linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Dan Williams <dan.j.williams@intel.com>
To: linux-mm@kvack.org
Cc: david@fromorbit.com, hch@lst.de, nvdimm@lists.linux.dev,
	akpm@linux-foundation.org, linux-fsdevel@vger.kernel.org
Subject: [PATCH v3 15/25] libnvdimm/pmem: Support pmem block devices without dax
Date: Fri, 14 Oct 2022 16:58:26 -0700	[thread overview]
Message-ID: <166579190607.2236710.1230996282258115812.stgit@dwillia2-xfh.jf.intel.com> (raw)
In-Reply-To: <166579181584.2236710.17813547487183983273.stgit@dwillia2-xfh.jf.intel.com>

In preparation for CONFIG_DAX growing a CONFIG_MMU dependency add
support for pmem to skip dax-device registration in the CONFIG_DAX=n
case.

alloc_dax() returns NULL in the CONFIG_DAX=n case, ERR_PTR() in the
failure case, and a dax-device in the success case.

dax_remove_host(), kill_dax() and put_dax() are safe to call if
setup_dax() returns 0 because it suceeded, or 0 because CONFIG_DAX=n.

Signed-off-by: Dan Williams <dan.j.williams@intel.com>
---
 drivers/nvdimm/Kconfig |    2 +-
 drivers/nvdimm/pmem.c  |   47 ++++++++++++++++++++++++++++++-----------------
 2 files changed, 31 insertions(+), 18 deletions(-)

diff --git a/drivers/nvdimm/Kconfig b/drivers/nvdimm/Kconfig
index 5a29046e3319..027acca1bac4 100644
--- a/drivers/nvdimm/Kconfig
+++ b/drivers/nvdimm/Kconfig
@@ -19,7 +19,7 @@ if LIBNVDIMM
 config BLK_DEV_PMEM
 	tristate "PMEM: Persistent memory block device support"
 	default LIBNVDIMM
-	select DAX
+	select DAX if MMU
 	select ND_BTT if BTT
 	select ND_PFN if NVDIMM_PFN
 	help
diff --git a/drivers/nvdimm/pmem.c b/drivers/nvdimm/pmem.c
index 7e88cd242380..068183ee9bf6 100644
--- a/drivers/nvdimm/pmem.c
+++ b/drivers/nvdimm/pmem.c
@@ -468,6 +468,32 @@ static const struct dev_pagemap_ops fsdax_pagemap_ops = {
 	.memory_failure		= pmem_pagemap_memory_failure,
 };
 
+static int setup_dax(struct pmem_device *pmem, struct gendisk *disk,
+		     struct nd_region *nd_region)
+{
+	struct dax_device *dax_dev;
+	int rc;
+
+	dax_dev = alloc_dax(pmem, &pmem_dax_ops);
+	if (IS_ERR(dax_dev))
+		return PTR_ERR(dax_dev);
+	if (!dax_dev)
+		return 0;
+	set_dax_nocache(dax_dev);
+	set_dax_nomc(dax_dev);
+	if (is_nvdimm_sync(nd_region))
+		set_dax_synchronous(dax_dev);
+	rc = dax_add_host(dax_dev, disk);
+	if (rc) {
+		kill_dax(dax_dev);
+		put_dax(dax_dev);
+		return rc;
+	}
+	dax_write_cache(dax_dev, nvdimm_has_cache(nd_region));
+	pmem->dax_dev = dax_dev;
+	return 0;
+}
+
 static int pmem_attach_disk(struct device *dev,
 		struct nd_namespace_common *ndns)
 {
@@ -477,7 +503,6 @@ static int pmem_attach_disk(struct device *dev,
 	struct resource *res = &nsio->res;
 	struct range bb_range;
 	struct nd_pfn *nd_pfn = NULL;
-	struct dax_device *dax_dev;
 	struct nd_pfn_sb *pfn_sb;
 	struct pmem_device *pmem;
 	struct request_queue *q;
@@ -578,24 +603,13 @@ static int pmem_attach_disk(struct device *dev,
 	nvdimm_badblocks_populate(nd_region, &pmem->bb, &bb_range);
 	disk->bb = &pmem->bb;
 
-	dax_dev = alloc_dax(pmem, &pmem_dax_ops);
-	if (IS_ERR(dax_dev)) {
-		rc = PTR_ERR(dax_dev);
-		goto out;
-	}
-	set_dax_nocache(dax_dev);
-	set_dax_nomc(dax_dev);
-	if (is_nvdimm_sync(nd_region))
-		set_dax_synchronous(dax_dev);
-	rc = dax_add_host(dax_dev, disk);
+	rc = setup_dax(pmem, disk, nd_region);
 	if (rc)
-		goto out_cleanup_dax;
-	dax_write_cache(dax_dev, nvdimm_has_cache(nd_region));
-	pmem->dax_dev = dax_dev;
+		goto out;
 
 	rc = device_add_disk(dev, disk, pmem_attribute_groups);
 	if (rc)
-		goto out_remove_host;
+		goto out_dax;
 	if (devm_add_action_or_reset(dev, pmem_release_disk, pmem))
 		return -ENOMEM;
 
@@ -607,9 +621,8 @@ static int pmem_attach_disk(struct device *dev,
 		dev_warn(dev, "'badblocks' notification disabled\n");
 	return 0;
 
-out_remove_host:
+out_dax:
 	dax_remove_host(pmem->disk);
-out_cleanup_dax:
 	kill_dax(pmem->dax_dev);
 	put_dax(pmem->dax_dev);
 out:



  parent reply	other threads:[~2022-10-14 23:58 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-10-14 23:56 [PATCH v3 00/25] Fix the DAX-gup mistake Dan Williams
2022-10-14 23:57 ` [PATCH v3 01/25] fsdax: Wait on @page not @page->_refcount Dan Williams
2022-10-14 23:57 ` [PATCH v3 02/25] fsdax: Use dax_page_idle() to document DAX busy page checking Dan Williams
2022-10-14 23:57 ` [PATCH v3 03/25] fsdax: Include unmapped inodes for page-idle detection Dan Williams
2022-10-14 23:57 ` [PATCH v3 04/25] fsdax: Introduce dax_zap_mappings() Dan Williams
2022-11-02 13:04   ` Aneesh Kumar K.V
2022-10-14 23:57 ` [PATCH v3 05/25] fsdax: Wait for pinned pages during truncate_inode_pages_final() Dan Williams
2022-10-14 23:57 ` [PATCH v3 06/25] fsdax: Validate DAX layouts broken before truncate Dan Williams
2022-10-14 23:57 ` [PATCH v3 07/25] fsdax: Hold dax lock over mapping insertion Dan Williams
2022-10-17 19:31   ` Jason Gunthorpe
2022-10-17 20:17     ` Dan Williams
2022-10-18  5:26       ` Christoph Hellwig
2022-10-18 17:30         ` Dan Williams
2022-10-14 23:57 ` [PATCH v3 08/25] fsdax: Update dax_insert_entry() calling convention to return an error Dan Williams
2022-10-14 23:57 ` [PATCH v3 09/25] fsdax: Rework for_each_mapped_pfn() to dax_for_each_folio() Dan Williams
2022-10-14 23:57 ` [PATCH v3 10/25] fsdax: Introduce pgmap_request_folios() Dan Williams
2022-10-17  6:31   ` Alistair Popple
2022-10-17 20:06     ` Dan Williams
2022-10-17 20:11       ` Jason Gunthorpe
2022-10-17 20:51         ` Dan Williams
2022-10-17 23:57           ` Jason Gunthorpe
2022-10-18  0:19             ` Dan Williams
2022-10-17 19:41   ` Jason Gunthorpe
2022-10-14 23:58 ` [PATCH v3 11/25] fsdax: Rework dax_insert_entry() calling convention Dan Williams
2022-10-14 23:58 ` [PATCH v3 12/25] fsdax: Cleanup dax_associate_entry() Dan Williams
2022-10-14 23:58 ` [PATCH v3 13/25] devdax: Minor warning fixups Dan Williams
2022-10-14 23:58 ` [PATCH v3 14/25] devdax: Fix sparse lock imbalance warning Dan Williams
2022-10-14 23:58 ` Dan Williams [this message]
2022-10-14 23:58 ` [PATCH v3 16/25] devdax: Move address_space helpers to the DAX core Dan Williams
2022-10-14 23:58 ` [PATCH v3 17/25] devdax: Sparse fixes for xarray locking Dan Williams
2022-10-14 23:58 ` [PATCH v3 18/25] devdax: Sparse fixes for vmfault_t / dax-entry conversions Dan Williams
2022-10-14 23:58 ` [PATCH v3 19/25] devdax: Sparse fixes for vm_fault_t in tracepoints Dan Williams
2022-10-14 23:58 ` [PATCH v3 20/25] devdax: add PUD support to the DAX mapping infrastructure Dan Williams
2022-10-14 23:59 ` [PATCH v3 21/25] devdax: Use dax_insert_entry() + dax_delete_mapping_entry() Dan Williams
2022-10-14 23:59 ` [PATCH v3 22/25] mm/memremap_pages: Replace zone_device_page_init() with pgmap_request_folios() Dan Williams
2022-10-17 19:17   ` Lyude Paul
2022-10-14 23:59 ` [PATCH v3 23/25] mm/memremap_pages: Initialize all ZONE_DEVICE pages to start at refcount 0 Dan Williams
2022-10-17  7:04   ` Alistair Popple
2022-10-17 19:48   ` Jason Gunthorpe
2022-10-14 23:59 ` [PATCH v3 24/25] mm/meremap_pages: Delete put_devmap_managed_page_refs() Dan Williams
2022-10-17  7:08   ` Alistair Popple
2022-10-14 23:59 ` [PATCH v3 25/25] mm/gup: Drop DAX pgmap accounting Dan Williams

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=166579190607.2236710.1230996282258115812.stgit@dwillia2-xfh.jf.intel.com \
    --to=dan.j.williams@intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=david@fromorbit.com \
    --cc=hch@lst.de \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=nvdimm@lists.linux.dev \
    /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: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox