svn commit: r368743 - head/tests/sys/net

Author: kp Date: Thu Dec 17 21:54:25 2020 New Revision: 368743 URL: https://svnweb.freebsd.org/changeset/base/368743 Log: net tests: Re-enable most if_clone tests All but one of these (tap_ipv6_up_stress) currently pass, so we should enable them so we don't regress. Modified: head/test

svn commit: r368666 - stable/12/tests/sys/netpfil/pf

Author: kp Date: Tue Dec 15 16:15:25 2020 New Revision: 368666 URL: https://svnweb.freebsd.org/changeset/base/368666 Log: MFC r368239: pf tests: Re-enable panicing tests We've fixed the vnet/epair cleanup race, so it is now safe to re-enable these tests. Sponsored by: Modirum MD

svn commit: r368665 - stable/12/sys/net

Author: kp Date: Tue Dec 15 16:13:32 2020 New Revision: 368665 URL: https://svnweb.freebsd.org/changeset/base/368665 Log: MFC r368238: net: Revert vnet/epair cleanup race mitigation Revert the mitigation code for the vnet/epair cleanup race (done in r365457). r368237 introduced a mor

svn commit: r368663 - stable/12/sys/net

Author: kp Date: Tue Dec 15 15:33:28 2020 New Revision: 368663 URL: https://svnweb.freebsd.org/changeset/base/368663 Log: MFC r368237: if: Fix panic when destroying vnet and epair simultaneously When destroying a vnet and an epair (with one end in the vnet) we often panicked. This wa

svn commit: r368656 - stable/12/sys/netpfil/pf

Author: kp Date: Tue Dec 15 08:29:45 2020 New Revision: 368656 URL: https://svnweb.freebsd.org/changeset/base/368656 Log: MFC r368588: pf: Allow net.pf.request_maxcount to be set from loader.conf Mark request_maxcount as RWTUN so we can set it both at runtime and from loader.conf. Th

svn commit: r368588 - head/sys/netpfil/pf

Author: kp Date: Sat Dec 12 20:14:39 2020 New Revision: 368588 URL: https://svnweb.freebsd.org/changeset/base/368588 Log: pf: Allow net.pf.request_maxcount to be set from loader.conf Mark request_maxcount as RWTUN so we can set it both at runtime and from loader.conf. This avoids usings g

svn commit: r368554 - stable/12/sys/net

Author: kp Date: Fri Dec 11 15:39:22 2020 New Revision: 368554 URL: https://svnweb.freebsd.org/changeset/base/368554 Log: MFC r368020, r368025: if: Protect V_ifnet in vnet_if_return() When we terminate a vnet (i.e. jail) we move interfaces back to their home vnet. We need to protect

svn commit: r368552 - stable/12/sys/net

Author: kp Date: Fri Dec 11 14:11:41 2020 New Revision: 368552 URL: https://svnweb.freebsd.org/changeset/base/368552 Log: MFC r368015: if: Remove ifnet_rwlock It no longer serves any purpose, as evidenced by the fact that we never take it without ifnet_sxlock. This differs slig

svn commit: r368488 - stable/12/tests/sys/netpfil/pf

Author: kp Date: Wed Dec 9 17:17:45 2020 New Revision: 368488 URL: https://svnweb.freebsd.org/changeset/base/368488 Log: MFC r368277: pf tests: Test case for bug #251414 Changing a table from not having counters to having counters (or vice versa) may trigger panics. PR:

svn commit: r368277 - head/tests/sys/netpfil/pf

Author: kp Date: Wed Dec 2 16:33:23 2020 New Revision: 368277 URL: https://svnweb.freebsd.org/changeset/base/368277 Log: pf tests: Test case for bug #251414 Changing a table from not having counters to having counters (or vice versa) may trigger panics. PR: 251414 MFC af

svn commit: r368239 - head/tests/sys/netpfil/pf

Author: kp Date: Tue Dec 1 16:44:36 2020 New Revision: 368239 URL: https://svnweb.freebsd.org/changeset/base/368239 Log: pf tests: Re-enable panicing tests We've fixed the vnet/epair cleanup race, so it is now safe to re-enable these tests. MFC after:2 weeks Sponsored by: Modi

svn commit: r368238 - head/sys/net

Author: kp Date: Tue Dec 1 16:34:43 2020 New Revision: 368238 URL: https://svnweb.freebsd.org/changeset/base/368238 Log: net: Revert vnet/epair cleanup race mitigation Revert the mitigation code for the vnet/epair cleanup race (done in r365457). r368237 introduced a more reliable fix.

svn commit: r368237 - head/sys/net

Author: kp Date: Tue Dec 1 16:23:59 2020 New Revision: 368237 URL: https://svnweb.freebsd.org/changeset/base/368237 Log: if: Fix panic when destroying vnet and epair simultaneously When destroying a vnet and an epair (with one end in the vnet) we often panicked. This was the result of th

svn commit: r368235 - stable/12/tests/sys/netpfil/pf/ioctl

Author: kp Date: Tue Dec 1 16:02:52 2020 New Revision: 368235 URL: https://svnweb.freebsd.org/changeset/base/368235 Log: MFC r367990: pf tests: provoke tag hashing panic Attempt to provoke the panic fixed in r367987. Modified: stable/12/tests/sys/netpfil/pf/ioctl/validation.c Direc

svn commit: r368101 - stable/12/tests/sys/netpfil/pf

newly added) +++ stable/12/tests/sys/netpfil/pf/src_track.sh Fri Nov 27 11:12:59 2020 (r368101, copy of r367869, head/tests/sys/netpfil/pf/src_track.sh) @@ -0,0 +1,66 @@ +# $FreeBSD$ +# +# SPDX-License-Identifier: BSD-2-Clause-FreeBSD +# +# Copyright (c) 2020 Kristof Provost

svn commit: r368100 - stable/12/sys/netpfil/pf

Author: kp Date: Fri Nov 27 11:12:10 2020 New Revision: 368100 URL: https://svnweb.freebsd.org/changeset/base/368100 Log: MFC r367867: pf: Fix incorrect assertion We never set PFRULE_RULESRCTRACK when calling pf_insert_src_node(). We do set PFRULE_SRCTRACK, so update the assertion to

svn commit: r368054 - stable/12/sys/net

Author: kp Date: Thu Nov 26 10:17:56 2020 New Revision: 368054 URL: https://svnweb.freebsd.org/changeset/base/368054 Log: Add missing NET_EPOCH_EXIT() to if_bridge In r367706 one return in bridge_input() did not NET_EPOCH_EXIT(), which caused seemingly random panics on the next use of NET

svn commit: r368025 - head/sys/net

Author: kp Date: Wed Nov 25 17:15:24 2020 New Revision: 368025 URL: https://svnweb.freebsd.org/changeset/base/368025 Log: if: Fix non-VIMAGE build if_link_ifnet() and if_unlink_ifnet() are needed even when VIMAGE is not enabled. MFC after:2 weeks Sponsored by: Modirum MDPay Mo

svn commit: r368020 - head/sys/net

Author: kp Date: Wed Nov 25 15:07:22 2020 New Revision: 368020 URL: https://svnweb.freebsd.org/changeset/base/368020 Log: if: Protect V_ifnet in vnet_if_return() When we terminate a vnet (i.e. jail) we move interfaces back to their home vnet. We need to protect our access to the V_ifnet C

svn commit: r368015 - head/sys/net

Author: kp Date: Wed Nov 25 10:56:38 2020 New Revision: 368015 URL: https://svnweb.freebsd.org/changeset/base/368015 Log: if: Remove ifnet_rwlock It no longer serves any purpose, as evidenced by the fact that we never take it without ifnet_sxlock. Sponsored by: Modirum MDPay Diffe

svn commit: r367990 - head/tests/sys/netpfil/pf/ioctl

Author: kp Date: Tue Nov 24 17:32:49 2020 New Revision: 367990 URL: https://svnweb.freebsd.org/changeset/base/367990 Log: pf tests: provoke tag hashing panic Attempt to provoke the panic fixed in r367987. MFC after:1 week Differential Revision:https://reviews.freebsd.org/

svn commit: r367869 - head/tests/sys/netpfil/pf

@@ +# $FreeBSD$ +# +# SPDX-License-Identifier: BSD-2-Clause-FreeBSD +# +# Copyright (c) 2020 Kristof Provost +# +# Redistribution and use in source and binary forms, with or without +# modification, are permitted provided that the following conditions +# are met: +# 1. Redistributions of source code

svn commit: r367867 - head/sys/netpfil/pf

Author: kp Date: Fri Nov 20 10:08:33 2020 New Revision: 367867 URL: https://svnweb.freebsd.org/changeset/base/367867 Log: pf: Fix incorrect assertion We never set PFRULE_RULESRCTRACK when calling pf_insert_src_node(). We do set PFRULE_SRCTRACK, so update the assertion to match. MFC a

svn commit: r367706 - stable/12/sys/net

Author: kp Date: Sun Nov 15 11:56:16 2020 New Revision: 367706 URL: https://svnweb.freebsd.org/changeset/base/367706 Log: MFC r366500: bridge: call member interface ioctl() without NET_EPOCH We're not allowed to hold NET_EPOCH while sleeping, so when we call ioctl() handlers for memb

svn commit: r367705 - stable/12/sys/net

Author: kp Date: Sun Nov 15 11:46:44 2020 New Revision: 367705 URL: https://svnweb.freebsd.org/changeset/base/367705 Log: bridge: epoch-ification Undo the revert (r363568). This commit still violates epoch rules by sleeping within NET_EPOCH. That will be resolved in the following commit.

Re: svn commit: r367321 - head/sys/amd64/linux

On 4 Nov 2020, at 11:45, Gordon Bergling wrote: Shouldn't such a commit be approved by someone? The committer’s guide seems clear that doc committers may fix comments: https://www.freebsd.org/doc/en_US.ISO8859-1/articles/committers-guide/committer.types.html doc committers may commit documen

svn commit: r367078 - head/sys/riscv/riscv

Author: kp Date: Tue Oct 27 12:44:49 2020 New Revision: 367078 URL: https://svnweb.freebsd.org/changeset/base/367078 Log: riscv: Minor cleanup in startup code - remove setting of register value which is not used until the next value is set - Use the L2_SHIFT constant when setting u

svn commit: r367058 - stable/12/tests/sys/netpfil/pf

Author: kp Date: Mon Oct 26 13:24:20 2020 New Revision: 367058 URL: https://svnweb.freebsd.org/changeset/base/367058 Log: MFC r366648: pf tests: Test that 'set skip on ' works on new group members There's a know issue where new group members don't get the 'set skip on' applied until

svn commit: r367057 - stable/12/sys/netpfil/pf

Author: kp Date: Mon Oct 26 13:23:40 2020 New Revision: 367057 URL: https://svnweb.freebsd.org/changeset/base/367057 Log: MFC r37: pf: do not remove kifs that are referenced by rules Even if a kif doesn't have an ifp or if_group pointer we still can't delete it if it's referenced

svn commit: r367056 - stable/12/sys/netpfil/pf

Author: kp Date: Mon Oct 26 13:22:55 2020 New Revision: 367056 URL: https://svnweb.freebsd.org/changeset/base/367056 Log: MFC r366647: pf: create a kif for flags If userspace tries to set flags (e.g. 'set skip on ') and doesn't exist we should create a kif so that we apply the flags

svn commit: r366667 - head/sys/netpfil/pf

Author: kp Date: Tue Oct 13 11:04:00 2020 New Revision: 37 URL: https://svnweb.freebsd.org/changeset/base/37 Log: pf: do not remove kifs that are referenced by rules Even if a kif doesn't have an ifp or if_group pointer we still can't delete it if it's referenced by a rule. In oth

svn commit: r366648 - head/tests/sys/netpfil/pf

Author: kp Date: Mon Oct 12 12:41:10 2020 New Revision: 366648 URL: https://svnweb.freebsd.org/changeset/base/366648 Log: pf tests: Test that 'set skip on ' works on new group members There's a know issue where new group members don't get the 'set skip on' applied until the rules are re-l

svn commit: r366647 - head/sys/netpfil/pf

Author: kp Date: Mon Oct 12 12:39:37 2020 New Revision: 366647 URL: https://svnweb.freebsd.org/changeset/base/366647 Log: pf: create a kif for flags If userspace tries to set flags (e.g. 'set skip on ') and doesn't exist we should create a kif so that we apply the flags when the does

svn commit: r366500 - head/sys/net

Author: kp Date: Tue Oct 6 19:19:56 2020 New Revision: 366500 URL: https://svnweb.freebsd.org/changeset/base/366500 Log: bridge: call member interface ioctl() without NET_EPOCH We're not allowed to hold NET_EPOCH while sleeping, so when we call ioctl() handlers for member interfaces we c

svn commit: r366461 - head/sbin/devfs

Author: kp Date: Mon Oct 5 19:26:54 2020 New Revision: 366461 URL: https://svnweb.freebsd.org/changeset/base/366461 Log: devfs.rules: unhide pf in vnet jails /dev/pf is usable in vnet jails, so don't hide the node there. We shouldn't expose /dev/pf in regular jails, as that gives them

svn commit: r366355 - head/sys/riscv/riscv

Author: kp Date: Fri Oct 2 07:30:11 2020 New Revision: 366355 URL: https://svnweb.freebsd.org/changeset/base/366355 Log: riscv: handle access faults in user mode Access faults in user mode are treated like TLB misses, which leads to an endless loop of faults. It's less serious than the s

svn commit: r366315 - in head/sys/riscv: include riscv

Author: kp Date: Thu Oct 1 15:04:55 2020 New Revision: 366315 URL: https://svnweb.freebsd.org/changeset/base/366315 Log: riscv: Add memmmap so we can mmap /dev/mem Reviewed by: mhorne Sponsored by: Axiado Differential Revision:https://reviews.freebsd.org/D26622 Modified: he

svn commit: r366284 - head/sys/riscv/riscv

Author: kp Date: Wed Sep 30 08:23:43 2020 New Revision: 366284 URL: https://svnweb.freebsd.org/changeset/base/366284 Log: riscv: Panic on PMP errors Load/store/fetch access exceptions always indicate a violation of a PMP rule. We can't treat those as page faults, because updating the page

svn commit: r365669 - releng/12.2/sys/net

Author: kp Date: Sat Sep 12 18:58:36 2020 New Revision: 365669 URL: https://svnweb.freebsd.org/changeset/base/365669 Log: MFC r365457: net: mitigate vnet / epair cleanup races There's a race where dying vnets move their interfaces back to their original vnet, and if_epair cleanup (wh

svn commit: r365659 - stable/12/sys/net

Author: kp Date: Sat Sep 12 12:45:31 2020 New Revision: 365659 URL: https://svnweb.freebsd.org/changeset/base/365659 Log: MFC r365457: net: mitigate vnet / epair cleanup races There's a race where dying vnets move their interfaces back to their original vnet, and if_epair cleanup (wh

svn commit: r365626 - in head/sys/cddl/dev: dtrace/riscv fbt/riscv

Author: kp Date: Fri Sep 11 09:15:49 2020 New Revision: 365626 URL: https://svnweb.freebsd.org/changeset/base/365626 Log: dtrace: fix fbt return probes on RISC-V Return values are passed in a0, so read it from there. We also pass a1 through to userspace, as the ABI allows small structs to

svn commit: r365457 - head/sys/net

Author: kp Date: Tue Sep 8 14:54:10 2020 New Revision: 365457 URL: https://svnweb.freebsd.org/changeset/base/365457 Log: net: mitigate vnet / epair cleanup races There's a race where dying vnets move their interfaces back to their original vnet, and if_epair cleanup (where deleting one i

svn commit: r365246 - head/sys/netgraph

Author: kp Date: Wed Sep 2 11:49:22 2020 New Revision: 365246 URL: https://svnweb.freebsd.org/changeset/base/365246 Log: ng_ether: Enter NET_EPOCH where required We must enter NET_EPOCH before calling ether_output_frame(). Several of the functions it calls (pfil_run_hooks, if_transmit) e

svn commit: r365245 - head/sys/riscv/riscv

Author: kp Date: Wed Sep 2 11:33:31 2020 New Revision: 365245 URL: https://svnweb.freebsd.org/changeset/base/365245 Log: riscv: very large dma mappings can cause integer overflow Fix the return type for _bus_dmamap_addseg(). Based on the same fix done for arm64 in r348571. Sponsored

svn commit: r364456 - stable/12/sys/netpfil/pf

Author: kp Date: Fri Aug 21 13:11:33 2020 New Revision: 364456 URL: https://svnweb.freebsd.org/changeset/base/364456 Log: MFC r355744: pf: Make request_maxcount runtime adjustable There's no reason for this to be a tunable. It's perfectly safe to change this at runtime. Modified:

svn commit: r363796 - in head: lib/libc/riscv/gen sys/riscv/include

= --- /dev/null 00:00:00 1970 (empty, because file is newly added) +++ head/lib/libc/riscv/gen/fpgetmask.c Mon Aug 3 12:48:51 2020 (r363796) @@ -0,0 +1,41 @@ +/*- + * Copyright (c) 2020 Axiado + * All rights reserved. + * + * This software was developed by Kristof Provost under + * s

Re: svn commit: r363568 - stable/12/sys/net

On 26 Jul 2020, at 21:21, mike tancsa wrote: Hi Kristof,     First off, thank you for all your efforts in pf and if_bridge.  I have trying to track down a problem with a golang app (sysutils/zrepl) that started acting up around the time the if_bridge stuff was commited (june 26th).  The probl

svn commit: r363568 - stable/12/sys/net

Author: kp Date: Sun Jul 26 17:44:03 2020 New Revision: 363568 URL: https://svnweb.freebsd.org/changeset/base/363568 Log: Revert bridge epochification Revert r363492, r363491, r363430, r363429 and r362650. The introduction of epoch in the network stack is incomplete in stable/12, and

svn commit: r363492 - stable/12/sys/net

Author: kp Date: Fri Jul 24 20:10:27 2020 New Revision: 363492 URL: https://svnweb.freebsd.org/changeset/base/363492 Log: bridge: Enter epoch for bridge_transmit() Just like the change done for bridge_input()/bridge_output() in r363430 we must enter epoch ourselves, because its coverage i

svn commit: r363491 - stable/12/sys/net

Author: kp Date: Fri Jul 24 20:09:52 2020 New Revision: 363491 URL: https://svnweb.freebsd.org/changeset/base/363491 Log: bridge: Fix mismerges from r360345 In r362650 we merged r360345. This required manual changes due to the differences in EPOCH macros between head and stable/12, and wa

svn commit: r363430 - stable/12/sys/net

Author: kp Date: Wed Jul 22 20:13:12 2020 New Revision: 363430 URL: https://svnweb.freebsd.org/changeset/base/363430 Log: bridge: Enter epoch for bridge_input()/bridge_output() In stable/12 epoch is not as wide as it is in head. The network stack isn't yet in epoch when bridge_input()/br

svn commit: r363429 - stable/12/sys/net

Author: kp Date: Wed Jul 22 19:43:55 2020 New Revision: 363429 URL: https://svnweb.freebsd.org/changeset/base/363429 Log: MFC r363308: bridge: Don't sleep during epoch While it doesn't trigger INVARIANTS or WITNESS on head it does in stable/12. There's also no reason for it, as we ca

Re: svn commit: r363308 - head/sys/net

On 18 Jul 2020, at 20:47, Konstantin Belousov wrote: On Sat, Jul 18, 2020 at 12:43:11PM +, Kristof Provost wrote: Author: kp Date: Sat Jul 18 12:43:11 2020 New Revision: 363308 URL: https://svnweb.freebsd.org/changeset/base/363308 Log: bridge: Don't sleep during epoch Whi

svn commit: r363308 - head/sys/net

Author: kp Date: Sat Jul 18 12:43:11 2020 New Revision: 363308 URL: https://svnweb.freebsd.org/changeset/base/363308 Log: bridge: Don't sleep during epoch While it doesn't trigger INVARIANTS or WITNESS on head it does in stable/12. There's also no reason for it, as we can easily report th

svn commit: r362977 - head/sys/riscv/riscv

Author: kp Date: Mon Jul 6 21:29:50 2020 New Revision: 362977 URL: https://svnweb.freebsd.org/changeset/base/362977 Log: riscv plic: Do not complete interrupts until the interrupt handler has run We cannot complete the interrupt (i.e. write to the claims/complete register until the inter

svn commit: r362853 - in head/sys/riscv: include riscv

Author: kp Date: Wed Jul 1 19:15:43 2020 New Revision: 362853 URL: https://svnweb.freebsd.org/changeset/base/362853 Log: riscv pmap: zero reserved pte bits in ppn The top 10 bits of a pte are reserved by specification[1] and are not part of the PPN. [1] 'Volume II: RISC-V Privileged

svn commit: r362852 - head/sys/riscv/riscv

Author: kp Date: Wed Jul 1 19:12:47 2020 New Revision: 362852 URL: https://svnweb.freebsd.org/changeset/base/362852 Log: riscv locore.S: load constant prior to loop A very minor micro-optimization; t0 is not clobbered between the loop top and bottom and there appear to be no other branch

svn commit: r362851 - head/sys/riscv/riscv

Author: kp Date: Wed Jul 1 19:11:02 2020 New Revision: 362851 URL: https://svnweb.freebsd.org/changeset/base/362851 Log: riscv: Log missing registers in dump_regs() If we panic we dump the registers for debugging. This is very useful, but it missed several registers (ra, sp, gp and tp).

svn commit: r362650 - stable/12/sys/net

Author: kp Date: Fri Jun 26 12:11:22 2020 New Revision: 362650 URL: https://svnweb.freebsd.org/changeset/base/362650 Log: MFC r360345: bridge: epoch-ification Run the bridge datapath under epoch, rather than under the BRIDGE_LOCK(). We still take the BRIDGE_LOCK() whenever we in

svn commit: r362649 - stable/12/sys/net

Author: kp Date: Fri Jun 26 10:08:57 2020 New Revision: 362649 URL: https://svnweb.freebsd.org/changeset/base/362649 Log: MFC r359641: bridge: Change lists to CK_LIST as a peparation for epochification Prepare the ground for a rework of the bridge locking approach. We will use an epo

svn commit: r362648 - stable/12/sys/net

Author: kp Date: Fri Jun 26 09:52:43 2020 New Revision: 362648 URL: https://svnweb.freebsd.org/changeset/base/362648 Log: MFC r358325: bridge: Move locking defines into if_bridge.c The locking defines for if_bridge used to live in if_bridgevar.h, but they're only ever used by the bri

svn commit: r362235 - head/contrib/llvm-project/clang/lib/Driver/ToolChains/Arch

Author: kp Date: Tue Jun 16 18:39:56 2020 New Revision: 362235 URL: https://svnweb.freebsd.org/changeset/base/362235 Log: llvm: Default to -mno-relax on RISC-V Compiling on a RISC-V system fails with 'relocation R_RISCV_ALIGN requires unimplemented linker relaxation; recompile with -mno-r

Re: svn commit: r362217 - head/stand/common

On 16 Jun 2020, at 19:11, Ed Maste wrote: On Tue, 16 Jun 2020 at 13:01, Ian Lepore wrote: As much as I prefer doing it this way, style(9) doesn't allow for variable declarations inside a for() statement (or even inside a local block, which is just too 1980s for me, but it is still our standa

svn commit: r362064 - stable/12/sys/netpfil/pf

Author: kp Date: Thu Jun 11 16:51:13 2020 New Revision: 362064 URL: https://svnweb.freebsd.org/changeset/base/362064 Log: MFC r357061: pf: Apply kif flags to new group members If we have a 'set skip on ' rule this flag it set on the group kif, but must also be set on all members. pfc

svn commit: r361893 - head/tests/sys/net

Author: kp Date: Sun Jun 7 13:53:02 2020 New Revision: 361893 URL: https://svnweb.freebsd.org/changeset/base/361893 Log: bridge tests: Re-enable STP test This test should no longer provoke large amounts of traffic, which can overwhelm single-core systems, preventing them from making prog

svn commit: r361761 - stable/12/sys/dev/bnxt

Author: kp Date: Wed Jun 3 18:09:31 2020 New Revision: 361761 URL: https://svnweb.freebsd.org/changeset/base/361761 Log: MFC r361279: bnxt: isc_nrxd_max and isc_ntxd_max must be powers of two Modified: stable/12/sys/dev/bnxt/bnxt.h stable/12/sys/dev/bnxt/if_bnxt.c Directory Properties

svn commit: r361762 - stable/11/sys/dev/bnxt

Author: kp Date: Wed Jun 3 18:09:32 2020 New Revision: 361762 URL: https://svnweb.freebsd.org/changeset/base/361762 Log: MFC r361279: bnxt: isc_nrxd_max and isc_ntxd_max must be powers of two Modified: stable/11/sys/dev/bnxt/bnxt.h stable/11/sys/dev/bnxt/if_bnxt.c Modified: stable/11

svn commit: r361701 - head/tests/sys/net

Author: kp Date: Mon Jun 1 19:26:16 2020 New Revision: 361701 URL: https://svnweb.freebsd.org/changeset/base/361701 Log: bridge tests: Avoid building a switching loop Enable STP before bringing the bridges up. This avoids a switching loop, which has a tendency to drown out progress in us

svn commit: r361279 - head/sys/dev/bnxt

Author: kp Date: Wed May 20 16:07:37 2020 New Revision: 361279 URL: https://svnweb.freebsd.org/changeset/base/361279 Log: bnxt: isc_nrxd_max and isc_ntxd_max must be powers of two Reviewed by: gallatin, rpokala MFC after:2 weeks Differential Revision:https://reviews.freebsd

Re: svn commit: r347418 - head/sys/net

On 19 May 2020, at 17:02, Andrew Gallatin wrote: On 2020-05-19 04:21, Kristof Provost wrote: The if_bnxt driver initialises |.isc_nrxd_max = {INT32_MAX, INT32_MAX, INT32_MAX},|, so presumably that’s the cause. I don’t know what a sane value would be though. I’ve defaulted to 4096 (because

Re: svn commit: r347418 - head/sys/net

On 10 May 2019, at 2:41, Eric Joyner wrote: Author: erj Date: Fri May 10 00:41:42 2019 New Revision: 347418 URL: https://svnweb.freebsd.org/changeset/base/347418 Log: iflib: use default ntxd and nrxd when user value is not power of 2 From Jake: A user may set a sysctl to override the defa

svn commit: r360935 - head/sys/opencrypto

Author: kp Date: Mon May 11 21:42:19 2020 New Revision: 360935 URL: https://svnweb.freebsd.org/changeset/base/360935 Log: opencrypto: Add missing ioctl exit SDTs The opencrypto ioctl code has very useful probe points at the various exit points. These allow us to figure out exactly why a r

svn commit: r360867 - stable/11/sys/netpfil/pf

Author: kp Date: Sun May 10 09:50:43 2020 New Revision: 360867 URL: https://svnweb.freebsd.org/changeset/base/360867 Log: MFC r360609: pf: Improve DIOCADDRULE validation We expect the addrwrap.p.dyn value to be set to NULL (and assert such), but do not verify it on input. Report

svn commit: r360868 - stable/12/sys/netpfil/pf

Author: kp Date: Sun May 10 09:50:44 2020 New Revision: 360868 URL: https://svnweb.freebsd.org/changeset/base/360868 Log: MFC r360609: pf: Improve DIOCADDRULE validation We expect the addrwrap.p.dyn value to be set to NULL (and assert such), but do not verify it on input. Report

svn commit: r360800 - stable/11/lib/libc/net

Author: kp Date: Thu May 7 21:14:12 2020 New Revision: 360800 URL: https://svnweb.freebsd.org/changeset/base/360800 Log: MFC r360231: libc: Shortcut if_indextoname() if index == 0 If the index we're trying to convert is 0 we can avoid a potentially expensive call to getifaddrs(). No

svn commit: r360799 - stable/12/lib/libc/net

Author: kp Date: Thu May 7 21:14:11 2020 New Revision: 360799 URL: https://svnweb.freebsd.org/changeset/base/360799 Log: MFC r360231: libc: Shortcut if_indextoname() if index == 0 If the index we're trying to convert is 0 we can avoid a potentially expensive call to getifaddrs(). No

svn commit: r360609 - head/sys/netpfil/pf

Author: kp Date: Sun May 3 16:09:35 2020 New Revision: 360609 URL: https://svnweb.freebsd.org/changeset/base/360609 Log: pf: Improve DIOCADDRULE validation We expect the addrwrap.p.dyn value to be set to NULL (and assert such), but do not verify it on input. Reported-by: syzbot+936

svn commit: r360607 - stable/12/sys/netpfil/pf

Author: kp Date: Sun May 3 16:06:17 2020 New Revision: 360607 URL: https://svnweb.freebsd.org/changeset/base/360607 Log: MFC r360344: pf: Improve input validation If we pass an anchor name which doesn't exist pfr_table_count() returns -1, which leads to an overflow in mallocarray()

svn commit: r360608 - stable/11/sys/netpfil/pf

Author: kp Date: Sun May 3 16:06:23 2020 New Revision: 360608 URL: https://svnweb.freebsd.org/changeset/base/360608 Log: MFC r360344: pf: Improve input validation If we pass an anchor name which doesn't exist pfr_table_count() returns -1, which leads to an overflow in mallocarray()

svn commit: r360347 - head/sys/netpfil/pf

Author: kp Date: Sun Apr 26 16:30:00 2020 New Revision: 360347 URL: https://svnweb.freebsd.org/changeset/base/360347 Log: pf: Virtualise pf_frag_mtx The pf_frag_mtx mutex protects the fragments queue. The fragments queue is virtualised already (i.e. per-vnet) so it makes no sense to block

svn commit: r360346 - head/tests/sys/net

Author: kp Date: Sun Apr 26 16:27:03 2020 New Revision: 360346 URL: https://svnweb.freebsd.org/changeset/base/360346 Log: bridge tests: Test for #216510 We used to have an issue with recursive locking with net.link.bridge.inherit_mac. This causes us to send an ARP request while we hold

svn commit: r360345 - head/sys/net

Author: kp Date: Sun Apr 26 16:22:35 2020 New Revision: 360345 URL: https://svnweb.freebsd.org/changeset/base/360345 Log: bridge: epoch-ification Run the bridge datapath under epoch, rather than under the BRIDGE_LOCK(). We still take the BRIDGE_LOCK() whenever we insert or delete ite

svn commit: r360344 - head/sys/netpfil/pf

Author: kp Date: Sun Apr 26 16:16:39 2020 New Revision: 360344 URL: https://svnweb.freebsd.org/changeset/base/360344 Log: pf: Improve input validation If we pass an anchor name which doesn't exist pfr_table_count() returns -1, which leads to an overflow in mallocarray() and thus a panic.

svn commit: r360343 - stable/11/sys/netpfil/pf

Author: kp Date: Sun Apr 26 16:13:51 2020 New Revision: 360343 URL: https://svnweb.freebsd.org/changeset/base/360343 Log: MFC r360098: pf: Improve ioctl() input validation Both DIOCCHANGEADDR and DIOCADDADDR take a struct pf_pooladdr from userspace. They failed to validate the dyn po

svn commit: r360342 - stable/12/sys/netpfil/pf

Author: kp Date: Sun Apr 26 16:13:50 2020 New Revision: 360342 URL: https://svnweb.freebsd.org/changeset/base/360342 Log: MFC r360098: pf: Improve ioctl() input validation Both DIOCCHANGEADDR and DIOCADDADDR take a struct pf_pooladdr from userspace. They failed to validate the dyn po

svn commit: r360341 - stable/12/sbin/pfctl

Author: kp Date: Sun Apr 26 16:06:09 2020 New Revision: 360341 URL: https://svnweb.freebsd.org/changeset/base/360341 Log: MFC r360096: pfctl: Remove unused variable Submitted by: Nick Rogers MFC after:1 week Sponsored by: RG Nets Modified: stable/12/sbin/pfctl/pfctl_parser.c

svn commit: r360299 - in stable/11/sys: kern net sys

Author: kp Date: Sat Apr 25 12:49:48 2020 New Revision: 360299 URL: https://svnweb.freebsd.org/changeset/base/360299 Log: MFC r360068: ethersubr: Make the mac address generation more robust If we create two (vnet) jails and create a bridge interface in each we end up with the same ma

svn commit: r360298 - in stable/12/sys: kern net sys

Author: kp Date: Sat Apr 25 12:49:48 2020 New Revision: 360298 URL: https://svnweb.freebsd.org/changeset/base/360298 Log: MFC r360068: ethersubr: Make the mac address generation more robust If we create two (vnet) jails and create a bridge interface in each we end up with the same ma

svn commit: r360268 - stable/11/sys/netpfil/pf

Author: kp Date: Fri Apr 24 15:27:56 2020 New Revision: 360268 URL: https://svnweb.freebsd.org/changeset/base/360268 Log: MFC r360042: pf: Do not allow negative ps_len in DIOCGETSTATES Userspace may pass a negative ps_len value to us, which causes an assertion failure in malloc().

svn commit: r360267 - stable/12/sys/netpfil/pf

Author: kp Date: Fri Apr 24 15:27:55 2020 New Revision: 360267 URL: https://svnweb.freebsd.org/changeset/base/360267 Log: MFC r360042: pf: Do not allow negative ps_len in DIOCGETSTATES Userspace may pass a negative ps_len value to us, which causes an assertion failure in malloc().

svn commit: r360231 - head/lib/libc/net

Author: kp Date: Thu Apr 23 21:16:51 2020 New Revision: 360231 URL: https://svnweb.freebsd.org/changeset/base/360231 Log: libc: Shortcut if_indextoname() if index == 0 If the index we're trying to convert is 0 we can avoid a potentially expensive call to getifaddrs(). No interface has an

Re: svn commit: r360068 - in head/sys: kern net sys

On 21 Apr 2020, at 4:34, Kyle Evans wrote: On Mon, Apr 20, 2020 at 9:14 PM Kyle Evans wrote: On Mon, Apr 20, 2020 at 8:15 PM Eric van Gyzen wrote: + sz = asprintf(&buf, M_TEMP, "%s-%s-%s", uuid, if_name(ifp), + jailname); + if (sz < 0) { + /* Fall back to a random mac addr

svn commit: r360099 - head/tests/sys/net

Author: kp Date: Sun Apr 19 16:30:49 2020 New Revision: 360099 URL: https://svnweb.freebsd.org/changeset/base/360099 Log: bridge tests: Ensure that bridges in different jails get different MAC addresses We used to have a problem where bridges created in different vnet jails would end up

svn commit: r360098 - head/sys/netpfil/pf

Author: kp Date: Sun Apr 19 16:10:20 2020 New Revision: 360098 URL: https://svnweb.freebsd.org/changeset/base/360098 Log: pf: Improve ioctl() input validation Both DIOCCHANGEADDR and DIOCADDADDR take a struct pf_pooladdr from userspace. They failed to validate the dyn pointer contained in

svn commit: r360097 - head/sbin/pfctl

Author: kp Date: Sun Apr 19 15:37:13 2020 New Revision: 360097 URL: https://svnweb.freebsd.org/changeset/base/360097 Log: pfctl: Call ifa_load() before ifa_grouplookup() ifa_grouplookup() uses the data loaded in ifa_load() (through is_a_group()), so we must call ifa_load() before we can

svn commit: r360096 - head/sbin/pfctl

Author: kp Date: Sun Apr 19 15:32:14 2020 New Revision: 360096 URL: https://svnweb.freebsd.org/changeset/base/360096 Log: pfctl: Remove unused variable Submitted by: Nick Rogers MFC after:1 week Sponsored by: RG Nets Modified: head/sbin/pfctl/pfctl_parser.c Modified: head/sbin/p

Re: svn commit: r360068 - in head/sys: kern net sys

On 19 Apr 2020, at 15:33, Ronald Klop wrote: On Sat, 18 Apr 2020 09:50:30 +0200, Kristof Provost wrote: Author: kp Date: Sat Apr 18 07:50:30 2020 New Revision: 360068 URL: https://svnweb.freebsd.org/changeset/base/360068 Log: ethersubr: Make the mac address generation more robust If we

svn commit: r360069 - head/sys/net

Author: kp Date: Sat Apr 18 08:00:58 2020 New Revision: 360069 URL: https://svnweb.freebsd.org/changeset/base/360069 Log: bridge: Simplify mac address generation Unconditionally use ether_gen_addr() to generate bridge mac addresses. This function is now less likely to generate duplicate

svn commit: r360068 - in head/sys: kern net sys

Author: kp Date: Sat Apr 18 07:50:30 2020 New Revision: 360068 URL: https://svnweb.freebsd.org/changeset/base/360068 Log: ethersubr: Make the mac address generation more robust If we create two (vnet) jails and create a bridge interface in each we end up with the same mac address on both

svn commit: r360043 - head/tests/sys/net

Author: kp Date: Fri Apr 17 14:57:15 2020 New Revision: 360043 URL: https://svnweb.freebsd.org/changeset/base/360043 Log: bridge tests: Test deleting a bridge with members Reviewed by: philip, emaste Sponsored by: The FreeBSD Foundation Differential Revision:https://reviews.fre

  1   2   3   4   5   6   7   8   >