Re: machdep.hlt_cpus not safe with ULE?
on 19/02/2011 14:36 Steven Hartland said the following: I'm trying to debug a possibly failing CPU, so I thought it would be easy just disable the cores using machdep.hlt_cpus and see if we see the panic's we've been seeing. The problem is it seems ULE doesnt properly support machdep.hlt_cpus and still schedules processes onto the halted cpus which obviously causes problems. Can anyone confirm this behaviour? Yes, your observations are correct. Please also see: http://www.freebsd.org/cgi/query-pr.cgi?pr=145385 Should machdep.hlt_cpus and I assume the logical counterpart never be used with ULE? -- Andriy Gapon ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: machdep.hlt_cpus not safe with ULE?
On Mon, Feb 21, 2011 at 8:29 AM, Andriy Gapon a...@freebsd.org wrote: on 19/02/2011 14:36 Steven Hartland said the following: I'm trying to debug a possibly failing CPU, so I thought it would be easy just disable the cores using machdep.hlt_cpus and see if we see the panic's we've been seeing. The problem is it seems ULE doesnt properly support machdep.hlt_cpus and still schedules processes onto the halted cpus which obviously causes problems. Can anyone confirm this behaviour? Yes, your observations are correct. Please also see: http://www.freebsd.org/cgi/query-pr.cgi?pr=145385 Should machdep.hlt_cpus and I assume the logical counterpart never be used with ULE? As a followup to this and based on discussions with other folks, the fact that it's using hlt to halt CPUs without rescheduling tasks / masking interrupts, etc is not good. So none of the *hlt* sysctls are really doing the right thing on x86. Thanks, -Garrett ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: machdep.hlt_cpus not safe with ULE?
- Original Message - From: Garrett Cooper gcoo...@freebsd.org As a followup to this and based on discussions with other folks, the fact that it's using hlt to halt CPUs without rescheduling tasks / masking interrupts, etc is not good. So none of the *hlt* sysctls are really doing the right thing on x86. Time to disable them until they are fixed properly then I would suggest? Regards Steve This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. In the event of misdirection, illegible or incomplete transmission please telephone +44 845 868 1337 or return the E.mail to postmas...@multiplay.co.uk. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: machdep.hlt_cpus not safe with ULE?
On Mon, Feb 21, 2011 at 12:30 PM, Steven Hartland kill...@multiplay.co.uk wrote: - Original Message - From: Garrett Cooper gcoo...@freebsd.org As a followup to this and based on discussions with other folks, the fact that it's using hlt to halt CPUs without rescheduling tasks / masking interrupts, etc is not good. So none of the *hlt* sysctls are really doing the right thing on x86. Time to disable them until they are fixed properly then I would suggest? Andriy's patch attached to the PR above does the right thing when first bringing up the system, but it's still broken with the sysctl case, so I would actually vote to disable the sysctls for now and commit his patch separately as it's better than the existing code is in that area. Thanks! -Garrett ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
machdep.hlt_cpus not safe with ULE?
I'm trying to debug a possibly failing CPU, so I thought it would be easy just disable the cores using machdep.hlt_cpus and see if we see the panic's we've been seeing. The problem is it seems ULE doesnt properly support machdep.hlt_cpus and still schedules processes onto the halted cpus which obviously causes problems. Can anyone confirm this behaviour? Should machdep.hlt_cpus and I assume the logical counterpart never be used with ULE? Regards Steve This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. In the event of misdirection, illegible or incomplete transmission please telephone +44 845 868 1337 or return the E.mail to postmas...@multiplay.co.uk. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: machdep.hlt_cpus not safe with ULE?
On Sat, 19 Feb 2011 12:36:57 - Steven Hartland kill...@multiplay.co.uk wrote: I'm trying to debug a possibly failing CPU, so I thought it would be easy just disable the cores using machdep.hlt_cpus and see if we see the panic's we've been seeing. The problem is it seems ULE doesnt properly support machdep.hlt_cpus and still schedules processes onto the halted cpus which obviously causes problems. Can anyone confirm this behaviour? Should machdep.hlt_cpus and I assume the logical counterpart never be used with ULE? Looking at the kernel source it appears that only sched_4bsd.c makes use of hlt_cpus_mask. -- Gary Jennejohn ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: machdep.hlt_cpus not safe with ULE?
- Original Message - From: Gary Jennejohn gljennj...@googlemail.com Looking at the kernel source it appears that only sched_4bsd.c makes use of hlt_cpus_mask. Given ULE is default do these need to be either removed totally or at least conditionally based on the scheduler choice as currently they are quite dangerous to a systems health? Regards Steve This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. In the event of misdirection, illegible or incomplete transmission please telephone +44 845 868 1337 or return the E.mail to postmas...@multiplay.co.uk. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org
Re: machdep.hlt_cpus not safe with ULE?
For reference I've found that an alternative is to set the following in loader.conf:- hint.lapic.2.disabled=1 hint.lapic.3.disabled=1 2 and 3 here are the apic numbers displayed by dmesg on boot for the cpu's Obviously this requires a reboot so no perfect for all uses but it does work for what we're testing. This e.mail is private and confidential between Multiplay (UK) Ltd. and the person or entity to whom it is addressed. In the event of misdirection, the recipient is prohibited from using, copying, printing or otherwise disseminating it or any information contained in it. In the event of misdirection, illegible or incomplete transmission please telephone +44 845 868 1337 or return the E.mail to postmas...@multiplay.co.uk. ___ freebsd-hackers@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-hackers To unsubscribe, send any mail to freebsd-hackers-unsubscr...@freebsd.org