b..811c399a3a76 100644
--- a/drivers/acpi/nfit/core.c
+++ b/drivers/acpi/nfit/core.c
@@ -2016,6 +2016,10 @@ static int acpi_nfit_register_dimms(struct acpi_nfit_desc *acpi_desc)
 			cmd_mask |= nfit_mem->dsm_mask & NVDIMM_STANDARD_CMDMASK;
 		}
 
+		/* Quirk to ignore LOCAL for labels on HYPERV DIMMs */
+		if (nfit_mem->family == NVDIMM_FAMILY_HYPERV)
+			set_bit(NDD_NOBLK, &flags);
+
 		if (test_bit(NFIT_MEM_LSR, &nfit_mem->flags)) {
 			set_bit(ND_CMD_GET_CONFIG_SIZE, &cmd_mask);
 			set_bit(ND_CMD_GET_CONFIG_DATA, &cmd_mask);
diff --git a/drivers/nvdimm/dimm_devs.c b/drivers/nvdimm/dimm_devs.c
index 4890310df874..553aa78abeee 100644
--- a/drivers/nvdimm/dimm_devs.c
+++ b/drivers/nvdimm/dimm_devs.c
@@ -11,6 +11,7 @@
  * General Public License for more details.
  */
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
+#include <linux/moduleparam.h>
 #include <linux/vmalloc.h>
 #include <linux/device.h>
 #include <linux/ndctl.h>
@@ -25,6 +26,10 @@
 
 static DEFINE_IDA(dimm_ida);
 
+static bool noblk;
+module_param(noblk, bool, 0444);
+MODULE_PARM_DESC(noblk, "force disable BLK / local alias support");
+
 /*
  * Retrieve bus and dimm handle and return if this bus supports
  * get_config_data commands
@@ -551,6 +556,8 @@ struct nvdimm *__nvdimm_create(struct nvdimm_bus *nvdimm_bus,
 
 	nvdimm->dimm_id = dimm_id;
 	nvdimm->provider_data = provider_data;
+	if (noblk)
+		flags |= 1 << NDD_NOBLK;
 	nvdimm->flags = flags;
 	nvdimm->cmd_mask = cmd_mask;
 	nvdimm->num_flush = num_flush;
diff --git a/drivers/nvdimm/label.c b/drivers/nvdimm/label.c
index 6d6e9a12150b..f3d753d3169c 100644
--- a/drivers/nvdimm/label.c
+++ b/drivers/nvdimm/label.c
@@ -392,6 +392,7 @@ int nd_label_reserve_dpa(struct nvdimm_drvdata *ndd)
 		return 0; /* no label, nothing to reserve */
 
 	for_each_clear_bit_le(slot, free, nslot) {
+		struct nvdimm *nvdimm = to_nvdimm(ndd->dev);
 		struct nd_namespace_label *nd_label;
 		struct nd_region *nd_region = NULL;
 		u8 label_uuid[NSLABEL_UUID_LEN];
@@ -406,6 +407,8 @@ int nd_label_reserve_dpa(struct nvdimm_drvdata *ndd)
 
 		memcpy(label_uuid, nd_label->uuid, NSLABEL_UUID_LEN);
 		flags = __le32_to_cpu(nd_label->flags);
+		if (test_bit(NDD_NOBLK, &nvdimm->flags))
+			flags &= ~NSLABEL_FLAG_LOCAL;
 		nd_label_gen_id(&label_id, label_uuid, flags);
 		res = nvdimm_allocate_dpa(ndd, &label_id,
 				__le64_to_cpu(nd_label->dpa),
diff --git a/drivers/nvdimm/namespace_devs.c b/drivers/nvdimm/namespace_devs.c
index 4b077555ac70..3677b0c4a33d 100644
--- a/drivers/nvdimm/namespace_devs.c
+++ b/drivers/nvdimm/namespace_devs.c
@@ -2492,6 +2492,12 @@ static int init_active_labels(struct nd_region *nd_region)
 			if (!label_ent)
 				break;
 			label = nd_label_active(ndd, j);
+			if (test_bit(NDD_NOBLK, &nvdimm->flags)) {
+				u32 flags = __le32_to_cpu(label->flags);
+
+				flags &= ~NSLABEL_FLAG_LOCAL;
+				label->flags = __cpu_to_le32(flags);
+			}
 			label_ent->label = label;
 
 			mutex_lock(&nd_mapping->lock);
diff --git a/drivers/nvdimm/region_devs.c b/drivers/nvdimm/region_devs.c
index e2818f94f292..3b58baa44b5c 100644
--- a/drivers/nvdimm/region_devs.c
+++ b/drivers/nvdimm/region_devs.c
@@ -1003,6 +1003,13 @@ static struct nd_region *nd_region_create(struct nvdimm_bus *nvdimm_bus,
 
 		if (test_bit(NDD_UNARMED, &nvdimm->flags))
 			ro = 1;
+
+		if (test_bit(NDD_NOBLK, &nvdimm->flags)
+				&& dev_type == &nd_blk_device_type) {
+			dev_err(&nvdimm_bus->dev, "%s: %s mapping%d is not BLK capable\n",
+					caller, dev_name(&nvdimm->dev), i);
+			return NULL;
+		}
 	}
 
 	if (dev_type == &nd_blk_device_type) {
diff --git a/include/linux/libnvdimm.h b/include/linux/libnvdimm.h
index 5440f11b0907..7da406ae3a2b 100644
--- a/include/linux/libnvdimm.h
+++ b/include/linux/libnvdimm.h
@@ -42,6 +42,8 @@ enum {
 	NDD_SECURITY_OVERWRITE = 3,
 	/*  tracking whether or not there is a pending device reference */
 	NDD_WORK_PENDING = 4,
+	/* ignore / filter NSLABEL_FLAG_LOCAL for this DIMM, i.e. no aliasing */
+	NDD_NOBLK = 5,
 
 	/* need to set a limit somewhere, but yes, this is likely overkill */
 	ND_IOCTL_MAX_BUFLEN = SZ_4M,