From: Joakim Tjernlund <joakim.tjernlund@transmode.se>
To: Scott Wood <scottwood@freescale.com>
Cc: Rex Feany <RFeany@mrv.com>,
"linuxppc-dev@ozlabs.org" <linuxppc-dev@ozlabs.org>
Subject: Re: [PATCH 4/8] 8xx: Fixup DAR from buggy dcbX instructions.
Date: Wed, 14 Oct 2009 21:05:02 +0200 [thread overview]
Message-ID: <OF64D89B73.6073827F-ONC125764F.00687BC4-C125764F.0068D4DC@transmode.se> (raw)
In-Reply-To: <20091014172003.GD21215@loki.buserror.net>
Scott Wood <scottwood@freescale.com> wrote on 14/10/2009 19:20:03:
>
> On Sun, Oct 11, 2009 at 06:35:08PM +0200, Joakim Tjernlund wrote:
> > This is an assembler version to fixup DAR not being set
> > by dcbX, icbi instructions. There are two versions, one
> > uses selfmodifing code, the other uses a
> > jump table but is much bigger(default).
> > ---
> > arch/powerpc/kernel/head_8xx.S | 146 +++++++++++++++++++++++++++++++++++++++-
> > 1 files changed, 145 insertions(+), 1 deletions(-)
> >
> > diff --git a/arch/powerpc/kernel/head_8xx.S b/arch/powerpc/kernel/head_8xx.S
> > index 093176c..9839e79 100644
> > --- a/arch/powerpc/kernel/head_8xx.S
> > +++ b/arch/powerpc/kernel/head_8xx.S
> > @@ -494,7 +494,8 @@ DataTLBError:
> >
> > mfspr r10, SPRN_DAR
> > cmpwi cr0, r10, 0x00f0
> > - beq- 2f /* must be a buggy dcbX, icbi insn. */
> > + beq- FixDAR /* must be a buggy dcbX, icbi insn. */
> > +DARFix: /* Return from dcbx instruction bug workaround, r10 holds value of DAR */
>
> Both FixDAR and DARFix? Could we make the labels a little clearer?
Yes, need to come up with better names :)
>
> > +/* This is the procedure to calculate the data EA for buggy dcbx,dcbi instructions
> > + * by decoding the registers used by the dcbx instruction and adding them.
> > + * DAR is set to the calculated address and r10 also holds the EA on exit.
> > + */
>
> How often does this happen? Could we just do it in C code after saving all
> the registers, and avoid the self modifying stuff (or the big switch
> statement equivalent)?
I had some problems with the C-version. I got lots of extra TLB errors for the same address
so I am not confident it will work in the long run.
BTW, you could add a test and printk in do_page_fault on address 0x000000f0.
if that ever hits there is a problem with dcbX fixup.
next prev parent reply other threads:[~2009-10-14 19:05 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-11 16:35 [PATCH 0/8] Fix 8xx MMU/TLB Joakim Tjernlund
2009-10-11 16:35 ` [PATCH 1/8] 8xx: invalidate non present TLBs Joakim Tjernlund
2009-10-11 16:35 ` [PATCH 2/8] 8xx: Update TLB asm so it behaves as linux mm expects Joakim Tjernlund
2009-10-11 16:35 ` [PATCH 3/8] 8xx: Tag DAR with 0x00f0 to catch buggy instructions Joakim Tjernlund
2009-10-11 16:35 ` [PATCH 4/8] 8xx: Fixup DAR from buggy dcbX instructions Joakim Tjernlund
2009-10-11 16:35 ` [PATCH 5/8] 8xx: dcbst sets store bit in DTLB error, workaround Joakim Tjernlund
2009-10-11 16:35 ` [PATCH 6/8] 8xx: Add missing Guarded setting in DTLB Error Joakim Tjernlund
2009-10-11 16:35 ` [PATCH 7/8] 8xx: Restore _PAGE_WRITETHRU Joakim Tjernlund
2009-10-11 16:35 ` [PATCH 8/8] 8xx: start using dcbX instructions in various copy routines Joakim Tjernlund
2009-10-11 21:26 ` [PATCH 7/8] 8xx: Restore _PAGE_WRITETHRU Benjamin Herrenschmidt
2009-10-11 22:21 ` Joakim Tjernlund
2009-10-11 22:45 ` Benjamin Herrenschmidt
2009-10-11 21:25 ` [PATCH 6/8] 8xx: Add missing Guarded setting in DTLB Error Benjamin Herrenschmidt
2009-10-11 22:19 ` Joakim Tjernlund
2009-10-11 22:44 ` Benjamin Herrenschmidt
2009-10-12 5:36 ` Joakim Tjernlund
2009-10-12 5:46 ` Benjamin Herrenschmidt
2009-10-12 6:59 ` Joakim Tjernlund
2009-10-14 17:02 ` [PATCH 5/8] 8xx: dcbst sets store bit in DTLB error, workaround Scott Wood
2009-10-14 17:20 ` [PATCH 4/8] 8xx: Fixup DAR from buggy dcbX instructions Scott Wood
2009-10-14 19:05 ` Joakim Tjernlund [this message]
2009-10-14 19:23 ` Scott Wood
2009-10-14 20:03 ` Joakim Tjernlund
2009-10-14 20:22 ` Scott Wood
2009-10-14 21:10 ` Joakim Tjernlund
2009-10-14 21:14 ` Scott Wood
2009-10-14 21:17 ` Benjamin Herrenschmidt
2009-10-14 21:41 ` Joakim Tjernlund
2009-10-14 21:52 ` Benjamin Herrenschmidt
2009-10-14 22:09 ` Joakim Tjernlund
2009-10-11 21:25 ` [PATCH 2/8] 8xx: Update TLB asm so it behaves as linux mm expects Benjamin Herrenschmidt
2009-10-14 16:57 ` Scott Wood
2009-10-14 16:56 ` [PATCH 1/8] 8xx: invalidate non present TLBs Scott Wood
2009-10-14 17:23 ` [PATCH 0/8] Fix 8xx MMU/TLB Scott Wood
2009-10-14 18:46 ` Joakim Tjernlund
2009-10-15 9:04 Joakim Tjernlund
2009-10-15 9:04 ` [PATCH 1/8] 8xx: invalidate non present TLBs Joakim Tjernlund
2009-10-15 9:04 ` [PATCH 2/8] 8xx: Update TLB asm so it behaves as linux mm expects Joakim Tjernlund
2009-10-15 9:04 ` [PATCH 3/8] 8xx: Tag DAR with 0x00f0 to catch buggy instructions Joakim Tjernlund
2009-10-15 9:04 ` [PATCH 4/8] 8xx: Fixup DAR from buggy dcbX instructions Joakim Tjernlund
2009-11-04 13:38 [PATCH 0/8] 8xx: Misc fixes for buggy insn Joakim Tjernlund
2009-11-04 13:38 ` [PATCH 1/8] 8xx: invalidate non present TLBs Joakim Tjernlund
2009-11-04 13:38 ` [PATCH 2/8] 8xx: Update TLB asm so it behaves as linux mm expects Joakim Tjernlund
2009-11-04 13:38 ` [PATCH 3/8] 8xx: Tag DAR with 0x00f0 to catch buggy instructions Joakim Tjernlund
2009-11-04 13:38 ` [PATCH 4/8] 8xx: Fixup DAR from buggy dcbX instructions Joakim Tjernlund
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=OF64D89B73.6073827F-ONC125764F.00687BC4-C125764F.0068D4DC@transmode.se \
--to=joakim.tjernlund@transmode.se \
--cc=RFeany@mrv.com \
--cc=linuxppc-dev@ozlabs.org \
--cc=scottwood@freescale.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;
as well as URLs for NNTP newsgroup(s).