Date:   Tue, 31 Jan 2023 09:32:37 +0100
From:   Peter Zijlstra <peterz@infradead.org>
To:     "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc:     Dave Hansen <dave.hansen@intel.com>,
        Borislav Petkov <bp@alien8.de>,
        Andy Lutomirski <luto@kernel.org>,
        Kuppuswamy Sathyanarayanan 
        <sathyanarayanan.kuppuswamy@linux.intel.com>,
        Thomas Gleixner <tglx@linutronix.de>,
        Elena Reshetova <elena.reshetova@intel.com>, x86@kernel.org,
        linux-coco@lists.linux.dev, linux-kernel@vger.kernel.org,
        kernel test robot <lkp@intel.com>
Subject: Re: [PATCH] x86/tdx: Do not corrupt frame-pointer in
 __tdx_hypercall()
Message-ID: <Y9jSJUwNlXSpV3uM@hirez.programming.kicks-ass.net>
References: <20230130135354.27674-1-kirill.shutemov@linux.intel.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20230130135354.27674-1-kirill.shutemov@linux.intel.com>
Precedence: bulk

On Mon, Jan 30, 2023 at 04:53:54PM +0300, Kirill A. Shutemov wrote:
> If compiled with CONFIG_FRAME_POINTER=y, objtool in not happy that
> __tdx_hypercall() messes up RBP.
> 
>   objtool: __tdx_hypercall+0x7f: return with modified stack frame
> 
> Rework the function to store TDX_HCALL_ flags on stack instead of RBP.
> 
> Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> Fixes: c30c4b2555ba ("x86/tdx: Refactor __tdx_hypercall() to allow pass down more arguments")
> Link: https://lore.kernel.org/all/202301290255.buUBs99R-lkp@intel.com
> Reported-by: kernel test robot <lkp@intel.com>
> Cc: Peter Zijlstra <peterz@infradead.org>
> ---
> 
> The patch is against tip/x86/tdx. tip/sched/core removes
> TDX_HCALL_ISSUE_STI. The trird hunk of the patch is not relevant
> after that.

Right, this should work. But it does leave me wondering, should we
perhaps strive to completely remove the flags thing and move to
__tdx_hypercall() and __tdx_hypercall_ret() or something? That is,
simply have two different functions, one with and one without return
data.

It should be trivial to generate that without actual code duplication.