[Discuss] FW: Kernel interpretation of pressing the reset button?
Cy Schubert
Cy.Schubert at cschubert.com
Thu Jun 14 13:04:59 PDT 2018
In message <4E6AE5F528744E479C4F2AC326FFF840024809DDB7 at EX10MBX01.intra.c
amosun.
bc.ca>, "Deryk Barker (Honorary Colleague)" writes:
> From: Discuss [discuss-bounces at vlug.org] on behalf of Cy Schubert [Cy.Schuber
> t at cschubert.com]
> Sent: Monday, June 11, 2018 5:46 PM
> To: discuss at vlug.org
> Subject: Re: [Discuss] Kernel interpretation of pressing the reset button?
>
> >In message <4E6AE5F528744E479C4F2AC326FFF840024809C631 at EX10MBX01.intra.c
> amosun.
> bc.ca>, "Deryk Barker (Honorary Colleague)" writes:
> >> Well, I *did* say unless things have changed.
> >>
> >> Certainly on the old PCI bus there was a line called RESET and *all* the b
> utt
> >> on did was assert this line. It was up to the CPU to notice it.
>
> > Can you explain to us how the CPU notices it?
>
> Presumably the same way it sees any other external signal: by sampling the pi
> n.
>
> >>
> >> Looking around there does not seem to be complete agreement these days and
> wi
> >> kipedia (I know, I know...) says:
> >>
> >> "On personal computers,[NB 1] the reset button clears the memory and reboo
> ts
> >> the machine forcibly."
> >>
> >> before, a few paragraphs later, saying"
> >>
> >> "Reset buttons or keypresses on any computer cannot erase the contents of
> sys
> >> tem RAM, which can only be done by powering the computer off and waiting a
> ppr
> >> oximately 10 seconds for residual current in the RAM chips to dissipate."
> >>
> >> So, the button either clears the memory or it doesn't...
>
> > When the reset pin is asserted and the CPU is reset, it starts
> > executing at FFF0:0000h. Motherboard manufactures are cagey about
> > whether the reset jumper clears memory or not. Does it matter if memory
> > is cleared or not? Remember all the registers are reset to zero. If you
> > say it does matter, please explain why it does. Remember again, all
> > registers are cleared.
>
> I did not claim that it mattered or, indeed, that it was or was not cleared,
> merely pointed out that the information available online is not consistent.
>
> Incidentally, according to https://en.wikipedia.org/wiki/Reset_vector the add
> ress held in the reset vector is different for different architectures (as on
> e would expect) and for the x86 series is FFFF0000h:FFF0h.
>
> Also, the reset line should be sampled by the various device controllers to r
> eset themselves.
>
> deryk
> _______________________________________________
> Discuss mailing list
> Discuss at vlug.org
> http://vlug.org/mailman/listinfo/discuss_vlug.org
>
> _______________________________________________
> Discuss mailing list
> Discuss at vlug.org
> http://vlug.org/mailman/listinfo/discuss_vlug.org
>
Intel, AMD, and the other hardware suppliers have excellent
documentation (in PDF format to). Yesterday AMD released 52 pages of
errata for Ryzen (also in PDF format) documenting hangs, panics,
crashes and random resets, all of which have been discussed on the
various Linux and BSD kernel lists. Go ahead and continue to check
wikipedia for a solution. I've wasted enough time here.
--
Cheers,
Cy Schubert <Cy.Schubert at cschubert.com>
FreeBSD UNIX: <cy at FreeBSD.org> Web: http://www.FreeBSD.org
The need of the many outweighs the greed of the few.
More information about the Discuss
mailing list