linux.git/Documentation/admin-guide/sysctl/net.rst, branch v6.0

net: Fix data-races around netdev_max_backlog.

2022-08-24T12:46:57+00:00

While reading netdev_max_backlog, it can be changed concurrently.
Thus, we need to add READ_ONCE() to its readers.

While at it, we remove the unnecessary spaces in the doc.

Fixes: 1da177e4c3f4 ("Linux-2.6.12-rc2")
Signed-off-by: Kuniyuki Iwashima 
Signed-off-by: David S. Miller

Documentation: add a description for net.core.high_order_alloc_disable

2022-07-09T03:15:59+00:00

A description is missing for the net.core.high_order_alloc_disable
option in admin-guide/sysctl/net.rst ; add it. The above sysctl option
was introduced by commit ce27ec60648d ("net: add high_order_alloc_disable
sysctl/static key").

Thanks to Eric for running again the benchmark cited in the above
commit, showing this knob is now mostly of historical importance.

Signed-off-by: Antoine Tenart 
Reviewed-by: Eric Dumazet 
Link: https://lore.kernel.org/r/20220707080245.180525-1-atenart@kernel.org
Signed-off-by: Jakub Kicinski

Merge git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net

2022-05-24T04:19:17+00:00

drivers/net/ethernet/cadence/macb_main.c
  5cebb40bc955 ("net: macb: Fix PTP one step sync support")
  138badbc21a0 ("net: macb: use NAPI for TX completion path")
https://lore.kernel.org/all/20220523111021.31489367@canb.auug.org.au/

net/smc/af_smc.c
  75c1edf23b95 ("net/smc: postpone sk_refcnt increment in connect()")
  3aba103006bc ("net/smc: align the connect behaviour with TCP")
https://lore.kernel.org/all/20220524114408.4bf1af38@canb.auug.org.au/

Signed-off-by: Jakub Kicinski

Documentation: add description for net.core.gro_normal_batch

2022-05-20T00:46:23+00:00

Describe it in admin-guide/sysctl/net.rst like other Network core options.
Users need to know gro_normal_batch for performance tuning.

Fixes: 323ebb61e32b ("net: use listified RX for handling GRO_NORMAL skbs")
Reported-by: Prijesh Patel 
Signed-off-by: Xin Long 
Reviewed-by: Edward Cree 
Link: https://lore.kernel.org/r/acf8a2c03b91bcde11f67ff89b6050089c0712a3.1652888963.git.lucien.xin@gmail.com
Signed-off-by: Jakub Kicinski

net: add skb_defer_max sysctl

2022-05-16T10:33:59+00:00

commit 68822bdf76f1 ("net: generalize skb freeing
deferral to per-cpu lists") added another per-cpu
cache of skbs. It was expected to be small,
and an IPI was forced whenever the list reached 128
skbs.

We might need to be able to control more precisely
queue capacity and added latency.

An IPI is generated whenever queue reaches half capacity.

Default value of the new limit is 64.

Signed-off-by: Eric Dumazet 
Signed-off-by: David S. Miller

txhash: Add txrehash sysctl description

2022-01-31T15:05:25+00:00

Update Documentation/admin-guide/sysctl/net.rst with txrehash usage
description.

Signed-off-by: Akhmat Karakotov 
Signed-off-by: David S. Miller

Merge tag 'powerpc-5.13-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux

2021-04-30T19:22:28+00:00

Pull powerpc updates from Michael Ellerman:

 - Enable KFENCE for 32-bit.

 - Implement EBPF for 32-bit.

 - Convert 32-bit to do interrupt entry/exit in C.

 - Convert 64-bit BookE to do interrupt entry/exit in C.

 - Changes to our signal handling code to use user_access_begin/end()
   more extensively.

 - Add support for time namespaces (CONFIG_TIME_NS)

 - A series of fixes that allow us to reenable STRICT_KERNEL_RWX.

 - Other smaller features, fixes & cleanups.

Thanks to Alexey Kardashevskiy, Andreas Schwab, Andrew Donnellan, Aneesh
Kumar K.V, Athira Rajeev, Bhaskar Chowdhury, Bixuan Cui, Cédric Le
Goater, Chen Huang, Chris Packham, Christophe Leroy, Christopher M.
Riedl, Colin Ian King, Dan Carpenter, Daniel Axtens, Daniel Henrique
Barboza, David Gibson, Davidlohr Bueso, Denis Efremov, dingsenjie,
Dmitry Safonov, Dominic DeMarco, Fabiano Rosas, Ganesh Goudar, Geert
Uytterhoeven, Geetika Moolchandani, Greg Kurz, Guenter Roeck, Haren
Myneni, He Ying, Jiapeng Chong, Jordan Niethe, Laurent Dufour, Lee
Jones, Leonardo Bras, Li Huafei, Madhavan Srinivasan, Mahesh Salgaonkar,
Masahiro Yamada, Nathan Chancellor, Nathan Lynch, Nicholas Piggin,
Oliver O'Halloran, Paul Menzel, Pu Lehui, Randy Dunlap, Ravi Bangoria,
Rosen Penev, Russell Currey, Santosh Sivaraj, Sebastian Andrzej Siewior,
Segher Boessenkool, Shivaprasad G Bhat, Srikar Dronamraju, Stephen
Rothwell, Thadeu Lima de Souza Cascardo, Thomas Gleixner, Tony Ambardar,
Tyrel Datwyler, Vaibhav Jain, Vincenzo Frascino, Xiongwei Song, Yang Li,
Yu Kuai, and Zhang Yunkai.

* tag 'powerpc-5.13-1' of git://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux: (302 commits)
  powerpc/signal32: Fix erroneous SIGSEGV on RT signal return
  powerpc: Avoid clang uninitialized warning in __get_user_size_allowed
  powerpc/papr_scm: Mark nvdimm as unarmed if needed during probe
  powerpc/kvm: Fix build error when PPC_MEM_KEYS/PPC_PSERIES=n
  powerpc/kasan: Fix shadow start address with modules
  powerpc/kernel/iommu: Use largepool as a last resort when !largealloc
  powerpc/kernel/iommu: Align size for IOMMU_PAGE_SIZE() to save TCEs
  powerpc/44x: fix spelling mistake in Kconfig "varients" -> "variants"
  powerpc/iommu: Annotate nested lock for lockdep
  powerpc/iommu: Do not immediately panic when failed IOMMU table allocation
  powerpc/iommu: Allocate it_map by vmalloc
  selftests/powerpc: remove unneeded semicolon
  powerpc/64s: remove unneeded semicolon
  powerpc/eeh: remove unneeded semicolon
  powerpc/selftests: Add selftest to test concurrent perf/ptrace events
  powerpc/selftests/perf-hwbreak: Add testcases for 2nd DAWR
  powerpc/selftests/perf-hwbreak: Coalesce event creation code
  powerpc/selftests/ptrace-hwbreak: Add testcases for 2nd DAWR
  powerpc/configs: Add IBMVNIC to some 64-bit configs
  selftests/powerpc: Add uaccess flush test
  ...

powerpc/bpf: Implement extended BPF on PPC32

2021-04-03T10:22:21+00:00

Implement Extended Berkeley Packet Filter on Powerpc 32

Test result with test_bpf module:

	test_bpf: Summary: 378 PASSED, 0 FAILED, [354/366 JIT'ed]

Registers mapping:

	[BPF_REG_0] = r11-r12
	/* function arguments */
	[BPF_REG_1] = r3-r4
	[BPF_REG_2] = r5-r6
	[BPF_REG_3] = r7-r8
	[BPF_REG_4] = r9-r10
	[BPF_REG_5] = r21-r22 (Args 9 and 10 come in via the stack)
	/* non volatile registers */
	[BPF_REG_6] = r23-r24
	[BPF_REG_7] = r25-r26
	[BPF_REG_8] = r27-r28
	[BPF_REG_9] = r29-r30
	/* frame pointer aka BPF_REG_10 */
	[BPF_REG_FP] = r17-r18
	/* eBPF jit internal registers */
	[BPF_REG_AX] = r19-r20
	[TMP_REG] = r31

As PPC32 doesn't have a redzone in the stack, a stack frame must always
be set in order to host at least the tail count counter.

The stack frame remains for tail calls, it is set by the first callee
and freed by the last callee.

r0 is used as temporary register as much as possible. It is referenced
directly in the code in order to avoid misusing it, because some
instructions interpret it as value 0 instead of register r0
(ex: addi, addis, stw, lwz, ...)

The following operations are not implemented:

		case BPF_ALU64 | BPF_DIV | BPF_X: /* dst /= src */
		case BPF_ALU64 | BPF_MOD | BPF_X: /* dst %= src */
		case BPF_STX | BPF_XADD | BPF_DW: /* *(u64 *)(dst + off) += src */

The following operations are only implemented for power of two constants:

		case BPF_ALU64 | BPF_MOD | BPF_K: /* dst %= imm */
		case BPF_ALU64 | BPF_DIV | BPF_K: /* dst /= imm */

Signed-off-by: Christophe Leroy 
Signed-off-by: Michael Ellerman 
Link: https://lore.kernel.org/r/61d8b149176ddf99e7d5cef0b6dc1598583ca202.1616430991.git.christophe.leroy@csgroup.eu

net: change netdev_unregister_timeout_secs min value to 1

2021-03-26T00:24:06+00:00

netdev_unregister_timeout_secs=0 can lead to printing the
"waiting for dev to become free" message every jiffy.
This is too frequent and unnecessary.
Set the min value to 1 second.

Also fix the merge issue introduced by
"net: make unregister netdev warning timeout configurable":
it changed "refcnt != 1" to "refcnt".

Signed-off-by: Dmitry Vyukov 
Suggested-by: Eric Dumazet 
Fixes: 5aa3afe107d9 ("net: make unregister netdev warning timeout configurable")
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Reviewed-by: Eric Dumazet 
Signed-off-by: David S. Miller

net: make unregister netdev warning timeout configurable

2021-03-24T00:22:50+00:00

netdev_wait_allrefs() issues a warning if refcount does not drop to 0
after 10 seconds. While 10 second wait generally should not happen
under normal workload in normal environment, it seems to fire falsely
very often during fuzzing and/or in qemu emulation (~10x slower).
At least it's not possible to understand if it's really a false
positive or not. Automated testing generally bumps all timeouts
to very high values to avoid flake failures.
Add net.core.netdev_unregister_timeout_secs sysctl to make
the timeout configurable for automated testing systems.
Lowering the timeout may also be useful for e.g. manual bisection.
The default value matches the current behavior.

Signed-off-by: Dmitry Vyukov 
Fixes: https://bugzilla.kernel.org/show_bug.cgi?id=211877
Cc: netdev@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: David S. Miller