From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Joakim Tjernlund <joakim.tjernlund@transmode.se>
Cc: Scott Wood <scottwood@freescale.com>,
"linuxppc-dev@ozlabs.org" <linuxppc-dev@ozlabs.org>,
Rex Feany <RFeany@mrv.com>
Subject: Re: [PATCH 0/6] PowerPc 8xx TLB/MMU fixes
Date: Tue, 06 Oct 2009 19:32:49 +1100 [thread overview]
Message-ID: <1254817969.6035.4.camel@pasglop> (raw)
In-Reply-To: <OFAEE03401.0C51FEE2-ONC1257647.002C29FE-C1257647.002C917A@transmode.se>
> > No, use get_user() not __get_user() or if you use the later, also use
> > access_ok(), and test the result in case it errors (if it does, you
> > probably want to just goto bad access and SEGV).
>
> OK, lets see what this gives us:
Hrm... did you change anything ? :-)
Ben.
> diff --git a/arch/powerpc/mm/fault.c b/arch/powerpc/mm/fault.c
> index c33c6de..1bf91d3 100644
> --- a/arch/powerpc/mm/fault.c
> +++ b/arch/powerpc/mm/fault.c
> @@ -153,7 +153,8 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
> #ifdef DEBUG_DCBX
> const char *istr = NULL;
>
> - insn = *((unsigned long *)regs->nip);
> + insn = 0;
> + __get_user(insn, (unsigned long __user *)regs->nip);
> if (((insn >> (31-5)) & 0x3f) == 31) {
> if (((insn >> 1) & 0x3ff) == 1014) /* dcbz ? 0x3f6 */
> istr = "dcbz";
> @@ -171,27 +172,32 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
> dar = regs->gpr[rb];
> if (ra)
> dar += regs->gpr[ra];
> - if (dar != address && address != 0x00f0 && trap == 0x300)
> + if (dar != address && trap == 0x300)
> printk(KERN_CRIT "%s: address:%lx, dar:%lx!\n", istr, address, dar);
> if (!strcmp(istr, "dcbst") && is_write) {
> printk(KERN_CRIT "dcbst R%ld,R%ld = %lx as a store, fixing!\n",
> ra, rb, dar);
> is_write = 0;
> }
> -
> +#if 0
> if (trap == 0x300 && address != dar) {
> __asm__ ("mtdar %0" : : "r" (dar));
> return 0;
> }
> +#endif
> }
> }
> #endif
> if (address == 0x00f0 && trap == 0x300) {
> - pte_t *ptep;
> + //pte_t *ptep;
>
> /* This is from a dcbX or icbi insn gone bad, these
> * insn do not set DAR so we have to do it here instead */
> - insn = *((unsigned long *)regs->nip);
> + if (get_user(insn, (unsigned long __user *)regs->nip)) {
> + printk(KERN_CRIT "get_user failed, NIP:%lx\n",
> + regs->nip);
> + goto bad_area_nosemaphore;
> + }
>
> ra = (insn >> (31-15)) & 0x1f; /* Reg RA */
> rb = (insn >> (31-20)) & 0x1f; /* Reg RB */
> @@ -206,7 +212,7 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
> trap, address, dar, error_code, istr);
> #endif
> address = dar;
> -#if 1
> +#if 0
> if (is_write && get_pteptr(mm, dar, &ptep, NULL)) {
> pte_t my_pte = *ptep;
>
> @@ -216,7 +222,7 @@ int __kprobes do_page_fault(struct pt_regs *regs, unsigned long address,
> }
> }
> #else
> - return 0;
> + //return 0;
> #endif
> }
> }
next prev parent reply other threads:[~2009-10-06 8:32 UTC|newest]
Thread overview: 49+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-05 12:16 [PATCH 0/6] PowerPc 8xx TLB/MMU fixes Joakim Tjernlund
2009-10-05 12:16 ` [PATCH 1/6] 8xx: DTLB Error must check for more errors Joakim Tjernlund
2009-10-05 12:16 ` [PATCH 2/6] 8xx, fault: Add some debug code to do_page_fault() Joakim Tjernlund
2009-10-05 12:16 ` [PATCH 3/6] 8xx: get rid of _PAGE_HWWRITE dependency in MMU Joakim Tjernlund
2009-10-05 12:16 ` [PATCH 4/6] 8xx: Tag DAR with 0x00f0 to catch buggy instructions Joakim Tjernlund
2009-10-05 12:16 ` [PATCH 5/6] 8xx: Fixup DAR from buggy dcbX instructions Joakim Tjernlund
2009-10-05 12:16 ` [PATCH 6/6] 8xx: start using dcbX instructions in various copy routines Joakim Tjernlund
2009-10-05 20:17 ` [PATCH 3/6] 8xx: get rid of _PAGE_HWWRITE dependency in MMU Benjamin Herrenschmidt
2009-10-05 21:25 ` Joakim Tjernlund
2009-10-05 21:37 ` Benjamin Herrenschmidt
2009-10-05 22:00 ` Joakim Tjernlund
2009-10-05 22:09 ` Benjamin Herrenschmidt
2009-10-05 22:55 ` Joakim Tjernlund
2009-10-05 23:15 ` Benjamin Herrenschmidt
2009-10-05 23:35 ` Joakim Tjernlund
2009-10-06 0:34 ` Benjamin Herrenschmidt
2009-10-06 6:15 ` Joakim Tjernlund
2009-10-06 6:45 ` Benjamin Herrenschmidt
2009-10-06 7:54 ` Joakim Tjernlund
2009-10-06 15:40 ` Joakim Tjernlund
2009-10-06 17:28 ` Joakim Tjernlund
2009-10-06 22:05 ` Joakim Tjernlund
2009-10-06 23:25 ` Benjamin Herrenschmidt
2009-10-07 1:07 ` Benjamin Herrenschmidt
2009-10-07 7:47 ` Joakim Tjernlund
2009-10-05 18:12 ` [PATCH 0/6] PowerPc 8xx TLB/MMU fixes Scott Wood
2009-10-05 18:27 ` Joakim Tjernlund
2009-10-05 20:09 ` Scott Wood
2009-10-05 21:04 ` Joakim Tjernlund
2009-10-05 21:31 ` Benjamin Herrenschmidt
2009-10-05 21:41 ` Joakim Tjernlund
2009-10-05 21:46 ` Scott Wood
2009-10-05 21:31 ` Scott Wood
2009-10-05 22:04 ` Rex Feany
2009-10-05 22:31 ` Joakim Tjernlund
2009-10-05 22:37 ` Benjamin Herrenschmidt
2009-10-05 22:58 ` Joakim Tjernlund
2009-10-05 23:49 ` Joakim Tjernlund
2009-10-06 1:52 ` Benjamin Herrenschmidt
2009-10-06 8:06 ` Joakim Tjernlund
2009-10-06 8:32 ` Benjamin Herrenschmidt [this message]
2009-10-06 10:58 ` Joakim Tjernlund
2009-10-06 11:06 ` Benjamin Herrenschmidt
2009-10-06 11:39 ` Joakim Tjernlund
2009-10-06 13:18 ` Joakim Tjernlund
2009-10-05 22:42 ` Rex Feany
2009-10-05 23:00 ` Joakim Tjernlund
2009-10-06 6:25 ` Joakim Tjernlund
2009-10-06 6:44 ` Benjamin Herrenschmidt
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=1254817969.6035.4.camel@pasglop \
--to=benh@kernel.crashing.org \
--cc=RFeany@mrv.com \
--cc=joakim.tjernlund@transmode.se \
--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).