IABSD.fr/src/sys/dev
Branch :
Log
| Author | Commit | Date | CI | Message |
|---|---|---|---|---|
 |
892a73ca | 2022-01-17 12:42:30 | change irq work from interrupt context back to process context like our 5.10 drm used (via a task) by changing from timeout_set() to timeout_set_proc() irq work is supposed to be in interrupt context but some path used on inteldrm with gen 9 graphics sleeps. Reported by Agnosto Dvonik on kaby lake and jcs@ and aja@ on comet lake. For Agnosto Dvonik this occured when running firefox or mpv, for jcs@ when playing a video in firefox, for aja@ after 5 min of running gnome. aja@ has run gnome for many hours with this change without problem. | |
|
f56254de | 2022-01-17 03:54:03 | unstub dma_fence_chain_find_seqno() for non zero seqno | |
|
08475498 | 2022-01-16 22:43:25 | drm/i915: Avoid bitwise vs logical OR warning in snb_wm_latency_quirk() From Nathan Chancellor 48d56b00c35266b00ec92aaf0db5b71ce1f27702 in linux 5.15.y/5.15.15 2e70570656adfe1c5d9a29940faa348d5f132199 in mainline linux | |
 |
3830ce0b | 2022-01-16 11:34:05 | modernise pcn_start with m_defrag and the "new" ifq pattern. aside from getting rid of a use of the ifq_deq_begin/commit/rollback stuff, it simplifies the code a lot. im also sick of having the diff in my tree. | |
|
c7f72189 | 2022-01-15 04:10:01 | bring back spelling fix from rev 1.138 lost in rev 1.140 pointed out by Brad | |
 |
36ce76ee | 2022-01-14 15:00:16 | We don't have 64-bit atomics on powerpc, but we don't really need them. So don't provide atomic64_cmpxchg() on powerpc and on other architectures make its implementation similar to atomic64_xchg(). This makes the tree build again on macppc. suggested by & ok jsg@ | |
|
5ca02815 | 2022-01-14 06:52:58 | update drm to linux 5.15.14 new hardware support includes Intel ehl/Elkhart Lake (embedded) jsl/Jasper Lake (atom) rkl/Rocket Lake (desktop) AMD van gogh APU (gfx1033) yellow carp / rembrandt APU (gfx1035?) Ryzen 6000 APU navy flounder / navi 22 (gfx1031) RX 6700, RX 6700 XT, RX 6700M, RX 6800M, RX 6850M XT dimgrey cavefish / navi 23 (gfx1032) Pro W6600, Pro W6600M, RX 6600, RX 6600 XT, RX 6600M, RX 6600S, RX 6650M, RX 6650M XT, RX 6700S, RX 6800S beige goby / navi 24 (gfx1034) RX 6500 XT, RX 6400, RX 6500M, RX 6300M Thanks to the OpenBSD Foundation for sponsoring this work niklas@ for helping with ttm and amdgpu and patrick@ for adapting rockchip drm. | |
 |
5e6f4d12 | 2022-01-12 19:55:12 | Extend the verbose logging. | |
 |
4582e297 | 2022-01-12 11:18:30 | Make acpi_getpropint() return uint64_t, as ACPI integers are in fact that wide and some _DSD properties depend on it. ok kettenis@ | |
|
548b81c5 | 2022-01-11 21:41:15 | Jasper Lake eMMC needs the same 0V quirk as Apollo Lake and Gemini Lake fixes accessing eMMC on Acer Swift 1 SF114-34 problem reported and fix tested by Sven Wolf | |
 |
c44af5cb | 2022-01-11 11:51:14 | add Synopsys Degisnware UART (dw-apb-uart) support To fix Allwinner H6's UART problem, need to add dw-apb-uart special code. ok kettenis@ | |
 |
a76e37e6 | 2022-01-11 10:34:13 | Remove KASSERT(0) and default switch case. No other sc_ncm_format switch has a default case and umb_ncm_setup_format() ensures that only 16 and 32bit formats are accepted. Fixes build error without DIAGNOSTIC set. Found by and OK robert@ | |
|
33d99505 | 2022-01-11 10:23:17 | fix RSB_DMCR_DEVICE_MODE_DATA value to enter RSB mode correctly ok kettenis@ patrick@ | |
|
1de6044a | 2022-01-11 06:23:05 | Add temporary verbose logging when remote coverage fails to attach. In the hopes of tracking down a rare bug seen on syzkaller. | |
|
c1a72b65 | 2022-01-11 06:00:41 | In revision 1.43 of kcov.c, the redundant conditional of checking for an exising kcov descriptor with the given device minor was removed since kcov is a cloning device; i.e. the device minor should always be unique. However, there's one edge case to still consider in which one thread have tracing enabled while another thread closes the same kcov descriptor. The kcov descriptor is kept alive until thread with tracing enabled exits to prevent usage after free. This does however cause the spec file layer above to flag the device minor as unused. Any subsequent open of /dev/kcov would trip on the assertion in kcovopen() until the thread with tracing enabled exits. Therefore unconditionally remove the kcov descriptor from the global list of active descriptors which is fine since the same kcov descriptor will later be freed in kcov_exit(). I have never seen this in the wild but realized while hunting another bug. | |
|
d6969a1f | 2022-01-11 00:37:23 | match on Intel Jasper Lake cavs/hda is pci class multimedia subclass audio so not automatically matched by azalia but confirmed to work after matched tested by Sven Wolf on Acer Swift 1 SF114-34 with Pentium Silver N6000 | |
|
caa2e0aa | 2022-01-10 23:41:12 | regen | |
|
4707620c | 2022-01-10 23:40:37 | add Intel Jasper Lake devices from Intel Pentium Silver and Intel Celeron Processors Datasheet 633935 | |
 |
91654044 | 2022-01-10 18:23:39 | Return ENOMEM on malloc errors to prevent use of uninitialized stack memory. Cleanup error handling while here. ok stsp@ visa@ | |
|
6a44dfb4 | 2022-01-10 04:47:53 | provide access to the hardware counters (MIB counters in the doc) via kstat | |
|
f86fe765 | 2022-01-10 04:11:13 | handle the status ring entries as 64bit words instead of a struct. the status ring entries are 8 bytes/64bit, and depending on the type of entry it has fields all over the place. this loads the descriptor with a single 64bit read, and then shifts and masks the bits out of it depending on the type of descriptor. this looks cleaner for the tx completions in particular. | |
|
4c58e589 | 2022-01-09 23:28:19 | Simplify igc_rxrinfo() to improve consistency and appease Coverity. CID 1510483 ok kevlo@ | |
|
4b1a56af | 2022-01-09 05:42:36 | spelling feedback and ok tb@ jmc@ ok ratchov@ | |
 |
64eea40c | 2022-01-08 22:54:49 | __LDPGSZ hasn't been used here since rev 1.23 (2013). Delete comment referring to it ok jsg@ | |
 |
9a56c644 | 2022-01-08 00:20:10 | Apply mpsafe changes from dwge(4) to dwxe(4): Rework the tx path to use the consumer and producer positions to work out the number of slots available, and to put packets on the ring until fewer than DWXE_NTXSEGS slots are left, making dwxe_start() and dwxe_txeof() work independently. While here, only write to DWXE_TX_CTL1 once per call to dwxe_start() rather than once per packet. Adjust the rx interrupt path to check the number of slots in use and return slots once per interrupt. Add interrupt and ifq barriers before taking the interface down. With all of this done, we can mark dwxe(4) mpsafe. tested on arm64 (a64 sopine) by mlarkin@ and armv7 (h2+) by me ok dlg@ | |
|
d4f6d54d | 2022-01-07 09:08:15 | fix aac build after -Wno-uninitialized was removed | |
|
ede0c0b9 | 2022-01-06 06:19:41 | regen | |
|
762450b1 | 2022-01-06 06:19:11 | ati 0x15e7 confirmed to be barcelo, 2022 ryzen 5000 apus uses the same green sardine firmware as cezanne | |
|
2bb0cdd1 | 2022-01-06 01:40:19 | drm/amdgpu: add support for IP discovery gc_info table v2 From Alex Deucher b8553330a07749e488d143b5704adf1042fd7c0a in linux 5.10.y/5.10.90 5e713c6afa34c0fd6f113bf7bb1c2847172d7b20 in mainline linux | |
|
ec2bd244 | 2022-01-06 01:37:46 | drm/amdgpu: When the VCN(1.0) block is suspended, powergating is explicitly enabled From chen gong 28863ffe21ff711d5109e3c208676258bdec3a1f in linux 5.10.y/5.10.90 b7865173cf6ae59942e2c69326a06e1c1df5ecf6 in mainline linux | |
|
64094178 | 2022-01-06 01:14:15 | unstub amdgpu_gem_force_release() | |
|
3852484d | 2022-01-05 18:54:20 | Use "bus-range" property to initialize the bus number configuration of the bridge when present on FDT platforms. Needed on platforms like the Apple M1 to make sure the PCI bus numbers match the IOMMU setup required by the device tree. ok patrick@ | |
 |
4b907771 | 2022-01-05 17:06:20 | Remove unused function arguments in iwm/iwx interrupt handlers. pointed out by + ok millert@ | |
 |
22060bbf | 2022-01-05 16:46:11 | Compensate for i386 pcitag_t union ok jsg kettenis | |
|
60a617df | 2022-01-05 16:33:42 | In iwx(4), fix wrong pointer assignment in iwx_bar_frame_release(). This bug caused the driver to read block ack request information sent by firmware from the wrong offset. The driver flushes buffered frames and moves its Rx block ack window based on this information. Possible consequences of this bug are packet loss or even stalled traffic if the Rx BA window gets out of sync between driver and firmware. Though this effect might get cancelled out when the driver re-syncs the BA window in its regular Rx code path. Spotted by Christian Ehrhardt. | |
|
83aa0ba6 | 2022-01-05 05:18:24 | rename ETHERTYPE_PAE to ETHERTYPE_EAPOL. everyone else seems to use ETHERTYPE_EAPOL, and as a bonus it also appears to be more correct. ok deraadt@ stsp@ | |
|
331b91e4 | 2022-01-05 03:53:26 | use the index provided by the txcompletion descriptor in txeof. this replaces the use of a register in txeof to figure out how much of the ring has been completed by the hardware. that register isn't reset when an interface is taken down and brought up again, which messes with the calculations of free slots on the ring, which in turn messes up the ability to transmit packets. this means the watchdog can't know where the hardware is up to anymore, so just restartthe chip if the watchdog fires. it seems to be important to start the tx ring in msk_init on the 0th ring entry that we use to reset the high address value to 0. we still fill the 0th descriptor, but we let the first msk_start call post it for us when a packet goes on the ring. a slight tweak and ok jmatthew@ | |
|
6543b26c | 2022-01-04 20:43:44 | struct pci_matchid is useful an upcoming userland program that wants to includes pcivar.h because it inspects vid/pid tables | |
|
fae5bdc0 | 2022-01-04 16:15:28 | Restrict the pci(4) ioctl interface to devices detected by the kernel. This fixes issues on the M1 Macs where the PCI probe done by Xorg breaks the WiFi chip. ok patrick@ | |
|
0df2aed8 | 2022-01-04 15:55:28 | fix length boundary checks for incoming packets in iwm/iwx The minimum length and the maximum length required were both too low, due to an error in accounting for the 4-byte packet length+flags header. Patch by Christian Ehrhardt | |
|
a048bec5 | 2022-01-04 15:53:57 | fix Rx Block Ack session validity checks in iwm(4) and iwx(4) I ported a NULL pointer check from iwlwifi rxmq.c which cannot happen in our version of this code. Instead we need to check the BA session ID to determine whether a BA session is valid. Patch by Christian Ehrhardt. | |
|
8a93f6f8 | 2022-01-04 13:40:58 | acpi_getprop() needs to actually make sure that we're looking at the correct property. While there adjust acpi_getpropint() as well to increase similarity with acpi_getprop(). ok kettenis@ | |
|
7797d4a0 | 2022-01-04 11:31:42 | remove the last part of driver for aic-6250 scsi on aviion | |
|
96f00877 | 2022-01-04 11:01:24 | remove files missed when sgi was removed ok visa@ | |
|
329157c6 | 2022-01-03 12:01:32 | the prism54 domain is long abandoned don't give any traffic to whoever registered it afterwards ok claudio@ stsp@ | |
|
f790b53d | 2022-01-03 09:48:41 | ansi | |
|
a7c97ccf | 2022-01-02 03:41:08 | addres -> address | |
|
b040ff10 | 2022-01-02 02:13:33 | preferense -> preference | |
|
f2e739fd | 2022-01-01 18:52:26 | Use correct defines for random seed magic/length. Spotted by Andreas Schnebinger | |
|
d3677684 | 2021-12-31 23:19:50 | spelling ok ratchov@ | |
|
86dca86f | 2021-12-31 11:24:24 | Interrups -> Interrupts | |
|
2e962c76 | 2021-12-31 09:24:18 | Newer Apple firmware on chipsets without a hardware RNG require the host to provide a buffer of random bytes to the device on initialization. | |
|
093575dd | 2021-12-31 08:15:47 | Fix up handling of IFF_ALLMULTI and call uaq_iff() from uaq_init(). from Brad | |
 |
54a76c43 | 2021-12-30 15:09:49 | Add Xiaomi MiniWif to list of devices supported by mtw(4). Works for me on amd64. ok hastings@ | |
|
61c8c0f0 | 2021-12-30 06:55:11 | Prevent a double free by assigning the new keymap and corresponding size after the allocation and initialization is done. Otherwise, a race is possible if malloc ends up sleeping. ok sashan@ Reported-by: syzbot+7f8224e9f1a3487caf25@syzkaller.appspotmail.com | |
|
ca9c4ded | 2021-12-30 00:25:41 | combine ifdef blocks from Ted Bullock | |
|
fe414f1d | 2021-12-29 18:48:45 | PHYs that we want to talk to over SGMII need to be switched into SGMII mode for the link to go up. This is relevant on RJ45 SFPs. ok sthen@ | |
|
0aed1edf | 2021-12-29 18:40:19 | Allow ohci(4) and ehci(4) to attach to acpi(4). ok kettenis@ | |
|
f11bc372 | 2021-12-29 17:52:22 | acpi_intr_disestablish() is available on all our ACPI platforms which support interrupts via ACPI, so there's need for an ifdef around it. ok kettenis@ | |
|
982627fc | 2021-12-29 07:15:13 | Do not allow send/receive of kcov descriptors as the file descriptor can be kept alive longer than expected causing syzkaller to no longer being able to enable remote coverage. ok visa@ Reported-by: syzbot+ab2016d729cda7b0d003@syzkaller.appspotmail.com | |
|
81a83fa0 | 2021-12-28 17:50:10 | The initial kd_lookup() in kcovopen() is redundant as kcov is a cloning device. Turn it into an assertion instead to better state the expected semantics of D_CLONE. Prompted after a discussion with visa@ | |
|
8bad4727 | 2021-12-27 17:56:03 | Not only BCM4378, but all PCIe core revisions >= 64 need to be accessed using the new sets of registers. | |
|
bb818c30 | 2021-12-27 17:32:14 | Map the chip ids used on Apple M1 Pro/Max and Apple T2 Macs to firmware names. | |
|
ea92f6c3 | 2021-12-27 17:12:34 | Support reading OTP information from a few more chips, necessary to learn firmare names on Apple M1 Pro/Max and Apple T2 Macs. | |
|
48b3e30f | 2021-12-27 15:38:25 | Yet another errno correction. | |
|
0813c1e9 | 2021-12-27 14:31:15 | Fix off-by-one in blob download, where in the case that the blob is evenly divisible by 1400, the last chunk isn't marked with an end flag. ok tobhe@ | |
|
64be38ba | 2021-12-27 14:28:13 | BCM4387 needs the D11 cores to be held in reset by us, the firmware will take care of releasing them, as otherwise initialization would fail some of the time. That chip also contains 3 of these, so make sure we reset all of them. Necessary on Apple M1 Pro/Max. | |
|
da5a4af2 | 2021-12-27 13:54:39 | Send TxCap and WiFi calibration blobs to the chip. | |
|
92fbeeb7 | 2021-12-27 12:03:59 | Switch module codename retrieval to use the newly proposed device tree bindings. | |
|
5ea0cc51 | 2021-12-27 10:59:20 | Handle trailing odd-sized 1024 byte blocks when calculating TCM RAM size. | |
|
1c0efb8b | 2021-12-27 10:56:55 | Bump rxpost and rxcomplete ring size to 1024 for newer chips. | |
|
8db7132f | 2021-12-26 20:50:17 | Add information about TCM rambase and how to check the SR capability for a few more chips. | |
|
a5d79199 | 2021-12-26 13:55:36 | Add address locators for the ACPI "bus" and use these to fix the order of the com(4) devices to match the traditional order one the ISA bus. ok patrick@, anton@ | |
 |
acaa8857 | 2021-12-24 06:53:18 | Support more mtw(4) devices. ASUS USB-N10 v2, D-Link DWA-127 rev B1, Edimax EW-7711UAn v2, various Ralink/MediaTek ids. ok stsp@ | |
|
9d2191b7 | 2021-12-24 06:19:24 | regen | |
|
2f4c4673 | 2021-12-24 06:18:11 | Add some more mtw(4) devices. ASUS USB-N10 v2, D-Link DWA-127 rev B1, Edimax EW-7711UAn v2, various Ralink/MediaTek ids. ok stsp@ | |
|
ad8c1ddc | 2021-12-23 12:14:15 | Get rid of unused next battery level argument. | |
 |
41a45098 | 2021-12-23 10:09:16 | Disk lock was held when returning to userland. Add a missing unlock in vnd ioctl error path. Reported-by: syzbot+6dde3fda33074a256318@syzkaller.appspotmail.com OK jsg@ anton@ | |
|
81e5db3b | 2021-12-23 01:39:44 | give et_setmulti() more chance of working ok claudio@ | |
|
1b8f8c22 | 2021-12-22 19:37:33 | Disable minimum power consumption in hostap mode. This improves connection reliability when bwfm is used as an access point. ok patrick@ | |
|
642ce627 | 2021-12-22 12:33:02 | drm/amdgpu: correct register access for RLC_JUMP_TABLE_RESTORE From Le Ma aec5897b277b13acd8f913d777654d4d092a24f1 in linux 5.10.y/5.10.88 f3a8076eb28cae1553958c629aecec479394bbe2 in mainline linux | |
|
71e30825 | 2021-12-22 12:30:38 | drm/amd/pm: fix a potential gpu_metrics_table memory leak From Lang Yu 222cebd995cdf11fe0d502749560f65e64990e55 in linux 5.10.y/5.10.88 aa464957f7e660abd554f2546a588f6533720e21 in mainline linux | |
|
08da896e | 2021-12-22 01:38:36 | %Lx -> %llx kernel printf does not support %Lx | |
|
57ec0946 | 2021-12-21 20:53:46 | Move checks on attach arguments from attach into match. ok anton@, deraadt@ | |
|
b73a164c | 2021-12-21 11:46:01 | knf nits | |
|
7f01505d | 2021-12-21 09:35:08 | Do not print "dt: 451 probes" at boot in dmesg. Btrace device dt(4) is enabled by default, this line does not provide much information. requested by kettenis@ deraadt@; OK mpi@ | |
|
731cc6c2 | 2021-12-21 07:44:22 | avoid returning uninitialised var in hidpp_send_report() ok anton@ | |
|
7a09fc77 | 2021-12-21 06:12:03 | Fix another vnd race pointed out by mpi@ and make sure to not unlock the vnode twice in the error path. Tested in snaps for a couple of days. | |
|
4d2db379 | 2021-12-21 06:11:16 | Ensure that the disk has been initialized after acquiring the lock and not before as we might end up sleeping while acquiring the lock, introducing a potential race. Tested in snaps for a couple of days. ok mpi@ Reported-by: syzbot+c87cdc2905b441c20d39@syzkaller.appspotmail.com | |
|
6c4edd67 | 2021-12-21 06:10:29 | Rename local variable intrfn to intr, matches what dev/fdt/com_fdt.c already does. | |
|
13d90461 | 2021-12-21 06:09:47 | Do not attach com at acpi when there's no address or irq present. Fixes a regression caused by the recent change to start attaching com at acpi as it turns out that Libreboot exposes console devices lacking crucial data in their acpi tables. The same console attaches fine over isa, therefore restore this behavior. Problem reported by <cipher-hearts at riseup dot net> on bugs@ ok deraadt@ kettenis@ | |
|
706a9991 | 2021-12-21 06:08:57 | errno overhaul, getting rid of some ambiguity. In the hopes of tracking down a rare but annoying problem related to remote coverage exposed by syzkaller. | |
|
cf997a48 | 2021-12-20 22:28:48 | Remove useless suser assert from dt(4). The ioctl(2) path checks the user anyway and close(2) may crash after setuid(2). Reported-by: syzbot+90e094f33d329fb2c3ab@syzkaller.appspotmail.com OK deraadt@ | |
|
b5553ee6 | 2021-12-20 19:24:32 | bus_dmamem_unmap() should not be called from interrupt context, so free and close flowrings using bwfm_do_async(). Reported by and ok kettenis@ | |
|
7a10236f | 2021-12-20 15:08:10 | Make iwm(4) attach to PCI devices with product ID 0x31dc. This device is part of the 9560 chip family. With a small device-specific quirk we can make iwm(4) attach and provide a working wifi interface. Problem reported and fix tested by Joao Victor. | |
|
b40d31ed | 2021-12-20 15:05:12 | regen | |
|
8e110848 | 2021-12-20 15:04:58 | Change a PCI device string: The "Gemini Lake CNVi" is an iwm(4) "AC 9560". ok hastings@ (who submitted the original string, based on a datasheet) | |
|
f225ca39 | 2021-12-20 14:54:37 | Make mtw(4) known to the config(8) framework. Not enabled yet. Pending firmware availability. ok stsp@ jmatthew@ | |
|
57a8187d | 2021-12-20 13:59:02 | Add mtw(4), a driver for MediaTek MT7601U wifi devices. Ported from run(4) with legacy chipsets removed. Not yet enabled in the build. ok stsp@ jmatthew@ | |
|
26d66e1b | 2021-12-20 04:21:32 | Rework the tx path to use the consumer and producer positions to work out the number of slots available, and to put packets on the ring until fewer than DWGE_NTXSEGS slots are left, making dwge_start() and dwge_txeof() work independently. While here, only write to GMAC_TX_POLL_DEMAND once per call to dwge_start() rather than once per packet. Adjust the rx interrupt path to check the number of slots in use and return slots once per interrupt. Add interrupt and ifq barriers before taking the interface down. With all of this done, we can mark dwge(4) mpsafe. ok dlg@ patrick@ |