stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* Patch "libnvdimm, dax: fix 1GB-aligned namespaces vs physical misalignment" has been added to the 4.9-stable tree
@ 2017-12-27 15:22 gregkh
  2017-12-27 15:46 ` Greg KH
  0 siblings, 1 reply; 3+ messages in thread
From: gregkh @ 2017-12-27 15:22 UTC (permalink / raw)
  To: dan.j.williams, gregkh, jane.chu; +Cc: stable, stable-commits


This is a note to let you know that I've just added the patch titled

    libnvdimm, dax: fix 1GB-aligned namespaces vs physical misalignment

to the 4.9-stable tree which can be found at:
    http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary

The filename of the patch is:
     libnvdimm-dax-fix-1gb-aligned-namespaces-vs-physical-misalignment.patch
and it can be found in the queue-4.9 subdirectory.

If you, or anyone else, feels it should not be added to the stable tree,
please let <stable@vger.kernel.org> know about it.


>From 41fce90f26333c4fa82e8e43b9ace86c4e8a0120 Mon Sep 17 00:00:00 2001
From: Dan Williams <dan.j.williams@intel.com>
Date: Mon, 4 Dec 2017 14:07:43 -0800
Subject: libnvdimm, dax: fix 1GB-aligned namespaces vs physical misalignment

From: Dan Williams <dan.j.williams@intel.com>

commit 41fce90f26333c4fa82e8e43b9ace86c4e8a0120 upstream.

The following namespace configuration attempt:

    # ndctl create-namespace -e namespace0.0 -m devdax -a 1G -f
    libndctl: ndctl_dax_enable: dax0.1: failed to enable
      Error: namespace0.0: failed to enable

    failed to reconfigure namespace: No such device or address

...fails when the backing memory range is not physically aligned to 1G:

    # cat /proc/iomem | grep Persistent
    210000000-30fffffff : Persistent Memory (legacy)

In the above example the 4G persistent memory range starts and ends on a
256MB boundary.

We handle this case correctly when needing to handle cases that violate
section alignment (128MB) collisions against "System RAM", and we simply
need to extend that padding/truncation for the 1GB alignment use case.

Fixes: 315c562536c4 ("libnvdimm, pfn: add 'align' attribute...")
Reported-and-tested-by: Jane Chu <jane.chu@oracle.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 drivers/nvdimm/pfn_devs.c |   15 ++++++++++++---
 1 file changed, 12 insertions(+), 3 deletions(-)

--- a/drivers/nvdimm/pfn_devs.c
+++ b/drivers/nvdimm/pfn_devs.c
@@ -562,6 +562,12 @@ static struct vmem_altmap *__nvdimm_setu
 	return altmap;
 }
 
+static u64 phys_pmem_align_down(struct nd_pfn *nd_pfn, u64 phys)
+{
+	return min_t(u64, PHYS_SECTION_ALIGN_DOWN(phys),
+			ALIGN_DOWN(phys, nd_pfn->align));
+}
+
 static int nd_pfn_init(struct nd_pfn *nd_pfn)
 {
 	u32 dax_label_reserve = is_nd_dax(&nd_pfn->dev) ? SZ_128K : 0;
@@ -617,13 +623,16 @@ static int nd_pfn_init(struct nd_pfn *nd
 	start = nsio->res.start;
 	size = PHYS_SECTION_ALIGN_UP(start + size) - start;
 	if (region_intersects(start, size, IORESOURCE_SYSTEM_RAM,
-				IORES_DESC_NONE) == REGION_MIXED) {
+				IORES_DESC_NONE) == REGION_MIXED
+			|| !IS_ALIGNED(start + resource_size(&nsio->res),
+				nd_pfn->align)) {
 		size = resource_size(&nsio->res);
-		end_trunc = start + size - PHYS_SECTION_ALIGN_DOWN(start + size);
+		end_trunc = start + size - phys_pmem_align_down(nd_pfn,
+				start + size);
 	}
 
 	if (start_pad + end_trunc)
-		dev_info(&nd_pfn->dev, "%s section collision, truncate %d bytes\n",
+		dev_info(&nd_pfn->dev, "%s alignment collision, truncate %d bytes\n",
 				dev_name(&ndns->dev), start_pad + end_trunc);
 
 	/*


Patches currently in stable-queue which might be from dan.j.williams@intel.com are

queue-4.9/libnvdimm-pfn-fix-start_pad-handling-for-aligned-namespaces.patch
queue-4.9/libnvdimm-dax-fix-1gb-aligned-namespaces-vs-physical-misalignment.patch
queue-4.9/acpi-nfit-fix-health-event-notification.patch

^ permalink raw reply	[flat|nested] 3+ messages in thread
* [4.9-stable PATCH 08/11] libnvdimm, dax: fix 1GB-aligned namespaces vs physical misalignment
@ 2018-02-23 22:06 Dan Williams
  2018-02-26 19:58 ` Patch "libnvdimm, dax: fix 1GB-aligned namespaces vs physical misalignment" has been added to the 4.9-stable tree gregkh
  0 siblings, 1 reply; 3+ messages in thread
From: Dan Williams @ 2018-02-23 22:06 UTC (permalink / raw)
  To: gregkh; +Cc: Jane Chu, linux-kernel, stable

commit 41fce90f26333c4fa82e8e43b9ace86c4e8a0120 upstream.

The following namespace configuration attempt:

    # ndctl create-namespace -e namespace0.0 -m devdax -a 1G -f
    libndctl: ndctl_dax_enable: dax0.1: failed to enable
      Error: namespace0.0: failed to enable

    failed to reconfigure namespace: No such device or address

...fails when the backing memory range is not physically aligned to 1G:

    # cat /proc/iomem | grep Persistent
    210000000-30fffffff : Persistent Memory (legacy)

In the above example the 4G persistent memory range starts and ends on a
256MB boundary.

We handle this case correctly when needing to handle cases that violate
section alignment (128MB) collisions against "System RAM", and we simply
need to extend that padding/truncation for the 1GB alignment use case.

Cc: <stable@vger.kernel.org>
Fixes: 315c562536c4 ("libnvdimm, pfn: add 'align' attribute...")
Reported-and-tested-by: Jane Chu <jane.chu@oracle.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/nvdimm/pfn_devs.c |   15 ++++++++++++---
 include/linux/kernel.h    |    1 +
 2 files changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/nvdimm/pfn_devs.c b/drivers/nvdimm/pfn_devs.c
index 42abdd2391c9..d6aa59ca68b9 100644
--- a/drivers/nvdimm/pfn_devs.c
+++ b/drivers/nvdimm/pfn_devs.c
@@ -563,6 +563,12 @@ static struct vmem_altmap *__nvdimm_setup_pfn(struct nd_pfn *nd_pfn,
 	return altmap;
 }
 
+static u64 phys_pmem_align_down(struct nd_pfn *nd_pfn, u64 phys)
+{
+	return min_t(u64, PHYS_SECTION_ALIGN_DOWN(phys),
+			ALIGN_DOWN(phys, nd_pfn->align));
+}
+
 static int nd_pfn_init(struct nd_pfn *nd_pfn)
 {
 	u32 dax_label_reserve = is_nd_dax(&nd_pfn->dev) ? SZ_128K : 0;
@@ -618,13 +624,16 @@ static int nd_pfn_init(struct nd_pfn *nd_pfn)
 	start = nsio->res.start;
 	size = PHYS_SECTION_ALIGN_UP(start + size) - start;
 	if (region_intersects(start, size, IORESOURCE_SYSTEM_RAM,
-				IORES_DESC_NONE) == REGION_MIXED) {
+				IORES_DESC_NONE) == REGION_MIXED
+			|| !IS_ALIGNED(start + resource_size(&nsio->res),
+				nd_pfn->align)) {
 		size = resource_size(&nsio->res);
-		end_trunc = start + size - PHYS_SECTION_ALIGN_DOWN(start + size);
+		end_trunc = start + size - phys_pmem_align_down(nd_pfn,
+				start + size);
 	}
 
 	if (start_pad + end_trunc)
-		dev_info(&nd_pfn->dev, "%s section collision, truncate %d bytes\n",
+		dev_info(&nd_pfn->dev, "%s alignment collision, truncate %d bytes\n",
 				dev_name(&ndns->dev), start_pad + end_trunc);
 
 	/*
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index bc6ed52a39b9..61054f12be7c 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -46,6 +46,7 @@
 #define REPEAT_BYTE(x)	((~0ul / 0xff) * (x))
 
 #define ALIGN(x, a)		__ALIGN_KERNEL((x), (a))
+#define ALIGN_DOWN(x, a)	__ALIGN_KERNEL((x) - ((a) - 1), (a))
 #define __ALIGN_MASK(x, mask)	__ALIGN_KERNEL_MASK((x), (mask))
 #define PTR_ALIGN(p, a)		((typeof(p))ALIGN((unsigned long)(p), (a)))
 #define IS_ALIGNED(x, a)		(((x) & ((typeof(x))(a) - 1)) == 0)

^ permalink raw reply related	[flat|nested] 3+ messages in thread
end of thread, other threads:[~2018-02-26 19:58 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-12-27 15:22 Patch "libnvdimm, dax: fix 1GB-aligned namespaces vs physical misalignment" has been added to the 4.9-stable tree gregkh
2017-12-27 15:46 ` Greg KH
  -- strict thread matches above, loose matches on Subject: below --
2018-02-23 22:06 [4.9-stable PATCH 08/11] libnvdimm, dax: fix 1GB-aligned namespaces vs physical misalignment Dan Williams
2018-02-26 19:58 ` Patch "libnvdimm, dax: fix 1GB-aligned namespaces vs physical misalignment" has been added to the 4.9-stable tree gregkh

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).