Received-SPF: pass (google.com: domain of linux-kernel+bounces-208650-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
Date: Mon, 10 Jun 2024 18:27:22 +0100
From: Catalin Marinas <catalin.marinas@arm.com>
To: Steven Price <steven.price@arm.com>
Cc: kvm@vger.kernel.org, kvmarm@lists.linux.dev,
	Suzuki K Poulose <suzuki.poulose@arm.com>,
	Marc Zyngier <maz@kernel.org>, Will Deacon <will@kernel.org>,
	James Morse <james.morse@arm.com>,
	Oliver Upton <oliver.upton@linux.dev>,
	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>
Subject: Re: [PATCH v3 09/14] arm64: Enable memory encrypt for Realms
Message-ID: <Zmc3euO2YGh-g9Th@arm.com>
References: <20240605093006.145492-1-steven.price@arm.com>
 <20240605093006.145492-10-steven.price@arm.com>
Precedence: bulk
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20240605093006.145492-10-steven.price@arm.com>

On Wed, Jun 05, 2024 at 10:30:01AM +0100, Steven Price wrote:
> +static int __set_memory_encrypted(unsigned long addr,
> +				  int numpages,
> +				  bool encrypt)
> +{
> +	unsigned long set_prot = 0, clear_prot = 0;
> +	phys_addr_t start, end;
> +	int ret;
> +
> +	if (!is_realm_world())
> +		return 0;
> +
> +	if (!__is_lm_address(addr))
> +		return -EINVAL;
> +
> +	start = __virt_to_phys(addr);
> +	end = start + numpages * PAGE_SIZE;
> +
> +	/*
> +	 * Break the mapping before we make any changes to avoid stale TLB
> +	 * entries or Synchronous External Aborts caused by RIPAS_EMPTY
> +	 */
> +	ret = __change_memory_common(addr, PAGE_SIZE * numpages,
> +				     __pgprot(0),
> +				     __pgprot(PTE_VALID));
> +
> +	if (encrypt) {
> +		clear_prot = PROT_NS_SHARED;
> +		ret = rsi_set_memory_range_protected(start, end);
> +	} else {
> +		set_prot = PROT_NS_SHARED;
> +		ret = rsi_set_memory_range_shared(start, end);
> +	}
> +
> +	if (ret)
> +		return ret;
> +
> +	set_prot |= PTE_VALID;
> +
> +	return __change_memory_common(addr, PAGE_SIZE * numpages,
> +				      __pgprot(set_prot),
> +				      __pgprot(clear_prot));
> +}

This works, does break-before-make and also rejects vmalloc() ranges
(for the time being).

One particular aspect I don't like is doing the TLBI twice. It's
sufficient to do it when you first make the pte invalid. We could guess
this in __change_memory_common() if set_mask has PTE_VALID. The call
sites are restricted to this file, just add a comment. An alternative
would be to add a bool flush argument to this function.

-- 
Catalin