From: "Radim Krčmář" <rkrcmar-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>
To: Tom Lendacky <thomas.lendacky-5C7GfCeVMHo@public.gmane.org>
Cc: linux-arch-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-efi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
linux-doc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
x86-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org,
linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org,
kasan-dev-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org,
linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org,
iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org,
Rik van Riel <riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
Arnd Bergmann <arnd-r2nGTMty4D4@public.gmane.org>,
Jonathan Corbet <corbet-T1hC0tSOHrs@public.gmane.org>,
Matt Fleming
<matt-mF/unelCI9GS6iBeEJttW/XRex20P6io@public.gmane.org>,
Joerg Roedel <joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>,
Konrad Rzeszutek Wilk
<konrad.wilk-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org>,
Paolo Bonzini <pbonzini-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
Larry Woodman <lwoodman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
Ingo Molnar <mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org>,
Borislav Petkov <bp-Gina5bIWoIWzQB+pC5nmwQ@public.gmane.org>,
Andy Lutomirski <luto-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>,
"H. Peter Anvin" <hpa-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org>,
Andrey Ryabinin
<aryabinin-5HdwGun5lf+gSpxsJD1C4w@public.gmane.org>,
Alexander Potapenko
<glider-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org>,
Thomas Gleixner <tgl>
Subject: Re: [RFC PATCH v3 13/20] x86: DMA support for memory encryption
Date: Tue, 15 Nov 2016 19:17:36 +0100 [thread overview]
Message-ID: <20161115181736.GA14060@potion> (raw)
In-Reply-To: <d5ebd13d-1278-8714-3f03-8ee7f04a2b38-5C7GfCeVMHo@public.gmane.org>
2016-11-15 11:02-0600, Tom Lendacky:
> On 11/15/2016 8:39 AM, Radim Krčmář wrote:
>> 2016-11-09 18:37-0600, Tom Lendacky:
>>> Since DMA addresses will effectively look like 48-bit addresses when the
>>> memory encryption mask is set, SWIOTLB is needed if the DMA mask of the
>>> device performing the DMA does not support 48-bits. SWIOTLB will be
>>> initialized to create un-encrypted bounce buffers for use by these devices.
>>>
>>> Signed-off-by: Tom Lendacky <thomas.lendacky-5C7GfCeVMHo@public.gmane.org>
>>> ---
>>> diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
>>> @@ -64,13 +66,15 @@ static struct dma_map_ops swiotlb_dma_ops = {
>>> * pci_swiotlb_detect_override - set swiotlb to 1 if necessary
>>> *
>>> * This returns non-zero if we are forced to use swiotlb (by the boot
>>> - * option).
>>> + * option). If memory encryption is enabled then swiotlb will be set
>>> + * to 1 so that bounce buffers are allocated and used for devices that
>>> + * do not support the addressing range required for the encryption mask.
>>> */
>>> int __init pci_swiotlb_detect_override(void)
>>> {
>>> int use_swiotlb = swiotlb | swiotlb_force;
>>>
>>> - if (swiotlb_force)
>>> + if (swiotlb_force || sme_me_mask)
>>> swiotlb = 1;
>>>
>>> return use_swiotlb;
>>
>> We want to return 1 even if only sme_me_mask is 1, because the return
>> value is used for detection. The following would be less obscure, IMO:
>>
>> if (swiotlb_force || sme_me_mask)
>> swiotlb = 1;
>>
>> return swiotlb;
>
> If we do that then all DMA would go through the swiotlb bounce buffers.
No, that is decided for example in swiotlb_map_page() and we need to
call pci_swiotlb_init() to register that function.
> By setting swiotlb to 1 we indicate that the bounce buffers will be
> needed for those devices that can't support the addressing range when
> the encryption bit is set (48 bit DMA). But if the device can support
> the addressing range we won't use the bounce buffers.
If we return 0 here, then pci_swiotlb_init() will not be called =>
dma_ops won't be set to swiotlb_dma_ops => we won't use bounce buffers.
>> We setup encrypted swiotlb and then decrypt it, but sometimes set it up
>> decrypted (late_alloc) ... why isn't the swiotlb set up decrypted
>> directly?
>
> When swiotlb is allocated in swiotlb_init(), it is too early to make
> use of the api to the change the page attributes. Because of this,
> the callback to make those changes is needed.
Thanks. (I don't know page table setup enough to see a lesser evil. :])
>>> @@ -541,7 +583,7 @@ static phys_addr_t
>>> map_single(struct device *hwdev, phys_addr_t phys, size_t size,
>>> enum dma_data_direction dir)
>>> {
>>> - dma_addr_t start_dma_addr = phys_to_dma(hwdev, io_tlb_start);
>>> + dma_addr_t start_dma_addr = swiotlb_phys_to_dma(hwdev, io_tlb_start);
>>
>> We have decrypted io_tlb_start before, so shouldn't its physical address
>> be saved without the sme bit? (Which changes a lot ...)
>
> I'm not sure what you mean here, can you elaborate a bit more?
The C-bit (sme bit) is a part of the physical address.
If we know that a certain physical page should be accessed as
unencrypted (the bounce buffer) then the C-bit is 0.
I'm wondering why we save the physical address with the C-bit set when
we know that it can't be accessed that way (because we remove it every
time).
The naming is a bit confusing, because physical addresses are actually
virtualized by SME -- maybe we should be calling them SME addresses?
next prev parent reply other threads:[~2016-11-15 18:17 UTC|newest]
Thread overview: 78+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-11-10 0:34 [RFC PATCH v3 00/20] x86: Secure Memory Encryption (AMD) Tom Lendacky
2016-11-10 0:34 ` [RFC PATCH v3 01/20] x86: Documentation for AMD Secure Memory Encryption (SME) Tom Lendacky
2016-11-10 10:51 ` Borislav Petkov
2016-11-14 17:15 ` Tom Lendacky
2016-11-10 0:34 ` [RFC PATCH v3 02/20] x86: Set the write-protect cache mode for full PAT support Tom Lendacky
2016-11-10 13:14 ` Borislav Petkov
2016-11-11 1:26 ` Kani, Toshimitsu
[not found] ` <1478827480.20881.142.camel-ZPxbGqLxI0U@public.gmane.org>
2016-11-14 16:51 ` Tom Lendacky
2016-11-10 0:35 ` [RFC PATCH v3 07/20] x86: Provide general kernel support for memory encryption Tom Lendacky
2016-11-10 0:36 ` [RFC PATCH v3 08/20] x86: Add support for early encryption/decryption of memory Tom Lendacky
[not found] ` <20161110003610.3280.22043.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-11-16 10:46 ` Borislav Petkov
[not found] ` <20161116104656.qz5wp33zzyja373r-fF5Pk5pvG8Y@public.gmane.org>
2016-11-16 19:22 ` Tom Lendacky
2016-11-10 0:36 ` [RFC PATCH v3 09/20] x86: Insure that boot memory areas are mapped properly Tom Lendacky
[not found] ` <20161110003620.3280.20613.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-11-17 12:20 ` Borislav Petkov
2016-11-19 18:12 ` Tom Lendacky
2016-11-10 0:37 ` [RFC PATCH v3 13/20] x86: DMA support for memory encryption Tom Lendacky
2016-11-15 14:39 ` Radim Krčmář
2016-11-15 17:02 ` Tom Lendacky
[not found] ` <d5ebd13d-1278-8714-3f03-8ee7f04a2b38-5C7GfCeVMHo@public.gmane.org>
2016-11-15 18:17 ` Radim Krčmář [this message]
2016-11-15 20:33 ` Tom Lendacky
2016-11-15 15:16 ` Michael S. Tsirkin
[not found] ` <20161115171443-mutt-send-email-mst-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org>
2016-11-15 18:29 ` Tom Lendacky
2016-11-15 19:16 ` Michael S. Tsirkin
2016-11-22 11:38 ` Borislav Petkov
2016-11-22 15:22 ` Michael S. Tsirkin
2016-11-22 15:41 ` Borislav Petkov
[not found] ` <20161122154137.z5vp3xcl5cpesuiz-fF5Pk5pvG8Y@public.gmane.org>
2016-11-22 20:41 ` Michael S. Tsirkin
2016-11-10 0:37 ` [RFC PATCH v3 15/20] x86: Check for memory encryption on the APs Tom Lendacky
2016-11-22 19:25 ` Borislav Petkov
[not found] ` <20161122192526.vg63jjhwsbjwex7i-fF5Pk5pvG8Y@public.gmane.org>
2016-11-29 18:00 ` Tom Lendacky
[not found] ` <20161110003426.3280.2999.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-11-10 0:34 ` [RFC PATCH v3 03/20] x86: Add the Secure Memory Encryption cpu feature Tom Lendacky
2016-11-11 11:53 ` Borislav Petkov
2016-11-10 0:35 ` [RFC PATCH v3 04/20] x86: Handle reduction in physical address size with SME Tom Lendacky
[not found] ` <20161110003513.3280.12104.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-11-15 12:10 ` Joerg Roedel
2016-11-15 12:14 ` Borislav Petkov
2016-11-15 14:40 ` Tom Lendacky
[not found] ` <a4cc5b07-89e1-aaa0-1977-1de95883ba62-5C7GfCeVMHo@public.gmane.org>
2016-11-15 15:33 ` Borislav Petkov
[not found] ` <20161115153338.a2cxmatnpqcgiaiy-fF5Pk5pvG8Y@public.gmane.org>
2016-11-15 16:06 ` Tom Lendacky
2016-11-15 16:33 ` Borislav Petkov
2016-11-15 17:08 ` Tom Lendacky
[not found] ` <20161115121456.f4slpk4i2jl3e2ke-fF5Pk5pvG8Y@public.gmane.org>
2016-11-15 21:22 ` Tom Lendacky
2016-11-15 21:33 ` Borislav Petkov
2016-11-15 22:01 ` Tom Lendacky
[not found] ` <20161115121035.GD24857-zLv9SwRftAIdnm+yROfE0A@public.gmane.org>
2016-11-15 14:32 ` Tom Lendacky
2016-11-10 0:35 ` [RFC PATCH v3 05/20] x86: Add Secure Memory Encryption (SME) support Tom Lendacky
2016-11-10 0:35 ` [RFC PATCH v3 06/20] x86: Add support to enable SME during early boot processing Tom Lendacky
[not found] ` <20161110003543.3280.99623.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-11-14 17:29 ` Borislav Petkov
2016-11-14 18:18 ` Tom Lendacky
2016-11-14 20:01 ` Borislav Petkov
2016-11-10 0:36 ` [RFC PATCH v3 10/20] Add support to access boot related data in the clear Tom Lendacky
2016-11-11 16:17 ` Kani, Toshimitsu
[not found] ` <1478880929.20881.148.camel-ZPxbGqLxI0U@public.gmane.org>
2016-11-14 16:24 ` Tom Lendacky
2016-11-17 15:55 ` Borislav Petkov
[not found] ` <20161117155543.vg3domfqm3bhp4f7-fF5Pk5pvG8Y@public.gmane.org>
2016-11-19 18:33 ` Tom Lendacky
2016-11-20 23:04 ` Borislav Petkov
[not found] ` <20161110003631.3280.73292.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-12-07 13:19 ` Matt Fleming
2016-12-09 14:26 ` Tom Lendacky
2016-11-10 0:36 ` [RFC PATCH v3 11/20] x86: Add support for changing memory encryption attribute Tom Lendacky
[not found] ` <20161110003655.3280.57333.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-11-17 17:39 ` Borislav Petkov
2016-11-19 18:48 ` Tom Lendacky
[not found] ` <6f1a16e4-5a84-20c0-4bd3-3be5ed933800-5C7GfCeVMHo@public.gmane.org>
2016-11-21 8:27 ` Borislav Petkov
2016-11-10 0:37 ` [RFC PATCH v3 12/20] x86: Decrypt trampoline area if memory encryption is active Tom Lendacky
[not found] ` <20161110003708.3280.29934.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-11-17 18:09 ` Borislav Petkov
2016-11-19 18:50 ` Tom Lendacky
2016-11-10 0:37 ` [RFC PATCH v3 14/20] iommu/amd: Disable AMD IOMMU " Tom Lendacky
2016-11-14 16:32 ` Joerg Roedel
2016-11-14 16:48 ` Tom Lendacky
2016-11-10 0:37 ` [RFC PATCH v3 16/20] x86: Do not specify encrypted memory for video mappings Tom Lendacky
2016-11-10 0:38 ` [RFC PATCH v3 17/20] x86/kvm: Enable Secure Memory Encryption of nested page tables Tom Lendacky
2016-11-10 0:38 ` [RFC PATCH v3 18/20] x86: Access the setup data through debugfs un-encrypted Tom Lendacky
2016-11-10 0:38 ` [RFC PATCH v3 19/20] x86: Add support to make use of Secure Memory Encryption Tom Lendacky
[not found] ` <20161110003826.3280.5546.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-11-24 12:50 ` Borislav Petkov
2016-11-29 18:40 ` Tom Lendacky
2016-11-10 0:38 ` [RFC PATCH v3 20/20] " Tom Lendacky
[not found] ` <20161110003838.3280.23327.stgit-qCXWGYdRb2BnqfbPTmsdiZQ+2ll4COg0XqFh9Ls21Oc@public.gmane.org>
2016-11-22 18:58 ` Borislav Petkov
2016-11-26 20:47 ` Borislav Petkov
2016-11-29 18:48 ` Tom Lendacky
2016-11-29 19:56 ` Borislav Petkov
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=20161115181736.GA14060@potion \
--to=rkrcmar-h+wxahxf7alqt0dzr+alfa@public.gmane.org \
--cc=arnd-r2nGTMty4D4@public.gmane.org \
--cc=aryabinin-5HdwGun5lf+gSpxsJD1C4w@public.gmane.org \
--cc=bp-Gina5bIWoIWzQB+pC5nmwQ@public.gmane.org \
--cc=corbet-T1hC0tSOHrs@public.gmane.org \
--cc=glider-hpIqsD4AKlfQT0dZR+AlfA@public.gmane.org \
--cc=hpa-YMNOUZJC4hwAvxtiuMwx3w@public.gmane.org \
--cc=iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA@public.gmane.org \
--cc=joro-zLv9SwRftAIdnm+yROfE0A@public.gmane.org \
--cc=kasan-dev-/JYPxA39Uh5TLH3MbocFFw@public.gmane.org \
--cc=konrad.wilk-QHcLZuEGTsvQT0dZR+AlfA@public.gmane.org \
--cc=kvm-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-arch-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-doc-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-efi-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org \
--cc=linux-mm-Bw31MaZKKs3YtjvyW6yDsg@public.gmane.org \
--cc=luto-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
--cc=lwoodman-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=matt-mF/unelCI9GS6iBeEJttW/XRex20P6io@public.gmane.org \
--cc=mingo-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=pbonzini-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=riel-H+wXaHxf7aLQT0dZR+AlfA@public.gmane.org \
--cc=thomas.lendacky-5C7GfCeVMHo@public.gmane.org \
--cc=x86-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org \
/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;
as well as URLs for NNTP newsgroup(s).