linux-coco.lists.linux.dev archive mirror
 help / color / mirror / Atom feed
From: Catalin Marinas <catalin.marinas@arm.com>
To: Steven Price <steven.price@arm.com>
Cc: kvm@vger.kernel.org, kvmarm@lists.linux.dev,
	Marc Zyngier <maz@kernel.org>, Will Deacon <will@kernel.org>,
	James Morse <james.morse@arm.com>,
	Oliver Upton <oliver.upton@linux.dev>,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Zenghui Yu <yuzenghui@huawei.com>,
	linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org, Joey Gouly <joey.gouly@arm.com>,
	Alexandru Elisei <alexandru.elisei@arm.com>,
	Christoffer Dall <christoffer.dall@arm.com>,
	Fuad Tabba <tabba@google.com>,
	linux-coco@lists.linux.dev,
	Ganapatrao Kulkarni <gankulkarni@os.amperecomputing.com>,
	Gavin Shan <gshan@redhat.com>,
	Shanker Donthineni <sdonthineni@nvidia.com>,
	Alper Gun <alpergun@google.com>
Subject: Re: [PATCH v5 14/19] arm64: Enforce bounce buffers for realm DMA
Date: Mon, 26 Aug 2024 13:39:32 +0300	[thread overview]
Message-ID: <ZsxbZMxOIP795qPM@arm.com> (raw)
In-Reply-To: <20240819131924.372366-15-steven.price@arm.com>

On Mon, Aug 19, 2024 at 02:19:19PM +0100, Steven Price wrote:
> Within a realm guest it's not possible for a device emulated by the VMM
> to access arbitrary guest memory. So force the use of bounce buffers to
> ensure that the memory the emulated devices are accessing is in memory
> which is explicitly shared with the host.
> 
> This adds a call to swiotlb_update_mem_attributes() which calls
> set_memory_decrypted() to ensure the bounce buffer memory is shared with
> the host. For non-realm guests or hosts this is a no-op.
> 
> Co-developed-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
> Signed-off-by: Steven Price <steven.price@arm.com>
> ---
> v3: Simplify mem_init() by using a 'flags' variable.
> ---
>  arch/arm64/kernel/rsi.c |  1 +
>  arch/arm64/mm/init.c    | 10 +++++++++-
>  2 files changed, 10 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/arm64/kernel/rsi.c b/arch/arm64/kernel/rsi.c
> index 5c2c977a50fb..69d8d9791c65 100644
> --- a/arch/arm64/kernel/rsi.c
> +++ b/arch/arm64/kernel/rsi.c
> @@ -6,6 +6,7 @@
>  #include <linux/jump_label.h>
>  #include <linux/memblock.h>
>  #include <linux/psci.h>
> +#include <linux/swiotlb.h>
>  
>  #include <asm/io.h>
>  #include <asm/rsi.h>
> diff --git a/arch/arm64/mm/init.c b/arch/arm64/mm/init.c
> index 9b5ab6818f7f..1d595b63da71 100644
> --- a/arch/arm64/mm/init.c
> +++ b/arch/arm64/mm/init.c
> @@ -41,6 +41,7 @@
>  #include <asm/kvm_host.h>
>  #include <asm/memory.h>
>  #include <asm/numa.h>
> +#include <asm/rsi.h>
>  #include <asm/sections.h>
>  #include <asm/setup.h>
>  #include <linux/sizes.h>
> @@ -369,8 +370,14 @@ void __init bootmem_init(void)
>   */
>  void __init mem_init(void)
>  {
> +	unsigned int flags = SWIOTLB_VERBOSE;
>  	bool swiotlb = max_pfn > PFN_DOWN(arm64_dma_phys_limit);
>  
> +	if (is_realm_world()) {
> +		swiotlb = true;
> +		flags |= SWIOTLB_FORCE;
> +	}
> +
>  	if (IS_ENABLED(CONFIG_DMA_BOUNCE_UNALIGNED_KMALLOC) && !swiotlb) {
>  		/*
>  		 * If no bouncing needed for ZONE_DMA, reduce the swiotlb
> @@ -382,7 +389,8 @@ void __init mem_init(void)
>  		swiotlb = true;
>  	}
>  
> -	swiotlb_init(swiotlb, SWIOTLB_VERBOSE);
> +	swiotlb_init(swiotlb, flags);
> +	swiotlb_update_mem_attributes();

IIRC Will mentioned on a previous version of this series: what do we do
with the kmalloc() minalign bouncing (or other bouncing)? I think this
would only work if the device is shared.

I'm more and more inclined to only support shared devices with this
series (no dev assignment) and make it a strict dependence on RMM 1.0.
Running it in a different configuration with private devices will fall
apart. With this condition, the patch looks fine:

Reviewed-by: Catalin Marinas <catalin.marinas@arm.com>

  reply	other threads:[~2024-08-26 10:39 UTC|newest]

Thread overview: 67+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-19 13:19 [PATCH v5 00/19] arm64: Support for running as a guest in Arm CCA Steven Price
2024-08-19 13:19 ` [PATCH v5 01/19] arm64: mm: Add top-level dispatcher for internal mem_encrypt API Steven Price
2024-08-26 10:00   ` Catalin Marinas
2024-08-19 13:19 ` [PATCH v5 02/19] arm64: mm: Add confidential computing hook to ioremap_prot() Steven Price
2024-08-26 10:01   ` Catalin Marinas
2024-08-19 13:19 ` [PATCH v5 03/19] arm64: rsi: Add RSI definitions Steven Price
2024-08-26 10:01   ` Catalin Marinas
2024-09-09  5:10   ` Gavin Shan
2024-09-09  9:12     ` Steven Price
2024-08-19 13:19 ` [PATCH v5 04/19] firmware/psci: Add psci_early_test_conduit() Steven Price
2024-08-23 13:29   ` Will Deacon
2024-08-30 15:54     ` Steven Price
2024-08-26 10:03   ` Catalin Marinas
2024-09-13 13:52     ` Suzuki K Poulose
2024-09-09 23:56   ` Gavin Shan
2024-08-19 13:19 ` [PATCH v5 05/19] arm64: Detect if in a realm and set RIPAS RAM Steven Price
2024-08-19 14:04   ` Suzuki K Poulose
2024-08-19 14:10     ` Steven Price
2024-09-09 15:15       ` Shanker Donthineni
2024-08-26 10:03   ` Catalin Marinas
2024-08-30 15:54     ` Steven Price
2024-09-10  0:09       ` Gavin Shan
2024-09-06 18:58   ` Shanker Donthineni
2024-08-19 13:19 ` [PATCH v5 06/19] arm64: realm: Query IPA size from the RMM Steven Price
2024-08-26 10:04   ` Catalin Marinas
2024-09-10  0:18   ` Gavin Shan
2024-08-19 13:19 ` [PATCH v5 07/19] arm64: rsi: Add support for checking whether an MMIO is protected Steven Price
2024-08-26 10:04   ` Catalin Marinas
2024-09-06  4:32   ` Gavin Shan
2024-09-06  4:52     ` Gavin Shan
2024-09-06 13:55     ` Steven Price
2024-09-08 23:53       ` Gavin Shan
2024-09-09  9:31         ` Steven Price
2024-09-10  3:51   ` Gavin Shan
2024-08-19 13:19 ` [PATCH v5 08/19] fixmap: Allow architecture overriding set_fixmap_io Steven Price
2024-08-19 13:19 ` [PATCH v5 09/19] fixmap: Pass down the full phys address for set_fixmap_io Steven Price
2024-08-26 10:05   ` Catalin Marinas
2024-08-19 13:19 ` [PATCH v5 10/19] arm64: Override set_fixmap_io Steven Price
2024-08-19 14:13   ` Suzuki K Poulose
2024-08-30 15:54     ` Steven Price
2024-08-19 13:19 ` [PATCH v5 11/19] arm64: rsi: Map unprotected MMIO as decrypted Steven Price
2024-08-19 14:11   ` Suzuki K Poulose
2024-08-30 15:54     ` Steven Price
2024-08-19 13:19 ` [PATCH v5 12/19] efi: arm64: Map Device with Prot Shared Steven Price
2024-08-26 10:13   ` Catalin Marinas
2024-09-09 13:55   ` Matias Ezequiel Vara Larsen
2024-09-10  4:15   ` Gavin Shan
2024-09-10  9:15     ` Suzuki K Poulose
2024-08-19 13:19 ` [PATCH v5 13/19] arm64: Make the PHYS_MASK_SHIFT dynamic Steven Price
2024-08-26 10:31   ` Catalin Marinas
2024-08-19 13:19 ` [PATCH v5 14/19] arm64: Enforce bounce buffers for realm DMA Steven Price
2024-08-26 10:39   ` Catalin Marinas [this message]
2024-08-19 13:19 ` [PATCH v5 15/19] arm64: mm: Avoid TLBI when marking pages as valid Steven Price
2024-08-26 10:41   ` Catalin Marinas
2024-08-19 13:19 ` [PATCH v5 16/19] arm64: Enable memory encrypt for Realms Steven Price
2024-08-26 10:46   ` Catalin Marinas
2024-08-19 13:19 ` [PATCH v5 17/19] irqchip/gic-v3-its: Share ITS tables with a non-trusted hypervisor Steven Price
2024-08-19 14:27   ` Marc Zyngier
2024-08-19 14:51   ` Suzuki K Poulose
2024-08-19 15:24     ` Marc Zyngier
2024-08-19 22:19       ` Suzuki K Poulose
2024-10-18  4:49   ` Shanker Donthineni
2024-08-19 13:19 ` [PATCH v5 18/19] irqchip/gic-v3-its: Rely on genpool alignment Steven Price
2024-08-19 13:19 ` [PATCH v5 19/19] virt: arm-cca-guest: TSM_REPORT support for realms Steven Price
2024-09-02  3:53   ` Aneesh Kumar K.V
2024-09-27 15:21     ` Steven Price
2024-09-09  4:13   ` Gavin Shan

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=ZsxbZMxOIP795qPM@arm.com \
    --to=catalin.marinas@arm.com \
    --cc=alexandru.elisei@arm.com \
    --cc=alpergun@google.com \
    --cc=christoffer.dall@arm.com \
    --cc=gankulkarni@os.amperecomputing.com \
    --cc=gshan@redhat.com \
    --cc=james.morse@arm.com \
    --cc=joey.gouly@arm.com \
    --cc=kvm@vger.kernel.org \
    --cc=kvmarm@lists.linux.dev \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-coco@lists.linux.dev \
    --cc=linux-kernel@vger.kernel.org \
    --cc=maz@kernel.org \
    --cc=oliver.upton@linux.dev \
    --cc=sdonthineni@nvidia.com \
    --cc=steven.price@arm.com \
    --cc=suzuki.poulose@arm.com \
    --cc=tabba@google.com \
    --cc=will@kernel.org \
    --cc=yuzenghui@huawei.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).