Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 01 Sep 2006 20:02:21 +1000
From:      Peter Chubb <peterc@gelato.unsw.edu.au>
To:        Robert Watson <rwatson@FreeBSD.org>
Cc:        marcel@FreeBSD.org, ia64@FreeBSD.org, Peter Chubb <peterc@gelato.unsw.edu.au>, ppc@FreeBSD.org
Subject:   Re: IA64, PPC system call path audit patches
Message-ID:  <87fyfbc3bm.wl%peterc@quokka.chubb.wattle.id.au>
In-Reply-To: <20060901095744.Q4921@fledge.watson.org>
References:  <20060901080402.W97485@fledge.watson.org> <87irk8at9i.wl%peterc@quokka.chubb.wattle.id.au> <20060901092636.E4921@fledge.watson.org> <87hczsasl5.wl%peterc@quokka.chubb.wattle.id.au> <20060901095744.Q4921@fledge.watson.org>

next in thread | previous in thread | raw e-mail | index | archive | help
>>>>> "Robert" == Robert Watson <rwatson@FreeBSD.org> writes:

Robert> On Fri, 1 Sep 2006, Peter Chubb wrote:

>>>>>>> "Robert" == Robert Watson <rwatson@FreeBSD.org> writes:
>>
Robert> On Fri, 1 Sep 2006, Peter Chubb wrote:
>>
>>>> You've only caught the IA64 slow path system call entries.  The
>>>> fast path is highly optimised assembly language inside
>>>> arch/ia64/kernel/fsys.S, that avoids doing a trap at all.
>>>> 
>>>> With a modern libc, syscall_via_break is only called for a very
>>>> few system calls.
>>
Robert> Hmm.  I'm confused by the above comment -- I'm catching system
Robert> calls on the kernel side of the system call invocation around
Robert> the system call, not on the libc side.  I only see two system
Robert> call demux points in the src/sys/ia64 tree:
>>  Sure.  Original libcs call the system call using break 0x10000,
>> which ends up in the code you saw.  Recent libcs call via a gate
>> page with an epc (execute privileged code) instruction that vectors
>> direcgtly to the syscall implementation.
>> 
Robert> ./ia32/ia32_trap.c: error = (*callp->sy_call)(td, args64);
Robert> ./ia64/trap.c: error = (*callp->sy_call)(td, args);
>>  Take a look in gate.S, symbol _kernel_syscall_via_epc
>> 
>> There's assembly language there that loads the function descriptor
>> from the table and branches to it.  THere are two kinds of system
>> call implementations: fast (implemented directly in assembly
>> language in fsys.S) and slow (the code in fsys.S `bubbles down'
>> into kernel space and then invokes the syscall directly.

Robert> As I read the epc_syscall code, it still passes through the
Robert> kernel syscall() function, which is instrumented in the patch.
Robert> Are you sure that the code does what you describe?  My ia64
Robert> assembly reading skills are weak to non-existent, but the
Robert> final branch in epc_syscall does seem to be to the C language
Robert> syscall path.

AAArrrrgg.  Sorry, I'm looking at Linux source (where the syscall path
is a *lot* more streamlined than the FreeBSD source).

My bad,  I'll go away and hide in a corner.

Peter C



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?87fyfbc3bm.wl%peterc>