Linux PARISC architecture development
 help / color / mirror / Atom feed
From: John David Anglin <dave.anglin@bell.net>
To: Mathieu Desnoyers <mathieu.desnoyers@efficios.com>,
	Helge Deller <deller@gmx.de>
Cc: linux-parisc <linux-parisc@vger.kernel.org>,
	James Bottomley <James.Bottomley@HansenPartnership.com>
Subject: Re: [PATCH] parisc: Fix syscall restarts (v2)
Date: Thu, 24 Dec 2015 11:51:54 -0500	[thread overview]
Message-ID: <567C22AA.1000700@bell.net> (raw)
In-Reply-To: <187718964.290846.1450973235666.JavaMail.zimbra@efficios.com>

On 2015-12-24 11:07 AM, Mathieu Desnoyers wrote:
>>> Is it valid to have unaligned instructions ? Does the architecture
>>> >>allow it, or it's a fumble and we should pr_warn ?
>> >
>> >How can it be unaligned? It's about u32...
> That would be an instruction that is volountarily offset
> from 1, 2, 3 bytes from 4-bytes multiples by the application.
> The only situation where I have seen this is in cases where
> applications are trying to play games with the debugger or
> disassembler and hide what they are doing: they can offset
> the start of a function like this, and therefore all the
> instructions within that function.
>
This is not possible on PA-RISC.  Indeed, user instruction addresses are 
always
offset by three.  There is no way to branch to an instruction that is 
offset.

The least significant two bits of an instruction address contain a 
priority level.
User code on linux and hpux executes at level 3.  The only way a user 
can change
privilege level is with a "gate" instruction (a special branch). Whether 
this is
permitted or not depends on page table permissions that the user can't 
change.
In practice, a level change is only allowed on the gateway page.

Dave

-- 
John David Anglin  dave.anglin@bell.net


      reply	other threads:[~2015-12-24 16:51 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-12-18 23:30 [PATCH] parisc: Fix syscall restarts Helge Deller
2015-12-20 13:59 ` Mathieu Desnoyers
2015-12-20 14:09   ` Mathieu Desnoyers
2015-12-20 15:49     ` Helge Deller
2015-12-20 16:50       ` James Bottomley
2015-12-20 20:35         ` Helge Deller
2015-12-21  8:03           ` James Bottomley
2015-12-21 14:39             ` Mathieu Desnoyers
2015-12-20 18:31       ` John David Anglin
2015-12-20 19:32         ` Helge Deller
2015-12-20 19:46           ` John David Anglin
2015-12-20 20:06             ` Helge Deller
2015-12-20 23:57             ` John David Anglin
2015-12-21 14:42         ` Mathieu Desnoyers
2015-12-21 15:12           ` John David Anglin
2015-12-20 19:39 ` John David Anglin
2015-12-20 19:48   ` Helge Deller
2015-12-20 20:01     ` John David Anglin
2015-12-20 20:18       ` Helge Deller
2015-12-20 20:45         ` John David Anglin
2015-12-20 20:14 ` John David Anglin
2015-12-20 20:19   ` Helge Deller
2015-12-20 20:21     ` Helge Deller
2015-12-20 20:53       ` John David Anglin
2015-12-21  9:19 ` [PATCH] parisc: Fix syscall restarts (v2) Helge Deller
2015-12-21 13:11   ` John David Anglin
2015-12-21 20:27   ` Mathieu Desnoyers
2015-12-21 20:54     ` Helge Deller
2015-12-24 16:07       ` Mathieu Desnoyers
2015-12-24 16:51         ` John David Anglin [this message]

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=567C22AA.1000700@bell.net \
    --to=dave.anglin@bell.net \
    --cc=James.Bottomley@HansenPartnership.com \
    --cc=deller@gmx.de \
    --cc=linux-parisc@vger.kernel.org \
    --cc=mathieu.desnoyers@efficios.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