Re: [PATCH v16 6/9] nvmet-passthru: Add passthru code to process commands
On 2020-07-26 9:41 a.m., Christoph Hellwig wrote:
> On Fri, Jul 24, 2020 at 12:33:51PM -0700, Keith Busch wrote:
>> On Fri, Jul 24, 2020 at 11:25:17AM -0600, Logan Gunthorpe wrote:
>>> + /*
>>> +* The passthru NVMe driver may have a limit on the number of segments
>>> +* which depends on the host's memory fragementation. To solve this,
>>> +* ensure mdts is limitted to the pages equal to the number of
>>
>> limited
>
> I've fixed this when applying.
>
>>> + /* don't support fuse commands */
>>> + id->fuses = 0;
>>
>> If a host were to set a fuse, the target should return an Invalid Field
>> error. Just to future-proof, rejecting commands with any flags set
>> (other than SGL, which you handled in patch 1/9) is probably what should
>> happen, like:
>>
>>> +u16 nvmet_parse_passthru_io_cmd(struct nvmet_req *req)
>>> +{
>>
>> if (req->cmd->common.flags & ~NVME_CMD_SGL_ALL)
>> return NVME_SC_INVALID_FIELD;
>>
>> Or maybe we could obviate the need for 1/9 with something like:
>>
>> req->cmd->common.flags &= ~NVME_CMD_SGL_ALL;
>> if (req->cmd->common.flags)
>> return NVME_SC_INVALID_FIELD;
>
> We'll also need this for the admin commands, but otherwise this sounds
> like a good idea. For now I've applied the series as-is, but an
> incremental patch for this would be nice.
Great, I can send one later in the week.
Thanks,
Logan
Re: [PATCH v16 6/9] nvmet-passthru: Add passthru code to process commands
On Fri, Jul 24, 2020 at 12:33:51PM -0700, Keith Busch wrote:
> On Fri, Jul 24, 2020 at 11:25:17AM -0600, Logan Gunthorpe wrote:
> > + /*
> > +* The passthru NVMe driver may have a limit on the number of segments
> > +* which depends on the host's memory fragementation. To solve this,
> > +* ensure mdts is limitted to the pages equal to the number of
>
> limited
I've fixed this when applying.
> > + /* don't support fuse commands */
> > + id->fuses = 0;
>
> If a host were to set a fuse, the target should return an Invalid Field
> error. Just to future-proof, rejecting commands with any flags set
> (other than SGL, which you handled in patch 1/9) is probably what should
> happen, like:
>
> > +u16 nvmet_parse_passthru_io_cmd(struct nvmet_req *req)
> > +{
>
> if (req->cmd->common.flags & ~NVME_CMD_SGL_ALL)
> return NVME_SC_INVALID_FIELD;
>
> Or maybe we could obviate the need for 1/9 with something like:
>
> req->cmd->common.flags &= ~NVME_CMD_SGL_ALL;
> if (req->cmd->common.flags)
> return NVME_SC_INVALID_FIELD;
We'll also need this for the admin commands, but otherwise this sounds
like a good idea. For now I've applied the series as-is, but an
incremental patch for this would be nice.
Re: [PATCH v16 6/9] nvmet-passthru: Add passthru code to process commands
On 2020-07-24 1:33 p.m., Keith Busch wrote:
> On Fri, Jul 24, 2020 at 11:25:17AM -0600, Logan Gunthorpe wrote:
>> +/*
>> + * The passthru NVMe driver may have a limit on the number of segments
>> + * which depends on the host's memory fragementation. To solve this,
>> + * ensure mdts is limitted to the pages equal to the number of
>
> limited
>
>> +/* don't support fuse commands */
>> +id->fuses = 0;
>
> If a host were to set a fuse, the target should return an Invalid Field
> error. Just to future-proof, rejecting commands with any flags set
> (other than SGL, which you handled in patch 1/9) is probably what should
> happen, like:
>
>> +u16 nvmet_parse_passthru_io_cmd(struct nvmet_req *req)
>> +{
>
> if (req->cmd->common.flags & ~NVME_CMD_SGL_ALL)
> return NVME_SC_INVALID_FIELD;
Yes, this seems like a good idea. I can add it.
> Or maybe we could obviate the need for 1/9 with something like:
>
> req->cmd->common.flags &= ~NVME_CMD_SGL_ALL;
> if (req->cmd->common.flags)
> return NVME_SC_INVALID_FIELD;
We used to clear the SGL flags in the target passthru code but Christoph
asked that it be done in the host code, hence patch 1/9.
Logan
Re: [PATCH v16 6/9] nvmet-passthru: Add passthru code to process commands
On Fri, Jul 24, 2020 at 11:25:17AM -0600, Logan Gunthorpe wrote:
> + /*
> + * The passthru NVMe driver may have a limit on the number of segments
> + * which depends on the host's memory fragementation. To solve this,
> + * ensure mdts is limitted to the pages equal to the number of
limited
> + /* don't support fuse commands */
> + id->fuses = 0;
If a host were to set a fuse, the target should return an Invalid Field
error. Just to future-proof, rejecting commands with any flags set
(other than SGL, which you handled in patch 1/9) is probably what should
happen, like:
> +u16 nvmet_parse_passthru_io_cmd(struct nvmet_req *req)
> +{
if (req->cmd->common.flags & ~NVME_CMD_SGL_ALL)
return NVME_SC_INVALID_FIELD;
Or maybe we could obviate the need for 1/9 with something like:
req->cmd->common.flags &= ~NVME_CMD_SGL_ALL;
if (req->cmd->common.flags)
return NVME_SC_INVALID_FIELD;
