| lists.openwall.net | lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC | |
|
Open Source and information security mailing list archives
| ||
|
Date: Mon, 26 Apr 2010 16:03:04 +0200
From: Eric Dumazet <eric.dumazet@...il.com>
To: hadi@...erus.ca
Cc: Changli Gao <xiaosuo@...il.com>,
"David S. Miller" <davem@...emloft.net>,
Tom Herbert <therbert@...gle.com>,
Stephen Hemminger <shemminger@...tta.com>,
netdev@...r.kernel.org
Subject: Re: [PATCH v6] net: batch skb dequeueing from softnet
input_pkt_queue
Le samedi 24 avril 2010 à 10:10 -0400, jamal a écrit :
> On Fri, 2010-04-23 at 18:02 -0400, jamal wrote:
>
> > Ive done a setup with the last patch from Changli + net-next - I will
> > post test results tomorrow AM.
>
> ok, annotated results attached.
>
> cheers,
> jamal
Jamal, I have a Nehalem setup now, and I can see
_raw_spin_lock_irqsave() abuse is not coming from network tree, but from
clockevents_notify()
My pktgen sends 1040989pps :
# Samples: 389707198131
#
# Overhead Command Shared Object Symbol
# ........ .............. ............................ ......
#
23.52% init [kernel.kallsyms] [k] _raw_spin_lock_irqsave
|
--- _raw_spin_lock_irqsave
|
|--94.74%-- clockevents_notify
| lapic_timer_state_broadcast
| acpi_idle_enter_bm
| cpuidle_idle_call
| cpu_idle
| start_secondary
|
|--4.10%-- tick_broadcast_oneshot_control
| tick_notify
| notifier_call_chain
| __raw_notifier_call_chain
| raw_notifier_call_chain
| clockevents_do_notify
| clockevents_notify
| lapic_timer_state_broadcast
| acpi_idle_enter_bm
| cpuidle_idle_call
| cpu_idle
| start_secondary
|
|--0.58%-- lapic_timer_state_broadcast
| acpi_idle_enter_bm
| cpuidle_idle_call
| cpu_idle
| start_secondary
--0.58%-- [...]
8.94% init [kernel.kallsyms] [k] acpi_os_read_port
|
--- acpi_os_read_port
|
|--99.55%-- acpi_hw_read_port
| acpi_hw_read
| acpi_hw_read_multiple
| acpi_hw_register_read
| acpi_read_bit_register
# Samples: 389233082962
#
# Overhead Command Shared Object Symbol
# ........ .............. ............................ ......
#
23.25% init [kernel.kallsyms] [k] _raw_spin_lock_irqsave
8.90% init [kernel.kallsyms] [k] acpi_os_read_port
2.93% init [kernel.kallsyms] [k] mwait_idle_with_hints
1.99% init [kernel.kallsyms] [k] schedule
1.94% udpsink [kernel.kallsyms] [k] schedule
1.73% swapper [kernel.kallsyms] [k] _raw_spin_lock_irqsave
1.48% init [kernel.kallsyms] [k] bnx2x_rx_int
1.47% init [kernel.kallsyms] [k] _raw_spin_unlock_irqrestore
1.44% init [kernel.kallsyms] [k] _raw_spin_lock
1.36% udpsink [kernel.kallsyms] [k] udp_recvmsg
1.05% udpsink [kernel.kallsyms] [k] __skb_recv_datagram
1.05% init [kernel.kallsyms] [k] __udp4_lib_lookup
1.04% udpsink [kernel.kallsyms] [k] copy_user_generic_string
1.04% udpsink [kernel.kallsyms] [k] __slab_free
0.99% init [kernel.kallsyms] [k] select_task_rq_fair
0.99% init [kernel.kallsyms] [k] try_to_wake_up
0.98% init [kernel.kallsyms] [k] task_rq_lock
0.93% init [kernel.kallsyms] [k] tick_broadcast_oneshot_control
0.89% init [kernel.kallsyms] [k] sock_queue_rcv_skb
0.89% udpsink [kernel.kallsyms] [k] sock_recv_ts_and_drops
0.88% udpsink [kernel.kallsyms] [k] kfree
0.79% swapper [kernel.kallsyms] [k] acpi_os_read_port
0.76% udpsink [kernel.kallsyms] [k] _raw_spin_lock_irqsave
0.73% udpsink [kernel.kallsyms] [k] inet_recvmsg
0.71% udpsink [vdso] [.] 0x000000ffffe431
0.65% udpsink [kernel.kallsyms] [k] sock_recvmsg
0.62% init [kernel.kallsyms] [k] gs_change
0.61% init [kernel.kallsyms] [k] enqueue_task_fair
0.61% init [kernel.kallsyms] [k] eth_type_trans
0.61% init [kernel.kallsyms] [k] sock_def_readable
0.60% udpsink [kernel.kallsyms] [k] _raw_spin_lock_bh
0.59% init [kernel.kallsyms] [k] ip_route_input
0.59% udpsink libpthread-2.3.4.so [.] __pthread_disable_asynccancel
0.56% init [kernel.kallsyms] [k] bnx2x_poll
0.56% udpsink [kernel.kallsyms] [k] __get_user_4
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Powered by blists - more mailing lists