kernel/git/next/linux-next.git/Documentation/usb, branch master

usb: gadget: f_fs: Introduce rw_proxy file descriptors

2026-07-10T13:00:47+00:00

Currently, FunctionFS exposes each USB endpoint as a separate, unidirectional file descriptor (e.g., `ep1` for IN, `ep2` for OUT). While this mirrors the underlying hardware structure, it forces userspace daemons implementing bidirectional protocols to manage multiple file descriptors. When dealing with legacy protocols which require exposing a single, bi-directional fd to userspace, this becomes problematic. This patch introduces the `FUNCTIONFS_RW_PROXY_EPS` UAPI flag. When passed in the descriptor header during initialization, FunctionFS provisions a "rw_proxy" bidirectional file descriptor (e.g., `ep1_rw`) alongside every pair of IN/OUT endpoints. Implementation details: - RW proxy files act as a pure VFS alias, proxying operations directly to the base ffs_epfile instances. A `read()` proxies to the OUT endpoint's file, and a `write()` proxies to the IN file. - Because operations are proxied natively, they reuse the underlying base endpoint's lock (`epfile->mutex`) and tracking state. This serializes concurrent I/O, preventing buffer corruption or races even if userspace mixes transfers across both the rw_proxy and base files while allowing full-duplex synchronous operations to occur concurrently without serializing on a single lock. - Control operations (like IOCTLs) and intentional stalls (via reverse-direction I/O) must still be issued on the base endpoints, as the rw_proxy returns `-ENOTTY` for IOCTLs and cannot trigger stalls. Assisted-by: Antigravity:gemini-3.1-pro Signed-off-by: Neill Kapron Link: https://patch.msgid.link/20260619040609.4010746-5-nkapron@google.com Signed-off-by: Greg Kroah-Hartman

usb: gadget: f_fs: Add zero-length packet ioctl

2026-07-10T13:00:45+00:00

When transferring data from a USB gadget to a host, a transfer is considered complete when the host receives a short packet (a packet smaller than wMaxPacketSize). If the total transfer length is an exact multiple of wMaxPacketSize, a Zero-Length Packet (ZLP) must be appended to signal the end of the transfer. FunctionFS currently provides no mechanism for userspace to instruct the kernel to set the `req->zero` flag on transfers. Userspace workarounds, such as manually submitting separate 0-byte requests, may not be available for legacy protocols which must maintain write behavior compatibility when moved to functionfs implementations. To resolve this, introduce a new ioctl, FUNCTIONFS_ENDPOINT_ENABLE_ZLP, which takes a pointer to a __u32 flag. When enabled, all subsequent transfers on that endpoint will have the `req->zero` flag set, allowing the underlying USB Device Controller (UDC) hardware to automatically append a ZLP only when mathematically required. For logical transfers chunked across multiple requests, userspace can dynamically toggle this flag, enabling it only prior to submitting the final chunk. The flag defaults to false to maintain backward compatibility. Once enabled, the state is persistent for the lifetime of the endpoint and will not be reset by opening or closing the endpoint file descriptors. Assisted-by: Gemini-CLI:gemini-3.1-pro-preview Signed-off-by: Neill Kapron Link: https://patch.msgid.link/20260619040609.4010746-4-nkapron@google.com Signed-off-by: Greg Kroah-Hartman

Merge tag 'usb-7.0-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb

2026-02-17T17:36:43+00:00

Pull USB / Thunderbolt updates from Greg KH: "Here is the "big" set of USB and Thunderbolt driver updates for 7.0-rc1. Overall more lines were removed than added, thanks to dropping the obsolete isp1362 USB host controller driver, always a nice change. Other than that, nothing major happening here, highlights are: - lots of dwc3 driver updates and new hardware support added - usb gadget function driver updates - usb phy driver updates - typec driver updates and additions - USB rust binding updates for syntax and formatting changes - more usb serial device ids added - other smaller USB core and driver updates and additions All of these have been in linux-next for a long time, with no reported problems" * tag 'usb-7.0-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/usb: (77 commits) usb: typec: ucsi: Add Thunderbolt alternate mode support usb: typec: hd3ss3220: Check if regulator needs to be switched usb: phy: tegra: parametrize PORTSC1 register offset usb: phy: tegra: parametrize HSIC PTS value usb: phy: tegra: return error value from utmi_wait_register usb: phy: tegra: cosmetic fixes dt-bindings: usb: renesas,usbhs: Add RZ/G3E SoC support usb: dwc2: fix resume failure if dr_mode is host usb: cdns3: fix role switching during resume usb: dwc3: gadget: Move vbus draw to workqueue context USB: serial: option: add Telit FN920C04 RNDIS compositions usb: dwc3: Log dwc3 address in traces usb: gadget: tegra-xudc: Add handling for BLCG_COREPLL_PWRDN usb: phy: tegra: add HSIC support usb: phy: tegra: use phy type directly usb: typec: ucsi: Enforce mode selection for cros_ec_ucsi usb: typec: ucsi: Support mode selection to activate altmodes usb: typec: Introduce mode_selection bit usb: typec: Implement mode selection usb: typec: Expose alternate mode priority via sysfs ...

Documentation: use a source-read extension for the index link boilerplate

2026-01-23T18:59:34+00:00

The root document usually has a special :ref:`genindex` link to the generated index. This is also the case for Documentation/index.rst. The other index.rst files deeper in the directory hierarchy usually don't. For SPHINXDIRS builds, the root document isn't Documentation/index.rst, but some other index.rst in the hierarchy. Currently they have a ".. only::" block to add the index link when doing SPHINXDIRS html builds. This is obviously very tedious and repetitive. The link is also added to all index.rst files in the hierarchy for SPHINXDIRS builds, not just the root document. Put the boilerplate in a sphinx-includes/subproject-index.rst file, and include it at the end of the root document for subproject builds in an ad-hoc source-read extension defined in conf.py. For now, keep having the boilerplate in translations, because this approach currently doesn't cover translated index link headers. Cc: Jonathan Corbet Cc: Mauro Carvalho Chehab Cc: Randy Dunlap Signed-off-by: Jani Nikula Tested-by: Mauro Carvalho Chehab Reviewed-by: Mauro Carvalho Chehab [jc: did s/doctree/kern_doc_dir/ ] Signed-off-by: Jonathan Corbet Message-ID: <20260123143149.2024303-1-jani.nikula@intel.com>

usb: gadget: f_sourcesink: Support maxburst configurability for bulk endpoints

2026-01-07T15:13:25+00:00

Add support to configure maxburst via configfs for bulk endpoints. Update gadget documentation describing the new configfs property. Signed-off-by: Krishna Kurapati Link: https://patch.msgid.link/20251227145224.2091397-1-krishna.kurapati@oss.qualcomm.com Signed-off-by: Greg Kroah-Hartman

usb: gadget: f_midi: allow customizing the USB MIDI interface string through configfs

2025-12-17T13:43:01+00:00

When using f_midi from configfs the USB MIDI interface string is hardcoded to 'MIDI function'. This USB string descriptor is used by some third-party OS or software to display the name of the MIDI device Since we add an additional string option a new macro block was created to factorize declarations Signed-off-by: Victor Krawiec Reviewed-by: Takashi Iwai Link: https://patch.msgid.link/20251209164006.143219-1-victor.krawiec@arturia.com Signed-off-by: Greg Kroah-Hartman

Merge tag 'docs-6.17' of git://git.lwn.net/linux

2025-07-31T15:36:51+00:00

Pull documentation updates from Jonathan Corbet: "It has been a relatively busy cycle for docs, especially the build system: - The Perl kernel-doc script was added to 2.3.52pre1 just after the turn of the millennium. Over the following 25 years, it accumulated a vast amount of cruft, all in a language few people want to deal with anymore. Mauro's Python replacement in 6.16 faithfully reproduced all of the cruft in the hope of avoiding regressions. Now that we have a more reasonable code base, though, we can work on cleaning it up; many of the changes this time around are toward that end. - A reorganization of the ext4 docs into the usual TOC format. - Various Chinese translations and updates. - A new script from Mauro to help with docs-build testing. - A new document for linked lists - A sweep through MAINTAINERS fixing broken GitHub git:// repository links. ...and lots of fixes and updates" * tag 'docs-6.17' of git://git.lwn.net/linux: (147 commits) scripts: add origin commit identification based on specific patterns sphinx: kernel_abi: fix performance regression with O= Documentation: core-api: entry: Replace deprecated KVM entry/exit functions docs: fault-injection: drop reference to md-faulty docs: document linked lists scripts: kdoc: make it backward-compatible with Python 3.7 docs: kernel-doc: emit warnings for ancient versions of Python Documentation/rtla: Describe exit status Documentation/rtla: Add include common_appendix.rst docs: kernel: Clarify printk_ratelimit_burst reset behavior Documentation: ioctl-number: Don't repeat macro names Documentation: ioctl-number: Shorten macros table Documentation: ioctl-number: Correct full path to papr-physical-attestation.h Documentation: ioctl-number: Extend "Include File" column width Documentation: ioctl-number: Fix linuxppc-dev mailto link overlayfs.rst: fix typos docs: kdoc: emit a warning for ancient versions of Python docs: kdoc: clean up check_sections() docs: kdoc: directly access the always-there KdocItem fields docs: kdoc: straighten up dump_declaration() ...

Documentation: treewide: Replace remaining spinics links with lore

2025-06-21T20:20:51+00:00

Long before introduction of lore.kernel.org, people would link to LKML threads on third-party archives (here spinics.net), which in some cases can be unreliable (as these were outside of kernel.org control). Replace links to them with lore counterparts (if any). Signed-off-by: Bagas Sanjaya Reviewed-by: Simon Horman Signed-off-by: Jonathan Corbet Link: https://lore.kernel.org/r/20250611065254.36608-2-bagasdotme@gmail.com

Documentation: usb: gadget: Wrap remaining usage snippets in literal code block

2025-06-19T10:25:32+00:00

Several configfs usage snippets forget to be formatted as literal code blocks. These were outputted in htmldocs output as normal paragraph instead. In particular, snippet for custom string descriptors as added in 15a7cf8caabee4 ("usb: gadget: configfs: Support arbitrary string descriptors") is shown as single combined paragraph, rather than two command lines. Wrap them like the rest of snippets. Fixes: 5e654a4655c3 ("Documentation/usb: gadget_configfs") Fixes: d80b5005c5dd ("docs: usb: convert documents to ReST") Fixes: 15a7cf8caabe ("usb: gadget: configfs: Support arbitrary string descriptors") Signed-off-by: Bagas Sanjaya Acked-by: Uwe Kleine-König Link: https://lore.kernel.org/r/20250610031705.32774-2-bagasdotme@gmail.com Signed-off-by: Greg Kroah-Hartman

docs: usb: gadget: Reindent numbered list

2025-06-19T10:25:23+00:00

Paragraphs that are part of a numbered list must be indented to render correctly in html. Do that right here. There are only whitespace changes in this patch. Signed-off-by: Uwe Kleine-König Reviewed-by: Bagas Sanjaya Link: https://lore.kernel.org/r/20250607224747.3653041-2-u.kleine-koenig@baylibre.com Signed-off-by: Greg Kroah-Hartman