x86, pmem: fix broken __copy_user_nocache cache-bypass assumptions
authorDan Williams <dan.j.williams@intel.com>
Thu, 6 Apr 2017 16:04:31 +0000 (09:04 -0700)
committerDan Williams <dan.j.williams@intel.com>
Wed, 12 Apr 2017 20:45:18 +0000 (13:45 -0700)
commit11e63f6d920d6f2dfd3cd421e939a4aec9a58dcd
tree83daf277166209e5b0bc01d0047e08ce3c4c17bb
parent956a4cd2c957acf638ff29951aabaa9d8e92bbc2
x86, pmem: fix broken __copy_user_nocache cache-bypass assumptions

Before we rework the "pmem api" to stop abusing __copy_user_nocache()
for memcpy_to_pmem() we need to fix cases where we may strand dirty data
in the cpu cache. The problem occurs when copy_from_iter_pmem() is used
for arbitrary data transfers from userspace. There is no guarantee that
these transfers, performed by dax_iomap_actor(), will have aligned
destinations or aligned transfer lengths. Backstop the usage
__copy_user_nocache() with explicit cache management in these unaligned
cases.

Yes, copy_from_iter_pmem() is now too big for an inline, but addressing
that is saved for a later patch that moves the entirety of the "pmem
api" into the pmem driver directly.

Fixes: 5de490daec8b ("pmem: add copy_from_iter_pmem() and clear_pmem()")
Cc: <stable@vger.kernel.org>
Cc: <x86@kernel.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Jeff Moyer <jmoyer@redhat.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Matthew Wilcox <mawilcox@microsoft.com>
Reviewed-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Signed-off-by: Toshi Kani <toshi.kani@hpe.com>
Signed-off-by: Dan Williams <dan.j.williams@intel.com>
arch/x86/include/asm/pmem.h