linux-kernel - Re: [PATCH 1/3] x86: allow kernel exception fixup for divide errors (#DE)

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <AANLkTikb1SN6YDsU0dPfTjDkLAZGOX27ccaiEc5OoQHC@mail.gmail.com>
Date:	Tue, 24 Aug 2010 07:22:00 -0400
From:	Brian Gerst <brgerst@...il.com>
To:	Avi Kivity <avi@...hat.com>
Cc:	x86@...nel.org, kvm@...r.kernel.org,
	Marcelo Tosatti <mtosatti@...hat.com>,
	linux-kernel@...r.kernel.org
Subject: Re: [PATCH 1/3] x86: allow kernel exception fixup for divide errors (#DE)

On Tue, Aug 24, 2010 at 7:10 AM, Avi Kivity <avi@...hat.com> wrote:
> KVM wants to emulate the DIV and IDIV instructions by executing them natively;
> this can cause a #DE to be raised.
>
> Allow the exception handling mechanism to process #DE exceptions so KVM can
> catch and process them.
>
> Signed-off-by: Avi Kivity <avi@...hat.com>
> ---
>  arch/x86/kernel/traps.c |   10 +++++++++-
>  1 files changed, 9 insertions(+), 1 deletions(-)
>
> diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
> index 725ef4d..dd313cf 100644
> --- a/arch/x86/kernel/traps.c
> +++ b/arch/x86/kernel/traps.c
> @@ -205,7 +205,8 @@ dotraplinkage void do_##name(struct pt_regs *regs, long error_code) \
>        do_trap(trapnr, signr, str, regs, error_code, &info);           \
>  }
>
> -DO_ERROR_INFO(0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->ip)
> +static DO_ERROR_INFO(0, SIGFPE, "divide error", divide_error_user,
> +                    FPE_INTDIV, regs->ip)
>  DO_ERROR(4, SIGSEGV, "overflow", overflow)
>  DO_ERROR(5, SIGSEGV, "bounds", bounds)
>  DO_ERROR_INFO(6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->ip)
> @@ -217,6 +218,13 @@ DO_ERROR(12, SIGBUS, "stack segment", stack_segment)
>  #endif
>  DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
>
> +dotraplinkage void do_divide_error(struct pt_regs *regs, long error_code)
> +{
> +       if (!user_mode_vm(regs) && fixup_exception(regs))
> +               return;
> +       do_divide_error_user(regs, error_code);
> +}
> +
>  #ifdef CONFIG_X86_64
>  /* Runs on IST stack */
>  dotraplinkage void do_stack_segment(struct pt_regs *regs, long error_code)

Kernel mode exceptions should already be handled by do_trap().  This
is unnecessary.

--
Brian Gerst
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/