From: Klaus Jensen <its@irrelevant.dk>
To: "Philippe Mathieu-Daudé" <philmd@redhat.com>
Cc: Fam Zheng <fam@euphon.net>, Kevin Wolf <kwolf@redhat.com>,
Damien Le Moal <Damien.LeMoal@wdc.com>,
"qemu-block@nongnu.org" <qemu-block@nongnu.org>,
Niklas Cassel <Niklas.Cassel@wdc.com>,
Dmitry Fomichev <Dmitry.Fomichev@wdc.com>,
Klaus Jensen <k.jensen@samsung.com>,
"qemu-devel@nongnu.org" <qemu-devel@nongnu.org>,
Maxim Levitsky <mlevitsk@redhat.com>,
Alistair Francis <Alistair.Francis@wdc.com>,
Keith Busch <kbusch@kernel.org>,
Matias Bjorling <Matias.Bjorling@wdc.com>
Subject: Re: [PATCH v8 07/11] hw/block/nvme: Support Zoned Namespace Command Set
Date: Wed, 4 Nov 2020 08:21:45 +0100 [thread overview]
Message-ID: <20201104072145.GA166505@apples.localdomain> (raw)
In-Reply-To: <b8d2a80b-4f7f-56eb-39dd-1e40f73c5a9b@redhat.com>
[-- Attachment #1: Type: text/plain, Size: 1605 bytes --]
On Nov 3 21:37, Philippe Mathieu-Daudé wrote:
> On 11/3/20 8:48 PM, Dmitry Fomichev wrote:
> >> -----Original Message-----
> >> From: Philippe Mathieu-Daudé <philmd@redhat.com>
> ...
> >>> typedef struct QEMU_PACKED NvmeCqe {
> >>> - uint32_t result;
> >>> - uint32_t rsvd;
> >>> + union {
> >>> + uint64_t result64;
> >>> + uint32_t result32;
> >>> + };
> >>
> >> When using packed structure you want to define all fields to
> >> avoid alignment confusion (and I'm surprised the compiler doesn't
> >> complain...). So this would be:
> >>
> >> union {
> >> uint64_t result64;
> >> struct {
> >> uint32_t result32;
> >> uint32_t rsvd32;
> >> };
> >> };
> >>
I align (hehe...) towards this. The amount of bit-juggling we need for
commands justify the need for separate NvmeCmd's, but in this case I
think an NvmeCqeZA is just unnecessary clutter. If the result value is
complex, the approach used by AERs is better I think:
typedef struct QEMU_PACKED NvmeAerResult {
uint8_t event_type;
uint8_t event_info;
uint8_t log_page;
uint8_t resv;
} NvmeAerResult;
NvmeAerResult *result = (NvmeAerResult *)&req->cqe.result32;
Since storing the Zone Append ALBA in the result64 isn't really a
complex operation, let's just assign it into that member directly.
(Addendum) That DW1 is command specific and no longer reserved is
defined by TP 4056 (Namespace Types) - not v1.4 or any of its revisions.
[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]
next prev parent reply other threads:[~2020-11-04 7:25 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-10-30 2:32 [PATCH v8 00/11] hw/block/nvme: Support Namespace Types and Zoned Namespace Command Set Dmitry Fomichev
2020-10-30 2:32 ` [PATCH v8 01/11] hw/block/nvme: Add Commands Supported and Effects log Dmitry Fomichev
2020-10-30 2:32 ` [PATCH v8 02/11] hw/block/nvme: Generate namespace UUIDs Dmitry Fomichev
2020-10-30 2:32 ` [PATCH v8 03/11] hw/block/nvme: Separate read and write handlers Dmitry Fomichev
2020-10-30 2:32 ` [PATCH v8 04/11] hw/block/nvme: Merge nvme_write_zeroes() with nvme_write() Dmitry Fomichev
2020-10-30 2:32 ` [PATCH v8 05/11] hw/block/nvme: Add support for Namespace Types Dmitry Fomichev
2020-10-30 2:32 ` [PATCH v8 06/11] hw/block/nvme: Support allocated CNS command variants Dmitry Fomichev
2020-11-04 7:48 ` Klaus Jensen
2020-10-30 2:32 ` [PATCH v8 07/11] hw/block/nvme: Support Zoned Namespace Command Set Dmitry Fomichev
2020-10-30 7:15 ` Philippe Mathieu-Daudé
2020-11-03 19:48 ` Dmitry Fomichev
2020-11-03 20:37 ` Philippe Mathieu-Daudé
2020-11-04 7:21 ` Klaus Jensen [this message]
2020-10-30 17:18 ` Niklas Cassel
2020-10-30 2:32 ` [PATCH v8 08/11] hw/block/nvme: Introduce max active and open zone limits Dmitry Fomichev
2020-10-30 2:32 ` [PATCH v8 09/11] hw/block/nvme: Support Zone Descriptor Extensions Dmitry Fomichev
2020-10-30 2:32 ` [PATCH v8 10/11] hw/block/nvme: Add injection of Offline/Read-Only zones Dmitry Fomichev
2020-10-30 2:32 ` [PATCH v8 11/11] hw/block/nvme: Document zoned parameters in usage text Dmitry Fomichev
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=20201104072145.GA166505@apples.localdomain \
--to=its@irrelevant.dk \
--cc=Alistair.Francis@wdc.com \
--cc=Damien.LeMoal@wdc.com \
--cc=Dmitry.Fomichev@wdc.com \
--cc=Matias.Bjorling@wdc.com \
--cc=Niklas.Cassel@wdc.com \
--cc=fam@euphon.net \
--cc=k.jensen@samsung.com \
--cc=kbusch@kernel.org \
--cc=kwolf@redhat.com \
--cc=mlevitsk@redhat.com \
--cc=philmd@redhat.com \
--cc=qemu-block@nongnu.org \
--cc=qemu-devel@nongnu.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: 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;
as well as URLs for NNTP newsgroup(s).