linux-kernel - Re: [PATCH v2 3/6] ext4: Use ext4_get_block

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20150703184824.GA13681@linux.intel.com>
Date:	Fri, 3 Jul 2015 14:48:24 -0400
From:	Matthew Wilcox <willy@...ux.intel.com>
To:	Theodore Ts'o <tytso@....edu>,
	Matthew Wilcox <matthew.r.wilcox@...el.com>,
	linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
	Alexander Viro <viro@...iv.linux.org.uk>,
	Andreas Dilger <adilger.kernel@...ger.ca>,
	linux-ext4@...r.kernel.org
Subject: Re: [PATCH v2 3/6] ext4: Use ext4_get_block_write() for DAX

On Fri, Jul 03, 2015 at 02:30:27PM -0400, Theodore Ts'o wrote:
> On Fri, Jul 03, 2015 at 10:40:40AM -0400, Matthew Wilcox wrote:
> > From: Matthew Wilcox <willy@...ux.intel.com>
> > 
> > DAX relies on the get_block function either zeroing newly allocated blocks
> > before they're findable by subsequent calls to get_block, or marking newly
> > allocated blocks as unwritten.  ext4_get_block() cannot create unwritten
> > extents, but ext4_get_block_write() can.
> 
> To be clear, this patch has no prerequistes or dependencies, right?
> That is, it would be fine if I take this through the ext4 git tree?
> Or is there a reason or a preference for carrying this patch
> somewhere else?

Right, no dependencies or prerequisites, completely independent of all
the other patches.

> Also, is there a way I can test the DAX functionality in ext4 using
> KVM?  If so, can you give me a cheat sheet about how I can do that?

I don't use KVM, but I can tell you what I do ... (additional explanation
added, not for Ted's benefit, but because people less familiar with
Linux than Ted is may happen upon this email for their own purposes).

In /etc/default/grub, I have this line:

GRUB_CMDLINE_LINUX="memmap=4G!4G"

        memmap=nn[KMG]!ss[KMG]
                        [KNL,X86] Mark specific memory as protected.
                        Region of memory to be used, from ss to ss+nn.
                        The memory region may be marked as e820 type 12 (0xc)
                        and is NVDIMM or ADR memory.


In my kernel config, I have:

CONFIG_X86_PMEM_LEGACY=y
CONFIG_LIBNVDIMM=y
CONFIG_BLK_DEV_PMEM=m

At boot, I "modprobe pmem".  On the desktop-class system I'm using as
my development machine, the BIOS doesn't clear RAM between boots (only
power cycles), so the partition table and ext4 filesystem stays good,
and all I have to do is:

mount -odax /dev/pmem0p1 /mnt/ram0/

Also my xfstests local.config:

TEST_DEV=/dev/pmem0p1
TEST_DIR=/mnt/ram0
SCRATCH_DEV=/dev/pmem0p2
SCRATCH_MNT=/mnt/ram1
TEST_FS_MOUNT_OPTS="-o dax"
EXT_MOUNT_OPTIONS="-o dax"
MKFS_OPTIONS="-b4096"

Hope I didn't forget anything.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/