From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jia He Date: Thu, 09 Jul 2020 02:06:28 +0000 Subject: [PATCH v3 5/6] device-dax: use fallback nid when numa_node is invalid Message-Id: <20200709020629.91671-6-justin.he@arm.com> List-Id: References: <20200709020629.91671-1-justin.he@arm.com> In-Reply-To: <20200709020629.91671-1-justin.he@arm.com> MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable To: Catalin Marinas , Will Deacon , Tony Luck , Fenghua Yu , Yoshinori Sato , Rich Felker , Dave Hansen , Andy Lutomirski , Peter Zijlstra , Thomas Gleixner , Ingo Molnar , Borislav Petkov , David Hildenbrand Cc: Kaly Xin , Michal Hocko , Dave Jiang , Baoquan He , linux-sh@vger.kernel.org, Vishal Verma , Masahiro Yamada , x86@kernel.org, Chuhong Yuan , linux-kernel@vger.kernel.org, Mike Rapoport , linux-mm@kvack.org, linux-nvdimm@lists.01.org, Jonathan Cameron , "H. Peter Anvin" , linux-ia64@vger.kernel.org, Dan Williams , Logan Gunthorpe , Andrew Morton , linux-arm-kernel@lists.infradead.org, Jia He numa_off is set unconditionally at the end of dummy_numa_init(), even with a fake numa node. ACPI detects node id as NUMA_NO_NODE(-1) in acpi_map_pxm_to_node() because it regards numa_off as turning off the numa node. Hence dev_dax->target_node is NUMA_NO_NODE on arm64 with fake numa. Without this patch, pmem can't be probed as a RAM device on arm64 if SRAT t= able isn't present: $ndctl create-namespace -fe namespace0.0 --mode=DEvdax --map=DEv -s 1g -a 6= 4K kmem dax0.0: rejecting DAX region [mem 0x240400000-0x2bfffffff] with invali= d node: -1 kmem: probe of dax0.0 failed with error -22 This fixes it by using fallback memory_add_physaddr_to_nid() as nid. Suggested-by: David Hildenbrand Signed-off-by: Jia He --- drivers/dax/kmem.c | 21 +++++++++++++-------- 1 file changed, 13 insertions(+), 8 deletions(-) diff --git a/drivers/dax/kmem.c b/drivers/dax/kmem.c index 275aa5f87399..218f66057994 100644 --- a/drivers/dax/kmem.c +++ b/drivers/dax/kmem.c @@ -31,22 +31,23 @@ int dev_dax_kmem_probe(struct device *dev) int numa_node; int rc; =20 + /* Hotplug starting at the beginning of the next block: */ + kmem_start =3D ALIGN(res->start, memory_block_size_bytes()); + /* * Ensure good NUMA information for the persistent memory. * Without this check, there is a risk that slow memory * could be mixed in a node with faster memory, causing - * unavoidable performance issues. + * unavoidable performance issues. Furthermore, fallback node + * id can be used when numa_node is invalid. */ numa_node =3D dev_dax->target_node; if (numa_node < 0) { - dev_warn(dev, "rejecting DAX region %pR with invalid node: %d\n", - res, numa_node); - return -EINVAL; + numa_node =3D memory_add_physaddr_to_nid(kmem_start); + dev_info(dev, "using nid %d for DAX region with undefined nid %pR\n", + numa_node, res); } =20 - /* Hotplug starting at the beginning of the next block: */ - kmem_start =3D ALIGN(res->start, memory_block_size_bytes()); - kmem_size =3D resource_size(res); /* Adjust the size down to compensate for moving up kmem_start: */ kmem_size -=3D kmem_start - res->start; @@ -100,15 +101,19 @@ static int dev_dax_kmem_remove(struct device *dev) resource_size_t kmem_start =3D res->start; resource_size_t kmem_size =3D resource_size(res); const char *res_name =3D res->name; + int numa_node =3D dev_dax->target_node; int rc; =20 + if (numa_node < 0) + numa_node =3D memory_add_physaddr_to_nid(kmem_start); + /* * We have one shot for removing memory, if some memory blocks were not * offline prior to calling this function remove_memory() will fail, and * there is no way to hotremove this memory until reboot because device * unbind will succeed even if we return failure. */ - rc =3D remove_memory(dev_dax->target_node, kmem_start, kmem_size); + rc =3D remove_memory(numa_node, kmem_start, kmem_size); if (rc) { any_hotremove_failed =3D true; dev_err(dev, --=20 2.17.1