When the script detects the need for an upgrade, it will
print either a warning or a note.

Let's change a little bit the order where messages will be
displayed, in order to make easier for the user to identify
the more important messages.

It should now be like this:

	Detected OS: Fedora release 31 (Thirty One).
	Sphinx version: 1.7.9

	Note: It is recommended at least Sphinx version 2.4.4 if you need PDF support.
	To upgrade Sphinx, use:

		/usr/bin/python3 -m venv sphinx_2.4.4
		. sphinx_2.4.4/bin/activate
		pip install -r ./Documentation/sphinx/requirements.txt

	If you want to exit the virtualenv, you can use:
		deactivate

All optional dependencies are met.
Needed package dependencies are met.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/20200421182758.04e0a53e@coco.lanSigned-off-by: NJonathan Corbet <corbet@lwn.net>

When python3 creates a venv, it adds python into it!

This causes any upgrade recommendation to look like this:

	/devel/v4l/docs/sphinx_1.7.9/bin/python3 -m venv sphinx_2.4.4
	. sphinx_2.4.4/bin/activate
	pip install -r ./Documentation/sphinx/requirements.txt

With is wrong (and it may not work). So, when recomending
an upgrade, exclude the venv dir from the search path, and
get the system's python.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/aa622ff71bebf6960fc0262fb90e7ebc7a999a02.1587478901.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

If one is running a Sphinx version older than what's recommended,
but there's already a newer working virtual env, change the
text, as it is just a matter of switching to the new venv, instead
of creating a new one from scratch.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/bcf79d0399a1c3444ca938dcdce599c3273980ab.1587478901.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

As requested by Jon, change the version check, in order to not
emit a warning if version is >= 1.7.9, but below 2.4.4.

After this patch, if someone used an older version, it will
say:

	./scripts/sphinx-pre-install
	Sphinx version 1.7.9
	Note: It is recommended at least Sphinx version 2.4.4 if you need PDF support.
	Detected OS: Fedora release 31 (Thirty One).

	To upgrade Sphinx, use:

		/devel/v4l/docs/sphinx_1.7.9/bin/python3 -m venv sphinx_2.4.4
		. sphinx_2.4.4/bin/activate
		pip install -r ./Documentation/sphinx/requirements.txt

	If you want to exit the virtualenv, you can use:
		deactivate

	All optional dependencies are met.
	Needed package dependencies are met.

If Sphinx is not detected at all, it
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/79584d317ba16f5d4f37801c5ee57cf04085f962.1587478901.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

If a venv doesn't contain Sphinx, or has an older Sphinx
version, ignore it.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/d11a00d88514e8a0357e1b0a05ebd518952a1d39.1587478901.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

The test driver is in tools/testing/nvdimm and cannot be selected by a
config option.
Signed-off-by: NMichal Suchanek <msuchanek@suse.de>
Link: https://lore.kernel.org/r/20200415211654.10827-1-msuchanek@suse.deSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Based on the implementation in kernel/sysctl.c (the proc_do_cad_pid()
function), kernel/reboot.c, and include/linux/sched/signal.h.
Signed-off-by: NStephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20200423183651.15365-1-steve@sk2.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

This fixes

	Documentation/virt/kvm/amd-memory-encryption.rst:76: WARNING: Inline literal start-string without end-string.

Fixes: 2da1ed62 ("KVM: SVM: document KVM_MEM_ENCRYPT_OP, let userspace detect if SEV is available")
Signed-off-by: NStephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20200424152637.120876-1-steve@sk2.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Following a merge fix-up, the literal block is introduced too early;
this patch merges the localhost mention with the introduction, fixing

	Documentation/filesystems/orangefs.rst:124: WARNING: Literal block expected; none found.
Signed-off-by: NStephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20200424153515.134500-1-steve@sk2.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Move away from the deprecated API and advertise the new one.
Signed-off-by: NWolfram Sang <wsa+renesas@sang-engineering.com>
Reviewed-by: NAlex Shi <alex.shi@linux.alibaba.com>
Link: https://lore.kernel.org/r/20200425100616.3363-1-wsa+renesas@sang-engineering.comSigned-off-by: NJonathan Corbet <corbet@lwn.net>

The page originally referenced to checkout Plan9 application and libraries
have been missing for quite some time and the development is carried out
in github and documented on this new site.
Signed-off-by: NJuan Manuel Méndez Rey <vejeta@gmail.com>
Link: https://lore.kernel.org/r/20200426015250.GA35090@camelotSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Add an entry to connect my email addresses.
Signed-off-by: NLeonardo Bras <leobras.c@gmail.com>
Link: https://lore.kernel.org/r/20200428024439.215806-1-leobras.c@gmail.comSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Signed-off-by: NFlavio Suligoi <f.suligoi@asem.it>
Link: https://lore.kernel.org/r/1588080745-21999-1-git-send-email-f.suligoi@asem.itSigned-off-by: NJonathan Corbet <corbet@lwn.net>

The documentation claims that two sched groups must not overlap. This is
no longer true, as overlapping sched groups are used on NUMA systems.
This change has been introduced by commit e3589f6c and was
documented by an in-code comment in commit 35a566e6.
Signed-off-by: NAdrian Freund <adrian@freund.io>
Link: https://lore.kernel.org/r/20200407130525.76663-1-adrian@freund.ioSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Add translation for the RISC-V maintenance guidelines as part
of the translation of things related to "process/"
Signed-off-by: NFederico Vaga <federico.vaga@vaga.pv.it>
Link: https://lore.kernel.org/r/20200405210647.24991-1-federico.vaga@vaga.pv.itSigned-off-by: NJonathan Corbet <corbet@lwn.net>

This file is close enough to being in rst format that I didn't feel
the need to alter it in any way.
Signed-off-by: NMatthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: NMichel Lespinasse <walken@google.com>
Link: https://lore.kernel.org/r/20200401173343.17472-1-willy@infradead.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Add kerneldocs comments to the pr_*() macros in printk.h.

Add a new rst node in the core-api manual describing the basic usage of
printk and the related macro aliases.
Signed-off-by: NRicardo Cañuelo <ricardo.canuelo@collabora.com>
Link: https://lore.kernel.org/r/20200403093617.18003-1-ricardo.canuelo@collabora.comSigned-off-by: NJonathan Corbet <corbet@lwn.net>

The commit d0684c8a ("watchdog: Make stop function optional")
made stop function not mandatory, but the comments
and the doc weren't reflected. Fix it to clarify.
Signed-off-by: NBumsik Kim <k.bumsik@gmail.com>
Reviewed-by: NGuenter Roeck <linux@roeck-us.net>
Link: https://lore.kernel.org/r/20200403031507.63487-1-k.bumsik@gmail.comSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Translations for the following documents in process/:
    - email-clients
    - management-style
Signed-off-by: NAlessia Mantegazza <amantegazza@vaga.pv.it>
Signed-off-by: NFederico Vaga <federico.vaga@vaga.pv.it>
Link: https://lore.kernel.org/r/20200413163457.52669-1-federico.vaga@vaga.pv.itSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Changes to make the text more formal and organized. The reasons are now cited and described at the same time.
Minor grammatical problems have also been fixed.
Signed-off-by: NCristian Souza <cristianmsbr@gmail.com>
Link: https://lore.kernel.org/r/20200411010201.GA22706@darkstarSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Habanalabs driver in misc exposes several IOCTLs to userspace. Document the
letter and IOCTLs number range in ioctl-number.rst.
Signed-off-by: NOded Gabbay <oded.gabbay@gmail.com>
Acked-by: NRandy Dunlap <rdunlap@infradead.org>
Link: https://lore.kernel.org/r/20200418174132.10597-1-oded.gabbay@gmail.comSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Fix a couple of typos: "as" -> "has" and "int" -> "in".
Signed-off-by: NGal Pressman <galpress@amazon.com>
Link: https://lore.kernel.org/r/20200420074115.23931-1-galpress@amazon.comSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Based on the implementation in kernel/bpf/syscall.c,
kernel/bpf/trampoline.c, include/linux/filter.h, and the documentation
in bpftool-prog.rst.
Signed-off-by: NStephen Kitt <steve@sk2.org>
Link: https://lore.kernel.org/r/20200315122648.20558-1-steve@sk2.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Mauro sez:

Patches 1 to 5 contain changes to the documentation toolset:

- The first 3 patches help to reduce a lot the number of reported
  kernel-doc issues, by making the tool more smart.

- Patches 4 and 5 are meant to partially address the PDF
  build, with now requires Sphinx version 2.4 or upper.

The remaining patches fix broken references detected by
this tool:

        ./scripts/documentation-file-ref-check

and address other random errors due to tags being mis-interpreted
or mis-used.

They are independent each other, but some may depend on
the kernel-doc improvements.

There are two ascii art drawings there. Use a block markup tag there
in order to get rid of those warnings:

	./lib/bitmap.c:189: WARNING: Unexpected indentation.
	./lib/bitmap.c:190: WARNING: Block quote ends without a blank line; unexpected unindent.
	./lib/bitmap.c:190: WARNING: Unexpected indentation.
	./lib/bitmap.c:191: WARNING: Line block ends without a blank line.

It should be noticed that there's actually a syntax violation
right now, as something like:

	/**
	 ...
	 @src:

will be handled as a definition for @src parameter, and not as
part of a diagram. So, we need to add something before it, in
order for this to be processed the way it should.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/1e2568fdfa838c1a0d8cc2a1d70dd4b6de99bfb1.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Adjust whitespaces and blank lines in order to get rid of this:

	./kernel/futex.c:491: WARNING: Definition list ends without a blank line; unexpected unindent.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/57788af7889161483e0c97f91c079cfb3986c4b3.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Use *foo makes the toolchain to think that this is an emphasis, causing
those warnings:

	./fs/inode.c:1609: WARNING: Inline emphasis start-string without end-string.
	./fs/inode.c:1609: WARNING: Inline emphasis start-string without end-string.
	./fs/inode.c:1615: WARNING: Inline emphasis start-string without end-string.

So, use, instead, ``*foo``, in order to mark it as a literal block.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/e8da46a0e57f2af6d63a0c53665495075698e28a.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

This warning:

	./include/uapi/linux/firewire-cdev.h:312: WARNING: Inline literal start-string without end-string.

is because %FOO doesn't work if there's a parenthesis at the
string (as a parenthesis may indicate a function). So, mark
the literal block using the alternate ``FOO`` syntax.
Acked-by: NStefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/9b2501a41eba27ccdd4603cac2353c0efba7a90a.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

The docs toolchain doesn't recognise this pattern:

	@link->[hw_]sata_spd_limit

As it can't really process it. So, instead, let's mark it with
a literal block markup:

	``link->[hw_]sata_spd_limit``

in order to get rid of the following warning:

	./drivers/ata/libata-core.c:5974: WARNING: Unknown target name: "hw".
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/9a21444df75c46095c4b1839d2061d19c9addcff.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

phy-rockchip-inno-usb2.txt was converted to yaml.

Fix the corresponding reference.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/287bd271f5c542e9d12a132a6b6a17672c9fd67c.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

The Sphinx build system for PDF is too complex and generate
lots of ancillary files, including one PDF file for each
image.

So, at the end, the main latex dir has 156 pdf files, instead
of the 71 ones that would match each generated book. That's
confusing and it makes harder to identify when something didn't
work.

So, instead, let's move the final PDF output(s) to a separate
dir. This way, the latex/ dir will have the temporary and the
final *.tex files, while the final pdf files that built ok
will be under the pdf/ directory.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/832752cbc9678a6e8d3d634bc3356d655d44684f.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

When generating the PDF output, the Documentation/i2c dir
will generate an i2c.pdf. The same happens with i2c.svg:
it will also produce a file with the same name, at the same dir.

This causes errors when building the PDF output. So, rename the
image to i2c_bus.svg.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Acked-by: NWolfram Sang <wsa@the-dreams.de>
Link: https://lore.kernel.org/r/ecf3d51909ce46b3e84a1df4b36f07d76989e5da.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

The User API chapter contains two sub-chapters. Mark them as
such.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Acked-by: NAndrew Donnellan <ajd@linux.ibm.com>
Link: https://lore.kernel.org/r/190d67397cd63e419de8d85b92e8018d48e8c345.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Changeset 32ced09d ("dt-bindings: serial: Convert slave-device bindings to json-schema")
moved a binding to json and updated the links.

Yet, one link was not changed, due to a merge conflict.

Update this one too.

Fixes: 32ced09d ("dt-bindings: serial: Convert slave-device bindings to json-schema")
Reviewed-by: NGeert Uytterhoeven <geert+renesas@glider.be>
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/9b1603e254d39c9607bfedefeedaafd2c44aeb19.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

The qcom-qusb2-phy.txt file was converted and renamed to yaml.
Update cross-reference accordingly.

Fixes: 8ce65d8d ("dt-bindings: phy: qcom,qusb2: Convert QUSB2 phy bindings to yaml")
Reviewed-by: NStephen Boyd <swboyd@chromium.org>
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/a055c564f2a79aa748064329d938db8b3c8edd58.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

The new free_page_reporting.rst file is not listed at the index.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/146432ae6965a2bb62c929a6b62f9d4010986622.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Instead of using "foo", let's use `foo`_, with is a ReST way of
saying that foo is a section of the document. With that, after
building the docs, an hyperlink is generated.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/f46b45f1aaec233217f2e0b0438bbd8cc16fe17b.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Several parts of this document define literals: ioctl names,
function calls, directory patches, etc. Mark those as literal
blocks, in order to improve its readability (both at text mode
and after parsed by Sphinx.

This fixes those two warnings:

	Documentation/admin-guide/mm/userfaultfd.rst:139: WARNING: Inline emphasis start-string without end-string.
	Documentation/admin-guide/mm/userfaultfd.rst:139: WARNING: Inline emphasis start-string without end-string.

produced during documentation build.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/2ae061761baf8fe00cdf8a7e6dae293756849a05.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

Currrently, two warnings are generated when building docs:

	./drivers/base/platform.c:136: WARNING: Unexpected indentation.
	./drivers/base/platform.c:214: WARNING: Unexpected indentation.

As examples are code blocks, they should use "::" markup. However,

	Example::

Is currently interpreted as a new section.

While we could fix kernel-doc to accept such new syntax, it is
easier to just replace it with:

	For Example::
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/564273815a76136fb5e453969b1012a786d99e28.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>

We need to add a blank line to avoid this warning:

	./include/linux/spi/spi.h:401: WARNING: Unexpected indentation.
Signed-off-by: NMauro Carvalho Chehab <mchehab+huawei@kernel.org>
Link: https://lore.kernel.org/r/1c701b3ac903dc0bc304dca958fbdee53bd38dc3.1586881715.git.mchehab+huawei@kernel.orgSigned-off-by: NJonathan Corbet <corbet@lwn.net>