aboutsummaryrefslogtreecommitdiff
path: root/sys/dev
Commit message (Collapse)AuthorAgeFilesLines
* virtio console: plug set-but-not-used varsMateusz Guzik2021-12-141-3/+3
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* vmx: plug set-but-not-used varMateusz Guzik2021-12-141-10/+0
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* sdhci: plug set-but-not-unused varsMateusz Guzik2021-12-141-2/+0
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* vtnet: plug set-but-not-used varsMateusz Guzik2021-12-141-5/+5
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* sk: plug set-but-not-used varsMateusz Guzik2021-12-141-2/+0
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* sys/dev/arcmsr: Update Areca RAID driver to fix some issues on ARC-1886.黃清隆2021-12-131-3/+38
| | | | | | | | | | | 1. Doorbell interrupt status may arrive lately when doorbell interrupt on ARC-1886. 2. System boot up hung when ARC-1886 with no volume created or no device attached. Many thanks to Areca for continuing to support FreeBSD. MFC after: 2 weeks
* ds3231: fix accidental breakage of !FDT in 1256067cAndriy Gapon2021-12-131-1/+2
| | | | MFC after: 10 days
* ds3231: allow configuration via hints on FDT systemsAndriy Gapon2021-12-131-3/+6
| | | | MFC after: 10 days
* pcf8591: remove write-only variablesAndriy Gapon2021-12-131-4/+1
| | | | MFC after: 1 week
* amdsbwd: always enable watchdog register decodingAndriy Gapon2021-12-131-12/+11
| | | | | | | | | | | This seems to be required even if the watchdog is accessed via the common MMIO space. Tested on: - Ryzen 3 3200U APU; - Ryzen 7 5800X CPU with X570 chipset. MFC after: 2 weeks
* hdda: Remove extra topo_unlockWarner Losh2021-12-101-1/+0
| | | | | Reported by: Taku YAMAMOTO Sponsored by: Netflix
* pcf85063: Fix real time clockHubert Mazur2021-12-101-5/+17
| | | | | Previosuly real time clock driver always set time format to 12-hour mode. Fix this by choosing hour mode depending on HW register.
* bxe: plug some of set-but-not-used varsMateusz Guzik2021-12-101-12/+1
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* hn: plug set-but-not-unused varsMateusz Guzik2021-12-101-4/+4
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* bfe: plug a set-but-not-unused varMateusz Guzik2021-12-101-2/+4
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* nfe: plug set-but-not-unused varsMateusz Guzik2021-12-101-6/+2
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* alc: plug set-but-not-unused varsMateusz Guzik2021-12-101-1/+1
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* asmc: plug set-but-not-unused varsMateusz Guzik2021-12-101-0/+4
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* xdma: Fix another -Wunused-but-set-variable warning previously missedJessica Clarke2021-12-101-3/+0
| | | | MFC after: 1 week
* xdma: Fix -Wunused-but-set-variable warningsJessica Clarke2021-12-102-14/+0
| | | | MFC after: 1 week
* ahc: Mark sysctls with CTLFLAG_MPSAFE.Alexander Motin2021-12-101-2/+2
| | | | MFC after: 2 weeks
* fxp: Mark sysctls with CTLFLAG_MPSAFE.Alexander Motin2021-12-101-2/+2
| | | | MFC after: 2 weeks
* wbwd: Mark sysctls with CTLFLAG_MPSAFE.Alexander Motin2021-12-101-3/+3
| | | | MFC after: 2 weeks
* hptiop: Remove Giant locking around ioctls.Alexander Motin2021-12-101-5/+0
| | | | | | The methods have their own locking and don't require Giant. MFC after: 2 weeks
* hpt27xx: Remove FreeBSD 9.x support including Giant.Alexander Motin2021-12-103-151/+17
| | | | MFC after: 2 weeks
* fxp: Remove unneeded Giant from bus_dma_tag_create().Alexander Motin2021-12-101-8/+5
| | | | | | All bus_dmamap_load() calls here use BUS_DMA_NOWAIT. MFC after: 2 weeks
* acpi_cpu: Replace Giant with bus_topo_lock.Alexander Motin2021-12-101-2/+2
|
* hda: Push giant / newbus topo lock down a layerWarner Losh2021-12-102-5/+22
| | | | | | | | | | Rather than picking up Giant at the start of these sysctl handlers, push acquiring Giant down a layer to protect walking the children and also to add/delete children for the reconfigure. Sponsored by: Netflix Reviewed by: mav, jhb Differential Revision: https://reviews.freebsd.org/D33057
* newbus: add bus_topo_assertWarner Losh2021-12-106-11/+12
| | | | | | | | | | Add bus_topo_assert() and implmement it as GIANT_REQUIRED for the moment. This will allow us to change more easily to a newbus-specific lock int he future. Sponsored by: Netflix Reviewed by: wulf, mav, jhb Differential Revision: https://reviews.freebsd.org/D31833
* Create wrapper for Giant taken for newbusWarner Losh2021-12-1028-103/+104
| | | | | | | | | | | Create a wrapper for newbus to take giant and for busses to take it too. bus_topo_lock() should be called before interacting with newbus routines and unlocked with bus_topo_unlock(). If you need the topology lock for some reason, bus_topo_mtx() will provide that. Sponsored by: Netflix Reviewed by: mav Differential Revision: https://reviews.freebsd.org/D31831
* ioat: plug set-but-not-unused varsMateusz Guzik2021-12-091-5/+2
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* crypto: Validate AES-GCM IV length in check_csp().John Baldwin2021-12-093-11/+0
| | | | | | | | | | | This centralizes the check for valid nonce lengths for AES-GCM. While here, remove some duplicate checks for valid AES-GCM tag lengths from ccp(4) and ccr(4). Reviewed by: markj Sponsored by: The FreeBSD Foundation Differential Revision: https://reviews.freebsd.org/D33194
* cxgbe(4): Update firmwares to 1.26.4.0Navdeep Parhar2021-12-095-3/+3
| | | | | | | | | | | | | | | | | | | | | | | (Rest is from the README that came with the firmware) Version : 1.26.4.0 Date : 12/02/2021 Fixes ----- BASE: - Fixed error on setting 25G speed on 100G copper with multiple FEC set in firmware commands. - Handle link of unknown optics modules by enabling module tx unconditionally. - Fixed link not coming up for 25G CRS phys. Firmware incorrectly tried to bring up the link in RS-FEC but as per IEEE spec, it must be BASER FEC. - Fixed an issue where firmware doesn't automatically retry next FEC if driver asks to bring up the link using RS-FEC and link doesn't come up. Obtained from: Chelsio Communications MFC after: 1 month Sponsored by: Chelsio Communications
* modules: increase MAXMODNAME and provide backward compatBjoern A. Zeeb2021-12-091-0/+1
| | | | | | | | | | | | | | | | | | | | | | With various firmware files used by graphics and wireless drivers we are exceeding the current 32 character module name (file path in kldxref) length. In order to overcome this issue bump it to the maximum path length for the next version. To be able to MFC provide backward compat support for another version of the struct as the offsets for the second half change due to the array size increase. MAXMODNAME being defined to MAXPATHLEN needs param.h to be included first. With only 7 modules (or LinuxKPI module.h) not doing that adjust them rather than including param.h in module.h [1]. Reported by: Greg V (greg unrelenting.technology) Sponsored by: The FreeBSD Foundation Suggested by: imp [1] MFC after: 10 days Reviewed by: imp (and others to different level) Differential Revision: https://reviews.freebsd.org/D32383
* fxp: plug set-but-not-used varsMateusz Guzik2021-12-091-2/+0
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* firewire: plug set-but-not-used varsMateusz Guzik2021-12-091-13/+31
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* pcm: plug set-but-not-used varsMateusz Guzik2021-12-091-3/+1
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* et: plug set-but-not-used varsMateusz Guzik2021-12-091-6/+0
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* amdtemp: plug set-but-not-used varsMateusz Guzik2021-12-091-2/+2
| | | | Sponsored by: Rubicon Communications, LLC ("Netgate")
* sdhci_fsl_fdt: Add support for HS200/HS400 modesHubert Mazur2021-12-081-2/+534
| | | | | | | | | | | | | | | | | | | | The controller requires some custom logic to perform MMC tuning and to later switch to HS400 mode. Implement it supplying mmcbr_tune and sdhci_set_uhs_timing devmethods respectivly. Since the latter is called unconditionally when the ios is updated we need to keep track of the tuning state in sc and execute the HS400 switch logic only when required. Two HS200/HS400 related errata were implemented. 1. In HS400 modes the clock divisors are limited to 4, 8, 12. Apply it by falling back to the closes, higher divider when needed. 2. Hardware tuning procedure can sometimes fails. If that is the case fallback to the software tuning. Reviewed by: manu, mw Obtained from: Semihalf Sponsored by: Alstom Group Differential revision: https://reviews.freebsd.org/D33320
* sdhci_fsl_fdt: Fix vccq reconfigurationArtur Rojek2021-12-081-14/+30
| | | | | | | | | | | | | | | | | | | As `vqmmc-supply` is an optional Device Tree property, don't skip vccq reconfiguration when the regulator is not specified. Instead, accept voltage specified by the `voltage-ranges` property. The actual voltage switch is done through a hw register in LS1028A. Add errata flag for other boards, as they are not supported. Return not supported error code when switching to 1.8V on affected platforms. Fixes: b08bf4c35ca ("sdhci_fsl_fdt: Skip vccq reconfiguration without regulator") Co-authored-by: Artur Rojek <ar@semihalf.com> Reviewed by: manu, mw Obtained from: Semihalf Sponsored by: Alstom Group Differential revision: https://reviews.freebsd.org/D33319
* sdhci_fsl_fdt: Add LS1012A compat stringHubert Mazur2021-12-081-0/+7
| | | | | | | | | | Add compat string for LS1012A board and fill SoC specific structure data. Reviewed by: manu, mw Obtained from: Semihalf Sponsored by: Alstom Group Differential Revision: https://reviews.freebsd.org/D33318
* e1000: fix interface capabilities managementVincenzo Maffione2021-12-081-19/+23
| | | | | | | | | | | | | | | | | The e1000 drivers (em, lem, igb) are currently looking at the iflib copies of the capabilities bitvectors (scctx->isc_capabilities and scctx->isc_capenable) rather than the ifnet ones (ifp->if_capabilities and ifp->if_capenable). However, the latter are the ones that are actually updated by ifconfig and that should be used by the drivers during interface operation. The former are set by the driver on interface attach (for iflib internal use) and should not be used anymore by the driver. This patch fixes the e1000 driver to use the correct bitvectors. PR: 260068 Reviewed by: markj MFC after: 2 weeks Differential Revision: https://reviews.freebsd.org/D33154
* sdhci_fsl_fdt: Implement pulse width detection errataHubert Mazur2021-12-081-2/+20
| | | | | | | | | | | Some boards do not detect pulse width reliably. Implement workaround by writing 0 to special register. Apply errata for board by adding flag to chosen soc specific data. Obtained from: Semihalf Sponsored by: Alstom Group Reviewed by: mw, manu Differential Revision: https://reviews.freebsd.org/D33222
* enetc: Serialize MDIO transactionsKornel Duleba2021-12-082-4/+22
| | | | | | | | | | In theory we can have multiple concurrent accesses to the MDIO bus, e.g. link status check tick and ifconfig request. In that case we need to make sure that all MDIO transaction are serialized. Obtained from: Semihalf Sponsored by: Alstom Group
* enetc: Fix VID/mcast address hash calculationKornel Duleba2021-12-081-2/+2
| | | | | | | | | The hash is calculated by XOR-ing together bits of the VID. Prepend a statement with "!!" to make sure that we're get bit 0 on both sides of the equation. Obtained from: Semihalf Sponsored by: Alstom Group
* ocs_fc: Remove unused function delarations.Ram Kishore Vegesna2021-12-071-2/+6
| | | | | | | Remove unused function declarations. Changes required for internal tool. Approved by: ken
* flex_spi: Don't try to destroy disk if it doesn't existKornel Duleba2021-12-071-1/+4
| | | | | | Try to stop and destroy the dist only if the driver has been successfully attached. Otherwise a kernel panic will be triggered from disk_destroy. The issue was observed on a board that missed SPI clock description in DT.
* ifnet_byindex() actually requires network epochGleb Smirnoff2021-12-061-0/+6
| | | | | | | | | | | | | Sweep over potentially unsafe calls to ifnet_byindex() and wrap them in epoch. Most of the code touched remains unsafe, as the returned pointer is being used after epoch exit. Mark that with a comment. Validate the index argument inside the function, reducing argument validation requirement from the callers and making V_if_index private to if.c. Reviewed by: melifaro Differential revision: https://reviews.freebsd.org/D33263
* nvme_sim: Only report PCI related stats when we canWarner Losh2021-12-061-9/+19
| | | | | | | | | | | For AHCI attached devices, we report the location and identification information of the AHCI controller that we're attached to. We also don't reprot link speed in that case, since we can't get to the PCIe config space registers to find that out. Sponsored by: Netflix Reviewed by: mav Differential Revision: https://reviews.freebsd.org/D33287
generated by cgit v1.3 (git 2.53.0) at 2026-05-13 02:00:46 +0000