* [PATCH v5 0/1] x86/test: Add a test for AMD SEV-ES #VC handling
@ 2022-02-08 16:26 Vasant Karasulli
2022-02-08 16:26 ` [PATCH v5 1/1] " Vasant Karasulli
0 siblings, 1 reply; 3+ messages in thread
From: Vasant Karasulli @ 2022-02-08 16:26 UTC (permalink / raw)
To: linux-kernel; +Cc: bp, jroedel, kvm, x86, thomas.lendacky, Vasant Karasulli
Hello!
This is the version 5 of the patch written to add a test for
AMD SEV-ES #VC handling. This version attempts to
address review comments to the previous version of the patch in
https://lore.kernel.org/kvm/YNRgHbPVGpLaByjH@zn.tnic/.
Following changes have been made to address the comments:
1. Changed the subject to x86/test from x86.
2. In file arch/x86/Kconfig.debug enclosed the config AMD_SEV_TEST_VC
inside the guard X86_TESTS.
3. Renamed the test file to sev-test-vc.c to indicate that
file includes all the #VC handling related tests.
4. Memory barries are apparently not necessary and have been removed.
Thanks,
Vasant
----------------------------------------------------------------------
arch/x86/Kbuild | 2 +
arch/x86/Kconfig.debug | 16 ++++
arch/x86/kernel/Makefile | 7 ++
arch/x86/tests/Makefile | 3 +
arch/x86/tests/sev-test-vc.c | 154 +++++++++++++++++++++++++++++++++++
5 files changed, 182 insertions(+)
create mode 100644 arch/x86/tests/Makefile
create mode 100644 arch/x86/tests/sev-test-vc.c
base-commit: dfd42facf1e4ada021b939b4e19c935dcdd55566
prerequisite-patch-id: 660950af15f9a972b1f71aa7702ebebe7dd9636b
prerequisite-patch-id: 9635b44cd178a2ee20d83acbc7a07101fd7e81a3
prerequisite-patch-id: 3a7cfa2a885ad20a9f325febcbe31ef53b41e0b1
prerequisite-patch-id: 5e48ede5d4df9de5df3d483874b4aa1711b5dcf0
prerequisite-patch-id: b66ebf185fceb6e8e98b466c974ed777f0200787
prerequisite-patch-id: d5bde21908fbada516dc757e29c6ad70926ab5bf
prerequisite-patch-id: ed38b7fc213fa66845dc25502f18b79e3dc907db
prerequisite-patch-id: 053546f9741643d742f5149c8441f8484e1918d1
prerequisite-patch-id: a20f9a6fd1ad2db0bae2550d5442d7275c481b01
prerequisite-patch-id: bde103392dd4c0a5c360446dbc8d2e5f82671575
prerequisite-patch-id: af44d5e0a6f51e1fe5cf6aef7d0be36e201663b4
prerequisite-patch-id: 952ebde20c8c084f43523f8230ced76beae08b3e
prerequisite-patch-id: 0b295c1545756626a9bca5bce8be4557742b7a56
prerequisite-patch-id: 98bf6777201eeab52938c28de1090387d454ef55
prerequisite-patch-id: d64628531375578e4bc6de3bcba1a3de658e28d8
prerequisite-patch-id: 752ea9954abeb624b803ad86e4974769f97ddd6e
--
2.32.0
^ permalink raw reply [flat|nested] 3+ messages in thread
* [PATCH v5 1/1] x86/test: Add a test for AMD SEV-ES #VC handling
2022-02-08 16:26 [PATCH v5 0/1] x86/test: Add a test for AMD SEV-ES #VC handling Vasant Karasulli
@ 2022-02-08 16:26 ` Vasant Karasulli
2022-03-17 13:21 ` Joerg Roedel
0 siblings, 1 reply; 3+ messages in thread
From: Vasant Karasulli @ 2022-02-08 16:26 UTC (permalink / raw)
To: linux-kernel
Cc: bp, jroedel, kvm, x86, thomas.lendacky, Varad Gautam,
Vasant Karasulli
From: Varad Gautam <varad.gautam@suse.com>
Add a KUnit based test to validate Linux's VC handling, and introduce
a new CONFIG_X86_TESTS to cover such tests. The test:
1. installs a kretprobe on the #VC handler (sev_es_ghcb_hv_call, to
access GHCB before/after the resulting VMGEXIT).
2. triggers an NAE.
3. checks that the kretprobe was hit with the right exit_code available
in GHCB.
Since relying on kprobes, the test does not cover NMI contexts.
Signed-off-by: Varad Gautam <varad.gautam@suse.com>
Signed-off-by: Vasant Karasulli <vkarasulli@suse.de>
---
arch/x86/Kbuild | 2 +
arch/x86/Kconfig.debug | 16 ++++
arch/x86/kernel/Makefile | 7 ++
arch/x86/tests/Makefile | 3 +
arch/x86/tests/sev-test-vc.c | 154 +++++++++++++++++++++++++++++++++++
5 files changed, 182 insertions(+)
create mode 100644 arch/x86/tests/Makefile
create mode 100644 arch/x86/tests/sev-test-vc.c
diff --git a/arch/x86/Kbuild b/arch/x86/Kbuild
index f384cb1a4f7a..90470c76866a 100644
--- a/arch/x86/Kbuild
+++ b/arch/x86/Kbuild
@@ -26,5 +26,7 @@ obj-y += net/
obj-$(CONFIG_KEXEC_FILE) += purgatory/
+obj-y += tests/
+
# for cleaning
subdir- += boot tools
diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index d3a6f74a94bd..e4f61af66816 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -279,3 +279,19 @@ endchoice
config FRAME_POINTER
depends on !UNWINDER_ORC && !UNWINDER_GUESS
bool
+
+config X86_TESTS
+ bool "Tests for x86"
+ help
+ This enables building the tests under arch/x86/tests.
+
+if X86_TESTS
+config AMD_SEV_TEST_VC
+ bool "Test for AMD SEV VC exception handling"
+ depends on AMD_MEM_ENCRYPT
+ select FUNCTION_TRACER
+ select KPROBES
+ select KUNIT
+ help
+ Enable KUnit-based testing for AMD SEV #VC exception handling.
+endif # X86_TESTS
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 6aef9ee28a39..69472a576909 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -24,6 +24,13 @@ CFLAGS_REMOVE_sev.o = -pg
CFLAGS_REMOVE_cc_platform.o = -pg
endif
+# AMD_SEV_TEST_VC registers a kprobe by function name. IPA-SRA creates
+# function copies and renames them to have an .isra suffix, which breaks kprobes'
+# lookup. Build with -fno-ipa-sra for the test.
+ifdef CONFIG_AMD_SEV_TEST_VC
+CFLAGS_sev.o += -fno-ipa-sra
+endif
+
KASAN_SANITIZE_head$(BITS).o := n
KASAN_SANITIZE_dumpstack.o := n
KASAN_SANITIZE_dumpstack_$(BITS).o := n
diff --git a/arch/x86/tests/Makefile b/arch/x86/tests/Makefile
new file mode 100644
index 000000000000..4beca64bd2aa
--- /dev/null
+++ b/arch/x86/tests/Makefile
@@ -0,0 +1,3 @@
+# SPDX-License-Identifier: GPL-2.0
+
+obj-$(CONFIG_AMD_SEV_TEST_VC) += sev-test-vc.o
diff --git a/arch/x86/tests/sev-test-vc.c b/arch/x86/tests/sev-test-vc.c
new file mode 100644
index 000000000000..eb5f790b64f5
--- /dev/null
+++ b/arch/x86/tests/sev-test-vc.c
@@ -0,0 +1,154 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Copyright (C) 2021 SUSE
+ *
+ * Author: Varad Gautam <varad.gautam@suse.com>
+ */
+
+#include <asm/cpufeature.h>
+#include <asm/debugreg.h>
+#include <asm/io.h>
+#include <asm/sev-common.h>
+#include <asm/svm.h>
+#include <kunit/test.h>
+#include <linux/kprobes.h>
+
+static struct kretprobe hv_call_krp;
+
+static int hv_call_krp_entry(struct kretprobe_instance *krpi,
+ struct pt_regs *regs)
+{
+ unsigned long ghcb_vaddr = regs_get_kernel_argument(regs, 0);
+ *((unsigned long *) krpi->data) = ghcb_vaddr;
+
+ return 0;
+}
+
+static int hv_call_krp_ret(struct kretprobe_instance *krpi,
+ struct pt_regs *regs)
+{
+ unsigned long ghcb_vaddr = *((unsigned long *) krpi->data);
+ struct ghcb *ghcb = (struct ghcb *) ghcb_vaddr;
+ struct kunit *test = current->kunit_test;
+
+ if (test && strstr(test->name, "sev_es_") && test->priv)
+ *((u64 *)test->priv) = ghcb->save.sw_exit_code;
+
+ return 0;
+}
+
+int sev_es_test_vc_init(struct kunit *test)
+{
+ int ret;
+
+ if (!cc_platform_has(CC_ATTR_GUEST_STATE_ENCRYPT)) {
+ kunit_info(test, "Not a SEV-ES guest. Skipping.");
+ ret = -EINVAL;
+ goto out;
+ }
+
+ memset(&hv_call_krp, 0, sizeof(hv_call_krp));
+ hv_call_krp.entry_handler = hv_call_krp_entry;
+ hv_call_krp.handler = hv_call_krp_ret;
+ hv_call_krp.maxactive = 100;
+ hv_call_krp.data_size = sizeof(unsigned long);
+ hv_call_krp.kp.symbol_name = "sev_es_ghcb_hv_call";
+ hv_call_krp.kp.addr = 0;
+
+ ret = register_kretprobe(&hv_call_krp);
+ if (ret) {
+ kunit_info(test, "Could not register kretprobe. Skipping.");
+ goto out;
+ }
+
+ test->priv = kunit_kzalloc(test, sizeof(u64), GFP_KERNEL);
+ if (!test->priv) {
+ ret = -ENOMEM;
+ kunit_info(test, "Could not allocate. Skipping.");
+ goto out;
+ }
+
+out:
+ return ret;
+}
+
+void sev_es_test_vc_exit(struct kunit *test)
+{
+ if (test->priv)
+ kunit_kfree(test, test->priv);
+
+ if (hv_call_krp.kp.addr)
+ unregister_kretprobe(&hv_call_krp);
+}
+
+#define check_op(kt, ec, op) \
+do { \
+ struct kunit *t = (struct kunit *) kt; \
+ op; \
+ KUNIT_EXPECT_EQ(t, (typeof(ec)) ec, \
+ *((typeof(ec) *)(t->priv))); \
+} while (0)
+
+static void sev_es_nae_cpuid(struct kunit *test)
+{
+ unsigned int cpuid_fn = 0x8000001f;
+
+ check_op(test, SVM_EXIT_CPUID, native_cpuid_eax(cpuid_fn));
+}
+
+static void sev_es_nae_wbinvd(struct kunit *test)
+{
+ check_op(test, SVM_EXIT_WBINVD, wbinvd());
+}
+
+static void sev_es_nae_msr(struct kunit *test)
+{
+ check_op(test, SVM_EXIT_MSR, __rdmsr(MSR_IA32_TSC));
+}
+
+static void sev_es_nae_dr7_rw(struct kunit *test)
+{
+ check_op(test, SVM_EXIT_WRITE_DR7,
+ native_set_debugreg(7, native_get_debugreg(7)));
+}
+
+static void sev_es_nae_ioio(struct kunit *test)
+{
+ unsigned long port = 0x80;
+ char val = 0;
+
+ check_op(test, SVM_EXIT_IOIO, val = inb(port));
+ check_op(test, SVM_EXIT_IOIO, outb(val, port));
+ check_op(test, SVM_EXIT_IOIO, insb(port, &val, sizeof(val)));
+ check_op(test, SVM_EXIT_IOIO, outsb(port, &val, sizeof(val)));
+}
+
+static void sev_es_nae_mmio(struct kunit *test)
+{
+ unsigned long lapic_ver_pa = 0xfee00030; /* APIC_DEFAULT_PHYS_BASE + APIC_LVR */
+ unsigned long __iomem *lapic = ioremap(lapic_ver_pa, 0x4);
+ unsigned long lapic_version = 0;
+
+ check_op(test, SVM_VMGEXIT_MMIO_READ, lapic_version = *lapic);
+ check_op(test, SVM_VMGEXIT_MMIO_WRITE, *lapic = lapic_version);
+
+ iounmap(lapic);
+}
+
+static struct kunit_case sev_es_vc_testcases[] = {
+ KUNIT_CASE(sev_es_nae_cpuid),
+ KUNIT_CASE(sev_es_nae_wbinvd),
+ KUNIT_CASE(sev_es_nae_msr),
+ KUNIT_CASE(sev_es_nae_dr7_rw),
+ KUNIT_CASE(sev_es_nae_ioio),
+ KUNIT_CASE(sev_es_nae_mmio),
+ {}
+};
+
+static struct kunit_suite sev_es_vc_test_suite = {
+ .name = "sev_es_test_vc",
+ .init = sev_es_test_vc_init,
+ .exit = sev_es_test_vc_exit,
+ .test_cases = sev_es_vc_testcases,
+};
+kunit_test_suite(sev_es_vc_test_suite);
--
2.32.0
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH v5 1/1] x86/test: Add a test for AMD SEV-ES #VC handling
2022-02-08 16:26 ` [PATCH v5 1/1] " Vasant Karasulli
@ 2022-03-17 13:21 ` Joerg Roedel
0 siblings, 0 replies; 3+ messages in thread
From: Joerg Roedel @ 2022-03-17 13:21 UTC (permalink / raw)
To: Vasant Karasulli
Cc: linux-kernel, bp, kvm, x86, thomas.lendacky, Varad Gautam
Hi Vasant,
thanks for submitting this.
On Tue, Feb 08, 2022 at 05:26:23PM +0100, Vasant Karasulli wrote:
> From: Varad Gautam <varad.gautam@suse.com>
>
> Add a KUnit based test to validate Linux's VC handling, and introduce
> a new CONFIG_X86_TESTS to cover such tests. The test:
> 1. installs a kretprobe on the #VC handler (sev_es_ghcb_hv_call, to
> access GHCB before/after the resulting VMGEXIT).
> 2. triggers an NAE.
> 3. checks that the kretprobe was hit with the right exit_code available
> in GHCB.
>
> Since relying on kprobes, the test does not cover NMI contexts.
>
> Signed-off-by: Varad Gautam <varad.gautam@suse.com>
> Signed-off-by: Vasant Karasulli <vkarasulli@suse.de>
> ---
> arch/x86/Kbuild | 2 +
> arch/x86/Kconfig.debug | 16 ++++
> arch/x86/kernel/Makefile | 7 ++
> arch/x86/tests/Makefile | 3 +
> arch/x86/tests/sev-test-vc.c | 154 +++++++++++++++++++++++++++++++++++
> 5 files changed, 182 insertions(+)
> create mode 100644 arch/x86/tests/Makefile
> create mode 100644 arch/x86/tests/sev-test-vc.c
Can we split this up a bit? Say into an initial patch which adds
arch/x86/tests/ and the Kconfig option and then three patches adding the
tests:
1. Infrastructure and instruction-based tests (cpuid, wbinvd)
2. Register access tests (MSR, DR7)
3. IO tests (mmio, ioio)
> +static void sev_es_nae_mmio(struct kunit *test)
> +{
> + unsigned long lapic_ver_pa = 0xfee00030; /* APIC_DEFAULT_PHYS_BASE + APIC_LVR */
Instead of the comment, please use the values directly to assign the
variable.
Regards,
--
Jörg Rödel
jroedel@suse.de
SUSE Software Solutions Germany GmbH
Maxfeldstr. 5
90409 Nürnberg
Germany
(HRB 36809, AG Nürnberg)
Geschäftsführer: Ivo Totev
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2022-03-17 13:21 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-02-08 16:26 [PATCH v5 0/1] x86/test: Add a test for AMD SEV-ES #VC handling Vasant Karasulli
2022-02-08 16:26 ` [PATCH v5 1/1] " Vasant Karasulli
2022-03-17 13:21 ` Joerg Roedel
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox