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]
Date: Wed, 14 Feb 2024 13:02:21 +0000
From: John Garry <john.g.garry@...cle.com>
To: Nilay Shroff <nilay@...ux.ibm.com>
Cc: alan.adamson@...cle.com, axboe@...nel.dk, brauner@...nel.org,
        bvanassche@....org, dchinner@...hat.com, djwong@...nel.org, hch@....de,
        jack@...e.cz, jbongio@...gle.com, jejb@...ux.ibm.com,
        kbusch@...nel.org, linux-block@...r.kernel.org,
        linux-fsdevel@...r.kernel.org, linux-kernel@...r.kernel.org,
        linux-nvme@...ts.infradead.org, linux-scsi@...r.kernel.org,
        linux-xfs@...r.kernel.org, martin.petersen@...cle.com,
        ming.lei@...hat.com, ojaswin@...ux.ibm.com, sagi@...mberg.me,
        tytso@....edu, viro@...iv.linux.org.uk
Subject: Re: [PATCH v3 14/15] nvme: Support atomic writes

On 14/02/2024 12:27, Nilay Shroff wrote:
>> Support reading atomic write registers to fill in request_queue
> 
>> properties.
> 
> 
> 
>> Use following method to calculate limits:
> 
>> atomic_write_max_bytes = flp2(NAWUPF ?: AWUPF)
> 

You still need to fix that mail client to not add extra blank lines.

>> atomic_write_unit_min = logical_block_size
> 
>> atomic_write_unit_max = flp2(NAWUPF ?: AWUPF)
> 
>> atomic_write_boundary = NABSPF
> 
> 
> 
> In case the device doesn't support namespace atomic boundary size (i.e. NABSPF
> 
> is zero) then while merging atomic block-IO we should allow merge.
> 
>   
> 
> For example, while front/back merging the atomic block IO, we check whether
> 
> boundary is defined or not. In case if boundary is not-defined (i.e. it's zero)
> 
> then we simply reject merging ateempt (as implemented in
> 
> rq_straddles_atomic_write_boundary()).

Are you sure about that? In rq_straddles_atomic_write_boundary(), if 
boundary == 0, then we return false, i.e. there is no boundary, so we 
can never be crossing it.

static bool rq_straddles_atomic_write_boundary(struct request *rq,
unsigned int front,
unsigned int back)
{
	unsigned int boundary = queue_atomic_write_boundary_bytes(rq->q);
	unsigned int mask, imask;
	loff_t start, end;

	if (!boundary)
		return false;

	...
}

And then will not reject a merge for that reason, like:

int ll_back_merge_fn(struct request *req, struct bio *bio, unsigned int 
nr_segs)
{
	...

	if (req->cmd_flags & REQ_ATOMIC) {
		if (rq_straddles_atomic_write_boundary(req,
			0, bio->bi_iter.bi_size)) {
			return 0;
		}
	}

	return ll_new_hw_segment(req, bio, nr_segs);
}


> 
> 
> 
> I am quoting this from NVMe spec (Command Set Specification, revision 1.0a,
> 
> Section 2.1.4.3) : "To ensure backwards compatibility, the values reported for
> 
> AWUN, AWUPF, and ACWU shall be set such that they  are  supported  even  if  a
> 
> write  crosses  an  atomic  boundary.  If  a  controller  does  not  guarantee
> 
> atomicity across atomic boundaries, the controller shall set AWUN, AWUPF, and
> 
> ACWU to 0h (1 LBA)."
> 
> 
> 

Thanks,
John


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ