linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Roman Kisel <romank@linux.microsoft.com>
To: "Arnd Bergmann" <arnd@arndb.de>,
	bhelgaas@google.com, "Borislav Petkov" <bp@alien8.de>,
	"Catalin Marinas" <catalin.marinas@arm.com>,
	"Conor Dooley" <conor+dt@kernel.org>,
	"Dave Hansen" <dave.hansen@linux.intel.com>,
	"Dexuan Cui" <decui@microsoft.com>,
	"Haiyang Zhang" <haiyangz@microsoft.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	krzk+dt@kernel.org, "Krzysztof Wilczyński" <kw@linux.com>,
	"K. Y. Srinivasan" <kys@microsoft.com>,
	"Lorenzo Pieralisi" <lpieralisi@kernel.org>,
	"Manivannan Sadhasivam" <manivannan.sadhasivam@linaro.org>,
	"Ingo Molnar" <mingo@redhat.com>, "Rob Herring" <robh@kernel.org>,
	ssengar@linux.microsoft.com,
	"Thomas Gleixner" <tglx@linutronix.de>,
	"Wei Liu" <wei.liu@kernel.org>, "Will Deacon" <will@kernel.org>,
	devicetree@vger.kernel.org,
	Linux-Arch <linux-arch@vger.kernel.org>,
	linux-arm-kernel@lists.infradead.org,
	linux-hyperv@vger.kernel.org, linux-kernel@vger.kernel.org,
	linux-pci@vger.kernel.org, x86@kernel.org
Cc: benhill@microsoft.com, bperkins@microsoft.com, sunilmut@microsoft.com
Subject: Re: [PATCH hyperv-next v4 1/6] arm64: hyperv: Use SMCCC to detect hypervisor presence
Date: Thu, 13 Feb 2025 15:23:56 -0800	[thread overview]
Message-ID: <593c22ca-6544-423d-84ee-7a06c6b8b5b9@linux.microsoft.com> (raw)
In-Reply-To: <1b14e3de-4d3e-420c-819c-31ffb2d448bd@app.fastmail.com>



On 2/11/2025 10:54 PM, Arnd Bergmann wrote:
> On Wed, Feb 12, 2025, at 02:43, Roman Kisel wrote:
>> +static bool hyperv_detect_via_smccc(void)
>> +{
>> +	struct arm_smccc_res res = {};
>> +
>> +	if (arm_smccc_1_1_get_conduit() != SMCCC_CONDUIT_HVC)
>> +		return false;
>> +	arm_smccc_1_1_hvc(ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID, &res);
>> +	if (res.a0 == SMCCC_RET_NOT_SUPPORTED)
>> +		return false;
>> +
>> +	return res.a0 == ARM_SMCCC_VENDOR_HYP_UID_HYPERV_REG_0 &&
>> +		res.a1 == ARM_SMCCC_VENDOR_HYP_UID_HYPERV_REG_1 &&
>> +		res.a2 == ARM_SMCCC_VENDOR_HYP_UID_HYPERV_REG_2 &&
>> +		res.a3 == ARM_SMCCC_VENDOR_HYP_UID_HYPERV_REG_3;
>> +}
> 
> I had to double-check that this function is safe to call on
> other hypervisors, at least when they follow the smccc spec.
> 
> Seeing that we have the same helper function checking for
> ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_* and there was another
> patch set adding a copy for gunyah, I wonder if we can
> put this into a drivers/firmware/smccc/smccc.c directly
> the same way we handle soc_id, and make it return a uuid_t,
> or perhaps take a constant uuid_t to compare against.

That would be very neat! I implemented the idea [1], please let me know
what you think. I can use that in the next version of the patch series
if looks good.

There is a function and a macro to make calling
the function easier. As the SMCCC header is used by the assembler, too,
hence I had to add __ASSEBLER__ guardrails. Another option could be to
pass four u32's not to use uuid_t so the header stays a healthy food
for the assembler :) For Gunyah, that would be

ARM_SMCCC_HYP_PRESENT(GUNYAH)

when using the change below.


 From f0d645e900c24f5be045b0f831f1e11494967b7f Mon Sep 17 00:00:00 2001
From: Roman Kisel <romank@linux.microsoft.com>
Date: Thu, 13 Feb 2025 15:10:45 -0800
Subject: [PATCH] drivers, smcc: Introduce arm_smccc_hyp_present

---
  arch/arm64/hyperv/mshyperv.c       | 18 +----------------
  drivers/firmware/smccc/kvm_guest.c |  9 +--------
  drivers/firmware/smccc/smccc.c     | 24 ++++++++++++++++++++++
  include/linux/arm-smccc.h          | 32 ++++++++++++++++++++++++++++++
  4 files changed, 58 insertions(+), 25 deletions(-)

diff --git a/arch/arm64/hyperv/mshyperv.c b/arch/arm64/hyperv/mshyperv.c
index 16e721d8e5df..0c5babe9e1ff 100644
--- a/arch/arm64/hyperv/mshyperv.c
+++ b/arch/arm64/hyperv/mshyperv.c
@@ -41,22 +41,6 @@ static bool hyperv_detect_via_acpi(void)
  #endif
  }

-static bool hyperv_detect_via_smccc(void)
-{
-	struct arm_smccc_res res = {};
-
-	if (arm_smccc_1_1_get_conduit() != SMCCC_CONDUIT_HVC)
-		return false;
-	arm_smccc_1_1_hvc(ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID, &res);
-	if (res.a0 == SMCCC_RET_NOT_SUPPORTED)
-		return false;
-
-	return res.a0 == ARM_SMCCC_VENDOR_HYP_UID_HYPERV_REG_0 &&
-		res.a1 == ARM_SMCCC_VENDOR_HYP_UID_HYPERV_REG_1 &&
-		res.a2 == ARM_SMCCC_VENDOR_HYP_UID_HYPERV_REG_2 &&
-		res.a3 == ARM_SMCCC_VENDOR_HYP_UID_HYPERV_REG_3;
-}
-
  static int __init hyperv_init(void)
  {
  	struct hv_get_vp_registers_output	result;
@@ -69,7 +53,7 @@ static int __init hyperv_init(void)
  	 *
  	 * In such cases, do nothing and return success.
  	 */
-	if (!hyperv_detect_via_acpi() && !hyperv_detect_via_smccc())
+	if (!hyperv_detect_via_acpi() && !ARM_SMCCC_HYP_PRESENT(HYPERV))
  		return 0;

  	/* Setup the guest ID */
diff --git a/drivers/firmware/smccc/kvm_guest.c 
b/drivers/firmware/smccc/kvm_guest.c
index f3319be20b36..ae37476cabc1 100644
--- a/drivers/firmware/smccc/kvm_guest.c
+++ b/drivers/firmware/smccc/kvm_guest.c
@@ -17,14 +17,7 @@ void __init kvm_init_hyp_services(void)
  	struct arm_smccc_res res;
  	u32 val[4];

-	if (arm_smccc_1_1_get_conduit() != SMCCC_CONDUIT_HVC)
-		return;
-
-	arm_smccc_1_1_invoke(ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID, &res);
-	if (res.a0 != ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_0 ||
-	    res.a1 != ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_1 ||
-	    res.a2 != ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_2 ||
-	    res.a3 != ARM_SMCCC_VENDOR_HYP_UID_KVM_REG_3)
+	if (!ARM_SMCCC_HYP_PRESENT(KVM))
  		return;

  	memset(&res, 0, sizeof(res));
diff --git a/drivers/firmware/smccc/smccc.c b/drivers/firmware/smccc/smccc.c
index a74600d9f2d7..86f75f44895f 100644
--- a/drivers/firmware/smccc/smccc.c
+++ b/drivers/firmware/smccc/smccc.c
@@ -67,6 +67,30 @@ s32 arm_smccc_get_soc_id_revision(void)
  }
  EXPORT_SYMBOL_GPL(arm_smccc_get_soc_id_revision);

+bool arm_smccc_hyp_present(const uuid_t *hyp_uuid)
+{
+	struct arm_smccc_res res = {};
+	struct {
+		u32 dwords[4]
+	} __packed res_uuid;
+
+	BUILD_BUG_ON(sizeof(res_uuid) != sizeof(uuid_t));
+
+	if (arm_smccc_1_1_get_conduit() != SMCCC_CONDUIT_HVC)
+		return false;
+	arm_smccc_1_1_hvc(ARM_SMCCC_VENDOR_HYP_CALL_UID_FUNC_ID, &res);
+	if (res.a0 == SMCCC_RET_NOT_SUPPORTED)
+		return false;
+
+	res_uuid.dwords[0] = res.a0;
+	res_uuid.dwords[1] = res.a1;
+	res_uuid.dwords[2] = res.a2;
+	res_uuid.dwords[3] = res.a3;
+
+	return uuid_equal((uuid_t *)&res_uuid, hyp_uuid);
+}
+EXPORT_SYMBOL_GPL(arm_smccc_hyp_present);
+
  static int __init smccc_devices_init(void)
  {
  	struct platform_device *pdev;
diff --git a/include/linux/arm-smccc.h b/include/linux/arm-smccc.h
index 67f6fdf2e7cd..63925506a0e5 100644
--- a/include/linux/arm-smccc.h
+++ b/include/linux/arm-smccc.h
@@ -7,6 +7,11 @@

  #include <linux/args.h>
  #include <linux/init.h>
+
+#ifndef __ASSEMBLER__
+#include <linux/uuid.h>
+#endif
+
  #include <uapi/linux/const.h>

  /*
@@ -333,6 +338,33 @@ s32 arm_smccc_get_soc_id_version(void);
   */
  s32 arm_smccc_get_soc_id_revision(void);

+#ifndef __ASSEMBLER__
+
+/**
+ * arm_smccc_hyp_present(const uuid_t *hyp_uuid)
+ *
+ * Returns `true` if the hypervisor advertises its presence via SMCCC.
+ *
+ * When the function returns `false`, the caller shall not assume that
+ * there is no hypervisor running. Instead, the caller must fall back to
+ * other approaches if any are available.
+ */
+bool arm_smccc_hyp_present(const uuid_t *hyp_uuid);
+
+#define ARM_SMCCC_HYP_PRESENT(HYP) 								\
+	({															\
+		const u32 uuid_as_dwords[4] = {							\
+			ARM_SMCCC_VENDOR_HYP_UID_ ## HYP ## _REG_0,			\
+			ARM_SMCCC_VENDOR_HYP_UID_ ## HYP ## _REG_1,			\
+			ARM_SMCCC_VENDOR_HYP_UID_ ## HYP ## _REG_2,			\
+			ARM_SMCCC_VENDOR_HYP_UID_ ## HYP ## _REG_3			\
+		};														\
+																\
+		arm_smccc_hyp_present((const uuid_t *)uuid_as_dwords);	\
+	})															\
+
+#endif /* !__ASSEMBLER__ */
+
  /**
   * struct arm_smccc_res - Result from SMC/HVC call
   * @a0-a3 result values from registers 0 to 3
-- 
2.43.0



> 
>        Arnd

-- 
Thank you,
Roman



  reply	other threads:[~2025-02-13 23:26 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-02-12  1:43 [PATCH hyperv-next v4 0/6] arm64: hyperv: Support Virtual Trust Level Boot Roman Kisel
2025-02-12  1:43 ` [PATCH hyperv-next v4 1/6] arm64: hyperv: Use SMCCC to detect hypervisor presence Roman Kisel
2025-02-12  6:54   ` Arnd Bergmann
2025-02-13 23:23     ` Roman Kisel [this message]
2025-02-14  8:05       ` Arnd Bergmann
2025-02-14 16:47         ` Roman Kisel
2025-02-24 23:22           ` Roman Kisel
2025-02-25  7:24             ` Arnd Bergmann
2025-02-25 22:25               ` Roman Kisel
2025-02-26 13:57                 ` Arnd Bergmann
2025-02-19 23:13   ` Michael Kelley
2025-02-20 16:34     ` Roman Kisel
2025-02-12  1:43 ` [PATCH hyperv-next v4 2/6] Drivers: hv: Enable VTL mode for arm64 Roman Kisel
2025-02-19 23:14   ` Michael Kelley
2025-02-20 16:36     ` Roman Kisel
2025-02-12  1:43 ` [PATCH hyperv-next v4 3/6] Drivers: hv: Provide arch-neutral implementation of get_vtl() Roman Kisel
2025-02-19 23:17   ` Michael Kelley
2025-02-12  1:43 ` [PATCH hyperv-next v4 4/6] dt-bindings: microsoft,vmbus: Add GIC and DMA coherence to the example Roman Kisel
2025-02-12  6:42   ` Krzysztof Kozlowski
2025-02-12 23:57     ` Roman Kisel
2025-02-13 20:50       ` Roman Kisel
2025-02-12  1:43 ` [PATCH hyperv-next v4 5/6] Drivers: hv: vmbus: Get the IRQ number from DeviceTree Roman Kisel
2025-02-19 23:20   ` Michael Kelley
2025-02-12  1:43 ` [PATCH hyperv-next v4 6/6] PCI: hv: Get vPCI MSI IRQ domain " Roman Kisel
2025-02-12 17:42   ` Bjorn Helgaas
2025-02-18 22:32     ` Roman Kisel
2025-02-19 23:51     ` Roman Kisel
2025-02-19 23:29   ` Michael Kelley
2025-02-20 16:41     ` Roman Kisel

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=593c22ca-6544-423d-84ee-7a06c6b8b5b9@linux.microsoft.com \
    --to=romank@linux.microsoft.com \
    --cc=arnd@arndb.de \
    --cc=benhill@microsoft.com \
    --cc=bhelgaas@google.com \
    --cc=bp@alien8.de \
    --cc=bperkins@microsoft.com \
    --cc=catalin.marinas@arm.com \
    --cc=conor+dt@kernel.org \
    --cc=dave.hansen@linux.intel.com \
    --cc=decui@microsoft.com \
    --cc=devicetree@vger.kernel.org \
    --cc=haiyangz@microsoft.com \
    --cc=hpa@zytor.com \
    --cc=krzk+dt@kernel.org \
    --cc=kw@linux.com \
    --cc=kys@microsoft.com \
    --cc=linux-arch@vger.kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-hyperv@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=lpieralisi@kernel.org \
    --cc=manivannan.sadhasivam@linaro.org \
    --cc=mingo@redhat.com \
    --cc=robh@kernel.org \
    --cc=ssengar@linux.microsoft.com \
    --cc=sunilmut@microsoft.com \
    --cc=tglx@linutronix.de \
    --cc=wei.liu@kernel.org \
    --cc=will@kernel.org \
    --cc=x86@kernel.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).