From: Dan Williams <dan.j.williams@intel.com> To: linux-nvdimm@lists.01.org Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>, Brice Goglin <Brice.Goglin@inria.fr>, Ard Biesheuvel <ard.biesheuvel@linaro.org>, "Rafael J. Wysocki" <rjw@rjwysocki.net>, Catalin Marinas <catalin.marinas@arm.com>, Will Deacon <will@kernel.org>, joao.m.martins@oracle.com, peterz@infradead.org, dave.hansen@linux.intel.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org, linux-acpi@vger.kernel.org, hch@lst.de Subject: [PATCH v2 10/22] ACPI: HMAT: Attach a device for each soft-reserved range Date: Sun, 12 Jul 2020 09:26:59 -0700 [thread overview] Message-ID: <159457121992.754248.18124727141742055701.stgit@dwillia2-desk3.amr.corp.intel.com> (raw) In-Reply-To: <159457116473.754248.7879464730875147365.stgit@dwillia2-desk3.amr.corp.intel.com> The hmem enabling in commit 'cf8741ac57ed ("ACPI: NUMA: HMAT: Register "soft reserved" memory as an "hmem" device")' only registered ranges to the hmem driver for each soft-reservation that also appeared in the HMAT. While this is meant to encourage platform firmware to "do the right thing" and publish an HMAT, the corollary is that platforms that fail to publish an accurate HMAT will strand memory from Linux usage. Additionally, the "efi_fake_mem" kernel command line option enabling will strand memory by default without an HMAT. Arrange for "soft reserved" memory that goes unclaimed by HMAT entries to be published as raw resource ranges for the hmem driver to consume. Include a module parameter to disable either this fallback behavior, or the hmat enabling from creating hmem devices. The module parameter requires the hmem device enabling to have unique name in the module namespace: "device_hmem". The driver depends on the architecture providing phys_to_target_node() which is only x86 via numa_meminfo() and arm64 via a generic memblock implementation. Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com> Cc: Brice Goglin <Brice.Goglin@inria.fr> Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Cc: Jeff Moyer <jmoyer@redhat.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will@kernel.org> Reviewed-by: Joao Martins <joao.m.martins@oracle.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> --- drivers/dax/Kconfig | 2 ++ drivers/dax/hmem/Makefile | 3 ++- drivers/dax/hmem/device.c | 35 +++++++++++++++++++++++++++++++++++ 3 files changed, 39 insertions(+), 1 deletion(-) diff --git a/drivers/dax/Kconfig b/drivers/dax/Kconfig index a229f45d34aa..557483ed2e9c 100644 --- a/drivers/dax/Kconfig +++ b/drivers/dax/Kconfig @@ -50,6 +50,8 @@ config DEV_DAX_HMEM config DEV_DAX_HMEM_DEVICES depends on DEV_DAX_HMEM + # requires phys_to_target_node() support + depends on X86_NUMA_KEEP_MEMINFO || MEMBLOCK_NUMA_INFO def_bool y config DEV_DAX_KMEM diff --git a/drivers/dax/hmem/Makefile b/drivers/dax/hmem/Makefile index a9d353d0c9ed..57377b4c3d47 100644 --- a/drivers/dax/hmem/Makefile +++ b/drivers/dax/hmem/Makefile @@ -1,5 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 obj-$(CONFIG_DEV_DAX_HMEM) += dax_hmem.o -obj-$(CONFIG_DEV_DAX_HMEM_DEVICES) += device.o +obj-$(CONFIG_DEV_DAX_HMEM_DEVICES) += device_hmem.o +device_hmem-y := device.o dax_hmem-y := hmem.o diff --git a/drivers/dax/hmem/device.c b/drivers/dax/hmem/device.c index b9dd6b27745c..cb6401c9e9a4 100644 --- a/drivers/dax/hmem/device.c +++ b/drivers/dax/hmem/device.c @@ -5,6 +5,9 @@ #include <linux/dax.h> #include <linux/mm.h> +static bool nohmem; +module_param_named(disable, nohmem, bool, 0444); + void hmem_register_device(int target_nid, struct resource *r) { /* define a clean / non-busy resource for the platform device */ @@ -17,6 +20,9 @@ void hmem_register_device(int target_nid, struct resource *r) struct memregion_info info; int rc, id; + if (nohmem) + return; + rc = region_intersects(res.start, resource_size(&res), IORESOURCE_MEM, IORES_DESC_SOFT_RESERVED); if (rc != REGION_INTERSECTS) @@ -63,3 +69,32 @@ void hmem_register_device(int target_nid, struct resource *r) out_pdev: memregion_free(id); } + +static __init int hmem_register_one(struct resource *res, void *data) +{ + /* + * If the resource is not a top-level resource it was already + * assigned to a device by the HMAT parsing. + */ + if (res->parent != &iomem_resource) { + pr_info("HMEM: skip %pr, already claimed\n", res); + return 0; + } + + hmem_register_device(phys_to_target_node(res->start), res); + + return 0; +} + +static __init int hmem_init(void) +{ + walk_iomem_res_desc(IORES_DESC_SOFT_RESERVED, + IORESOURCE_MEM, 0, -1, NULL, hmem_register_one); + return 0; +} + +/* + * As this is a fallback for address ranges unclaimed by the ACPI HMAT + * parsing it must be at an initcall level greater than hmat_init(). + */ +late_initcall(hmem_init); _______________________________________________ Linux-nvdimm mailing list -- linux-nvdimm@lists.01.org To unsubscribe send an email to linux-nvdimm-leave@lists.01.org
WARNING: multiple messages have this Message-ID (diff)
From: Dan Williams <dan.j.williams@intel.com> To: linux-nvdimm@lists.01.org Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com>, Brice Goglin <Brice.Goglin@inria.fr>, Ard Biesheuvel <ard.biesheuvel@linaro.org>, "Rafael J. Wysocki" <rjw@rjwysocki.net>, Jeff Moyer <jmoyer@redhat.com>, Catalin Marinas <catalin.marinas@arm.com>, Will Deacon <will@kernel.org>, Joao Martins <joao.m.martins@oracle.com>, peterz@infradead.org, vishal.l.verma@intel.com, dave.hansen@linux.intel.com, linux-mm@kvack.org, linux-kernel@vger.kernel.org, linux-acpi@vger.kernel.org, hch@lst.de, joao.m.martins@oracle.com Subject: [PATCH v2 10/22] ACPI: HMAT: Attach a device for each soft-reserved range Date: Sun, 12 Jul 2020 09:26:59 -0700 [thread overview] Message-ID: <159457121992.754248.18124727141742055701.stgit@dwillia2-desk3.amr.corp.intel.com> (raw) In-Reply-To: <159457116473.754248.7879464730875147365.stgit@dwillia2-desk3.amr.corp.intel.com> The hmem enabling in commit 'cf8741ac57ed ("ACPI: NUMA: HMAT: Register "soft reserved" memory as an "hmem" device")' only registered ranges to the hmem driver for each soft-reservation that also appeared in the HMAT. While this is meant to encourage platform firmware to "do the right thing" and publish an HMAT, the corollary is that platforms that fail to publish an accurate HMAT will strand memory from Linux usage. Additionally, the "efi_fake_mem" kernel command line option enabling will strand memory by default without an HMAT. Arrange for "soft reserved" memory that goes unclaimed by HMAT entries to be published as raw resource ranges for the hmem driver to consume. Include a module parameter to disable either this fallback behavior, or the hmat enabling from creating hmem devices. The module parameter requires the hmem device enabling to have unique name in the module namespace: "device_hmem". The driver depends on the architecture providing phys_to_target_node() which is only x86 via numa_meminfo() and arm64 via a generic memblock implementation. Cc: Jonathan Cameron <Jonathan.Cameron@huawei.com> Cc: Brice Goglin <Brice.Goglin@inria.fr> Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org> Cc: "Rafael J. Wysocki" <rjw@rjwysocki.net> Cc: Jeff Moyer <jmoyer@redhat.com> Cc: Catalin Marinas <catalin.marinas@arm.com> Cc: Will Deacon <will@kernel.org> Reviewed-by: Joao Martins <joao.m.martins@oracle.com> Signed-off-by: Dan Williams <dan.j.williams@intel.com> --- drivers/dax/Kconfig | 2 ++ drivers/dax/hmem/Makefile | 3 ++- drivers/dax/hmem/device.c | 35 +++++++++++++++++++++++++++++++++++ 3 files changed, 39 insertions(+), 1 deletion(-) diff --git a/drivers/dax/Kconfig b/drivers/dax/Kconfig index a229f45d34aa..557483ed2e9c 100644 --- a/drivers/dax/Kconfig +++ b/drivers/dax/Kconfig @@ -50,6 +50,8 @@ config DEV_DAX_HMEM config DEV_DAX_HMEM_DEVICES depends on DEV_DAX_HMEM + # requires phys_to_target_node() support + depends on X86_NUMA_KEEP_MEMINFO || MEMBLOCK_NUMA_INFO def_bool y config DEV_DAX_KMEM diff --git a/drivers/dax/hmem/Makefile b/drivers/dax/hmem/Makefile index a9d353d0c9ed..57377b4c3d47 100644 --- a/drivers/dax/hmem/Makefile +++ b/drivers/dax/hmem/Makefile @@ -1,5 +1,6 @@ # SPDX-License-Identifier: GPL-2.0 obj-$(CONFIG_DEV_DAX_HMEM) += dax_hmem.o -obj-$(CONFIG_DEV_DAX_HMEM_DEVICES) += device.o +obj-$(CONFIG_DEV_DAX_HMEM_DEVICES) += device_hmem.o +device_hmem-y := device.o dax_hmem-y := hmem.o diff --git a/drivers/dax/hmem/device.c b/drivers/dax/hmem/device.c index b9dd6b27745c..cb6401c9e9a4 100644 --- a/drivers/dax/hmem/device.c +++ b/drivers/dax/hmem/device.c @@ -5,6 +5,9 @@ #include <linux/dax.h> #include <linux/mm.h> +static bool nohmem; +module_param_named(disable, nohmem, bool, 0444); + void hmem_register_device(int target_nid, struct resource *r) { /* define a clean / non-busy resource for the platform device */ @@ -17,6 +20,9 @@ void hmem_register_device(int target_nid, struct resource *r) struct memregion_info info; int rc, id; + if (nohmem) + return; + rc = region_intersects(res.start, resource_size(&res), IORESOURCE_MEM, IORES_DESC_SOFT_RESERVED); if (rc != REGION_INTERSECTS) @@ -63,3 +69,32 @@ void hmem_register_device(int target_nid, struct resource *r) out_pdev: memregion_free(id); } + +static __init int hmem_register_one(struct resource *res, void *data) +{ + /* + * If the resource is not a top-level resource it was already + * assigned to a device by the HMAT parsing. + */ + if (res->parent != &iomem_resource) { + pr_info("HMEM: skip %pr, already claimed\n", res); + return 0; + } + + hmem_register_device(phys_to_target_node(res->start), res); + + return 0; +} + +static __init int hmem_init(void) +{ + walk_iomem_res_desc(IORES_DESC_SOFT_RESERVED, + IORESOURCE_MEM, 0, -1, NULL, hmem_register_one); + return 0; +} + +/* + * As this is a fallback for address ranges unclaimed by the ACPI HMAT + * parsing it must be at an initcall level greater than hmat_init(). + */ +late_initcall(hmem_init);
next prev parent reply other threads:[~2020-07-12 16:43 UTC|newest] Thread overview: 100+ messages / expand[flat|nested] mbox.gz Atom feed top 2020-07-12 16:26 [PATCH v2 00/22] device-dax: Support sub-dividing soft-reserved ranges Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-12 16:26 ` [PATCH v2 01/22] x86/numa: Cleanup configuration dependent command-line options Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-12 16:26 ` [PATCH v2 02/22] x86/numa: Add 'nohmat' option Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-12 16:58 ` Randy Dunlap 2020-07-12 16:58 ` Randy Dunlap 2020-07-12 16:26 ` [PATCH v2 03/22] efi/fake_mem: Arrange for a resource entry per efi_fake_mem instance Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-12 16:26 ` [PATCH v2 04/22] ACPI: HMAT: Refactor hmat_register_target_device to hmem_register_device Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-12 16:26 ` [PATCH v2 05/22] resource: Report parent to walk_iomem_res_desc() callback Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-12 16:26 ` [PATCH v2 06/22] x86: Move NUMA_KEEP_MEMINFO and related definition to x86-internals Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-12 16:26 ` [PATCH v2 07/22] numa: Introduce a generic memory_add_physaddr_to_nid() Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-13 6:58 ` Mike Rapoport 2020-07-13 6:58 ` Mike Rapoport 2020-07-13 15:42 ` Dan Williams 2020-07-13 15:42 ` Dan Williams 2020-07-13 15:42 ` Dan Williams 2020-07-12 16:26 ` [PATCH v2 08/22] memblock: Introduce a generic phys_addr_to_target_node() Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-13 7:03 ` Mike Rapoport 2020-07-13 7:03 ` Mike Rapoport 2020-07-13 15:48 ` Dan Williams 2020-07-13 15:48 ` Dan Williams 2020-07-13 15:48 ` Dan Williams 2020-07-14 1:36 ` Justin He 2020-07-14 1:36 ` Justin He 2020-07-12 16:26 ` [PATCH v2 09/22] arm64: Convert to generic memblock for numa-info Dan Williams 2020-07-12 16:26 ` Dan Williams 2020-07-12 16:26 ` Dan Williams [this message] 2020-07-12 16:26 ` [PATCH v2 10/22] ACPI: HMAT: Attach a device for each soft-reserved range Dan Williams 2020-07-12 16:27 ` [PATCH v2 11/22] device-dax: Drop the dax_region.pfn_flags attribute Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 12/22] device-dax: Move instance creation parameters to 'struct dev_dax_data' Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 13/22] device-dax: Make pgmap optional for instance creation Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 14/22] device-dax: Kill dax_kmem_res Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 15/22] device-dax: Add an allocation interface for device-dax instances Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 16/22] device-dax: Introduce 'seed' devices Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 17/22] drivers/base: Make device_find_child_by_name() compatible with sysfs inputs Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 17:09 ` Greg Kroah-Hartman 2020-07-12 17:09 ` Greg Kroah-Hartman 2020-07-13 15:39 ` Dan Williams 2020-07-13 15:39 ` Dan Williams 2020-07-13 15:39 ` Dan Williams 2020-07-13 15:52 ` Greg Kroah-Hartman 2020-07-13 15:52 ` Greg Kroah-Hartman 2020-07-13 16:09 ` Dan Williams 2020-07-13 16:09 ` Dan Williams 2020-07-13 16:09 ` Dan Williams 2020-07-13 16:12 ` Greg Kroah-Hartman 2020-07-13 16:12 ` Greg Kroah-Hartman 2020-07-13 16:36 ` Dan Williams 2020-07-13 16:36 ` Dan Williams 2020-07-13 16:36 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 18/22] device-dax: Add resize support Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 19/22] mm/memremap_pages: Convert to 'struct range' Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-13 16:36 ` Ralph Campbell 2020-07-13 16:36 ` Ralph Campbell 2020-07-13 16:54 ` Dan Williams 2020-07-13 16:54 ` Dan Williams 2020-07-13 16:54 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 20/22] mm/memremap_pages: Support multiple ranges per invocation Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 16:27 ` [PATCH v2 21/22] device-dax: Add dis-contiguous resource support Dan Williams 2020-07-12 16:27 ` Dan Williams 2020-07-12 16:28 ` [PATCH v2 22/22] device-dax: Introduce 'mapping' devices Dan Williams 2020-07-12 16:28 ` Dan Williams 2020-07-16 13:18 ` Joao Martins 2020-07-16 13:18 ` Joao Martins 2020-07-16 16:00 ` Dan Williams 2020-07-16 16:00 ` Dan Williams 2020-07-16 16:00 ` Dan Williams 2020-07-16 19:04 ` Joao Martins 2020-07-16 19:04 ` Joao Martins 2020-07-16 17:29 ` [PATCH v1 0/4] device-dax: Further improvements to subdivision Joao Martins 2020-07-16 17:29 ` Joao Martins 2020-07-16 17:29 ` [PATCH v1 1/4] device-dax: Make align a per-device property Joao Martins 2020-07-16 17:29 ` Joao Martins 2020-07-16 17:29 ` [PATCH v1 2/4] device-dax: Add an 'align' attribute Joao Martins 2020-07-16 17:29 ` Joao Martins 2020-07-31 14:52 ` Dan Williams 2020-07-31 14:52 ` Dan Williams 2020-07-31 14:52 ` Dan Williams 2020-07-16 17:29 ` [PATCH v1 3/4] dax/hmem: Introduce dax_hmem.region_idle parameter Joao Martins 2020-07-16 17:29 ` Joao Martins 2020-07-16 17:29 ` [PATCH v1 4/4] device-dax: Add a range mapping allocation attribute Joao Martins 2020-07-16 17:29 ` Joao Martins
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=159457121992.754248.18124727141742055701.stgit@dwillia2-desk3.amr.corp.intel.com \ --to=dan.j.williams@intel.com \ --cc=Brice.Goglin@inria.fr \ --cc=Jonathan.Cameron@huawei.com \ --cc=ard.biesheuvel@linaro.org \ --cc=catalin.marinas@arm.com \ --cc=dave.hansen@linux.intel.com \ --cc=hch@lst.de \ --cc=joao.m.martins@oracle.com \ --cc=linux-acpi@vger.kernel.org \ --cc=linux-kernel@vger.kernel.org \ --cc=linux-mm@kvack.org \ --cc=linux-nvdimm@lists.01.org \ --cc=peterz@infradead.org \ --cc=rjw@rjwysocki.net \ --cc=will@kernel.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes, see mirroring instructions on how to clone and mirror all data and code used by this external index.