Messages in this thread | | | From | Kevin Hilman <> | Subject | Re: [PATCH RFC 0/41] RCU commits for 3.4 | Date | Thu, 02 Feb 2012 14:47:14 -0800 |
| |
Hi Paul,
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com> writes:
> This patchset makes RCU more aggressive about entering dyntick-idle mode > in order to improve energy efficiency,
I gave this series a spin on an ARM/SMP platform (OMAP4) and the boot hung (tail end of boot log below.) I then realized that had isolcpus=1 on the cmdline from a previous experiment. Removing isolcpus from the cmdline got me back to a working boot.
IIRC, some want to remove isolcpus, so I'm not sure if it's expected to work with your series, but thought you'd want to know.
Kevin
[ 1.105834] Serial: 8250/16550 driver, 4 ports, IRQ sharing enabled [ 1.353942] omap_uart.0: ttyO0 at MMIO 0x4806a000 (irq = 104) is a OMAP UART0 [ 1.462707] omap_uart.1: ttyO1 at MMIO 0x4806c000 (irq = 105) is a OMAP UART1 [ 1.548614] omap_uart.2: ttyO2 at MMIO 0x48020000 (irq = 106) is a OMAP UART2 [ 1.556304] console [ttyO2] enabled, bootconsole disabled [ 1.556304] console [ttyO2] enabled, bootconsole disabled [ 61.587493] INFO: rcu_sched detected stalls on CPUs/tasks: { 1} (detected by 0, t=7682 jiffies) [ 61.596649] INFO: Stall ended before state dump start
| |