From: Dave Hansen <dave.hansen@intel.com>
To: Kuppuswamy Sathyanarayanan
<sathyanarayanan.kuppuswamy@linux.intel.com>,
Peter Zijlstra <peterz@infradead.org>,
Andy Lutomirski <luto@kernel.org>
Cc: Andi Kleen <ak@linux.intel.com>,
Kirill Shutemov <kirill.shutemov@linux.intel.com>,
Kuppuswamy Sathyanarayanan <knsathya@kernel.org>,
Dan Williams <dan.j.williams@intel.com>,
Raj Ashok <ashok.raj@intel.com>,
Sean Christopherson <seanjc@google.com>,
linux-kernel@vger.kernel.org
Subject: Re: [RFC v1 26/26] x86/kvm: Use bounce buffers for TD guest
Date: Thu, 1 Apr 2021 14:17:54 -0700 [thread overview]
Message-ID: <1741d632-0ed0-6a5b-efae-4fdb2689e20f@intel.com> (raw)
In-Reply-To: <72ebe1575ff79bce7be38b8093aeb41032562a22.1612563142.git.sathyanarayanan.kuppuswamy@linux.intel.com>
On 2/5/21 3:38 PM, Kuppuswamy Sathyanarayanan wrote:
> From: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
>
> TDX doesn't allow to perform DMA access to guest private memory.
> In order for DMA to work properly in TD guest, user SWIOTLB bounce
> buffers.
>
> Move AMD SEV initialization into common code and adopt for TDX.
This would be best if it can draw a parallel between TDX and SEV.
> arch/x86/kernel/pci-swiotlb.c | 2 +-
> arch/x86/kernel/tdx.c | 3 +++
> arch/x86/mm/mem_encrypt.c | 44 -------------------------------
> arch/x86/mm/mem_encrypt_common.c | 45 ++++++++++++++++++++++++++++++++
> 4 files changed, 49 insertions(+), 45 deletions(-)
>
> diff --git a/arch/x86/kernel/pci-swiotlb.c b/arch/x86/kernel/pci-swiotlb.c
> index c2cfa5e7c152..020e13749758 100644
> --- a/arch/x86/kernel/pci-swiotlb.c
> +++ b/arch/x86/kernel/pci-swiotlb.c
> @@ -49,7 +49,7 @@ int __init pci_swiotlb_detect_4gb(void)
> * buffers are allocated and used for devices that do not support
> * the addressing range required for the encryption mask.
> */
> - if (sme_active())
> + if (sme_active() || is_tdx_guest())
> swiotlb = 1;
>
> return swiotlb;
> diff --git a/arch/x86/kernel/tdx.c b/arch/x86/kernel/tdx.c
> index f51a19168adc..ccb9401bd706 100644
> --- a/arch/x86/kernel/tdx.c
> +++ b/arch/x86/kernel/tdx.c
> @@ -9,6 +9,7 @@
> #include <asm/vmx.h>
> #include <asm/insn.h>
> #include <linux/sched/signal.h> /* force_sig_fault() */
> +#include <linux/swiotlb.h>
>
> #ifdef CONFIG_KVM_GUEST
> #include "tdx-kvm.c"
> @@ -472,6 +473,8 @@ void __init tdx_early_init(void)
>
> legacy_pic = &null_legacy_pic;
>
> + swiotlb_force = SWIOTLB_FORCE;
Dumb question time. But, what is the difference between
swiotlb = 1;
and
swiotlb_force = SWIOTLB_FORCE;
It would be nice of the patch to enable me to be a lazy reviewer.
> cpuhp_setup_state(CPUHP_AP_ONLINE_DYN, "tdx:cpu_hotplug",
> NULL, tdx_cpu_offline_prepare);
>
> diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c
> index 11a6a7b3af7e..7fbbb2f3d426 100644
> --- a/arch/x86/mm/mem_encrypt.c
> +++ b/arch/x86/mm/mem_encrypt.c
Should we be renaming this to amd_mem_encrypt.c or something?
...
> - */
> - if (sev_active())
> - static_branch_enable(&sev_enable_key);
> -
> - print_mem_encrypt_feature_info();
> -}
> -
> diff --git a/arch/x86/mm/mem_encrypt_common.c b/arch/x86/mm/mem_encrypt_common.c
> index b6d93b0c5dcf..6f3d90d4d68e 100644
> --- a/arch/x86/mm/mem_encrypt_common.c
> +++ b/arch/x86/mm/mem_encrypt_common.c
> @@ -10,6 +10,7 @@
> #include <linux/mm.h>
> #include <linux/mem_encrypt.h>
> #include <linux/dma-mapping.h>
> +#include <linux/swiotlb.h>
>
> /* Override for DMA direct allocation check - ARCH_HAS_FORCE_DMA_UNENCRYPTED */
> bool force_dma_unencrypted(struct device *dev)
> @@ -36,3 +37,47 @@ bool force_dma_unencrypted(struct device *dev)
>
> return false;
> }
> +
> +static void print_mem_encrypt_feature_info(void)
> +{
This function is now named wrong IMNHO. If it's about AMD only, it
needs AMD in the name.
> + pr_info("AMD Memory Encryption Features active:");
> +
> + /* Secure Memory Encryption */
> + if (sme_active()) {
> + /*
> + * SME is mutually exclusive with any of the SEV
> + * features below.
> + */
> + pr_cont(" SME\n");
> + return;
> + }
> +
> + /* Secure Encrypted Virtualization */
> + if (sev_active())
> + pr_cont(" SEV");
> +
> + /* Encrypted Register State */
> + if (sev_es_active())
> + pr_cont(" SEV-ES");
> +
> + pr_cont("\n");
> +}
I'm really tempted to say this needs to be off in arch/x86/kernel/cpu/amd.c
> +/* Architecture __weak replacement functions */
> +void __init mem_encrypt_init(void)
> +{
> + if (!sme_me_mask && !is_tdx_guest())
> + return;
The direct check of sme_me_mask looks odd now. What does this *MEAN*?
Are we looking to jump out of here if no memory encryption is enabled?
I'd much rather this look more like:
if (!x86_memory_encryption())
return;
> + /* Call into SWIOTLB to update the SWIOTLB DMA buffers */
> + swiotlb_update_mem_attributes();
> + /*
> + * With SEV, we need to unroll the rep string I/O instructions.
> + */
> + if (sev_active())
> + static_branch_enable(&sev_enable_key);
> +
> + if (!is_tdx_guest())
> + print_mem_encrypt_feature_info();
> +}
next prev parent reply other threads:[~2021-04-01 21:17 UTC|newest]
Thread overview: 161+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-02-06 3:02 Test Email sathyanarayanan.kuppuswamy
2021-02-05 23:38 ` [RFC v1 00/26] Add TDX Guest Support Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 01/26] x86/paravirt: Introduce CONFIG_PARAVIRT_XL Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 02/26] x86/cpufeatures: Add TDX Guest CPU feature Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 03/26] x86/cpufeatures: Add is_tdx_guest() interface Kuppuswamy Sathyanarayanan
2021-04-01 21:08 ` Dave Hansen
2021-04-01 21:15 ` Kuppuswamy, Sathyanarayanan
2021-04-01 21:19 ` Dave Hansen
2021-04-01 22:25 ` Kuppuswamy, Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 04/26] x86/tdx: Get TD execution environment information via TDINFO Kuppuswamy Sathyanarayanan
2021-02-08 10:00 ` Peter Zijlstra
2021-02-08 19:10 ` Kuppuswamy, Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 05/26] x86/traps: Add #VE support for TDX guest Kuppuswamy Sathyanarayanan
2021-02-08 10:20 ` Peter Zijlstra
2021-02-08 16:23 ` Andi Kleen
2021-02-08 16:33 ` Peter Zijlstra
2021-02-08 16:46 ` Sean Christopherson
2021-02-08 16:59 ` Peter Zijlstra
2021-02-08 19:05 ` Kuppuswamy, Sathyanarayanan
2021-02-08 16:46 ` Andi Kleen
2021-02-12 19:20 ` Dave Hansen
2021-02-12 19:47 ` Andy Lutomirski
2021-02-12 20:06 ` Sean Christopherson
2021-02-12 20:17 ` Dave Hansen
2021-02-12 20:37 ` Sean Christopherson
2021-02-12 20:46 ` Dave Hansen
2021-02-12 20:54 ` Sean Christopherson
2021-02-12 21:06 ` Dave Hansen
2021-02-12 21:37 ` Sean Christopherson
2021-02-12 21:47 ` Andy Lutomirski
2021-02-12 21:48 ` Dave Hansen
2021-02-14 19:33 ` Andi Kleen
2021-02-14 19:54 ` Andy Lutomirski
2021-02-12 20:20 ` Andy Lutomirski
2021-02-12 20:44 ` Sean Christopherson
2021-02-05 23:38 ` [RFC v1 06/26] x86/tdx: Add HLT " Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 07/26] x86/tdx: Wire up KVM hypercalls Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 08/26] x86/tdx: Add MSR support for TDX guest Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 09/26] x86/tdx: Handle CPUID via #VE Kuppuswamy Sathyanarayanan
2021-02-05 23:42 ` Andy Lutomirski
2021-02-07 14:13 ` Kirill A. Shutemov
2021-02-07 16:01 ` Dave Hansen
2021-02-07 20:29 ` Kirill A. Shutemov
2021-02-07 22:31 ` Dave Hansen
2021-02-07 22:45 ` Andy Lutomirski
2021-02-08 17:10 ` Sean Christopherson
2021-02-08 17:35 ` Andy Lutomirski
2021-02-08 17:47 ` Sean Christopherson
2021-03-18 21:30 ` [PATCH v1 1/1] x86/tdx: Add tdcall() and tdvmcall() helper functions Kuppuswamy Sathyanarayanan
2021-03-19 16:55 ` Sean Christopherson
2021-03-19 17:42 ` Kuppuswamy, Sathyanarayanan
2021-03-19 18:22 ` Dave Hansen
2021-03-19 19:58 ` Kuppuswamy, Sathyanarayanan
2021-03-26 23:38 ` [PATCH v2 1/1] x86/tdx: Add __tdcall() and __tdvmcall() " Kuppuswamy Sathyanarayanan
2021-04-20 17:36 ` Dave Hansen
2021-04-20 19:20 ` Kuppuswamy, Sathyanarayanan
2021-04-20 19:59 ` Dave Hansen
2021-04-20 23:12 ` Kuppuswamy, Sathyanarayanan
2021-04-20 23:42 ` Dave Hansen
2021-04-23 1:09 ` Kuppuswamy, Sathyanarayanan
2021-04-23 1:21 ` Dave Hansen
2021-04-23 1:35 ` Andi Kleen
2021-04-23 15:15 ` Sean Christopherson
2021-04-23 15:28 ` Dan Williams
2021-04-23 15:38 ` Andi Kleen
2021-04-23 15:50 ` Sean Christopherson
2021-04-23 15:47 ` Andi Kleen
2021-04-23 18:18 ` Kuppuswamy, Sathyanarayanan
2021-04-20 23:53 ` Dan Williams
2021-04-20 23:59 ` Kuppuswamy, Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 10/26] x86/io: Allow to override inX() and outX() implementation Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 11/26] x86/tdx: Handle port I/O Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 12/26] x86/tdx: Handle in-kernel MMIO Kuppuswamy Sathyanarayanan
2021-04-01 19:56 ` Dave Hansen
2021-04-01 22:26 ` Sean Christopherson
2021-04-01 22:53 ` Dave Hansen
2021-02-05 23:38 ` [RFC v1 13/26] x86/tdx: Handle MWAIT, MONITOR and WBINVD Kuppuswamy Sathyanarayanan
2021-02-05 23:43 ` Andy Lutomirski
2021-02-05 23:54 ` Kuppuswamy, Sathyanarayanan
2021-02-06 1:05 ` Andy Lutomirski
2021-03-27 0:18 ` [PATCH v1 1/1] " Kuppuswamy Sathyanarayanan
2021-03-27 2:40 ` Andy Lutomirski
2021-03-27 3:40 ` Kuppuswamy, Sathyanarayanan
2021-03-27 16:03 ` Andy Lutomirski
2021-03-27 22:54 ` [PATCH v2 " Kuppuswamy Sathyanarayanan
2021-03-29 17:14 ` Dave Hansen
2021-03-29 21:55 ` Kuppuswamy, Sathyanarayanan
2021-03-29 22:02 ` Dave Hansen
2021-03-29 22:09 ` Kuppuswamy, Sathyanarayanan
2021-03-29 22:12 ` Dave Hansen
2021-03-29 22:42 ` Kuppuswamy, Sathyanarayanan
2021-03-29 23:16 ` [PATCH v3 " Kuppuswamy Sathyanarayanan
2021-03-29 23:23 ` Andy Lutomirski
2021-03-29 23:37 ` Kuppuswamy, Sathyanarayanan
2021-03-29 23:42 ` Sean Christopherson
2021-03-29 23:58 ` Andy Lutomirski
2021-03-30 2:04 ` Andi Kleen
2021-03-30 2:58 ` Andy Lutomirski
2021-03-30 15:14 ` Sean Christopherson
2021-03-30 16:37 ` Andy Lutomirski
2021-03-30 16:57 ` Sean Christopherson
2021-04-07 15:24 ` Andi Kleen
2021-03-31 21:09 ` [PATCH v4 " Kuppuswamy Sathyanarayanan
2021-03-31 21:49 ` Dave Hansen
2021-03-31 22:29 ` Kuppuswamy, Sathyanarayanan
2021-03-31 21:53 ` Sean Christopherson
2021-03-31 22:00 ` Dave Hansen
2021-03-31 22:06 ` Sean Christopherson
2021-03-31 22:11 ` Dave Hansen
2021-03-31 22:28 ` Kuppuswamy, Sathyanarayanan
2021-03-31 22:32 ` Sean Christopherson
2021-03-31 22:34 ` Dave Hansen
2021-04-01 3:28 ` Andi Kleen
2021-04-01 3:46 ` Dave Hansen
2021-04-01 4:24 ` Andi Kleen
2021-04-01 4:51 ` [PATCH v5 " Kuppuswamy Sathyanarayanan
2021-03-29 23:39 ` [PATCH v3 " Sean Christopherson
2021-03-29 23:38 ` Dave Hansen
2021-03-30 4:56 ` [PATCH v1 " Xiaoyao Li
2021-03-30 15:00 ` Andi Kleen
2021-03-30 15:10 ` Dave Hansen
2021-03-30 17:02 ` Kuppuswamy, Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 14/26] ACPI: tables: Add multiprocessor wake-up support Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 15/26] x86/boot: Add a trampoline for APs booting in 64-bit mode Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 16/26] x86/boot: Avoid #VE during compressed boot for TDX platforms Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 17/26] x86/boot: Avoid unnecessary #VE during boot process Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 18/26] x86/topology: Disable CPU hotplug support for TDX platforms Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 19/26] x86/tdx: Forcefully disable legacy PIC for TDX guests Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 20/26] x86/tdx: Introduce INTEL_TDX_GUEST config option Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 21/26] x86/mm: Move force_dma_unencrypted() to common code Kuppuswamy Sathyanarayanan
2021-04-01 20:06 ` Dave Hansen
2021-04-06 15:37 ` Kirill A. Shutemov
2021-04-06 16:11 ` Dave Hansen
2021-04-06 16:37 ` Kirill A. Shutemov
2021-02-05 23:38 ` [RFC v1 22/26] x86/tdx: Exclude Shared bit from __PHYSICAL_MASK Kuppuswamy Sathyanarayanan
2021-04-01 20:13 ` Dave Hansen
2021-04-06 15:54 ` Kirill A. Shutemov
2021-04-06 16:12 ` Dave Hansen
2021-02-05 23:38 ` [RFC v1 23/26] x86/tdx: Make pages shared in ioremap() Kuppuswamy Sathyanarayanan
2021-04-01 20:26 ` Dave Hansen
2021-04-06 16:00 ` Kirill A. Shutemov
2021-04-06 16:14 ` Dave Hansen
2021-02-05 23:38 ` [RFC v1 24/26] x86/tdx: Add helper to do MapGPA TDVMALL Kuppuswamy Sathyanarayanan
2021-02-05 23:38 ` [RFC v1 25/26] x86/tdx: Make DMA pages shared Kuppuswamy Sathyanarayanan
2021-04-01 21:01 ` Dave Hansen
2021-04-06 16:31 ` Kirill A. Shutemov
2021-04-06 16:38 ` Dave Hansen
2021-04-06 17:16 ` Sean Christopherson
2021-02-05 23:38 ` [RFC v1 26/26] x86/kvm: Use bounce buffers for TD guest Kuppuswamy Sathyanarayanan
2021-04-01 21:17 ` Dave Hansen [this message]
2021-02-06 3:04 ` Test Email sathyanarayanan.kuppuswamy
2021-02-06 6:24 ` [RFC v1 00/26] Add TDX Guest Support sathyanarayanan.kuppuswamy
2021-03-31 21:38 ` Kuppuswamy, Sathyanarayanan
2021-04-02 0:02 ` Dave Hansen
2021-04-02 2:48 ` Andi Kleen
2021-04-02 15:27 ` Dave Hansen
2021-04-02 21:32 ` Andi Kleen
2021-04-03 16:26 ` Dave Hansen
2021-04-03 17:28 ` Andi Kleen
2021-04-04 15:02 ` Dave Hansen
2021-04-12 17:24 ` Dan Williams
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=1741d632-0ed0-6a5b-efae-4fdb2689e20f@intel.com \
--to=dave.hansen@intel.com \
--cc=ak@linux.intel.com \
--cc=ashok.raj@intel.com \
--cc=dan.j.williams@intel.com \
--cc=kirill.shutemov@linux.intel.com \
--cc=knsathya@kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=peterz@infradead.org \
--cc=sathyanarayanan.kuppuswamy@linux.intel.com \
--cc=seanjc@google.com \
/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).