linux-kernel - Re: [PATCH] dma-direct: Force unencrypted DMA under SME for certain DMA masks

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]

Date:   Wed, 24 Jul 2019 17:34:26 +0000
From:   "Lendacky, Thomas" <Thomas.Lendacky@....com>
To:     Robin Murphy <robin.murphy@....com>,
        "Kirill A. Shutemov" <kirill@...temov.name>
CC:     "iommu@...ts.linux-foundation.org" <iommu@...ts.linux-foundation.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        "x86@...nel.org" <x86@...nel.org>, Christoph Hellwig <hch@....de>,
        Marek Szyprowski <m.szyprowski@...sung.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Ingo Molnar <mingo@...hat.com>, Borislav Petkov <bp@...en8.de>,
        "H . Peter Anvin" <hpa@...or.com>,
        Dave Hansen <dave.hansen@...ux.intel.com>,
        Andy Lutomirski <luto@...nel.org>,
        Peter Zijlstra <peterz@...radead.org>,
        Lianbo Jiang <lijiang@...hat.com>
Subject: Re: [PATCH] dma-direct: Force unencrypted DMA under SME for certain
 DMA masks

On 7/24/19 12:06 PM, Robin Murphy wrote:
> On 24/07/2019 17:42, Lendacky, Thomas wrote:
>> On 7/24/19 10:55 AM, Kirill A. Shutemov wrote:
>>> On Wed, Jul 10, 2019 at 07:01:19PM +0000, Lendacky, Thomas wrote:
>>>> @@ -351,6 +355,32 @@ bool sev_active(void)
>>>>   }
>>>>   EXPORT_SYMBOL(sev_active);
>>>>   +/* Override for DMA direct allocation check -
>>>> ARCH_HAS_FORCE_DMA_UNENCRYPTED */
>>>> +bool force_dma_unencrypted(struct device *dev)
>>>> +{
>>>> +    /*
>>>> +     * For SEV, all DMA must be to unencrypted addresses.
>>>> +     */
>>>> +    if (sev_active())
>>>> +        return true;
>>>> +
>>>> +    /*
>>>> +     * For SME, all DMA must be to unencrypted addresses if the
>>>> +     * device does not support DMA to addresses that include the
>>>> +     * encryption mask.
>>>> +     */
>>>> +    if (sme_active()) {
>>>> +        u64 dma_enc_mask = DMA_BIT_MASK(__ffs64(sme_me_mask));
>>>> +        u64 dma_dev_mask = min_not_zero(dev->coherent_dma_mask,
>>>> +                        dev->bus_dma_mask);
>>>> +
>>>> +        if (dma_dev_mask <= dma_enc_mask)
>>>> +            return true;
>>>
>>> Hm. What is wrong with the dev mask being equal to enc mask? IIUC, it
>>> means that device mask is wide enough to cover encryption bit, doesn't it?
>>
>> Not really...  it's the way DMA_BIT_MASK works vs bit numbering. Let's say
>> that sme_me_mask has bit 47 set. __ffs64 returns 47 and DMA_BIT_MASK(47)
>> will generate a mask without bit 47 set (0x7fffffffffff). So the check
>> will catch anything that does not support at least 48-bit DMA.
> 
> Couldn't that be expressed as just:
> 
>     if (sme_me_mask & dma_dev_mask == sme_me_mask)

Actually !=, but yes, it could have been done like that, I just didn't
think of it.

Thanks,
Tom

> 
> ?
> 
> Robin.