syzbot

 sign-in | mailing list | source | docs
🐞 Open [167]
🐞 Fixed [299]
🐞 Invalid [988]
≡ Crashes
Kernel Health Bugs/Month Bug Lifetimes Fuzzing
💬 Send us feedback

Title Rank 🛈 Repro Cause bisect Fix bisect Count Last Reported Patched Closed Patch
assert "flags & (PR_WAITOK | PR_NOWAIT)" failed in subr_pool.c -1 2 35d 35d 3/3 33d eab2e406eb7b pfr_attach_table() needs wait flag.
panic: maddr rwlock ADDR: enter read deadlock 2 2529 71d 72d 3/3 71d 2b6fc957b059 Backout: Protect IGMP and MLD6 fast timer with rwlock.
witness: shared lock of (rwlock) maddr while exclusively locked -1 1910 71d 72d 3/3 71d 2b6fc957b059 Backout: Protect IGMP and MLD6 fast timer with rwlock.
panic: excl->share 2 98 71d 72d 3/3 71d 2b6fc957b059 Backout: Protect IGMP and MLD6 fast timer with rwlock.
protection_fault: in6_addmulti (5) -1 1 102d 102d 3/3 72d 80bc9799356e Protect IGMP and MLD6 fast timer with rwlock.
kernel: privileged instruction fault trap, code=NUM (2) -1 1 98d 98d 3/3 86d dca1e142a338 Fix race in vmm(4) where
uvm_fault: vm_terminate (2) -1 2 97d 114d 3/3 89d 10e7775f0707 Fix race in vmm(4) vm termination path.
assert "part >= NUM && part < MAXPARTITIONS" failed in subr_disk.c -1 117 121d 122d 3/3 119d b5bf9d3d0aeb bogus asserts; ok krw gnezdo
panic: mtx ADDR: locking against myself (2) 2 338 214d 370d 3/3 213d 5e82fd317c62 Prevent lock recursion in error path.
panic: malloc: allocation too large, type = NUM, size = ADDR (5) 2 1 236d 236d 3/3 227d 753a922a6fb4 Set the upper boundary of 'kern.seminfo.semopm' variable to "(MALLOC_MAX / sizeof(struct sembuf))". Otherwise the greater value could exceed the MALLOC_MAX limit and cause panic.
uvm_fault: dt_ioctl_record_stop -1 1306 228d 344d 3/3 228d 5b6ddc98dee7 Fix sleeping race in dt(4) ioctl(2).
assert "nlevel >= IPL_NONE" failed in intr.c (3) -1 1 281d 281d 3/3 275d b64faffc3616 run ifq_barrier against all ifqs in if_detach, not just if_snd.
assert "d->bd_in_uiomove == NUM" failed in bpf.c (3) -1 1 281d 281d 3/3 275d b64faffc3616 run ifq_barrier against all ifqs in if_detach, not just if_snd.
assert "(TAILQ_NEXT(inp, inp_queue) == NULL) || (TAILQ_NEXT(inp, inp_queue) == _Q_INVALID)" failed in in_pcb.c -1 16 277d 277d 3/3 276d 6bb594b93bf9 Fix use-after-free of inpcb.
protection_fault: in_pcb_iterator -1 4 277d 277d 3/3 276d 6bb594b93bf9 Fix use-after-free of inpcb.
witness: userret: sysctl -1 4 283d 284d 3/3 282d d8a43137eb27 Add forgotten NET_UNLOCK_SHARED() in FILLINPTABLE().
witness: thread exiting with locks held (3) -1 3 282d 284d 3/3 282d d8a43137eb27 Add forgotten NET_UNLOCK_SHARED() in FILLINPTABLE().
panic: mutex ADDR not held in in_pcb_iterator 2 5 285d 286d 3/3 285d b9215d49b9b3 Add forgotten mtx_enter() in FILLINPTABLE().
uvm_fault: ether_frm_ctloutput -1 3 318d 319d 3/3 318d 7d4a6d53c3d0 handle NULL for the option data mbufs in the ethernet setsockopt handling
panic: assertwaitok: non-zero mutex count: NUM (3) 2 1 336d 336d 3/3 334d 42337cd26dad Release `sb_mtx' mutex(9) while doing sleeping m_copym(..., M_WAIT) in soreceive() and somove(). It is possible in both places. We copy only `len' bytes from the single mbuf(9) pointed by `m'. The `len' is always less than m->m_len. The m->m_len could only grow while `sb_mtx' is unlocked, but concurrent thread will not override our chunk of m->m_data. There is no difference with the lockless uiomove(mtod(m)).
assert "va >= entry->start" failed in uvm_fault.c (2) -1 171 366d 406d 3/3 350d 818e6b572963 Make vslock(9) similar to mlock(2): silently ignore non-mapped ranges.
panic: uvm_fault_unwire_locked: address not in map (4) 2 C 638 366d 1443d 3/3 350d 818e6b572963 Make vslock(9) similar to mlock(2): silently ignore non-mapped ranges.
uvm_fault: rtrequest (4) -1 1 394d 394d 3/3 387d 45a541306066 Move kassert from resolve to add case in rtrequest().
panic: malloc: allocation too large, type = NUM, size = ADDR (3) 2 58 403d 406d 3/3 403d 910ed27a3d72 Limit net.bpf.maxbufsize sysctl(8) to a value that malloc(9) can handle. Introduce MALLOC_MAX definition to keep this value in sync and use it system wide.
openbsd test error: SYZFATAL: image testing failed w/o kernel bug -1 47 407d 413d 3/3 405d db5d28f093d6 Bump kcov buffer size limit
assert "nlevel >= IPL_NONE" failed in intr.c (2) -1 122 483d 741d 3/3 483d a921796a245d make sure bpfsdetach is holding a bpf_d ref when invalidating stuff.
pool: double put: shmpl -1 12 496d 570d 3/3 496d 4a445d448ac2 Prevent a double free by unlinking the descriptor before unmaping & freeing it.
panic: kernel diagnostic assertion "uvm_page_owner_locked_p(pg) 2 70 510d 513d 3/3 499d 0535051c104d Check if the mapping for an vm_map_entry exists while holding its lock.
panic: free: size too large NUM > NUM (ADDR) type sysctl 2 1 539d 539d 3/3 537d dd2b8016139a Fix sleeping race during malloc in sysctl hw.disknames.
assert "to_ticks >= NUM" failed in kern_timeout.c -1 1 595d 595d 3/3 587d 2293e68203ec restrict the maximum wait time you can set via BIOCSWTIMEOUT to 5 minutes.
multicore boot error: uvm_fault (2) -1 12 814d 814d 3/3 611d ef9354f58d69 sync
panic: lock (rwlock) solock not locked 2 C 33 1289d 1290d 3/3 614d ef9354f58d69 sync
witness: reversal: lock order data missing (3) -1 8192 614d 1462d 3/3 614d ef9354f58d69 sync
panic: vmxon failed 2 syz 3 703d 708d 3/3 702d ddfb6951b469 vmm: protect vmm activation with the vmm_softc rwlock.
uvm_fault: igmp_leavegroup (5) -1 C 12 704d 705d 3/3 704d 4733ced31b52 Prevent changing interface loopback flag from userland.
panic: unix: lock not held 2 C 2 716d 716d 3/3 710d e679c1f37540 Allow listen(2) only on sockets of type SOCK_STREAM or SOCK_SEQPACKET. listen(2) man(1) page clearly prohibits sockets of other types.
panic: rw_enter: unix locking against myself 2 C 2 733d 733d 3/3 728d d1ea0a7c7de9 Do UNP_CONNECTING and UNP_BINDING flags check in uipc_listen() and return EINVAL if set. This prevents concurrent solisten() thread to make this socket listening while socket is unlocked.
panic: malformed IPv4 option passed to ip_optcopy (3) 2 syz 40 741d 742d 3/3 741d c8a0ef6cb991 Validate IPv4 packet options in divert output.
assert "to->to_kclock == KCLOCK_UPTIME" failed in kern_timeout.c (3) -1 1 764d 764d 3/3 751d 0a34672fd384 timeout: make to_kclock validation more rigorous
kqueue: knote !ACTIVE (3) -1 1 771d 771d 3/3 769d d618cc99d763 Invert broken check of panic string in if_linkstate().
assert "sin6tosa(&ro->ro_dst)->sa_family == AF_INET6" failed in in6_src.c -1 C 218 773d 775d 3/3 773d 4a2c3c9ecbcf In in_pcbrtentry() add missing return of in6_pcbrtentry() value.
uvm_fault: refcnt_take -1 1 788d 788d 3/3 787d 470ec98d6191 Backout priterator() for walking allprocess list.
protection_fault: fill_file (3) -1 1 797d 797d 3/3 791d d221342b6cd4 Introduce priterator(), the `ps_list' iterator. Some of `allprocess' list walkthroughs have context switch within, so make exit1() wait until the last reference released.
witness: reversal: solock pf_lock -1 2 795d 802d 3/3 794d 66bd633ef3ce Use domain name for socket lock.
witness: reversal: pf_lock solock -1 syz 23 795d 802d 3/3 794d 66bd633ef3ce Use domain name for socket lock.
assert "dtlookup(unit) == NULL" failed in dt_dev.c (3) -1 1 805d 805d 3/3 803d c57c5c683707 Prevent simultaneous dt(4) open.
assert "__mp_lock_held(&sched_lock, curcpu()) == NUM" failed in kern_lock.c (4) -1 1 819d 819d 3/3 818d 540e94b5c59a soreceive() must not hold mutex when calling sblock().
panic: assertwaitok: non-zero mutex count: NUM 2 C 10 818d 819d 3/3 818d 540e94b5c59a soreceive() must not hold mutex when calling sblock().
uvm_fault: rtrequest (2) -1 C 15 854d 854d 3/3 853d a08e228de7c0 Fix rt_setgate() error handling.
assert "!_kernel_lock_held()" failed in kern_fork.c (3) -1 107 868d 1518d 3/3 868d 6d6c8141fa4d Adjust KERNEL_ASSERT_UNLOCKED() to not assert during a panic.
assert "!_kernel_lock_held()" failed in uvm_map.c (2) -1 13 903d 1120d 3/3 868d 6d6c8141fa4d Adjust KERNEL_ASSERT_UNLOCKED() to not assert during a panic.
assert "((len + sizeof(long) - NUM) &~ (sizeof(long) - NUM)) <= M_SIZE(m)" failed in uipc_mbuf.c -1 C 2 881d 881d 3/3 878d 7b4d35e0a60b Avoid assertion failure when splitting mbuf cluster.
multicore boot error: witness: lock_object uninitialized: ADDR -1 108 890d 893d 3/3 890d 33aa5105d87e Revert commitid: KtmyJEoS0WWxmlZ5 --- Protect interface queues with read once and mutex.
assert "p->p_stat == SONPROC || p->p_stat == SSLEEP || p->p_stat == SSTOP" failed in kern_synch.c -1 syz 10 912d 941d 3/3 909d de92f022b000 Before coredump or in pledge_fail use SINGLE_UNWIND to stop all threads.
assert "type != NULL" failed in subr_witness.c -1 syz 1 924d 924d 3/3 915d ed07db5bc17e Revert commitid: yfAefyNWibUyjkU2, ESyyH5EKxtrXGkS6 and itscfpFvJLOj8mHB;
assert "nlevel >= IPL_NONE" failed in intr.c -1 C 259 915d 924d 3/3 915d ed07db5bc17e Revert commitid: yfAefyNWibUyjkU2, ESyyH5EKxtrXGkS6 and itscfpFvJLOj8mHB;
witness: reversal: &sched_lock &pr->ps_mtx -1 syz 170 918d 920d 3/3 918d 4ab8b759f72e Fix SCHED_LOCK() leak in single_thread_set()
kernel: integer divide fault trap, code=NUM (4) -1 C 6 1003d 1059d 3/3 997d 6d14abdcc245 Avoid division by 0 in m_pool_used
panic: rw_enter: rtlck locking against myself 2 5 1144d 1144d 3/3 1143d aa5e72b7343d Revert the `rt_lock' rwlock(9) diff to fix the recursive rwlock(9) acquisition.
panic: acquiring blockable sleep lock with spinlock or critical section held (kernel_lock) &kernel_lock (7) 2 syz 9 1171d 1187d 3/3 1159d 9cf4e548dcbe Use solock() instead solock_shared() within sys_getsockopt(). Otherwise we acquiring kernel lock with mutex(9) held. This partially reverts rev 1.205 of sys/kern/uipc_syscalls.c. Shared solock() is still fine for getsockname(2) and getpeername(2).
pool: free list modified: semapl -1 syz 2 1317d 1317d 3/3 1265d 5bf1588a93fe Fix memory corruptions with sysv semaphores due to sleeps in copyin, copyout and malloc. During a sleep another thread could delete the semaphore (and possibly allocate another one at the same location with different permissions) which would lead to an invalid access after wake up. Therefore check the semaphore pointer, the sequence, the permissions and some values in seminfo after each sleep. OK bluhm@ Reported-by: syzbot+60ba811fe2e8a6b0f975@syzkaller.appspotmail.com
panic: solock: lock not held 2 C 19 1289d 1289d 3/3 1289d 536333f5835e Use pru_send function to check socket splicing compatibility. Only checking socket type is not sufficient as it could splice together unix and inet sockets resulting in crashes. As splicing is about sending, the same send function looks like a good criteria. Reported-by: syzbot+fc6901d63d858d5dd00a@syzkaller.appspotmail.com Reported-by: syzbot+0e026f1bf8b259c6395e@syzkaller.appspotmail.com OK gnezdo@
assert "uvn->u_obj.uo_refs == NUM" failed in uvm_vnode.c -1 syz 8 1305d 1356d 3/3 1293d ce1ab8dd6b18 Lock vmobjlock then check u_flags & UVM_VNODE_VALID in uvn_attach
kernel: integer divide fault trap, code=NUM (3) -1 C 157 1294d 1316d 3/3 1294d b18f9d9c6b70 Syzkaller found a missing input validation in pipex mppe keylenbits. Reported-by: syzbot+2eca95b271d07ab91b43@syzkaller.appspotmail.com tested yasuoka@; OK mvs@ yasuoka@
panic: tcp_output 2 C 762 1295d 1296d 3/3 1295d 355f588a8749 Sendmsg could crash in tcp_output due to a missing check after the introduction of tcp_send. OK mvs@, bluhm@, gnezdo@ Reported-by: syzbot+e859fd353c90eeac26f8@syzkaller.appspotmail.com
no output from test machine (7) -1 C 124310 1322d 1945d 3/3 1322d fa90ac5c787b Add a pool for the allocation of the pf_anchor struct. It was possible to exhaust kernel memory by repeatedly calling pfioctl DIOCXBEGIN with different anchor names. OK bluhm@ Reported-by: syzbot+9dd98cbce69e26f0fc11@syzkaller.appspotmail.com
witness: userret: sendmsg -1 C 28 1353d 1353d 3/3 1353d 635da3731041 Unlock peer in the SOCK_STREAM and SOCK_SEQPACKET error path.
assert "__mp_lock_held(&sched_lock, curcpu()) == NUM" failed in kern_lock.c -1 C 860 1356d 1467d 3/3 1356d 1690b19512cc Move the deep check back into the loop. There are ways that even though we're deep the code will SSTOP and sleep and then on wakeup we need to recheck the deep conditions. Issue analyzed and OK by mpi@
pool: free list modified: fdescpl (2) -1 C 6 1364d 1409d 3/3 1358d dc399801d5c5 kqueue: Clear task when closing kqueue
protection_fault: pf_anchor_global_RB_REMOVE -1 73 1397d 1484d 3/3 1359d 30d709c8a85c Allow waiting during ktable allocation in pf_ioctl.
uvm_fault: pf_anchor_global_RB_INSERT_COLOR (2) -1 1 1360d 1360d 3/3 1359d 30d709c8a85c Allow waiting during ktable allocation in pf_ioctl.
assert "TAILQ_EMPTY(&lock->lf_blkhd)" failed in vfs_lockf.c (2) -1 C 2 1389d 1389d 3/3 1383d ee8a225221fd Fix ambiguity with lock range end
panic: m_copydata: null mbuf (2) 2 syz 2 1396d 1396d 3/3 1393d 0271abd8e494 In pf the kernel paniced if IP options in packet within ICMP payload were truncated. Drop such packets instead. Reported-by: syzbot+91abd3aa2fdfe900f9ce@syzkaller.appspotmail.com OK sashan@ claudio@
multicore build error (15) -1 6 1411d 1411d 3/3 1411d ba537df4ec28 Using mutex initializer for static variable does not compile with witness. Make ratecheck mutex global. Reported-by: syzbot+9864ba1338526d0e8aca@syzkaller.appspotmail.com
witness: reversal: pf_lock netlock (3) -1 syz 25 1417d 1429d 3/3 1417d f3a753b5d089 Release PF und NET lock before calling copyout for DIOCIGETIFACES. OK sashan@ Reported-by: syzbot+b6afd166e314799e3809@syzkaller.appspotmail.com
protection_fault: lf_advlock -1 C 2 1465d 1509d 3/3 1418d c30ab30fe47b Remove the lock if an identical overlapping one is already present. Preventing a use after free discovered by syzkaller.
protection_fault: lf_findoverlap -1 C 2 1432d 1464d 3/3 1418d c30ab30fe47b Remove the lock if an identical overlapping one is already present. Preventing a use after free discovered by syzkaller.
assert "TAILQ_EMPTY(&lock->lf_blkhd)" failed in vfs_lockf.c -1 1 1432d 1432d 3/3 1418d c30ab30fe47b Remove the lock if an identical overlapping one is already present. Preventing a use after free discovered by syzkaller.
protection_fault: lf_clearlock -1 C 1 1434d 1434d 3/3 1418d c30ab30fe47b Remove the lock if an identical overlapping one is already present. Preventing a use after free discovered by syzkaller.
witness: reversal: pf_lock netlock (2) -1 C 2 1438d 1438d 3/3 1436d 2e34be423f16 Release PF und NET lock before calling copyin for DIOCXROLLBACK. OK bluhm@ Reported-by: syzbot+2945769fc3e6fd9ee413@syzkaller.appspotmail.com
witness: reversal: pf_lock netlock -1 C 19 1449d 1501d 3/3 1438d 82d8999861ab Release PF und NET lock before calling copyin and copyout for DIOCXBEGIN. OK bluhm@ OK sashan@ Reported-by: syzbot+b22ec16c5bf937578937@syzkaller.appspotmail.com
assert "next != NULL && next->start <= entry->end" failed in uvm_fault.c (4) -1 C 9 1442d 1456d 3/3 1442d f77c87828076 Replace KASSERT in uvm_fault_unwire_locked() with code that handles the case where not all pages are wired. The KASSERT can be triggered in multi-threaded applications when a thread calling munmap(2) races another thread that invokes sysctl(2). Properly written code shouldn't do this, but making the kernel crash in this case is a bit harsh.
openbsd build error (17) -1 2 1449d 1449d 3/3 1449d Put call to vmx_remote_vmclear() under #ifdef MULTIPROCESSOR to unbreak build of amd64 GENERIC
uvm_fault: igmp_leavegroup (3) -1 37 1449d 1541d 3/3 1449d a134c703e536 if_detach() does if_remove(ifp); NET_LOCK(); rti_delete(). New igmp groups may join while sleeping in interface destruction. In this case if_get() in igmp_joingroup() fails and rti_fill() is not called. Then inm->inm_rti may be NULL. This is the condition when syzkaller crashes in igmp_leavegroup(). Pass the ifp the current CPU is already holding down to igmp_joingroup() and igmp_leavegroup() to avoid half constructed igmp groups. Calling if_get() in caller and callee makes no sense anyway. Reported-by: syzbot+146823a676b7bea83649@syzkaller.appspotmail.com OK denis@
openbsd boot error: uvm_fault -1 9 1453d 1454d 3/3 1453d 475618162a2c Revert previous. Breaks probing native IDE devices.
multicore boot error: uvm_fault -1 21 1453d 1454d 3/3 1453d 475618162a2c Revert previous. Breaks probing native IDE devices.
panic:p a anciqc:uikerrinngel dbialgocnkoastbliec a s s e r t i o n "! _k er neslle_lepo lcokc_khe l dw ( )i 2 1 1455d 1455d 3/3 1453d d25fea59a0de For raw IP packets rip_input() traverses the loop of all PCBs. From there it calls sbappendaddr() while holding the raw table mutex. This ends in sorwakeup() where we finally grab the kernel lock while holding a mutex. Witness detects this misuse. Use the same solution as for PCB notify. Collect the affected PCBs in a temporary list. The list is protected by exclusive net lock. syzbot+ebe3f03a472fecf5e42e@syzkaller.appspotmail.com OK claudio@
panic: acquiring blockable sleep lock with spinlock or critical section held (kernel_lock) &kernel_lock (6) 2 C 80 1454d 1455d 3/3 1453d 2be5be2c1f9e For raw IPv6 packets rip6_input() traverses the loop of all PCBs. From there it calls sbappendaddr() while holding the raw6 table mutex. This ends in sorwakeup() where we finally grab the kernel lock while holding a mutex. Witness detects this misuse. Use the same solution as for PCB notify. Collect the affected PCBs in a temporary list. The list is protected by exclusive net lock. Reported-by: syzbot+5b2679ee9be0895d26f9@syzkaller.appspotmail.com OK claudio@
panic: apcaqnuiicr:in g b l o kckearbnelle d isalgeneops t ic a s s leorctkio n " w!_iktehr nselp_inlolocckk_ 2 1 1455d 1455d 3/3 1453d d25fea59a0de For raw IP packets rip_input() traverses the loop of all PCBs. From there it calls sbappendaddr() while holding the raw table mutex. This ends in sorwakeup() where we finally grab the kernel lock while holding a mutex. Witness detects this misuse. Use the same solution as for PCB notify. Collect the affected PCBs in a temporary list. The list is protected by exclusive net lock. syzbot+ebe3f03a472fecf5e42e@syzkaller.appspotmail.com OK claudio@
panic: acquiring blockable sleep lock with spinlock or critical section held (pkanerinc:e l _l o c kke)r &nkeerl ndiea 2 1 1455d 1455d 3/3 1453d d25fea59a0de For raw IP packets rip_input() traverses the loop of all PCBs. From there it calls sbappendaddr() while holding the raw table mutex. This ends in sorwakeup() where we finally grab the kernel lock while holding a mutex. Witness detects this misuse. Use the same solution as for PCB notify. Collect the affected PCBs in a temporary list. The list is protected by exclusive net lock. syzbot+ebe3f03a472fecf5e42e@syzkaller.appspotmail.com OK claudio@
panic: apcaquniirci: ng bkloerckneabll de i a g n o st ic a s s er ti o n " ! _ ke r n e l_ l oc k _ h sel 2 1 1454d 1454d 3/3 1453d 2be5be2c1f9e For raw IPv6 packets rip6_input() traverses the loop of all PCBs. From there it calls sbappendaddr() while holding the raw6 table mutex. This ends in sorwakeup() where we finally grab the kernel lock while holding a mutex. Witness detects this misuse. Use the same solution as for PCB notify. Collect the affected PCBs in a temporary list. The list is protected by exclusive net lock. Reported-by: syzbot+5b2679ee9be0895d26f9@syzkaller.appspotmail.com OK claudio@
panic: acquiring blockable sleep lock with spinlock or critical section held (kernelp_alonicck): &kkerernnelel d_ilao 2 1 1455d 1455d 3/3 1453d d25fea59a0de For raw IP packets rip_input() traverses the loop of all PCBs. From there it calls sbappendaddr() while holding the raw table mutex. This ends in sorwakeup() where we finally grab the kernel lock while holding a mutex. Witness detects this misuse. Use the same solution as for PCB notify. Collect the affected PCBs in a temporary list. The list is protected by exclusive net lock. syzbot+ebe3f03a472fecf5e42e@syzkaller.appspotmail.com OK claudio@
panic: acquiring blockable sleep lock with spinlock or critical section held (kernel_lock) &kernel_lock (5) 2 C 104 1455d never 3/3 1455d d25fea59a0de For raw IP packets rip_input() traverses the loop of all PCBs. From there it calls sbappendaddr() while holding the raw table mutex. This ends in sorwakeup() where we finally grab the kernel lock while holding a mutex. Witness detects this misuse. Use the same solution as for PCB notify. Collect the affected PCBs in a temporary list. The list is protected by exclusive net lock. syzbot+ebe3f03a472fecf5e42e@syzkaller.appspotmail.com OK claudio@
panic: acquiring blockable sleep lock with spinlock or critical section held (kernel_lock) &kernel_lock (4) 2 C 249 1455d 1456d 3/3 1455d 63abc0ec39b5 For multicast and broadcast packets udp_input() traverses the loop of all UDP PCBs. From there it calls udp_sbappend() while holding the UDP table mutex. This ends in sorwakeup() where we finally grab the kernel lock while holding a mutex. Witness detects this misuse. Use the same solution as for PCB notify. Collect the affected PCBs in a temporary list. The list is protected by exclusive net lock. Reported-by: syzbot+7596cb96fb9f3c9d6f4f@syzkaller.appspotmail.com OK sashan@
witness: reversal: vmmaplk fdlock -1 syz 22 1465d 1465d 3/3 1462d 447db83cf4f0 Revert holding a read lock on the map while copying out data during sysctl(2).
witness: exclusive lock of (rwlock) vmmaplk while share locked -1 C 9 1465d 1465d 3/3 1462d 447db83cf4f0 Revert holding a read lock on the map while copying out data during sysctl(2).
witness: reversal: netlock vmmaplk -1 C 166 1465d 1465d 3/3 1462d 447db83cf4f0 Revert holding a read lock on the map while copying out data during sysctl(2).
panic: kmmaplk: lock not shared 2 C 15 1465d 1466d 3/3 1465d 447db83cf4f0 Revert holding a read lock on the map while copying out data during sysctl(2).
assert "next != NULL && next->start <= entry->end" failed in uvm_fault.c (3) -1 syz 11 1466d 1481d 3/3 1465d 5f69141a284a Hold a read lock on the map while copying out data during a sysctl(2) call to prevent another thread from unmapping the memory and triggering an assertion or even corrupting random physical memory pages.
uvm_fault: socreate -1 C 106 1479d 1480d 3/3 1479d 532245610f13 Reported-by: syzbot+1b5b209ce506db4d411d@syzkaller.appspotmail.com Revert the pr_usrreqs move: syzkaller found a NULL pointer deref and I won't be available to monitor for followup issues for a bit
assert "sc->sc_dev == NUM" failed in if_tun.c (2) -1 1 1486d 1486d 3/3 1479d 1c9104c31d3f have another go at fixing assert "sc->sc_dev == NUM" failed.
assert "m->m_pkthdr.pf.prio <= IFQ_MAXPRIO" failed in ifq.c -1 syz 2 1495d 1495d 3/3 1489d 2960c8affbaf check pf rule "set prio" values consistently.
assert "sc->sc_dev == NUM" failed in if_tun.c -1 24 1490d 1522d 3/3 1489d 156bbf72d5e4 prevent (re)opening of tun/tap interfaces that are being destroyed.
assert "len >= NUM && !M_READONLY(m)" failed in uipc_mbuf.c -1 syz 153 1491d 1500d 3/3 1491d 4be097b86868 The length value in bpf_movein() is casted to from size_t to u_int and then rounded before checking. Put the same check before the calculations to avoid overflow. Reported-by: syzbot+6f29d23eca959c5a9705@syzkaller.appspotmail.com OK claudio@
kernel: integer divide fault trap, code=NUM (2) -1 C 8 1497d 1500d 3/3 1495d 43496f5abc1b let pfattach() to also initialize pf_default_rule_new to avoid div-by-zero in pf_purge()
panic: unhandled af (4) 2 syz 11 1514d 1543d 3/3 1511d 035d4f5430cb An af-to pf rule must have an address family naf to use after translation. Make stricter sanity checks in pf ioctl to avoid later crashes during packet processing. Reported-by: syzbot+0ef9190e7d0195496d0d@syzkaller.appspotmail.com OK sashan@
uvm_fault: x86_ipi_db (7) -1 4 1537d 1537d 3/3 1537d 7945134bbda7 Use a distinct variable while iterating the list of existing devices.
uvm_fault: pppacopen -1 C 92 1537d 1537d 3/3 1537d 7945134bbda7 Use a distinct variable while iterating the list of existing devices.
assert "TAILQ_EMPTY(&ifp->if_addrhooks)" failed in if.c (2) -1 1 1539d 1539d 3/3 1537d 188a0a692db5 Prevent concurrent access to incomplete or dying `sc' caused by sleep points in pppacopen() and pppacclose() paths. Use the same "sc_ready" logic we use for 'pppx_if' structure.
panic: free: size too small NUM <= NUM / NUM (ADDR) type devbuf 2 1 1540d 1540d 3/3 1537d 61c8c0f0fd3e 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.
SYZFAIL: ioctl remote attach failed (3) -1 13 1538d 1543d 3/3 1538d 982627fcf222 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.
assert "(rule != NULL) && (rule->ruleset != NULL)" failed in pf_ioctl.c -1 C 7 1541d 2084d 3/3 1541d 5a50f165b8ad DIOCHANGERRULE ioctl must set pointer to ruleset in rule it inserts.
witness: userret: ioctl (3) -1 syz 2 1545d 1545d 3/3 1544d 41a45098bac2 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@
uvm_fault: sysctl_diskinit -1 1 1551d 1551d 3/3 1544d 278923c07fda While malloc sleeps, the disk list could change during sysctl. Then allocated memory could be too short for the list of disks. Retry allocating enough space until it did not change. The disk list and duid memory are protected by kernel lock. Use asserts to mark this explicitly. Reported-by: syzbot+807423f6868bbfb836bc@syzkaller.appspotmail.com OK anton@ mpi@
uvm_fault: pfsync_state_import -1 C 10 2032d 2102d 3/3 1545d d13e571b26fd Remove ptr_array from struct pf_ruleset
panic: uvm_fault_unwire_locked: address not in map (3) 2 1 1548d 1548d 3/3 1546d 65315d4d4359 Fix a typo in mlock(2) error path triggering a double-free.
panic: vndioctl: null vp 2 C 3 1551d 1552d 3/3 1546d 4d2db379fd33 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.
panic: malloc: allocation too large, type = NUM, size = ADDR 2 C 3 1552d 1553d 3/3 1546d 0b7f5aec6317 Let malloc return an error as opposed of panicking when sysctl kern.shminfo.shmseg is set to something ridiculously large.
openbsd build error (15) -1 37 1547d 1609d 3/3 1546d revert vnode: remove VLOCKSWORK and check locking when vop_islocked != nullop (both kernel and userland bits)
assert "suser(curproc) == NUM" failed in dt_dev.c -1 C 7 1547d 1547d 3/3 1547d cf997a482905 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@
uvm_fault: uvm_fault_lower -1 C 583 1547d 1551d 3/3 1547d f08b8e936de9 Do not try to unlock a NULL object.
kernel: integer divide fault trap, code=NUM -1 syz 2 1560d 1560d 3/3 1551d 38bfd041cb0f fix zero division found by syzkaller. The sanity checks in pf(4) ioctls are not powerful enough to detect invalid port ranges (or even invalid rules). syzkaller does not use pfctl(8), it uses ioctl(2) to pass some random chunk of memory as a rule to pf(4). Fix adds explicit check for 0 divider to pf_get_transaddr(). It should make syzkaller happy without disturbing anyone else.
uvm_fault: igmp_leavegroup (2) -1 1 1555d 1555d 3/3 1551d 5af2eee02794 Syzkaller found a dereference in igmp_leavegroup() where inm->inm_rti is NULL. It should be set in rti_fill(), but is not if malloc(9) fails. There is no rollback after malloc failure so the field stays uninitialized. The code is only called from ioctl, setsockopt or a task. Malloc should wait instead of failing, otherwise syscalls would be unreliable. While there also put an M_WAIT in the init code. During init malloc must not fail. OK mvs@ Reported-by: syzbot+e22326057ccf34908d78@syzkaller.appspotmail.com
uvm_fault: nd6_dad_duplicated (2) -1 76 1554d 1560d 3/3 1554d ce9bd2eccc20 nd6_dad_ns_input() could trigger a NULL deref in nd6_dad_duplicated(). It checks dp in two of three places. One check got lost in revision 1.83. Do a dp == NULL once at the beginning. OK jsg@ Reported-by: syzbot+88c0ce914a0b10b7e1c8@syzkaller.appspotmail.com
witness: userret: ioctl (2) -1 1 1560d 1560d 3/3 1560d aa5628258068 Add missing kernel unlock in error path.
panic: rw_enter: dklk locking against myself 2 C 6 1622d 1650d 3/3 1618d placing the same vnd underneath a vnd (with VNDIOCSET) is a lock violation, but other circumstances are also bad, so let's block all vnd on top of vnd. While here, fix some toctou multiple-copyin of the path, and restructure the ioctl defer all softc updates to the end. ok mpi
multicore test error: timed out (2) -1 66 1632d 1634d 3/3 1632d 11ad9ee57f54 regen
openbsd build error (14) -1 2 1634d 1634d 3/3 1632d 11ad9ee57f54 regen
openbsd test error: timed out (2) -1 33 1632d 1634d 3/3 1632d 11ad9ee57f54 regen
multicore build error (13) -1 3 1634d 1634d 3/3 1632d 11ad9ee57f54 regen
assert "_kernel_lock_held()" failed in uvm_vnode.c -1 4 1657d 1658d 3/3 1651d 8e114673a836 vmm(4): grab kernel lock before vmspace init
panic: acquiring blockable sleep lock with spinlock or critical section held (kernel_lock) &kernel_lock (3) 2 C 2 1653d 1656d 3/3 1651d 8e114673a836 vmm(4): grab kernel lock before vmspace init
assert "_kernel_lock_held()" failed in uvm_device.c -1 7 1656d 1658d 3/3 1651d 8e114673a836 vmm(4): grab kernel lock before vmspace init
uvm_fault: rtm_output -1 C 14 1651d 1651d 3/3 1651d d2084f835172 Fix NULL pointer dereference introduced by previous commit.
panic: vcpulock: lock not held 2 1 1658d 1658d 3/3 1652d 5f3d69798ad5 vmm(4): fix vcpu locking issues reported by syzbot
panic: lock (rwlock) vcpulock not locked 2 C 1 1657d 1657d 3/3 1652d 5f3d69798ad5 vmm(4): fix vcpu locking issues reported by syzbot
panic: rw_enter: vcpulock locking against myself 2 1 1656d 1656d 3/3 1652d 5f3d69798ad5 vmm(4): fix vcpu locking issues reported by syzbot
assert "_kernel_lock_held()" failed in uvm_map.c -1 C 11 1656d 1658d 3/3 1655d 8e114673a836 vmm(4): grab kernel lock before vmspace init
panic: acquiring blockable sleep lock with spinlock or critical section held (kernel_lock) &kernel_lock (2) 2 C 152 1657d 2378d 3/3 1657d 1be9dae600aa Prevent lock ordering issue by raising ipl level of vcpu_pool to IPL_MPFLOOR.
panic: mutex ADDR not held in knote_dequeue 2 C 24 1739d 1740d 3/3 1739d baa47be05351 Remember to lock kqueue mutex in filt_timermodify().
witness: reversal: lock order data missing -1 C 667 1768d 1838d 3/3 1768d 8b594b45f4b5 Move ktrfds() below fdpunlock(). This fixes lock order issue between vn_lock(9) and fdplock().
panic: acquiring blockable sleep lock with spinlock or critical section held (rwlock) solock 2 C 4 1785d 1785d 3/3 1784d Revert per-socket `so_lock' rwlock(9) and use it to protect routing (PF_ROUTE) sockets. There is a locking issue with timeouts that needs to be fixed. Requested by deraadt@
SYZFAIL: tun: can't open device -1 syz 4 1838d 1838d 3/3 1832d 43dfcaac64e4 Issuing FIOSETOWN and TIOCSPGRP ioctl commands on a tun(4) device leaks device references causing a hang while trying to remove the same interface since the reference count will never reach zero. Instead of returning, break out of the switch in order to ensure that tun_put() gets called.
assert "frag->fr_firstoff[index] != NULL" failed in pf_norm.c -1 syz 4 1850d 1869d 3/3 1848d be7274bff2cd When cutting of the head of an overlapping fragment during pf reassembly, reinsert the fragment into the lookup table with correct index. Reported-by: syzbot+d043455a5346f726f1c4@syzkaller.appspotmail.com OK claudio@
kernel: protection fault trap, code=0 (8) -1 syz 1348 1860d 2049d 3/3 1860d 44a45654aa78 Interface group names must fit into IFNAMSIZ and be unique. But the kernel made the unique check before trunkating with strlcpy(). So there could be two interface groups with the same name. The kif is created by a name lookup. The trunkated names are equal, so there was only one kif owned by both groups. When the groups got destroyed, the single kif was removed twice from the RB tree. Check length of group name before doing the unique check. The empty group name was allowed and is now invalid. Reported-by: syzbot+f47e8296ebd559f9bbff@syzkaller.appspotmail.com OK deraadt@ gnezdo@ anton@ mvs@ claudio@
panic: rw_enter: pf_state_lock locking against myself 2 C 8 1861d 1861d 3/3 1860d d7220220b7ed pfsync_state_import() must not be called with the pf state lock held, since the actual modification of the state table is done by a call to pf_state_insert(), which takes the pf state lock itself. Other calls to pfsync_state_import() also only have the pf lock.
panic: vrele: v_writecount != 0 (2) 2 C 1955 1866d 1866d 3/3 1866d df61468f8652 Revert previous commit. The vnode returned by ptm_vn_open() is open and can not simply be vrele()-ed on error. The code currently depends on closef() to do the cleanup.
assert "TAILQ_EMPTY(&kq->kq_head)" failed in kern_event.c -1 C 2 1875d 1875d 3/3 1874d 715db9d67ba3 kqueue: Fix termination assert
uvm_fault: pf_addr_compare -1 C 347 1880d 1881d 3/3 1880d c34fe1b3cf88 An invalid packet may not have set src and dst in packet descriptor. Add a NULL check to prevent crash in pflog(4) introduced in previous commit. Reported-by: syzbot+c6d2f2ad34b822bce98a@syzkaller.appspotmail.com
uvm_fault: m_copyback -1 69 1898d 2046d 3/3 1881d 2cbebc019f52 pflog(4) tried to log the translated packet with rdr-to, nat-to, and af-to addresses and ports applied. Therefore it created a mbuf chain on the stack with a partial copy. This is too complicated for IP options, extension header, NAT46 af-to, and fragmented mbuf chains. It even caused a crash in syzkaller. Usually the length checks in pf_setup_pdesc() rejected the faked mbuf and the goto copy logged the packet unmodified. Remove the pflog_mtap() function and call bpf_mtap_hdr() directly. As the old buggy code was bypassed in most cases, tcpdump(8) output of pflog does not change. Uncondionally log the unmodified packet. Reported-by: syzbot+947e89e06ac3fec187d0@syzkaller.appspotmail.com OK sashan@
kernel: double fault trap, code=0 (4) -1 C 493 1905d 2162d 3/3 1892d c11d76984715 Syzkaller has found a stack overflow in socket splicing. Broadcast packets were resent through simplex broadcast delivery and socket splicing. Although there is an M_LOOP check in somove(9), it did not take effect. if_input_local() cleared the M_BCAST and M_MCAST flags with m_resethdr(). As if_input_local() is used for broadcast and multicast delivery, it was a mistake to delete them. Keep the M_BCAST and M_MCAST mbuf flags when packets are reinjected into the network stack. Reported-by: syzbot+a43ace363f1b663238f8@syzkaller.appspotmail.com OK anton@; discussed with claudio@
kernel: integer divide fault trap, code=0 -1 syz 4 2079d 2107d 3/3 1915d 39c2a1337a94 Reject rules with invalid port ranges
openbsd test error: uvm_fault: spllower -1 6 1930d 1930d 3/3 1923d 3ba77c9295b2 Revert previous extension of the SCHED_LOCK(), the state isn't passed down.
multicore test error: uvm_fault: spllower -1 12 1930d 1930d 3/3 1923d 3ba77c9295b2 Revert previous extension of the SCHED_LOCK(), the state isn't passed down.
uvm_fault: wsevent_fini (3) -1 C 2 1928d 1928d 3/3 1926d 996a5b4d63fa Fix yet another wscons race. In the same subsystem, the following properties must always hold true:
no output from test machine (6) -1 C 2779 1945d 1967d 3/3 1945d 0124df67671a Fix handling of MSG_PEEK in soreceive() for the case where an empty mbuf is encountered in a seqpacket socket.
no output from test machine (5) -1 C 49467 1967d 2064d 3/3 1967d d6d1940044d6 Fix a deadlock between uvn_io() and uvn_flush(). While faulting on a page backed by a vnode, uvn_io() will end up being called in order to populate newly allocated pages using I/O on the backing vnode. Before performing the I/O, newly allocated pages are flagged as busy by uvn_get(), that is before uvn_io() tries to lock the vnode. Such pages could then end up being flushed by uvn_flush() which already has acquired the vnode lock. Since such pages are flagged as busy, uvn_flush() will wait for them to be flagged as not busy. This will never happens as uvn_io() cannot make progress until the vnode lock is released.
witness: userret: ioctl -1 C 3 1972d 1972d 3/3 1971d 610c242e11d3 - missing NET_UNLOCK() in pf_ioctl.c error path
panic: spl assertion failure in yield 2 C 12 2099d 2117d 3/3 1972d - move NET_LOCK() further down in pf_ioctl.c. Also move memory allocations outside of NET_LOCK()/PF_LOCK() scope in easy spots.
openbsd build error (11) -1 44 1973d 1977d 3/3 1972d a524e041107f timeout(9): fix compilation under NKCOV
multicore build error (8) -1 88 1973d 1977d 3/3 1972d a524e041107f timeout(9): fix compilation under NKCOV
panic: malloc: allocation too large, type = 98, size = ADDR 2 C 2 1989d 1989d 3/3 1988d Recent changes for PROT_NONE pages to not count against resource limits, failed to note this also guarded against heavy amap allocations in the MAP_SHARED case. Bring back the checks for MAP_SHARED from semarie, ok kettenis https://syzkaller.appspot.com/bug?extid=d80de26a8db6c009d060
openbsd build error (10) -1 3 2010d 2010d 3/3 2008d Unbreak tree. Instead of passing struct process to siginit() just pass the struct sigacts since that is the only thing that is modified by siginit.
multicore build error (7) -1 5 2010d 2011d 3/3 2008d Unbreak tree. Instead of passing struct process to siginit() just pass the struct sigacts since that is the only thing that is modified by siginit.
panic: syn_cache_insert: cacheoverflow: impossible (2) 2 C 6 2045d 2091d 3/3 2023d 7c72bba2fa26 Convert tcp_sysctl to sysctl_bounded_args
assert "curproc->p_kd == NULL" failed in kcov.c (2) -1 1 2035d 2035d 3/3 2023d ece33e2f6ca2 Before clearing the kcov descriptor associated with a thread make sure no other thread is currently within a remote section. Otherwise, the remote subsystem could end up in a broken state where it doesn't reset the necessary bits upon leaving the remote section.
assert "curproc->p_kd == NULL" failed in kcov.c -1 2 2040d 2041d 3/3 2040d 2fa3abdda1a4 When detaching common remote coverage, do not clear any fields. Instead, let kr_free() do the work. Otherwise a thread currently inside a remote section could end up not decrementing the number of ongoing sections while exiting the same remote section.
uvm_fault: pfi_address_add -1 syz 1359 2060d 2117d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
pool: free list modified: phpool -1 1 2091d 2091d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
panic: sbdrop 2 syz 183 2051d 2351d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
uvm_fault: soreceive -1 syz 240 2060d 2325d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
panic: ifa_update_broadaddr does not support dynamic length (2) 2 13 2054d 2204d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
uvm_fault: in6ifa_ifpforlinklocal -1 1 2105d 2105d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
uvm_fault: pool_do_put (2) -1 syz 53 2061d 2304d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
kernel: protection fault trap, code=0 (7) -1 syz 774692 2049d 2293d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
uvm_fault: in_delmulti -1 syz 375431 2050d 2336d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
panic: tcp_output: template len != hdrlen - optlen 2 syz 487 2051d 2363d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
soreceive 1a -1 syz 510 2052d 2209d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
uvm_fault (2) -1 syz 1 2102d 2102d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
pool: cpu free list modified: mbufpl -1 syz 15863 2049d 2336d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
pool: free list modified: art_heap4 (2) -1 4 2092d 2122d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
soreceive 3 -1 syz 755 2049d 2209d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
uvm_fault: vio_rxeof -1 syz 15700 2060d 2371d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
uvm_fault: ifa_update_broadaddr -1 3275 2060d 2304d 3/3 2049d efa3c3dd644f Validate input given to ioctl(SIOCAIFADDR_IN6) like NetBSD already does. Fixes a bunch of panics reported by syzkaller.
multicore build error (6) -1 4 2055d 2068d 3/3 2052d Revert r1.170. dlg and jmatthew simultaneously fixed this the correct way.
openbsd build error (9) -1 1 2068d 2068d 3/3 2052d Revert r1.170. dlg and jmatthew simultaneously fixed this the correct way.
uvm_fault: wsevent_fini (2) -1 1 2076d 2076d 3/3 2055d 25f2901d1c23 Fix yet another panic in which wsevent_fini() ends up being called with NULL. This one is a race caused by clearing the me_evp member before calling routines that could end up sleeping.
no output from test machine (3) -1 C 25293 2071d 2085d 3/3 2071d A pty write containing VDISCARD, VREPRINT, or various retyping cases of VERASE would perform (sometimes irrelevant) compute in the kernel which can be heavy (especially with our insufficient tty subsystem locking). Use tsleep_nsec for 1 tick in such circumstances to yield cpu, and also bring interruptability to ptcwrite() https://syzkaller.appspot.com/bug?extid=462539bc18fef8fc26cc ok kettenis millert, discussions with greg and anton
panic: unhandled af (2) 2 C 22 2088d 2114d 3/3 2087d 38e8113e1d44 state import should accept AF_INET/AF_INET6 only
panic: syn_cache_insert: bucketoverflow: impossible 2 C 5 2100d 2118d 3/3 2097d 2b10bfc1e665 Refuse to set 0 or a negative value for net.inet.tcp.synbucketlimit.
panic: syn_cache_insert: cacheoverflow: impossible 2 5 2097d 2116d 3/3 2097d 2b10bfc1e665 Refuse to set 0 or a negative value for net.inet.tcp.synbucketlimit.
no output from test machine -1 C 551400 2138d 2713d 3/3 2138d 7bb4371dccb1 Do not wait indefinitely for flushing when closing a tty.
panic: m_copydata: null mbuf 2 C 396 2142d 2329d 3/3 2140d 574b3a4fa98d Do sanity checks in ip6_pullexthdr() preventing a panic in m_copydata(9).
assert "!ISSET(rt->rt_flags, RTF_LOCAL)" failed in nd6.c -1 1 2205d 2205d 3/3 2154d 8e6c5245c1d8 Never update the ND entry (cache) corresponding to a RTF_LOCAL route.
uvm_fault: pfi_dynaddr_remove -1 C 17 2263d 2267d 3/3 2156d 3d97bff14298 fix insufficient input sanitization in pf_rulecopyin() and pf_pool_copyin()
uvm_fault: pfr_detach_table -1 C 12 2263d 2266d 3/3 2156d 3d97bff14298 fix insufficient input sanitization in pf_rulecopyin() and pf_pool_copyin()
openbsd build error (8) -1 2 2161d 2161d 3/3 2161d 20c8eb7cf336 Add bse(4) device to unbreak build.
multicore build error (5) -1 4 2161d 2161d 3/3 2161d 20c8eb7cf336 Add bse(4) device to unbreak build.
panic: rw_enter: netlock locking against myself 2 syz 2 2167d 2167d 3/3 2164d 27427a72e313 In sosplice(), temporarily release the socket lock before calling FRELE() as the last reference could be dropped which in turn will cause soclose() to be called where the socket lock is unconditionally acquired. Note that this is only a problem for sockets protected by the non-recursive NET_LOCK() right now.
assert "p == curproc" failed in vfs_vops.c -1 C 187 2176d 2177d 3/3 2173d fc5a743df3a9 Revert previous, syzkaller found a way to trigger the KASSERT().
uvm_fault: strlcpy (2) -1 1 2185d 2185d 3/3 2178d 9fcf6ed4d02d Prevent out of bounds read in strlcpy due to vcp_name not being NUL-terminated.
panic: vputonfreelist: lock count is not zero 2 C 2 2208d 2209d 3/3 2180d 2a9890d8c8d9 Relax the lockcount assertion in vputonfreelist(). Back when I fixed several problems with the vnode exclusive lock implementation, I overlooked the fact that a vnode can be in a state where the usecount is zero while the holdcount still being positive. There could still be threads waiting on the vnode lock in uvn_io() as long as the holdcount is positive.
uvm_fault: wsevent_fini -1 1 2189d 2189d 3/3 2183d be78d62e13f6 Ensure that me_evp is still NULL before assignment during open of wscons devices. This condition is checked early on during open but since the same routine could end up sleeping before assigning me_evp, a race against adding the same wscons device to a wsmux could be lost. This in turn can cause a NULL deference during close.
kernel: double fault trap, code=0 (3) -1 C 69 2195d 2204d 3/3 2195d aa1987fe7ce4 Fix unlimited recursion caused by local outbound bcast/mcast packet sent via spliced socket.
pool: free list modified: aobjpl -1 C 3 2293d 2305d 3/3 2203d c5a231fb6c1e Grab a reference for the shared memory segment before calling uvm_map() as the same function could end up putting the thread to sleep. Allowing another thread to free the shared memory segment, which in turns causes a use-after-free.
openbsd boot error: uvm_fault: softclock -1 33 2204d 2205d 3/3 2204d previous commit accidentally aliased two unique timeouts hit by millert
multicore boot error: uvm_fault: softclock -1 66 2204d 2205d 3/3 2204d previous commit accidentally aliased two unique timeouts hit by millert
kqueue: knote !QUEUED -1 syz 8 2219d 2255d 3/3 2219d 8c4786361ce1 Raise SPL when updating kn_status. Otherwise the field can become inconsistent if knote_acquire() or knote_release() is preempted by an interrupt that modifies the same knote.
panic: unhandled af 2 C 21 2262d 2264d 3/3 2258d 4804479228fb Check address family of pf ioctl(2) DIOCNATLOOK parameter at kernel entry instead of calling panic() due to unhandled af. Reported-by: syzbot+92be143c2dd1746cf2af@syzkaller.appspotmail.com from Benjamin Baier
panic: attempt to execute user address 2 syz 1539 2269d 2363d 3/3 2265d f30ff743e528 Fix some races in kqueue_register().
witness: reversal: vmmaplk inode -1 C 103124 2280d 2607d 3/3 2280d Use separate rwlock initializations for userland ("vmspace") and kernel maps. This lets witness know that these really are different classes avoiding false positives when detecting lock order reversals.
kernel: protection fault trap, code=0 (6) -1 C 55024 2293d 2378d 3/3 2293d 0f83bb56e561 Fix a bad offset calculation in uvm_share.
panic: uvm_mapent_clone: no space in map for entry in empty map 2 C 12 2294d 2300d 3/3 2294d 0f83bb56e561 Fix a bad offset calculation in uvm_share.
panic: amap_pp_adjref: negative reference count 2 C 98 2304d 2356d 3/3 2294d 0f83bb56e561 Fix a bad offset calculation in uvm_share.
uvm_fault: amap_pp_adjref -1 1 2356d 2356d 3/3 2294d 0f83bb56e561 Fix a bad offset calculation in uvm_share.
uvm_fault: uvm_unmap_remove (2) -1 C 7836 2294d 2377d 3/3 2294d 3c82c0b2df98 Fix uvm_unmap_remove panic when tearing down VMs.
panic: uvmspace_fork: no space in map for entry in empty map 2 C 450 2302d 2356d 3/3 2294d 0f83bb56e561 Fix a bad offset calculation in uvm_share.
assert "TAILQ_EMPTY(&ifp->if_addrhooks)" failed in if.c -1 2 2314d 2316d 3/3 2309d 9e254176dfa1 take care to avoid a race when creating the same interface.
panic: ifa_update_broadaddr does not support dynamic length 2 syz 6780 2321d 2379d 3/3 2321d b36fd3da6fde Do propper kernel input validation for in_control() ioctl(2) SIOCGIFADDR, SIOCGIFNETMASK, SIOCGIFDSTADDR, SIOCGIFBRDADDR, SIOCSIFADDR, SIOCSIFNETMASK, SIOCSIFDSTADDR, and SIOCSIFBRDADDR. Name in_ioctl_set_ifaddr() consistently. Use in_sa2sin() to validate inet address. Combine if_addrlist loops and add comment. Although netmask is not a inet address, length must be valid. Reported-by: syzbot+5fc6da002fc4e8d994be@syzkaller.appspotmail.com OK visa@
pool: free list modified: shmpl -1 C 22 2476d 2571d 3/3 2330d d13730a27993 Copy in the user-supplied buffer in shmctl(2) before looking up the shared memory segment. Otherwise, if copyin ends up sleeping it allows another thread to remove the same segment leading to a use-after-free.
malloc: free list modified: devbuf -1 syz 5 2337d 2344d 3/3 2337d put bpfdesc reference counting back, revert change introduced in 1.175 as: BPF: remove redundant reference counting of filedescriptors
uvm_fault: kqueue_scan -1 1 2359d 2359d 3/3 2337d put bpfdesc reference counting back, revert change introduced in 1.175 as: BPF: remove redundant reference counting of filedescriptors
uvm_fault: witness_checkorder -1 syz 2 2340d 2340d 3/3 2337d put bpfdesc reference counting back, revert change introduced in 1.175 as: BPF: remove redundant reference counting of filedescriptors
uvm_fault: filt_bpfrdetach -1 syz 1 2337d 2337d 3/3 2337d put bpfdesc reference counting back, revert change introduced in 1.175 as: BPF: remove redundant reference counting of filedescriptors
uvm_fault: bpfioctl -1 C 14 2345d 2429d 3/3 2337d put bpfdesc reference counting back, revert change introduced in 1.175 as: BPF: remove redundant reference counting of filedescriptors
assert "ps->ps_uvncount == 0" failed in kern_unveil.c -1 syz 226 2364d 2632d 3/3 2357d a239dbafd306 Only increment the ps_uvncount counter when a path is successfully added to the corresponding red-black tree; meaning the path was not already present in the tree. This prevents an assertion to trigger in unveil_destroy() later on when the process exits.
assert "ln != NULL" failed in nd6.c -1 1 2383d 2383d 3/3 2360d bdbfbec5cea8 Do more sanity checks when accepting socket addresses in routing messages from user land. Inspect length field early in rtm_xaddrs(). Strings must be NUL terminated. The socket address type and length depend on the routing message type. Currently checks are not super strict to avoid too much user land fallout. OK mpi@ Reported-by: syzbot+638dbf7851da8e255af5@syzkaller.appspotmail.com
assert "ifa == rt->rt_ifa" failed in nd6.c -1 9 2360d 2361d 3/3 2360d bdbfbec5cea8 Do more sanity checks when accepting socket addresses in routing messages from user land. Inspect length field early in rtm_xaddrs(). Strings must be NUL terminated. The socket address type and length depend on the routing message type. Currently checks are not super strict to avoid too much user land fallout. OK mpi@ Reported-by: syzbot+638dbf7851da8e255af5@syzkaller.appspotmail.com
uvm_fault: strlcpy -1 18 2368d 2488d 3/3 2366d bdbfbec5cea8 Do more sanity checks when accepting socket addresses in routing messages from user land. Inspect length field early in rtm_xaddrs(). Strings must be NUL terminated. The socket address type and length depend on the routing message type. Currently checks are not super strict to avoid too much user land fallout. OK mpi@ Reported-by: syzbot+638dbf7851da8e255af5@syzkaller.appspotmail.com
uvm_fault: uvm_unmap_remove -1 C 780 2377d 2396d 3/3 2377d 00ba8250173b vm_teardown() must be serialized since it modifies the global vmm_softc structure. Therefore grab the appropriate lock before calling the same function. This issue has been known for a while and reported before but lacking a way to easily reproduce it; until syzkaller came up with a reproducer.
panic: malloc: allocation too large, type = 2, size = ADDR (2) 2 C 16842 2379d 2396d 3/3 2379d 225e50e8a961 Do not decrement the number of VMs counter twice in one of vm_create() error paths. If creation of the first VM fails, the counter will wrap around to a huge value. The same value could later be passed to malloc() through vm_get_info() causing a panic.
kernel: protection fault trap, code=0 (5) -1 C 607 2385d 2459d 3/3 2384d a42056240bd9 Fix a route use after free in IPv6 multicast route. Move the mrt6_mcast6_del() out of the rtable_walk(). This avoids recursion to prevent stack overflow. Also it allows freeing the route outside of the walk. Now mrt6_mcast_del() frees the route only when it is deleted from the routing table. If that fails, it must not be freed. After the route is returned by mf6c_find(), it is reference counted. Then we need a rtfree(), but not in the other case. Name mrt6_mcast_add() and mrt6_mcast_del() consistently. Move rt_timer_remove_all() into mrt6_mcast_del(). Reported-by: syzbot+af7d510593d74c825960@syzkaller.appspotmail.com OK mpi@
panic: acquiring blockable sleep lock with spinlock or critical section held (kernel_lock) &kernel_lock 2 33 2393d 2398d 3/3 2393d d3f3cb99fa83 Do not use the flow of the first fragment to store ECN information. Handle the ECN in the fragment queue. Reported-by: syzbot+0aa80b25e9041001cac8@syzkaller.appspotmail.com fix from FreeBSD; OK claudio@
uvm_fault: frag6_input -1 24 2393d 2397d 3/3 2393d d3f3cb99fa83 Do not use the flow of the first fragment to store ECN information. Handle the ECN in the fragment queue. Reported-by: syzbot+0aa80b25e9041001cac8@syzkaller.appspotmail.com fix from FreeBSD; OK claudio@
witness: userret: returning with the following locks held: -1 C 315 2393d 2604d 3/3 2393d b808994cf339 When a thread tries to exclusively lock a vnode, the same thread must ensure that any other thread currently trying to acquire the underlying vnode lock has observed that the same vnode is about to be exclusively locked. Such threads must then sleep until the exclusive lock has been released and then try to acquire the lock again. Otherwise, exclusive access to the vnode cannot be guaranteed.
panic: vput: ref cnt 2 C 6 2407d 2408d 3/3 2405d d627fa5cc364 Serialize access to the vnode pointers associated with acct(2) system accounting. Prevents a race where the acct thread and the acct(2) syscall both tries to close a vnode.
assert "timo || _kernel_lock_held()" failed in kern_synch.c -1 C 2 2407d 2408d 3/3 2407d 93e05fce3bab Revert unlock of lseek(2) since vn_lock() could end up calling tsleep() which is not allowed without holding the kernel lock. Otherwise, wakeups could be lost.
panic: vrele: v_writecount != 0 2 C 51 2415d 2421d 3/3 2415d 3e253b4759f0 Favor vn_close() in the error path of diskmapioctl() since side-effects caused by calling vn_open() with write permissions must be reverted. Otherwise, the vfs subsystem could panic while releasing the last vnode reference if the writecount is still positive.
panic: vput: v_writecount != 0 (2) 2 C 112 2415d 2421d 3/3 2415d 3e253b4759f0 Favor vn_close() in the error path of diskmapioctl() since side-effects caused by calling vn_open() with write permissions must be reverted. Otherwise, the vfs subsystem could panic while releasing the last vnode reference if the writecount is still positive.
assert "cpipe->pipe_buffer.cnt == 0" failed in sys_pipe.c -1 C 8 2442d 2442d 3/3 2442d 40f8ed5eebb0 backout the unlock of pipe(2) and pipe2(2)
kernel: protection fault trap, code=0 (4) -1 C 759 2460d 2573d 3/3 2459d cf34c7c30780 Prevent recursions by not deleting entries inside rtable_walk(9).
uvm_fault: arp_rtrequest -1 C 79 2478d 2489d 3/3 2467d 4cb088386ee5 In arp_rtrequest and nd6_rtrequest return early if the RTF_MPLS flag is set. These mpls routes use the rt_llinfo structure to store the MPLS label and would confuse the arp and nd6 code. OK bluhm@ anton@
assert "(rt->rt_flags & RTF_MPATH) || mrt->rt_priority != prio" failed in rtable.c -1 C 143 2479d 2492d 3/3 2467d ff10691ed095 Copy the user provided sockaddr into a normalized sockaddr in rtrequest() before adding it to the routing table. The rtable code is doing memcmp() of those rt_dest sockaddrs so it is important that they are stored in a canonical form. To do this struct domain is extended to include the sockaddr size for this address family. OK bluhm@ anton@
uvm_fault: mrt6_ioctl -1 C 2 2489d 2489d 3/3 2477d a8f2b5c7d3d3 Add missing NULL check for the protocol control block (pcb) pointer in mrt{6,}_ioctl. Calling shutdown(2) on the socket prior to the ioctl command can cause it to be NULL.
uvm_fault: mrt_ioctl -1 C 2 2478d 2478d 3/3 2477d a8f2b5c7d3d3 Add missing NULL check for the protocol control block (pcb) pointer in mrt{6,}_ioctl. Calling shutdown(2) on the socket prior to the ioctl command can cause it to be NULL.
uvm_fault: pckbc_start (2) -1 syz 2 2490d 2490d 3/3 2478d bc79b6e32eb3 Prevent corruption of the pckbc command queue. If multiple synchronous commands are in flight and all corresponding threads are sleeping waiting for a response, the first command to timeout will clear the command queue. The remaining threads once awake will then try to remove a dequeued command from the queue, leading to corruption. Instead, remove commands from the queue before waking up the sleeping thread. A quirk is still needed to handle the case where tsleep() returns successfully during suspend.
assert "_kernel_lock_held()" failed in kern_event.c -1 C 11 2508d 2511d 3/3 2478d 1be240a95e4a Removed all diagnostic, calling printf() here might create a recursion.
witness: reversal: &pr->ps_mtx &sched_lock -1 C 141 2480d 2481d 3/3 2480d 17b25159f963 Revert to using the SCHED_LOCK() to protect time accounting.
uvm_fault: wsmux_do_ioctl (4) -1 C 3 2493d 2517d 3/3 2490d 78fe050fe549 A problem fixed in wskbd is also present in wsmux. Repeating the previous commit message:
panic: malloc: out of space in kmem_map 2 97 2543d 2608d 3/3 2491d During fuzzing, one or many fuzzing processes are often stuck waiting on memory from the subproc malloc subsystem which is exhausted. Attempt to circumvent such scenarios by allocation the kcov coverage buffer using km_alloc() instead.
uvm_fault: rtm_report -1 C 2 2516d 2516d 3/3 2500d 1b18c0494f67 Make rt_mpls_set() be more strict in what it accepts. Also ensure that the RTF_MPLS can't be toggled without rt_mpls_set() being called. While RTF_MPLS is part of RTF_FMASK it should be excluded from the flags and mask when they are applied to the route since toggling it requires a call to rt_mpls_set(). OK bluhm@
syzkaller: testing failed: failed to run ["go" "test" "-short" "./..."]: exit status 1 -1 36 2501d 2503d 3/3 2501d fbb8d265a895 Restore previous behavior of limiting deadlock detection to posix-style locks.
multicore build error (3) -1 4 2505d 2505d 3/3 2505d always rearm completion queues as leaving them disarmed would lead to rx or tx completion stalling.
openbsd build error (5) -1 2 2505d 2505d 3/3 2505d always rearm completion queues as leaving them disarmed would lead to rx or tx completion stalling.
panic: timeout_add: to_ticks < 0 (3) 2 1 2528d 2528d 3/3 2517d 1af424413523 Lower the accepted upper bound for bd_rtout to INT_MAX in order to prevent passing negative values to timeout_add().
panic: mtx ADDR: locking against myself 2 C 3 2522d 2522d 3/3 2520d vmm(4): remove a debug printf that was causing lock issues (it was being called from an IPI routine).
panic: timeout_add: to_ticks < 0 (2) 2 syz 12 2539d 2540d 3/3 2538d bf2018479c9a Reject negative and too large timeouts passed to BIOCSRTIMEOUT. Since the timeout converted to ticks is later passed timeout_add(), it could cause a panic if the timeout is negative.
uvm_fault: rtable_satoplen (2) -1 syz 101 2542d 2551d 3/3 2538d 575ef11475ca Add a more strict rtm_hdrlen size check. Make sure that at least struct rt_msghdr bytes are passed in. Also return a failure from rtm_xaddrs() if rti_addrs has bad flags or run out of space. Ok bluhm@
uvm_fault: memcpy -1 C 460 2542d 2562d 3/3 2542d 575ef11475ca Add a more strict rtm_hdrlen size check. Make sure that at least struct rt_msghdr bytes are passed in. Also return a failure from rtm_xaddrs() if rti_addrs has bad flags or run out of space. Ok bluhm@
panic: malloc: allocation too large, type = 2, size = ADDR 2 C 914 2543d 2606d 3/3 2543d fd7c80607c62 Restrict the number of allowed wsmux devices, just like wskbd and wsmouse already does. Otherwise, malloc could panic if the device minor is sufficiently large.
openbsd build error (2) -1 1 2547d 2547d 3/3 2546d 6baecefef8fe Tweak previous: include <sys/stdint.h> for INT64_MAX/INT64_MIN.
assert "tname->un_flags & UNVEIL_USERSET" failed in kern_unveil.c -1 C 1447 2546d 2549d 3/3 2546d f4c23aa848ae Remove this assert, I can't do this here with UNVEIL_INSPECT added aggressively today. Hopefully post release a glorious flensing will remove UNVEIL_INSPECT anyway
uvm_fault: rtable_satoplen -1 158 2551d 2568d 3/3 2551d fab4809e7ec2 Make sure pointer is within bounds before dereferencing it.
witness: acquiring duplicate lock of same type: "&sc->sc_lock" -1 C 231 2581d 2582d 3/3 2581d 1f8a38b155bf When adding a wsmux device to an existing wsmux device using ioctl(WSMUXIO_ADD_DEVICE), two distinct locks of the same type are acquired. Thus, witness will emit warning. Since acquiring two different locks of the same type is harmless in this context, relax the witness check by flagging the locks as RWL_DUPOK.
panic: timeout_add: to_ticks < 0 2 C 66 2581d 2586d 3/3 2581d 3cfc9cae129c Reject negative input from userland in spkrioctl(). One of the arguments are later passed to timeout_add() which panics if the given ticks are negative. While here, clamp arguments in pcppi_bell() in order to prevent overflow.
uvm_fault: pckbc_start -1 1 2583d 2583d 3/3 2582d 0c0bf6318018 Serialize access to the list of attached child devices belonging to a wsmux. When invoking wsevsrc_* functions on a attached child device, underlying driver can sleep; this introduces a race where another thread is able to modify the list leading to all kinds of corruptions.
uvm_fault: wsmux_detach_sc -1 syz 10 2585d 2596d 3/3 2582d 0c0bf6318018 Serialize access to the list of attached child devices belonging to a wsmux. When invoking wsevsrc_* functions on a attached child device, underlying driver can sleep; this introduces a race where another thread is able to modify the list leading to all kinds of corruptions.
panic: vmmaplk: lock not shared 2 C 16 2593d 2605d 3/3 2593d Always refault if relocking maps fails after IO. This fixes a regression
panic: pool_p_free: semupl free list modified: page ADDR; item addr ADDR; offset 0x0=ADDR 2 C 5 2594d 2713d 3/3 2593d When freeing the sem_undo structure in semundo_adjust(), update the
pool: free list modified: semupl -1 C 6 2600d 2713d 3/3 2593d When freeing the sem_undo structure in semundo_adjust(), update the
pool: double put: mbufpl -1 6 2659d 2661d 3/3 2593d Avoid an mbuf double free in the oob soreceive() path. In the
uvm_fault: wsmux_do_ioctl (2) -1 C 17 2601d 2605d 3/3 2600d In wskbdclose(), use the same logic as in wskbdopen() to determine if
kernel: protection fault trap, code=0 (3) -1 C 3 2605d 2605d 3/3 2605d Validate the user-supplied device index given to WSMUXIO_ADD_DEVICE. The same
uvm_fault: VOP_ACCESS -1 393 2609d 2652d 3/3 2609d namei can return a null dvp on success. check this before access.
kernel: protection fault trap, code=0 (2) -1 syz 109 2610d 2634d 3/3 2610d Introduce a dedicated entry point data structure for file locks. This new data
panic: malformed IPv4 option passed to ip_optcopy (2) 2 C 149 2616d 2619d 3/3 2613d Bring back the ip_pcbopts() refactor. Pad the option buffer and therefor the mbuf to the next word length as it is required by the standard. Also use the correct offset from the input mbuf. OK visa@, input & OK bluhm@
panic: uvm_fault_unwire_locked: address not in map 2 C 2 2638d 2638d 3/3 2622d Hold a read lock on the map while doing the actual device I/O during in
assert "next != NULL && next->start <= entry->end" failed in uvm_fault.c -1 C 2 2636d 2636d 3/3 2622d Hold a read lock on the map while doing the actual device I/O during in
panic: malformed IPv4 option passed to ip_optcopy 2 C 10 2636d 2645d 3/3 2625d Validate the version, and all length fields of IP packets passed to a raw socket
panic: m_zero: M_READONLY 2 C 3 2630d 2630d 3/3 2625d It is possible to call m_zero with a read-only cluster. In that case just
assert "__mp_lock_held(&sched_lock, curcpu()) == 0" failed in kern_lock.c -1 C 17 2627d 2638d 3/3 2626d ec412da11be4 Fix unsafe use of ptsignal() in mi_switch().
uvm_fault: m_free -1 12 2636d 2659d 3/3 2626d 54e30ac1a804 Fix mbuf releated crashes in switch(4). They have been found by syzkaller as pool corruption panic. It is unclear which bug caused what, but it should be better now. - Check M_PKTHDR with assertion before accessing m_pkthdr. - Do not access oh_length without m_pullup(). - After checking if there is space at the end of the mbuf, don't overwrite the data at the beginning. Append the new content. - Do not set m_len and m_pkthdr.len when it is unclear whether the ofp_error header fits at all. Use m_makespace() to adjust the mbuf. Reported-by: syzbot+6efc0a9d5b700b54392e@syzkaller.appspotmail.com test akoshibe@; OK claudio@
pool: free list modified: mbufpl -1 syz 13 2645d 2659d 3/3 2626d 54e30ac1a804 Fix mbuf releated crashes in switch(4). They have been found by syzkaller as pool corruption panic. It is unclear which bug caused what, but it should be better now. - Check M_PKTHDR with assertion before accessing m_pkthdr. - Do not access oh_length without m_pullup(). - After checking if there is space at the end of the mbuf, don't overwrite the data at the beginning. Append the new content. - Do not set m_len and m_pkthdr.len when it is unclear whether the ofp_error header fits at all. Use m_makespace() to adjust the mbuf. Reported-by: syzbot+6efc0a9d5b700b54392e@syzkaller.appspotmail.com test akoshibe@; OK claudio@
panic: pr_find_pagehead: mbufpl: incorrect page 2 3 2640d 2656d 3/3 2626d 54e30ac1a804 Fix mbuf releated crashes in switch(4). They have been found by syzkaller as pool corruption panic. It is unclear which bug caused what, but it should be better now. - Check M_PKTHDR with assertion before accessing m_pkthdr. - Do not access oh_length without m_pullup(). - After checking if there is space at the end of the mbuf, don't overwrite the data at the beginning. Append the new content. - Do not set m_len and m_pkthdr.len when it is unclear whether the ofp_error header fits at all. Use m_makespace() to adjust the mbuf. Reported-by: syzbot+6efc0a9d5b700b54392e@syzkaller.appspotmail.com test akoshibe@; OK claudio@
panic: pool_cache_item_magic_check: mcl64k cpu free list modified: item addr ADDR+24 ADDR!=ADDR 2 1 2657d 2657d 3/3 2626d 54e30ac1a804 Fix mbuf releated crashes in switch(4). They have been found by syzkaller as pool corruption panic. It is unclear which bug caused what, but it should be better now. - Check M_PKTHDR with assertion before accessing m_pkthdr. - Do not access oh_length without m_pullup(). - After checking if there is space at the end of the mbuf, don't overwrite the data at the beginning. Append the new content. - Do not set m_len and m_pkthdr.len when it is unclear whether the ofp_error header fits at all. Use m_makespace() to adjust the mbuf. Reported-by: syzbot+6efc0a9d5b700b54392e@syzkaller.appspotmail.com test akoshibe@; OK claudio@
uvm_fault: switchread -1 1 2654d 2654d 3/3 2626d 54e30ac1a804 Fix mbuf releated crashes in switch(4). They have been found by syzkaller as pool corruption panic. It is unclear which bug caused what, but it should be better now. - Check M_PKTHDR with assertion before accessing m_pkthdr. - Do not access oh_length without m_pullup(). - After checking if there is space at the end of the mbuf, don't overwrite the data at the beginning. Append the new content. - Do not set m_len and m_pkthdr.len when it is unclear whether the ofp_error header fits at all. Use m_makespace() to adjust the mbuf. Reported-by: syzbot+6efc0a9d5b700b54392e@syzkaller.appspotmail.com test akoshibe@; OK claudio@
uvm_fault: switchwrite -1 syz 7 2636d 2656d 3/3 2626d 54e30ac1a804 Fix mbuf releated crashes in switch(4). They have been found by syzkaller as pool corruption panic. It is unclear which bug caused what, but it should be better now. - Check M_PKTHDR with assertion before accessing m_pkthdr. - Do not access oh_length without m_pullup(). - After checking if there is space at the end of the mbuf, don't overwrite the data at the beginning. Append the new content. - Do not set m_len and m_pkthdr.len when it is unclear whether the ofp_error header fits at all. Use m_makespace() to adjust the mbuf. Reported-by: syzbot+6efc0a9d5b700b54392e@syzkaller.appspotmail.com test akoshibe@; OK claudio@
panic: pool_cache_item_magic_check: mbufpl cpu free list modified: item addr ADDR+16 0x0!=ADDR 2 1 2647d 2647d 3/3 2629d 54e30ac1a804 Fix mbuf releated crashes in switch(4). They have been found by syzkaller as pool corruption panic. It is unclear which bug caused what, but it should be better now. - Check M_PKTHDR with assertion before accessing m_pkthdr. - Do not access oh_length without m_pullup(). - After checking if there is space at the end of the mbuf, don't overwrite the data at the beginning. Append the new content. - Do not set m_len and m_pkthdr.len when it is unclear whether the ofp_error header fits at all. Use m_makespace() to adjust the mbuf. Reported-by: syzbot+6efc0a9d5b700b54392e@syzkaller.appspotmail.com test akoshibe@; OK claudio@
panic: pr_find_pagehead: mbufpl: page header missing 2 C 10 2646d 2659d 3/3 2635d Fix mbuf releated crashes in switch(4). They have been found by
pool: free list modified: mcl2k -1 C 4 2644d 2698d 3/3 2643d Replace a wrong poor mans m_trailingspace() with the real thing. The mbuf
panic: receive 3: so ADDR, so_type 1, m ADDR, m_type 6 2 C 18 2645d 2654d 3/3 2644d When using MSG_WAITALL, soreceive() can sleep while processing the receive buffer of a stream socket. Then a new pair of control and data mbuf can be appended to the mbuf queue. In this case, terminate the loop with a short read to prevent a panic. Userland should read the control message with the next system call. OK claudio@ deraadt@
panic: kernel diagnostic assertion "M_DATABUF(m) + M_SIZE(m) >= (m->m_data + m->m_len)" failed: file "/syzkaller/manager 2 C 7 2654d 2660d 3/3 2644d 49729d6ed45f In PRU_DISCONNECT don't fall through into PRU_ABORT since the latter frees the inpcb apart from the disconnect. Just call soisdisconnected() and clear the inp->inp_faddr since the socket is still valid after a disconnect. Problem found by syzkaller via Greg Steuck OK visa@ Fixes: Reported-by: syzbot+2cd350dfe5c96f6469f2@syzkaller.appspotmail.com Reported-by: syzbot+139ac2d7d3d60162334b@syzkaller.appspotmail.com Reported-by: syzbot+02168317bd0156c13b69@syzkaller.appspotmail.com Reported-by: syzbot+de8d2459ecf4cdc576a1@syzkaller.appspotmail.com
pool: double put: lockfpl -1 1 2721d 2713d 3/3 2644d Rework previous lockf fix; bluhm@ noticed a regress failure during consecutive runs. This is a second attempt in which the lockf structure is turned into a doubly linked list which makes it easier to ensure correctness during list insertion and deletion.
uvm_fault: killjobc -1 1 2649d 2649d 3/3 2645d When no child devices are attached to a wsmux device, make sure to return an
uvm_fault: wsmux_do_ioctl -1 2 2675d 2713d 3/3 2645d Utilize sigio with wscons. The old behavior of always making the process group of the process who opens the device the default recipient of sigio is removed as a side-effect of this change. Issuing ioctl(FIOSETOWN) is therefore mandatory in order to receive sigio, which is more consistent with other subsystems supporting sigio.
uvm_fault: sogetopt -1 C 2 2662d 2662d 1/3 2659d In PRU_DISCONNECT don't fall through into PRU_ABORT since the latter frees
kernel: protection fault trap, code=0 -1 C 16 2659d 2664d 1/3 2659d In PRU_DISCONNECT don't fall through into PRU_ABORT since the latter frees
uvm_fault: ip_ctloutput -1 C 11 2659d 2663d 1/3 2659d In PRU_DISCONNECT don't fall through into PRU_ABORT since the latter frees
uvm_fault: ip_pcbopts -1 C 6 2659d 2663d 1/3 2659d In PRU_DISCONNECT don't fall through into PRU_ABORT since the latter frees