From: Craig Hada <chada@cup.hp.com>
To: hppa-linux@thepuffingroup.com
Subject: Re: [hppa-linux] Qn on PA-RISC Page tables and TLB
Date: Mon, 29 Mar 1999 08:01:20 -0800 [thread overview]
Message-ID: <199903291601.IAA09728@psalpha2.cup.hp.com> (raw)
In-Reply-To: Your message of "Sun, 28 Mar 1999 12:37:05 PST." <Pine.LNX.3.96.990328122620.118A-100000@quark.ix.netcom.com>
>
> On Thu, 25 Mar 1999, Craig Hada wrote:
> > >
> >
> > In the absense of a harware TLB walker, the page directory format can be
> > structured by software to fit its needs. However, the TLB miss handler
> > must reformat the bits to fit the format of the TLB insert instructions.
> > To get maximum performace from the system, the TLB miss handlers must be
> > made very efficient. The format of the page directory entry in the PA-RISC
> > Arch 1.1 book minimizes the work of the TLB miss handlers by aligning the
> > fields to match the format of the TLB insert instructions.
> > >
> > >
> Is it necessary that the virtual to physical mappings in TLB be inserted
> only in a tlb miss fault. Can I insert entries in the anticipation that
> there will be a fault soon, for example when I first enable
> virtual mode ? If not, there is a potential chicken and egg problem
> because I will get a TLB miss fault as soon as I turn virtual mode on and
> TLB miss fault handler has to run in virtual mode with translations
> on.....and there aint any tranlations yet inserted ...you get the pitcure
>
> -pkd
>
>
Under HP-UX, the TLB miss handler runs in real mode (data and code translation
disabled) so there is no need to have a virtual translation for the code
representing the TLB miss handler or the page directory (both are equivalently
mapped). If a translation is not found in the page directory, then you enter
virtual mode to handle the page fault.
You can speculatively insert entries into the TLB but you also may be
replacing entries that are soon to be used. Since the TLB replacement
algorithm is not documented, you are probably better off just inserting
the missed upon translation.
-Craig
>
> -------------------------------------------------------------------------
> To unsubscribe: send e-mail to hppa-linux-request@thepuffingroup.com with
> `unsubscribe' as the subject.
next prev parent reply other threads:[~1999-03-29 16:07 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
1999-03-25 7:10 [hppa-linux] Qn on PA-RISC Page tables and TLB Kumar
1999-03-25 17:18 ` Grant Grundler
1999-03-25 20:31 ` Craig Hada
1999-03-28 20:37 ` Kumar
1999-03-28 22:52 ` Michael Shalayeff
1999-03-29 16:01 ` Craig Hada [this message]
-- strict thread matches above, loose matches on Subject: below --
1999-03-25 15:59 Bjorn Helgaas
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=199903291601.IAA09728@psalpha2.cup.hp.com \
--to=chada@cup.hp.com \
--cc=hppa-linux@thepuffingroup.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox