- 16 May, 2014 12 commits
-
-
Jarno Rajahalme authored
We already extract the TCP flags for the key, might as well use that for stats. Signed-off-by:
Jarno Rajahalme <jrajahalme@nicira.com> Acked-by:
Pravin B Shelar <pshelar@nicira.com> Signed-off-by:
Jesse Gross <jesse@nicira.com>
-
Jarno Rajahalme authored
The 'output' argument of the ovs_nla_put_flow() is the one from which the bits are written to the netlink attributes. For SCTP we accidentally used the bits from the 'swkey' instead. This caused the mask attributes to include the bits from the actual flow key instead of the mask. Signed-off-by:
-
Jarno Rajahalme authored
Keep kernel flow stats for each NUMA node rather than each (logical) CPU. This avoids using the per-CPU allocator and removes most of the kernel-side OVS locking overhead otherwise on the top of perf reports and allows OVS to scale better with higher number of threads. With 9 handlers and 4 revalidators netperf TCP_CRR test flow setup rate doubles on a server with two hyper-threaded physical CPUs (16 logical cores each) compared to the current OVS master. Tested with non-trivial flow table with a TCP port match rule forcing all new connections with unique port numbers to OVS userspace. The IP addresses are still wildcarded, so the kernel flows are not considered as exact match 5-tuple flows. This type of flows can be expected to appear in large numbers as the result of more effective wildcarding made possible by improvements in OVS userspace flow classifier. Perf results for this test (master): Events: 305K cycles + 8.43% ovs-vswitchd [kernel.kallsyms] [k] mutex_spin_on_owner + 5.64% ovs-vswitchd [kernel.kallsyms] [k] __ticket_spin_lock + 4.75% ovs-vswitchd ovs-vswitchd [.] find_match_wc + 3.32% ovs-vswitchd libpthread-2.15.so [.] pthread_mutex_lock + 2.61% ovs-vswitchd [kernel.kallsyms] [k] pcpu_alloc_area + 2.19% ovs-vswitchd ovs-vswitchd [.] flow_hash_in_minimask_range + 2.03% swapper [kernel.kallsyms] [k] intel_idle + 1.84% ovs-vswitchd libpthread-2.15.so [.] pthread_mutex_unlock + 1.64% ovs-vswitchd ovs-vswitchd [.] classifier_lookup + 1.58% ovs-vswitchd libc-2.15.so [.] 0x7f4e6 + 1.07% ovs-vswitchd [kernel.kallsyms] [k] memset + 1.03% netperf [kernel.kallsyms] [k] __ticket_spin_lock + 0.92% swapper [kernel.kallsyms] [k] __ticket_spin_lock ... And after this patch: Events: 356K cycles + 6.85% ovs-vswitchd ovs-vswitchd [.] find_match_wc + 4.63% ovs-vswitchd libpthread-2.15.so [.] pthread_mutex_lock + 3.06% ovs-vswitchd [kernel.kallsyms] [k] __ticket_spin_lock + 2.81% ovs-vswitchd ovs-vswitchd [.] flow_hash_in_minimask_range + 2.51% ovs-vswitchd libpthread-2.15.so [.] pthread_mutex_unlock + 2.27% ovs-vswitchd ovs-vswitchd [.] classifier_lookup + 1.84% ovs-vswitchd libc-2.15.so [.] 0x15d30f + 1.74% ovs-vswitchd [kernel.kallsyms] [k] mutex_spin_on_owner + 1.47% swapper [kernel.kallsyms] [k] intel_idle + 1.34% ovs-vswitchd ovs-vswitchd [.] flow_hash_in_minimask + 1.33% ovs-vswitchd ovs-vswitchd [.] rule_actions_unref + 1.16% ovs-vswitchd ovs-vswitchd [.] hindex_node_with_hash + 1.16% ovs-vswitchd ovs-vswitchd [.] do_xlate_actions + 1.09% ovs-vswitchd ovs-vswitchd [.] ofproto_rule_ref + 1.01% netperf [kernel.kallsyms] [k] __ticket_spin_lock ... There is a small increase in kernel spinlock overhead due to the same spinlock being shared between multiple cores of the same physical CPU, but that is barely visible in the netperf TCP_CRR test performance (maybe ~1% performance drop, hard to tell exactly due to variance in the test results), when testing for kernel module throughput (with no userspace activity, handful of kernel flows). On flow setup, a single stats instance is allocated (for the NUMA node 0). As CPUs from multiple NUMA nodes start updating stats, new NUMA-node specific stats instances are allocated. This allocation on the packet processing code path is made to never block or look for emergency memory pools, minimizing the allocation latency. If the allocation fails, the existing preallocated stats instance is used. Also, if only CPUs from one NUMA-node are updating the preallocated stats instance, no additional stats instances are allocated. This eliminates the need to pre-allocate stats instances that will not be used, also relieving the stats reader from the burden of reading stats that are never used. Signed-off-by:
-
Jarno Rajahalme authored
The 5-tuple optimization becomes unnecessary with a later per-NUMA node stats patch. Remove it first to make the changes easier to grasp. Signed-off-by:
-
Joe Perches authored
It's slightly smaller/faster for some architectures. Signed-off-by:
Joe Perches <joe@perches.com> Signed-off-by:
-
Joe Perches authored
Add "openvswitch: " prefix to OVS_NLERR output to match the other OVS_NLERR output of datapath.c Signed-off-by:
-
Joe Perches authored
Each use of pr_<level>_once has a per-site flag. Some of the OVS_NLERR messages look as if seeing them multiple times could be useful, so use net_ratelimit() instead of pr_info_once. Signed-off-by:
-
Daniele Di Proietto authored
This is necessary, since u64 is not unsigned long long in all architectures: u64 could be also uint64_t. Signed-off-by:
Daniele Di Proietto <daniele.di.proietto@gmail.com> Signed-off-by:
-
Daniele Di Proietto authored
This function must cast a const value to a non const value. By adding an uintptr_t cast the warning is suppressed. To avoid the cast (proper solution) several function signatures must be changed. Signed-off-by:
-
Daniele Di Proietto authored
This change, firstly, avoids declaring the formal parameter const, since it is treated as non const. (to avoid -Wcast-qual) Secondly, it cast the pointer from void* to u8*, since it is used in arithmetic (to avoid -Wpointer-arith) Signed-off-by:
-
Daniele Di Proietto authored
In few functions, const formal parameters are assigned or cast to non-const. These changes suppress warnings if compiled with -Wcast-qual. Signed-off-by:
-
dingtianhong authored
The port->count was used to count the number of macvlan devs in the same port, but the list vlans could play the same role to do that, so free the port if the list vlans is empty and no need to use the parameter count. Signed-off-by:
Ding Tianhong <dingtianhong@huawei.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
- 15 May, 2014 24 commits
-
-
Julia Lawall authored
Netdev_priv is an accessor function, and has no purpose if its result is not used. A simplified version of the semantic match that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ local idexpression x; @@ -x = netdev_priv(...); ... when != x // </smpl> Signed-off-by:
Julia Lawall <Julia.Lawall@lip6.fr> Signed-off-by:
-
Julia Lawall authored
Netdev_priv is an accessor function, and has no purpose if its result is not used. A simplified version of the semantic match that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ local idexpression x; @@ -x = netdev_priv(...); ... when != x // </smpl> Signed-off-by:
-
Julia Lawall authored
Netdev_priv is an accessor function, and has no purpose if its result is not used. A simplified version of the semantic match that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ local idexpression x; @@ -x = netdev_priv(...); ... when != x // </smpl> Signed-off-by:
-
Julia Lawall authored
Netdev_priv is an accessor function, and has no purpose if its result is not used. A simplified version of the semantic match that fixes this problem is as follows: (http://coccinelle.lip6.fr/) // <smpl> @@ local idexpression x; @@ -x = netdev_priv(...); ... when != x // </smpl> Signed-off-by:
-
Florian Fainelli authored
Packets need to be at least 64 bytes to enter the switch port logic, including the FCS, otherwise they will be discarded as RUNT packets. With packets having Broadcom tags, the 4-bytes tag is first stripped off the packet, and the packet length is then checked, so we need to make sure that the packet length with FCS is at least 64 bytes. Signed-off-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
-
Florian Fainelli authored
The link adjustment callback can be called as frequently as desired by the PHY library, as such, let's avoid doing a Read/Modify/Write sequence if nothing changed, which is more than likely since we are interfaced with a switch device. Signed-off-by:
-
Joe Perches authored
These should not have trailing semicolons so remove them. Signed-off-by:
-
David S. Miller authored
Alexei Starovoitov says: ==================== internal BPF jit for x64 and JITed seccomp Internal BPF JIT compiler for x86_64 replaces classic BPF JIT. Use it in seccomp and in tracing filters (sent as separate patch) ==================== Signed-off-by: -
Alexei Starovoitov authored
Take advantage of internal BPF JIT 05-sim-long_jumps.c of libseccomp was used as micro-benchmark: seccomp_rule_add_exact(ctx,... seccomp_rule_add_exact(ctx,... rc = seccomp_load(ctx); for (i = 0; i < 10000000; i++) syscall(...); $ sudo sysctl net.core.bpf_jit_enable=1 $ time ./bench real 0m2.769s user 0m1.136s sys 0m1.624s $ sudo sysctl net.core.bpf_jit_enable=0 $ time ./bench real 0m5.825s user 0m1.268s sys 0m4.548s Signed-off-by:Alexei Starovoitov <ast@plumgrid.com> Signed-off-by:
-
Alexei Starovoitov authored
Maps all internal BPF instructions into x86_64 instructions. This patch replaces original BPF x64 JIT with internal BPF x64 JIT. sysctl net.core.bpf_jit_enable is reused as on/off switch. Performance: 1. old BPF JIT and internal BPF JIT generate equivalent x86_64 code. No performance difference is observed for filters that were JIT-able before Example assembler code for BPF filter "tcpdump port 22" original BPF -> old JIT: original BPF -> internal BPF -> new JIT: 0: push %rbp 0: push %rbp 1: mov %rsp,%rbp 1: mov %rsp,%rbp 4: sub $0x60,%rsp 4: sub $0x228,%rsp 8: mov %rbx,-0x8(%rbp) b: mov %rbx,-0x228(%rbp) // prologue 12: mov %r13,-0x220(%rbp) 19: mov %r14,-0x218(%rbp) 20: mov %r15,-0x210(%rbp) 27: xor %eax,%eax // clear A c: xor %ebx,%ebx 29: xor %r13,%r13 // clear X e: mov 0x68(%rdi),%r9d 2c: mov 0x68(%rdi),%r9d 12: sub 0x6c(%rdi),%r9d 30: sub 0x6c(%rdi),%r9d 16: mov 0xd8(%rdi),%r8 34: mov 0xd8(%rdi),%r10 3b: mov %rdi,%rbx 1d: mov $0xc,%esi 3e: mov $0xc,%esi 22: callq 0xffffffffe1021e15 43: callq 0xffffffffe102bd75 27: cmp $0x86dd,%eax 48: cmp $0x86dd,%rax 2c: jne 0x0000000000000069 4f: jne 0x000000000000009a 2e: mov $0x14,%esi 51: mov $0x14,%esi 33: callq 0xffffffffe1021e31 56: callq 0xffffffffe102bd91 38: cmp $0x84,%eax 5b: cmp $0x84,%rax 3d: je 0x0000000000000049 62: je 0x0000000000000074 3f: cmp $0x6,%eax 64: cmp $0x6,%rax 42: je 0x0000000000000049 68: je 0x0000000000000074 44: cmp $0x11,%eax 6a: cmp $0x11,%rax 47: jne 0x00000000000000c6 6e: jne 0x0000000000000117 49: mov $0x36,%esi 74: mov $0x36,%esi 4e: callq 0xffffffffe1021e15 79: callq 0xffffffffe102bd75 53: cmp $0x16,%eax 7e: cmp $0x16,%rax 56: je 0x00000000000000bf 82: je 0x0000000000000110 58: mov $0x38,%esi 88: mov $0x38,%esi 5d: callq 0xffffffffe1021e15 8d: callq 0xffffffffe102bd75 62: cmp $0x16,%eax 92: cmp $0x16,%rax 65: je 0x00000000000000bf 96: je 0x0000000000000110 67: jmp 0x00000000000000c6 98: jmp 0x0000000000000117 69: cmp $0x800,%eax 9a: cmp $0x800,%rax 6e: jne 0x00000000000000c6 a1: jne 0x0000000000000117 70: mov $0x17,%esi a3: mov $0x17,%esi 75: callq 0xffffffffe1021e31 a8: callq 0xffffffffe102bd91 7a: cmp $0x84,%eax ad: cmp $0x84,%rax 7f: je 0x000000000000008b b4: je 0x00000000000000c2 81: cmp $0x6,%eax b6: cmp $0x6,%rax 84: je 0x000000000000008b ba: je 0x00000000000000c2 86: cmp $0x11,%eax bc: cmp $0x11,%rax 89: jne 0x00000000000000c6 c0: jne 0x0000000000000117 8b: mov $0x14,%esi c2: mov $0x14,%esi 90: callq 0xffffffffe1021e15 c7: callq 0xffffffffe102bd75 95: test $0x1fff,%ax cc: test $0x1fff,%rax 99: jne 0x00000000000000c6 d3: jne 0x0000000000000117 d5: mov %rax,%r14 9b: mov $0xe,%esi d8: mov $0xe,%esi a0: callq 0xffffffffe1021e44 dd: callq 0xffffffffe102bd91 // MSH e2: and $0xf,%eax e5: shl $0x2,%eax e8: mov %rax,%r13 eb: mov %r14,%rax ee: mov %r13,%rsi a5: lea 0xe(%rbx),%esi f1: add $0xe,%esi a8: callq 0xffffffffe1021e0d f4: callq 0xffffffffe102bd6d ad: cmp $0x16,%eax f9: cmp $0x16,%rax b0: je 0x00000000000000bf fd: je 0x0000000000000110 ff: mov %r13,%rsi b2: lea 0x10(%rbx),%esi 102: add $0x10,%esi b5: callq 0xffffffffe1021e0d 105: callq 0xffffffffe102bd6d ba: cmp $0x16,%eax 10a: cmp $0x16,%rax bd: jne 0x00000000000000c6 10e: jne 0x0000000000000117 bf: mov $0xffff,%eax 110: mov $0xffff,%eax c4: jmp 0x00000000000000c8 115: jmp 0x000000000000011c c6: xor %eax,%eax 117: mov $0x0,%eax c8: mov -0x8(%rbp),%rbx 11c: mov -0x228(%rbp),%rbx // epilogue cc: leaveq 123: mov -0x220(%rbp),%r13 cd: retq 12a: mov -0x218(%rbp),%r14 131: mov -0x210(%rbp),%r15 138: leaveq 139: retq On fully cached SKBs both JITed functions take 12 nsec to execute. BPF interpreter executes the program in 30 nsec. The difference in generated assembler is due to the following: Old BPF imlements LDX_MSH instruction via sk_load_byte_msh() helper function inside bpf_jit.S. New JIT removes the helper and does it explicitly, so ldx_msh cost is the same for both JITs, but generated code looks longer. New JIT has 4 registers to save, so prologue/epilogue are larger, but the cost is within noise on x64. Old JIT checks whether first insn clears A and if not emits 'xor %eax,%eax'. New JIT clears %rax unconditionally. 2. old BPF JIT doesn't support ANC_NLATTR, ANC_PAY_OFFSET, ANC_RANDOM extensions. New JIT supports all BPF extensions. Performance of such filters improves 2-4 times depending on a filter. The longer the filter the higher performance gain. Synthetic benchmarks with many ancillary loads see 20x speedup which seems to be the maximum gain from JIT Notes: . net.core.bpf_jit_enable=2 + tools/net/bpf_jit_disasm is still functional and can be used to see generated assembler . there are two jit_compile() functions and code flow for classic filters is: sk_attach_filter() - load classic BPF bpf_jit_compile() - try to JIT from classic BPF sk_convert_filter() - convert classic to internal bpf_int_jit_compile() - JIT from internal BPF seccomp and tracing filters will just call bpf_int_jit_compile() Signed-off-by: -
Alexei Starovoitov authored
Split bpf_jit_compile() into two functions to improve readability of for(pass++) loop. The change follows similar style of JIT compilers for arm, powerpc, s390 The body of new do_jit() was not reformatted to reduce noise in this patch, since the following patch replaces most of it. Tested with BPF testsuite. Signed-off-by:
-
David S. Miller authored
Phoebe Buckheister says: ==================== 802154: some cleanups and fixes This series adds some definitions for 802.15.4 header fields that were missing, changes 6lowpan fragmentation to be aware of security headers and fixes 802.15.4 datagram socket sendmsg(), which was entirely incompliant to date. Also a few minor changes to mac_cb handling, mark a single-use function static, and correctly check for EMSGSIZE conditions during wpan_header_create. Changes since v1: * rename mac_cb_alloc to mac_cb_init * catch all error cases of sendmsg() instead of only !conn && msg_name * redo 6lowpan fragmentation to not clone lower layer headers ==================== Signed-off-by: -
Phoebe Buckheister authored
This function is only used within the same translation unit, so mark it static. Signed-off-by:
Phoebe Buckheister <phoebe.buckheister@itwm.fraunhofer.de> Signed-off-by:
-
Phoebe Buckheister authored
802.15.4 datagram sockets do not currently have a compliant sendmsg(). The destination address supplied is always ignored, and in unconnected mode, packets are broadcast instead of dropped with -EDESTADDRREQ. This patch fixes 802.15.4 dgram sockets to be compliant, i.e. !conn && !msg_name => -EDESTADDRREQ !conn && msg_name => send to msg_name conn && !msg_name => send to connected conn && msg_name => -EISCONN Signed-off-by:
-
Phoebe Buckheister authored
Currently, 6lowpan creates one 802.15.4 MAC header for the original packet the device was given by upper layers and reuses this header for all fragments, if fragmentation is required. This also reuses frame sequence numbers, which must not happen. 6lowpan also has issues with fragmentation in the presence of security headers, since those may imply the presence of trailing fields that are not accounted for by the fragmentation code right now. Fix both of these issues by properly allocating fragment skbs with headromm and tailroom as specified by the underlying device, create one header for each skb instead of reusing the original header, let the underlying device do the rest. Signed-off-by:
-
Phoebe Buckheister authored
The current mac_cb handling of ieee802154 is rather awkward and limited. Decompose the single flags field into multiple fields with the meanings of each subfield of the flags field to make future extensions (for example, link-layer security) easier. Also don't set the frame sequence number in upper layers, since that's a thing the MAC is supposed to set on frame transmit - we set it on header creation, but assuming that upper layers do not blindly duplicate our headers, this is fine. Signed-off-by:
-
Phoebe Buckheister authored
The current WPAN header creation code checks for EMSGSIZE conditions, but does not account for the MIC field that link layer security may add at the end of the frame. Now that we can accurately calculate the maximum payload size of packets, use that to check for EMSGSIZE conditions. Signed-off-by:
-
Phoebe Buckheister authored
When dealing with 802.15.4, one often has to know the maximum payload size for a given packet. This depends on many factors, one of which is whether or not a security header is present in the frame. These definitions and functions provide an easy way for any upper layer to calculate the maximum payload size for a packet. The first obvious user for this is 6lowpan, which duplicates this calculation and gets it partially wrong because it ignores security headers. Signed-off-by:
-
Markus Lottmann authored
This unifies the behaviour with other network device drivers and allows for a matching of the PCI device path in UDEV rules. Signed-off-by:
Markus Lottmann <markus.lottmann1986@gmail.com> Signed-off-by:
-
David S. Miller authored
> include/net/ip.h:211:5: warning: "CONFIG_SYSCTL" is not defined [-Wundef] > #if CONFIG_SYSCTL > ^ Reported-by:
Stephen Rothwell <sfr@canb.auug.org.au> Signed-off-by:
-
David S. Miller authored
George Cherian says: ==================== TI CPSW Cleanup This series does some minimal cleanups. -Conversion of pr_*() to dev_*() -Convert kzalloc to devm_kzalloc. ==================== Signed-off-by: -
George Cherian authored
Convert kzalloc() to devm_kzalloc(). Signed-off-by:
George Cherian <george.cherian@ti.com> Reviewed-by:
Felipe Balbi <balbi@ti.com> Signed-off-by:
-
George Cherian authored
Convert the lone pr_err() to dev_err() call. Signed-off-by:
-
George Cherian authored
Convert all pr_*() calls to dev_*() calls. No functional changes. Signed-off-by:
-
- 14 May, 2014 4 commits
-
-
Darek Marcinkiewicz authored
Sparse was reporting quite a few warnings for the driver. Those get fixed by this patch. Signed-off-by:
Dariusz Marcinkiewicz <reksio@newterm.pl> Signed-off-by:
-
Joe Perches authored
Missing a colon on definition use is a bit odd so change the macro for the 32 bit case to declare an __attribute__((unused)) and __deprecated variable. The __deprecated attribute will cause gcc to emit an error if the variable is actually used. Signed-off-by:
-
Jingoo Han authored
Use devm_ioremap_resource() because devm_request_and_ioremap() is obsoleted by devm_ioremap_resource(). Signed-off-by:
Jingoo Han <jg1.han@samsung.com> Acked-by:
-
David S. Miller authored
Or Gerlitz says: ==================== Mellanox driver update 2014-05-12 This patchset introduce some small bug fixes: Eyal fixed some compilation and syntactic checkers warnings. Ido fixed a coruption in user priority mapping when changing number of channels. Shani fixed some other problems when modifying MAC address. Yuval fixed a problem when changing IRQ affinity during high traffic - IRQ changes got effective only after the first pause in traffic. This patchset was tested and applied over commit 93dccc59: "mdio_bus: fix devm_mdiobus_alloc_size export" Changes from V1: - applied feedback from Dave to use true/false and not 0/1 in patch 1/9 - removed the patch from Noa which adddressed a bug in flow steering table when using a bond device, as the fix might need to be in the bonding driver, this is now dicussed in the netdev thread "bonding directly changes underlying device address" Changes from V0: - Patch 1/9 - net/mlx4_core: Enforce irq affinity changes immediatly - Moved the new members to a hot cache line as Eric suggested ==================== Signed-off-by:
-
