From: Marc Herbert <marc.herbert@linux.intel.com>
To: Dave Jiang <dave.jiang@intel.com>,
Alison Schofield <alison.schofield@intel.com>
Cc: Dan Williams <dan.j.williams@intel.com>,
nvdimm@lists.linux.dev, linux-cxl@vger.kernel.org
Subject: Re: [NDCTL PATCH v6 4/4] cxl/test: Add test for cxl features device
Date: Thu, 22 May 2025 11:25:04 -0700 [thread overview]
Message-ID: <acfac209-430b-486a-b468-3b98575b8b12@linux.intel.com> (raw)
In-Reply-To: <fc21a089-efa0-41b4-bc3e-70f6e7423719@intel.com>
>>>>>
>>>>> +uuid_dep = dependency('uuid', required: false)
>>>>> +if get_option('fwctl').enabled() and uuid_dep.found()
>>>>> + fwctl = executable('fwctl', 'fwctl.c',
>>>>> + dependencies : libcxl_deps,
>>>>> + include_directories : root_inc,
>>>>> + )
>>>>> + cxl_features = find_program('cxl-features.sh')
>>>>> + tests += [
>>>>> + [ 'cxl-features.sh', cxl_features, 'cxl' ],
>>>>> + ]
>>>>> +endif
>>>>
>>>> Is the fwctl feature enabled fuss still needed now that the UAPI headers
>>>> are vendored locally? Seems the test will quickly SKIP if fwctl dev not
>>>> found. I kind of like the idea of seeing a 'SKIP' and knowing the test
>>>> didn't run than seeing nothing at all in the test output.
>>>
>>> This gives the option to disable fwctl if needed. Also there is a libuuid dependency since we use uuid lib calls.
>>
>> Is it possible to go anywhere without uuid?
>>
>> [snip]
>>
>> It looks like a hard requirement to me.
>>
>> UUIDs are rarely ever "optional". It could be required only
>> by some optional tool but that does not seem to be the case here
>
> Apparently it's not an requirement for kernel build. As my feature user header caused issues due to lack of libuuid package.
I think I wasn't clear, let me be more explicit: I think you should drop
all the code with "uuid_dep" above. It would not just simplify the code
(uuid_dep.found() is always true), but it would also get rid of
some (unfounded) reader's anxiety about an (im)possible system
configuration where fwctl is not tested without even printing a "SKIP".
It never hurts to express dependencies, but I think it's also bad to
look like you can opt out when you cannot.
prev parent reply other threads:[~2025-05-22 18:25 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2025-05-09 16:39 [NDCTL PATCH v6 0/4] ndctl: Add support and test for CXL Features support Dave Jiang
2025-05-09 16:39 ` [NDCTL PATCH v6 1/4] cxl: Add cxl_bus_get_by_provider() Dave Jiang
2025-05-09 16:39 ` [NDCTL PATCH v6 2/4] cxl: Enumerate major/minor of FWCTL char device Dave Jiang
2025-05-09 16:39 ` [NDCTL PATCH v6 3/4] ndctl: Add features.h from kernel UAPI Dave Jiang
2025-05-09 16:39 ` [NDCTL PATCH v6 4/4] cxl/test: Add test for cxl features device Dave Jiang
2025-05-15 18:24 ` Alison Schofield
2025-05-15 18:28 ` Dave Jiang
2025-05-22 17:37 ` Marc Herbert
2025-05-22 17:46 ` Dave Jiang
2025-05-22 18:25 ` Marc Herbert [this message]
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=acfac209-430b-486a-b468-3b98575b8b12@linux.intel.com \
--to=marc.herbert@linux.intel.com \
--cc=alison.schofield@intel.com \
--cc=dan.j.williams@intel.com \
--cc=dave.jiang@intel.com \
--cc=linux-cxl@vger.kernel.org \
--cc=nvdimm@lists.linux.dev \
/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