summaryrefslogtreecommitdiff
path: root/sys
AgeCommit message (Collapse)Author
2023-06-27Use shared net lock for DIOCGETIFACESKlemens Nanni
snmpd(8) and 'pfctl -s Interfaces' dump pf's internal list of interfaces. pf's internal interface list is completely protected by the pf lock, pf lock assertions since pf_if.c r1.110 from over a week ago support this. pfi_*() iterate over net lock protected if_groups lists, but only to read, so downgrade from exclusive write net lock to a shared read-only one. Feedback mvs OK sashan
2023-06-27Remove net lock from DIOC{SET,CLR}IFFLAGKlemens Nanni
pf.conf's 'set skip on ifN' and 'pfctl -F all|Reset' set and clear flags, PFI_IFLAG_SKIP being the only flag. Nothing else in base uses these ioctls and internal state is protected by the pf lock already. OK sashan
2023-06-27Attach 0x51f1 devices to iwx(4) and fix params used for 0x7a70 devices.Stefan Sperling
from reyk@
2023-06-27regenStefan Sperling
2023-06-27add 0x51f1 iwx(4) PCI device ID; from reyk@Stefan Sperling
2023-06-27amd64: MCOUNT_EXIT: restore interrupts, don't unconditionally reenable themScott Soule Cheloha
This bug can cause all sorts of problems, but in particular it was most easily reproduced as a double fault in the syscall return path on this CPU model: Intel(R) Core(TM) i7-8650U CPU @ 1.90GHz Tons of help from guenther@ in narrowing down the root cause. Fix tweaked by guenther@. Additional input from deraadt@ and kettenis@. ok guenther@
2023-06-27Return error if the USB request to get the sample rate fails.Alexandre Ratchov
found by mlarkin
2023-06-26timeout_hardclock_update: provide initial value for automatic variablesScott Soule Cheloha
2023-06-26Revert unrelated change that sneaked into the pf_ioctl.c commit.Claudio Jeker
2023-06-26Close all pf transactions before opening a new one in DIOCGETRULES.Claudio Jeker
Processes like snmpd or systat open pf(4) once and then issue many DIOCGETRULES calls over their runtime. This accumulates many pf_trans structs over their lifetime. At some point the kernel runs out of memory because of that. By closing all transactions before creating a new one, long living processes do no longer leak transactions. This probably needs further refinement once more transactions types are added but for now this solves the problem. Problem found by florian@ OK sashan@ kn@
2023-06-25Provide kstats based on the byte and packet counters available in someJonathan Matthew
dwge(4) implementations. The counters are all 32 bit, so enable reset-on-read and accumulate them into 64 bit software counters, and enable the MMC interrupts that indicate one or more counters is halfway to overflowing. Tested on an RK3399, which has the counters, and an Allwinner A20, which doesn't. ok dlg@
2023-06-24Calculate inet PCB SIP hash without table mutex.Alexander Bluhm
Goal is to run UDP input in parallel. Btrace kstack analysis shows that SIP hash for PCB lookup is quite expensive. When running in parallel, there is also lock contention on the PCB table mutex. It results in better performance to calculate the hash value before taking the mutex. The hash secret has to be constant as hash calculation must not depend on values protected by the table mutex. Do not reseed anymore when hash table gets resized. Analysis also shows that asserting a rw_lock while holding a mutex is a bit expensive. Just remove the netlock assert. OK dlg@ mvs@
2023-06-24Add initial support for StarFive VisionFive V2 to stfclock(4).Joel Sing
This adds initial support for the syscrg and pll clocks on the StarFive VisionFive V2 JH7110 SoC. ok kettenis@
2023-06-23Avoid division by 0 in m_pool_usedGreg Steuck
OK dlg@ Reported-by: syzbot+a377d5cd833c2343429a@syzkaller.appspotmail.com
2023-06-23ksmn(4): Support thermal sensors on Ryzen 9 79xxMike Larkin
ok deraadt, brynet
2023-06-22Revert "clockintr_cpu_init: initialize starting offsets with ↵Scott Soule Cheloha
clockintr_stagger()" octeon machines do not increment the global variable "ncpus" (init_main.c) in the same spot as other platforms, which violates the KASSERT in clockintr_stagger(), causing a panic. We need to bring octeon's behavior into alignment with every other platform before proceeding with this patch. Reported and debugged by bluhm@.
2023-06-22Instead of tsleep on lbolt do a tsleep with a 1 second timeout.Claudio Jeker
Result is the same and gets rid of a lbolt use. OK miod@
2023-06-22drm/amdgpu: Don't set struct drm_driver.output_poll_changedJonathan Gray
From Thomas Zimmermann 4b4cae8e4b37c7ad4d9721dbda060a62e49e3c17 in linux-6.1.y/6.1.35 0e3172bac3f43759719384403fe2d1e4c61f87e0 in mainline linux
2023-06-22drm/amdgpu: add missing radeon secondary PCI IDJonathan Gray
From Alex Deucher 21863dc45aed8d68a2573cce032ecbc8fe7c5bf5 in linux-6.1.y/6.1.35 e61f67749b351c19455ce3085af2ae9af80023bc in mainline linux
2023-06-22drm/amd/pm: workaround for compute workload type on some skusJonathan Gray
From Kenneth Feng 6f5b5ce9397d91b0a8bc8d15dc8fd72bc0aacaa4 in linux-6.1.y/6.1.35 7ca302d488f80cf4529620acc1c545f9022d8bb8 in mainline linux
2023-06-22drm/amd: Tighten permissions on VBIOS flashing attributesJonathan Gray
From Mario Limonciello b69a10df900386871703a4cc3be974485f52ff7c in linux-6.1.y/6.1.35 7ab1a4913d0051cf5196ef7987b5fa42c25e13b6 in mainline linux
2023-06-22drm/amd: Make sure image is written to trigger VBIOS image update flowJonathan Gray
From Mario Limonciello b2706d862b6577b8218bdef1c238e175624b906d in linux-6.1.y/6.1.35 3eb1a3a04056ba3df3205e169b8acc9da0c65a94 in mainline linux
2023-06-22drm/amd/display: edp do not add non-edid timingsJonathan Gray
From Hersen Wu ee8c6580c3106b0061c7cdff1804e5a1f1e1d7fe in linux-6.1.y/6.1.35 e749dd10e5f292061ad63d2b030194bf7d7d452c in mainline linux
2023-06-22drm/amdgpu: vcn_4_0 set instance 0 init sched score to 1Jonathan Gray
From Sonny Jiang 5d1fdfb3d185a9726478ba174235eff7c16f1101 in linux-6.1.y/6.1.35 9db5ec1ceb5303398ec4f899d691073d531257c3 in mainline linux
2023-06-22drm:amd:amdgpu: Fix missing buffer object unlock in failure pathJonathan Gray
From Sukrut Bellary 8ca9880735b0a33a4ff54328d6c5eb60fe2b0a6d in linux-6.1.y/6.1.35 60ecaaf54886b0642d5c4744f7fbf1ff0d6b3e42 in mainline linux
2023-06-22drm: panel-orientation-quirks: Change Air's quirk to support Air PlusJonathan Gray
From Maya Matuszczyk c45aed74318e65b8da938a593c124ad2ee1a1e66 in linux-6.1.y/6.1.35 1aa7f416175619e0286fddc5fc44e968b06bf2aa in mainline linux
2023-06-21Remove extra trailing whitespaceMike Larkin
ok stsp
2023-06-21Revert "schedcpu, uvm_meter(9): make uvm_meter() an independent timeout"Scott Soule Cheloha
Sometimes causes boot hang after mounting root partition. Thread 1: https://marc.info/?l=openbsd-misc&m=168736497407357&w=2 Thread 2: https://marc.info/?l=openbsd-misc&m=168737429214370&w=2
2023-06-21call rk3328 not rk3288 functions in rk3328_pinctrl()Jonathan Gray
tested by and ok kettenis@
2023-06-20schedcpu, uvm_meter(9): make uvm_meter() an independent timeoutScott Soule Cheloha
uvm_meter(9) should not base its periodic uvm_loadav() call on the UTC clock. It also no longer needs to periodically wake up proc0 because proc0 doesn't do any work. schedcpu() itself may change or go away, but as kettenis@ notes we probably can't completely remove the concept of a "load average" from OpenBSD, given its long Unix heritage. So, (1) remove the uvm_meter() call from schedcpu(), (2) make uvm_meter() an independent timeout started alongside schedcpu() during scheduler_start(), and (3) delete the vestigial periodic proc0 wakeup. With input from deraadt@, kettenis@, and claudio@. deraadt@ cautions that this change may confuse administrators who hold the load average in high regard. Thread: https://marc.info/?l=openbsd-tech&m=168710929409153&w=2 general agreement with this direction from kettenis@ ok claudio@
2023-06-20Fix definition of EMMC_DLL_TXCLK register offset.Mark Kettenis
ok dlg@
2023-06-20drm/amd/display: Reduce sdp bw after urgent to 90%Jonathan Gray
From Alvin Lee 8953be60ec33cba8289a855a886f64508f8c2c0c in linux-6.1.y/6.1.34 e1a600208286c197c2696e51fc313e49889315bd in mainline linux
2023-06-20drm/amd/pm: Fix power context allocation in SMU13Jonathan Gray
From Lijo Lazar 8695a443ad8f38e239b2080f02c4d739d6e33686 in linux-6.1.y/6.1.34 1d13c49cf4e246b218d71873f1bb1bbd376aa10e in mainline linux
2023-06-20drm/amdgpu: change reserved vram info printJonathan Gray
From YiPeng Chai 8e143bae25cdbab69ea3e2e7bb6faf7711ce7583 in linux-6.1.y/6.1.34 dac652220ba0e5a2ef2da2a47a60b60aea333fdb in mainline linux
2023-06-20drm/amdgpu: fix xclk freq on CHIP_STONEYJonathan Gray
From Chia-I Wu 34419aa0b448b0eee941102793893e0e256abda1 in linux-6.1.y/6.1.34 b447b079cf3a9971ea4d31301e673f49612ccc18 in mainline linux
2023-06-20drm/amd/pm: conditionally disable pcie lane switching for some ↵Jonathan Gray
sienna_cichlid SKUs From Evan Quan 416ba40ff3c4ecafbed19029c9e9884a7fee3da5 in linux-6.1.y/6.1.34 38e4ced804796c5725e2a52ec3601951552c4a97 in mainline linux
2023-06-20drm/i915/gt: Use the correct error value when kernel_context() failsJonathan Gray
From Andi Shyti 8d42c563e4ddbe2d0d71249f137b02daa66972db in linux-6.1.y/6.1.34 40023959dbab3c6ad56fa7213770e63d197b69fb in mainline linux
2023-06-20drm/amdgpu: fix Null pointer dereference error in amdgpu_device_recover_vramJonathan Gray
From Horatio Zhang c5a17f3247bd7f6c2e22678dbfcd73832f487e3f in linux-6.1.y/6.1.34 2a1eb1a343208ce7d6839b73d62aece343e693ff in mainline linux
2023-06-20drm/i915/selftests: Add some missing error propagationJonathan Gray
From Tvrtko Ursulin 76eef453a675794e98400c8889446b1503a9dfa6 in linux-6.1.y/6.1.34 79d0150d2d983a4f6efee676cea06027f586fcd0 in mainline linux
2023-06-20drm/i915/selftests: Stop using kthread_stop()Jonathan Gray
From Tvrtko Ursulin 4e7f1f6da79e1acda58b738e34ff972821d13072 in linux-6.1.y/6.1.34 6407cf533217e09dfd895e64984c3f1ee3802373 in mainline linux
2023-06-20drm/i915: Use 18 fast wake AUX sync lenJonathan Gray
From Jouni Hogander 5f285409c91bb97d78b05477d57a921ba1156465 in linux-6.1.y/6.1.34 2d6f2f79e06571d41eb1223abebe9097511c9544 in mainline linux
2023-06-20drm/i915: Explain the magic numbers for AUX SYNC/precharge lengthJonathan Gray
From Ville Syrjala 7bf7bebdc20df3d53a6c7af583d704e811fed85d in linux-6.1.y/6.1.34 26bfc3f36f2104c174dfc72415547d5c28ef3f1c in mainline linux
2023-06-19Fix RK3588 support. Makes dwmmc(4) actually work.Mark Kettenis
ok patrick@
2023-06-19Add another reset for the PCIe controller on the RK3588.Mark Kettenis
ok dlg@
2023-06-19drm/amd/display: Have Payload Properly Created After ResumeJonathan Gray
From Fangzhi Zuo 4042d7ad40f1c3182225461c62926f217a3d0ede in linux-6.1.y/6.1.33 482e6ad9adde69d9da08864b4ccf4dfd53edb2f0 in mainline linux
2023-06-19drm/amd/pm: reverse mclk and fclk clocks levels for renoirJonathan Gray
From Tim Huang 8db2ea7b80d6539c2c3d4e2ca96ca1bc56b982d9 in linux-6.1.y/6.1.33 55e02c14f9b5fd973ba32a16a715baa42617f9c6 in mainline linux
2023-06-19drm/amd/pm: reverse mclk and fclk clocks levels for yellow carpJonathan Gray
From Tim Huang f05f3fcc782445b707b998859d9a03410ecbdb73 in linux-6.1.y/6.1.33 f1373a97a41f429e0095d4be388092ffa3c1a157 in mainline linux
2023-06-19drm/amd/pm: reverse mclk clocks levels for SMU v13.0.5Jonathan Gray
From Tim Huang 1c729bd5b30fe8178c53aeebf6503ea38966f769 in linux-6.1.y/6.1.33 c1d35412b3e826ae8119e3fb5f51dd0fa5b6b567 in mainline linux
2023-06-19drm/amd/pm: resolve reboot exception for si olandJonathan Gray
From Guchun Chen 0f8f233ed76754b0c9262eb2e82f8529da0bef16 in linux-6.1.y/6.1.33 e490d60a2f76bff636c68ce4fe34c1b6c34bbd86 in mainline linux
2023-06-19drm/amd/pm: reverse mclk and fclk clocks levels for vangoghJonathan Gray
From Tim Huang e0a0f5d2ba593bbf155a41890f675390fd312e2c in linux-6.1.y/6.1.33 bfc03568d9d81332382c73a1985a90c4506bd36c in mainline linux
generated by cgit v1.2.3 (git 2.25.1) at 2024-12-17 07:32:54 +0000