[PATCH v2 00/10] selftests: kvm: s390: Add s390x ucontrol selftests

[PATCH v2 00/10] selftests: kvm: s390: Add s390x ucontrol selftests

[Christoph Schlameuss]

This patch series adds a selftest suite to validate the s390x
architecture specific ucontrol KVM interface.

When creating a VM on s390x it is possible to create it as userspace
controlled VM or in short ucontrol VM.
These VMs delegates the management of the VM to userspace instead
of handling most events within the kernel. Consequently the userspace
has to manage interrupts, memory allocation etc.

Before this patch set this functionality lacks any public test cases.
It is desirable to add test cases for this interface to be able to
reduce the risk of breaking changes in the future.

In order to provision a ucontrol VM the kernel needs to be compiled with
the CONFIG_KVM_S390_UCONTROL enabled. The users with sys_admin capability
can then create a new ucontrol VM providing the KVM_VM_S390_UCONTROL
parameter to the KVM_CREATE_VM ioctl.

The kernels existing selftest helper functions can only be partially be
reused for these tests.

The test cases cover existing special handling of ucontrol VMs within the
implementation and basic VM creation and handling cases:
* Reject setting HPAGE when VM is ucontrol
* Assert KVM_GET_DIRTY_LOG is rejected
* Assert KVM_S390_VM_MEM_LIMIT_SIZE is rejected
* Assert state of initial SIE flags setup by the kernel
* Run simple program in VM with and without DAT
* Assert KVM_EXIT_S390_UCONTROL exit on not mapped memory access
* Assert functionality of storage keys in ucontrol VM

Running the test cases requires sys_admin capabilities to start the
ucontrol VM.
This can be achieved by running as root or with a command like:

    sudo setpriv --reuid nobody --inh-caps -all,+sys_admin \
      --ambient-caps -all,+sys_admin --bounding-set -all,+sys_admin \
      ./ucontrol_test

The patch set does also contain some code cleanup / consolidation of
architecture specific defines that are now used in multiple test cases.

---

V1 -> V2:
- add ucontrol to s390 debug config (new patch)
- PATCH 2: changed atomic_t to __u32 (thanks Claudio)
- PATCH 4: reformatted comment in FIXTURE_SETUP(uc_kvm)
- PATCH 5: refactored to display 8 byte blocks + more internal reuse
           (thanks Claudio)
- PATCH 7: make use of more declarative defines instead of magic values
- PATCH 8: make use of more declarative defines instead of magic values
           (thanks Claudio)
- PATCH 9: add reference to fix verified by the test case


Christoph Schlameuss (10):
  selftests: kvm: s390: Define page sizes in shared header
  selftests: kvm: s390: Add kvm_s390_sie_block definition for userspace
    tests
  selftests: kvm: s390: Add s390x ucontrol test suite with hpage test
  selftests: kvm: s390: Add test fixture and simple VM setup tests
  selftests: kvm: s390: Add debug print functions
  selftests: kvm: s390: Add VM run test case
  selftests: kvm: s390: Add uc_map_unmap VM test case
  selftests: kvm: s390: Add uc_skey VM test case
  selftests: kvm: s390: Verify reject memory region operations for
    ucontrol VMs
  s390: Enable KVM_S390_UCONTROL config in debug_defconfig

 arch/s390/Kconfig.debug                       |   3 +
 tools/testing/selftests/kvm/.gitignore        |   1 +
 tools/testing/selftests/kvm/Makefile          |   1 +
 .../selftests/kvm/include/s390x/debug_print.h |  69 ++
 .../selftests/kvm/include/s390x/processor.h   |   5 +
 .../testing/selftests/kvm/include/s390x/sie.h | 240 +++++++
 .../selftests/kvm/lib/s390x/processor.c       |  10 +-
 tools/testing/selftests/kvm/s390x/cmma_test.c |   7 +-
 tools/testing/selftests/kvm/s390x/config      |   2 +
 .../testing/selftests/kvm/s390x/debug_test.c  |   4 +-
 tools/testing/selftests/kvm/s390x/memop.c     |   4 +-
 tools/testing/selftests/kvm/s390x/tprot.c     |   5 +-
 .../selftests/kvm/s390x/ucontrol_test.c       | 614 ++++++++++++++++++
 13 files changed, 949 insertions(+), 16 deletions(-)
 create mode 100644 tools/testing/selftests/kvm/include/s390x/debug_print.h
 create mode 100644 tools/testing/selftests/kvm/include/s390x/sie.h
 create mode 100644 tools/testing/selftests/kvm/s390x/config
 create mode 100644 tools/testing/selftests/kvm/s390x/ucontrol_test.c


base-commit: 66ebbdfdeb093e097399b1883390079cd4c3022b
-- 
2.45.2

[PATCH v2 01/10] selftests: kvm: s390: Define page sizes in shared header

[Christoph Schlameuss]

Multiple test cases need page size and shift definitions.
By moving the definitions to a single architecture specific header we
limit the repetition.

Make use of PAGE_SIZE, PAGE_SHIFT and PAGE_MASK defines in existing
code.

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
---
 tools/testing/selftests/kvm/include/s390x/processor.h |  5 +++++
 tools/testing/selftests/kvm/lib/s390x/processor.c     | 10 +++++-----
 tools/testing/selftests/kvm/s390x/cmma_test.c         |  7 ++++---
 tools/testing/selftests/kvm/s390x/memop.c             |  4 +---
 tools/testing/selftests/kvm/s390x/tprot.c             |  5 ++---
 5 files changed, 17 insertions(+), 14 deletions(-)

diff --git a/tools/testing/selftests/kvm/include/s390x/processor.h b/tools/testing/selftests/kvm/include/s390x/processor.h
index 255c9b990f4c..481bd2fd6a32 100644
--- a/tools/testing/selftests/kvm/include/s390x/processor.h
+++ b/tools/testing/selftests/kvm/include/s390x/processor.h
@@ -21,6 +21,11 @@
 #define PAGE_PROTECT	0x200		/* HW read-only bit  */
 #define PAGE_NOEXEC	0x100		/* HW no-execute bit */
 
+/* Page size definitions */
+#define PAGE_SHIFT 12
+#define PAGE_SIZE BIT_ULL(PAGE_SHIFT)
+#define PAGE_MASK (~(PAGE_SIZE - 1))
+
 /* Is there a portable way to do this? */
 static inline void cpu_relax(void)
 {
diff --git a/tools/testing/selftests/kvm/lib/s390x/processor.c b/tools/testing/selftests/kvm/lib/s390x/processor.c
index 4ad4492eea1d..20cfe970e3e3 100644
--- a/tools/testing/selftests/kvm/lib/s390x/processor.c
+++ b/tools/testing/selftests/kvm/lib/s390x/processor.c
@@ -14,7 +14,7 @@ void virt_arch_pgd_alloc(struct kvm_vm *vm)
 {
 	vm_paddr_t paddr;
 
-	TEST_ASSERT(vm->page_size == 4096, "Unsupported page size: 0x%x",
+	TEST_ASSERT(vm->page_size == PAGE_SIZE, "Unsupported page size: 0x%x",
 		    vm->page_size);
 
 	if (vm->pgd_created)
@@ -79,7 +79,7 @@ void virt_arch_pg_map(struct kvm_vm *vm, uint64_t gva, uint64_t gpa)
 	}
 
 	/* Fill in page table entry */
-	idx = (gva >> 12) & 0x0ffu;		/* page index */
+	idx = (gva >> PAGE_SHIFT) & 0x0ffu;		/* page index */
 	if (!(entry[idx] & PAGE_INVALID))
 		fprintf(stderr,
 			"WARNING: PTE for gpa=0x%"PRIx64" already set!\n", gpa);
@@ -91,7 +91,7 @@ vm_paddr_t addr_arch_gva2gpa(struct kvm_vm *vm, vm_vaddr_t gva)
 	int ri, idx;
 	uint64_t *entry;
 
-	TEST_ASSERT(vm->page_size == 4096, "Unsupported page size: 0x%x",
+	TEST_ASSERT(vm->page_size == PAGE_SIZE, "Unsupported page size: 0x%x",
 		    vm->page_size);
 
 	entry = addr_gpa2hva(vm, vm->pgd);
@@ -103,7 +103,7 @@ vm_paddr_t addr_arch_gva2gpa(struct kvm_vm *vm, vm_vaddr_t gva)
 		entry = addr_gpa2hva(vm, entry[idx] & REGION_ENTRY_ORIGIN);
 	}
 
-	idx = (gva >> 12) & 0x0ffu;		/* page index */
+	idx = (gva >> PAGE_SHIFT) & 0x0ffu;		/* page index */
 
 	TEST_ASSERT(!(entry[idx] & PAGE_INVALID),
 		    "No page mapping for vm virtual address 0x%lx", gva);
@@ -168,7 +168,7 @@ struct kvm_vcpu *vm_arch_vcpu_add(struct kvm_vm *vm, uint32_t vcpu_id)
 	struct kvm_sregs sregs;
 	struct kvm_vcpu *vcpu;
 
-	TEST_ASSERT(vm->page_size == 4096, "Unsupported page size: 0x%x",
+	TEST_ASSERT(vm->page_size == PAGE_SIZE, "Unsupported page size: 0x%x",
 		    vm->page_size);
 
 	stack_vaddr = __vm_vaddr_alloc(vm, stack_size,
diff --git a/tools/testing/selftests/kvm/s390x/cmma_test.c b/tools/testing/selftests/kvm/s390x/cmma_test.c
index b39033844756..e32dd59703a0 100644
--- a/tools/testing/selftests/kvm/s390x/cmma_test.c
+++ b/tools/testing/selftests/kvm/s390x/cmma_test.c
@@ -17,16 +17,17 @@
 #include "kvm_util.h"
 #include "kselftest.h"
 #include "ucall_common.h"
+#include "processor.h"
 
 #define MAIN_PAGE_COUNT 512
 
 #define TEST_DATA_PAGE_COUNT 512
 #define TEST_DATA_MEMSLOT 1
-#define TEST_DATA_START_GFN 4096
+#define TEST_DATA_START_GFN PAGE_SIZE
 
 #define TEST_DATA_TWO_PAGE_COUNT 256
 #define TEST_DATA_TWO_MEMSLOT 2
-#define TEST_DATA_TWO_START_GFN 8192
+#define TEST_DATA_TWO_START_GFN (2 * PAGE_SIZE)
 
 static char cmma_value_buf[MAIN_PAGE_COUNT + TEST_DATA_PAGE_COUNT];
 
@@ -66,7 +67,7 @@ static void guest_dirty_test_data(void)
 		"	lghi 5,%[page_count]\n"
 		/* r5 += r1 */
 		"2:	agfr 5,1\n"
-		/* r2 = r1 << 12 */
+		/* r2 = r1 << PAGE_SHIFT */
 		"1:	sllg 2,1,12(0)\n"
 		/* essa(r4, r2, SET_STABLE) */
 		"	.insn rrf,0xb9ab0000,4,2,1,0\n"
diff --git a/tools/testing/selftests/kvm/s390x/memop.c b/tools/testing/selftests/kvm/s390x/memop.c
index f2df7416be84..4374b4cd2a80 100644
--- a/tools/testing/selftests/kvm/s390x/memop.c
+++ b/tools/testing/selftests/kvm/s390x/memop.c
@@ -16,6 +16,7 @@
 #include "kvm_util.h"
 #include "kselftest.h"
 #include "ucall_common.h"
+#include "processor.h"
 
 enum mop_target {
 	LOGICAL,
@@ -226,9 +227,6 @@ static void memop_ioctl(struct test_info info, struct kvm_s390_mem_op *ksmo,
 
 #define CHECK_N_DO(f, ...) ({ f(__VA_ARGS__, CHECK_ONLY); f(__VA_ARGS__); })
 
-#define PAGE_SHIFT 12
-#define PAGE_SIZE (1ULL << PAGE_SHIFT)
-#define PAGE_MASK (~(PAGE_SIZE - 1))
 #define CR0_FETCH_PROTECTION_OVERRIDE	(1UL << (63 - 38))
 #define CR0_STORAGE_PROTECTION_OVERRIDE	(1UL << (63 - 39))
 
diff --git a/tools/testing/selftests/kvm/s390x/tprot.c b/tools/testing/selftests/kvm/s390x/tprot.c
index 7a742a673b7c..12d5e1cb62e3 100644
--- a/tools/testing/selftests/kvm/s390x/tprot.c
+++ b/tools/testing/selftests/kvm/s390x/tprot.c
@@ -9,9 +9,8 @@
 #include "kvm_util.h"
 #include "kselftest.h"
 #include "ucall_common.h"
+#include "processor.h"
 
-#define PAGE_SHIFT 12
-#define PAGE_SIZE (1 << PAGE_SHIFT)
 #define CR0_FETCH_PROTECTION_OVERRIDE	(1UL << (63 - 38))
 #define CR0_STORAGE_PROTECTION_OVERRIDE	(1UL << (63 - 39))
 
@@ -151,7 +150,7 @@ static enum stage perform_next_stage(int *i, bool mapped_0)
 		 * instead.
 		 * In order to skip these tests we detect this inside the guest
 		 */
-		skip = tests[*i].addr < (void *)4096 &&
+		skip = tests[*i].addr < (void *)PAGE_SIZE &&
 		       tests[*i].expected != TRANSL_UNAVAIL &&
 		       !mapped_0;
 		if (!skip) {
-- 
2.45.2

[PATCH v2 02/10] selftests: kvm: s390: Add kvm_s390_sie_block definition for userspace tests

[Christoph Schlameuss]

Subsequent tests do require direct manipulation of the SIE instruction.
This commit introduces the SIE definition for the selftests.

There are already definitions of this within the kernel.
This differs in two ways.
* This is the first definition of this in userspace.
* This does not require atomicity for the flags.

With the userspace definition of the SIE block layout now being present
we can reuse the values in other tests where applicable.

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
---
 .../testing/selftests/kvm/include/s390x/sie.h | 240 ++++++++++++++++++
 .../testing/selftests/kvm/s390x/debug_test.c  |   4 +-
 2 files changed, 242 insertions(+), 2 deletions(-)
 create mode 100644 tools/testing/selftests/kvm/include/s390x/sie.h

diff --git a/tools/testing/selftests/kvm/include/s390x/sie.h b/tools/testing/selftests/kvm/include/s390x/sie.h
new file mode 100644
index 000000000000..160acd4a1db9
--- /dev/null
+++ b/tools/testing/selftests/kvm/include/s390x/sie.h
@@ -0,0 +1,240 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/*
+ * Definition for kernel virtual machines on s390.
+ *
+ * Adapted copy of struct definition kvm_s390_sie_block from
+ * arch/s390/include/asm/kvm_host.h for use in userspace selftest programs.
+ *
+ * Copyright IBM Corp. 2008, 2024
+ *
+ * Authors:
+ *  Christoph Schlameuss <schlameuss@linux.ibm.com>
+ *  Carsten Otte <cotte@de.ibm.com>
+ */
+
+#ifndef SELFTEST_KVM_SIE_H
+#define SELFTEST_KVM_SIE_H
+
+#include <linux/types.h>
+
+struct kvm_s390_sie_block {
+#define CPUSTAT_STOPPED    0x80000000
+#define CPUSTAT_WAIT       0x10000000
+#define CPUSTAT_ECALL_PEND 0x08000000
+#define CPUSTAT_STOP_INT   0x04000000
+#define CPUSTAT_IO_INT     0x02000000
+#define CPUSTAT_EXT_INT    0x01000000
+#define CPUSTAT_RUNNING    0x00800000
+#define CPUSTAT_RETAINED   0x00400000
+#define CPUSTAT_TIMING_SUB 0x00020000
+#define CPUSTAT_SIE_SUB    0x00010000
+#define CPUSTAT_RRF        0x00008000
+#define CPUSTAT_SLSV       0x00004000
+#define CPUSTAT_SLSR       0x00002000
+#define CPUSTAT_ZARCH      0x00000800
+#define CPUSTAT_MCDS       0x00000100
+#define CPUSTAT_KSS        0x00000200
+#define CPUSTAT_SM         0x00000080
+#define CPUSTAT_IBS        0x00000040
+#define CPUSTAT_GED2       0x00000010
+#define CPUSTAT_G          0x00000008
+#define CPUSTAT_GED        0x00000004
+#define CPUSTAT_J          0x00000002
+#define CPUSTAT_P          0x00000001
+	__u32 cpuflags;			/* 0x0000 */
+	__u32: 1;			/* 0x0004 */
+	__u32 prefix : 18;
+	__u32: 1;
+	__u32 ibc : 12;
+	__u8	reserved08[4];		/* 0x0008 */
+#define PROG_IN_SIE BIT(0)
+	__u32	prog0c;			/* 0x000c */
+	union {
+		__u8	reserved10[16];	/* 0x0010 */
+		struct {
+			__u64	pv_handle_cpu;
+			__u64	pv_handle_config;
+		};
+	};
+#define PROG_BLOCK_SIE	BIT(0)
+#define PROG_REQUEST	BIT(1)
+	__u32	prog20;			/* 0x0020 */
+	__u8	reserved24[4];		/* 0x0024 */
+	__u64	cputm;			/* 0x0028 */
+	__u64	ckc;			/* 0x0030 */
+	__u64	epoch;			/* 0x0038 */
+	__u32	svcc;			/* 0x0040 */
+#define LCTL_CR0	0x8000
+#define LCTL_CR6	0x0200
+#define LCTL_CR9	0x0040
+#define LCTL_CR10	0x0020
+#define LCTL_CR11	0x0010
+#define LCTL_CR14	0x0002
+	__u16   lctl;			/* 0x0044 */
+	__s16	icpua;			/* 0x0046 */
+#define ICTL_OPEREXC	0x80000000
+#define ICTL_PINT	0x20000000
+#define ICTL_LPSW	0x00400000
+#define ICTL_STCTL	0x00040000
+#define ICTL_ISKE	0x00004000
+#define ICTL_SSKE	0x00002000
+#define ICTL_RRBE	0x00001000
+#define ICTL_TPROT	0x00000200
+	__u32	ictl;			/* 0x0048 */
+#define ECA_CEI		0x80000000
+#define ECA_IB		0x40000000
+#define ECA_SIGPI	0x10000000
+#define ECA_MVPGI	0x01000000
+#define ECA_AIV		0x00200000
+#define ECA_VX		0x00020000
+#define ECA_PROTEXCI	0x00002000
+#define ECA_APIE	0x00000008
+#define ECA_SII		0x00000001
+	__u32	eca;			/* 0x004c */
+#define ICPT_INST	0x04
+#define ICPT_PROGI	0x08
+#define ICPT_INSTPROGI	0x0C
+#define ICPT_EXTREQ	0x10
+#define ICPT_EXTINT	0x14
+#define ICPT_IOREQ	0x18
+#define ICPT_WAIT	0x1c
+#define ICPT_VALIDITY	0x20
+#define ICPT_STOP	0x28
+#define ICPT_OPEREXC	0x2C
+#define ICPT_PARTEXEC	0x38
+#define ICPT_IOINST	0x40
+#define ICPT_KSS	0x5c
+#define ICPT_MCHKREQ	0x60
+#define ICPT_INT_ENABLE	0x64
+#define ICPT_PV_INSTR	0x68
+#define ICPT_PV_NOTIFY	0x6c
+#define ICPT_PV_PREF	0x70
+	__u8	icptcode;		/* 0x0050 */
+	__u8	icptstatus;		/* 0x0051 */
+	__u16	ihcpu;			/* 0x0052 */
+	__u8	reserved54;		/* 0x0054 */
+#define IICTL_CODE_NONE		 0x00
+#define IICTL_CODE_MCHK		 0x01
+#define IICTL_CODE_EXT		 0x02
+#define IICTL_CODE_IO		 0x03
+#define IICTL_CODE_RESTART	 0x04
+#define IICTL_CODE_SPECIFICATION 0x10
+#define IICTL_CODE_OPERAND	 0x11
+	__u8	iictl;			/* 0x0055 */
+	__u16	ipa;			/* 0x0056 */
+	__u32	ipb;			/* 0x0058 */
+	__u32	scaoh;			/* 0x005c */
+#define FPF_BPBC	0x20
+	__u8	fpf;			/* 0x0060 */
+#define ECB_GS		0x40
+#define ECB_TE		0x10
+#define ECB_SPECI	0x08
+#define ECB_SRSI	0x04
+#define ECB_HOSTPROTINT	0x02
+#define ECB_PTF		0x01
+	__u8	ecb;			/* 0x0061 */
+#define ECB2_CMMA	0x80
+#define ECB2_IEP	0x20
+#define ECB2_PFMFI	0x08
+#define ECB2_ESCA	0x04
+#define ECB2_ZPCI_LSI	0x02
+	__u8	ecb2;			/* 0x0062 */
+#define ECB3_AISI	0x20
+#define ECB3_AISII	0x10
+#define ECB3_DEA	0x08
+#define ECB3_AES	0x04
+#define ECB3_RI		0x01
+	__u8	ecb3;			/* 0x0063 */
+#define ESCA_SCAOL_MASK ~0x3fU
+	__u32	scaol;			/* 0x0064 */
+	__u8	sdf;			/* 0x0068 */
+	__u8	epdx;			/* 0x0069 */
+	__u8	cpnc;			/* 0x006a */
+	__u8	reserved6b;		/* 0x006b */
+	__u32	todpr;			/* 0x006c */
+#define GISA_FORMAT1 0x00000001
+	__u32	gd;			/* 0x0070 */
+	__u8	reserved74[12];		/* 0x0074 */
+	__u64	mso;			/* 0x0080 */
+	__u64	msl;			/* 0x0088 */
+	__u64	psw_mask;		/* 0x0090 */
+	__u64	psw_addr;		/* 0x0098 */
+	__u64	gg14;			/* 0x00a0 */
+	__u64	gg15;			/* 0x00a8 */
+	__u8	reservedb0[8];		/* 0x00b0 */
+#define HPID_KVM	0x4
+#define HPID_VSIE	0x5
+	__u8	hpid;			/* 0x00b8 */
+	__u8	reservedb9[7];		/* 0x00b9 */
+	union {
+		struct {
+			__u32	eiparams;	/* 0x00c0 */
+			__u16	extcpuaddr;	/* 0x00c4 */
+			__u16	eic;		/* 0x00c6 */
+		};
+		__u64	mcic;			/* 0x00c0 */
+	} __packed;
+	__u32	reservedc8;		/* 0x00c8 */
+	union {
+		struct {
+			__u16	pgmilc;		/* 0x00cc */
+			__u16	iprcc;		/* 0x00ce */
+		};
+		__u32	edc;			/* 0x00cc */
+	} __packed;
+	union {
+		struct {
+			__u32	dxc;		/* 0x00d0 */
+			__u16	mcn;		/* 0x00d4 */
+			__u8	perc;		/* 0x00d6 */
+			__u8	peratmid;	/* 0x00d7 */
+		};
+		__u64	faddr;			/* 0x00d0 */
+	} __packed;
+	__u64	peraddr;		/* 0x00d8 */
+	__u8	eai;			/* 0x00e0 */
+	__u8	peraid;			/* 0x00e1 */
+	__u8	oai;			/* 0x00e2 */
+	__u8	armid;			/* 0x00e3 */
+	__u8	reservede4[4];		/* 0x00e4 */
+	union {
+		__u64	tecmc;		/* 0x00e8 */
+		struct {
+			__u16	subchannel_id;	/* 0x00e8 */
+			__u16	subchannel_nr;	/* 0x00ea */
+			__u32	io_int_parm;	/* 0x00ec */
+			__u32	io_int_word;	/* 0x00f0 */
+		};
+	} __packed;
+	__u8	reservedf4[8];		/* 0x00f4 */
+#define CRYCB_FORMAT_MASK	0x00000003
+#define CRYCB_FORMAT0		0x00000000
+#define CRYCB_FORMAT1		0x00000001
+#define CRYCB_FORMAT2		0x00000003
+	__u32	crycbd;			/* 0x00fc */
+	__u64	gcr[16];		/* 0x0100 */
+	union {
+		__u64	gbea;		/* 0x0180 */
+		__u64	sidad;
+	};
+	__u8    reserved188[8];		/* 0x0188 */
+	__u64   sdnxo;			/* 0x0190 */
+	__u8    reserved198[8];		/* 0x0198 */
+	__u32	fac;			/* 0x01a0 */
+	__u8	reserved1a4[20];	/* 0x01a4 */
+	__u64	cbrlo;			/* 0x01b8 */
+	__u8	reserved1c0[8];		/* 0x01c0 */
+#define ECD_HOSTREGMGMT	0x20000000
+#define ECD_MEF		0x08000000
+#define ECD_ETOKENF	0x02000000
+#define ECD_ECC		0x00200000
+	__u32	ecd;			/* 0x01c8 */
+	__u8	reserved1cc[18];	/* 0x01cc */
+	__u64	pp;			/* 0x01de */
+	__u8	reserved1e6[2];		/* 0x01e6 */
+	__u64	itdba;			/* 0x01e8 */
+	__u64   riccbd;			/* 0x01f0 */
+	__u64	gvrd;			/* 0x01f8 */
+} __packed __aligned(512);
+
+#endif /* SELFTEST_KVM_SIE_H */
diff --git a/tools/testing/selftests/kvm/s390x/debug_test.c b/tools/testing/selftests/kvm/s390x/debug_test.c
index 84313fb27529..ad8095968601 100644
--- a/tools/testing/selftests/kvm/s390x/debug_test.c
+++ b/tools/testing/selftests/kvm/s390x/debug_test.c
@@ -2,12 +2,12 @@
 /* Test KVM debugging features. */
 #include "kvm_util.h"
 #include "test_util.h"
+#include "sie.h"
 
 #include <linux/kvm.h>
 
 #define __LC_SVC_NEW_PSW 0x1c0
 #define __LC_PGM_NEW_PSW 0x1d0
-#define ICPT_INSTRUCTION 0x04
 #define IPA0_DIAG 0x8300
 #define PGM_SPECIFICATION 0x06
 
@@ -85,7 +85,7 @@ static void test_step_pgm_diag(void)
 	vm = test_step_int_1(&vcpu, test_step_pgm_diag_guest_code,
 			     __LC_PGM_NEW_PSW, new_psw);
 	TEST_ASSERT_KVM_EXIT_REASON(vcpu, KVM_EXIT_S390_SIEIC);
-	TEST_ASSERT_EQ(vcpu->run->s390_sieic.icptcode, ICPT_INSTRUCTION);
+	TEST_ASSERT_EQ(vcpu->run->s390_sieic.icptcode, ICPT_INST);
 	TEST_ASSERT_EQ(vcpu->run->s390_sieic.ipa & 0xff00, IPA0_DIAG);
 	vcpu_ioctl(vcpu, KVM_S390_IRQ, &irq);
 	vcpu_run(vcpu);
-- 
2.45.2

[PATCH v2 03/10] selftests: kvm: s390: Add s390x ucontrol test suite with hpage test

[Christoph Schlameuss]

Add test suite to validate the s390x architecture specific ucontrol KVM
interface.

Make use of the selftest test harness.

* uc_cap_hpage testcase verifies that a ucontrol VM cannot be run with
  hugepages.

To allow testing of the ucontrol interface the kernel needs a
non-default config containing CONFIG_KVM_S390_UCONTROL.
This config needs to be set to built-in (y) as this cannot be built as
module.

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
---
 tools/testing/selftests/kvm/.gitignore        |  1 +
 tools/testing/selftests/kvm/Makefile          |  1 +
 tools/testing/selftests/kvm/s390x/config      |  2 +
 .../selftests/kvm/s390x/ucontrol_test.c       | 78 +++++++++++++++++++
 4 files changed, 82 insertions(+)
 create mode 100644 tools/testing/selftests/kvm/s390x/config
 create mode 100644 tools/testing/selftests/kvm/s390x/ucontrol_test.c

diff --git a/tools/testing/selftests/kvm/.gitignore b/tools/testing/selftests/kvm/.gitignore
index 6d9381d60172..f2a30a58cd71 100644
--- a/tools/testing/selftests/kvm/.gitignore
+++ b/tools/testing/selftests/kvm/.gitignore
@@ -5,3 +5,4 @@
 !*.h
 !*.S
 !*.sh
+!config
diff --git a/tools/testing/selftests/kvm/Makefile b/tools/testing/selftests/kvm/Makefile
index 48d32c5aa3eb..b3dc0a5bf0d4 100644
--- a/tools/testing/selftests/kvm/Makefile
+++ b/tools/testing/selftests/kvm/Makefile
@@ -186,6 +186,7 @@ TEST_GEN_PROGS_s390x += s390x/tprot
 TEST_GEN_PROGS_s390x += s390x/cmma_test
 TEST_GEN_PROGS_s390x += s390x/debug_test
 TEST_GEN_PROGS_s390x += s390x/shared_zeropage_test
+TEST_GEN_PROGS_s390x += s390x/ucontrol_test
 TEST_GEN_PROGS_s390x += demand_paging_test
 TEST_GEN_PROGS_s390x += dirty_log_test
 TEST_GEN_PROGS_s390x += guest_print_test
diff --git a/tools/testing/selftests/kvm/s390x/config b/tools/testing/selftests/kvm/s390x/config
new file mode 100644
index 000000000000..23270f2d679f
--- /dev/null
+++ b/tools/testing/selftests/kvm/s390x/config
@@ -0,0 +1,2 @@
+CONFIG_KVM=y
+CONFIG_KVM_S390_UCONTROL=y
diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
new file mode 100644
index 000000000000..a706a9f2b9ea
--- /dev/null
+++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
@@ -0,0 +1,78 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * Test code for the s390x kvm ucontrol interface
+ *
+ * Copyright IBM Corp. 2024
+ *
+ * Authors:
+ *  Christoph Schlameuss <schlameuss@linux.ibm.com>
+ */
+#include "kselftest_harness.h"
+#include "kvm_util.h"
+
+#include <linux/capability.h>
+#include <linux/sizes.h>
+
+#define SYS_ADMIN_CAP 0x200000
+
+/* so directly declare capget to check caps without libcap */
+int capget(cap_user_header_t header, cap_user_data_t data);
+
+/**
+ * In order to create user controlled virtual machines on S390,
+ * check KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL
+ * as privileged user (SYS_ADMIN).
+ */
+void require_ucontrol_admin(void)
+{
+	struct __user_cap_data_struct data[_LINUX_CAPABILITY_U32S_3];
+	struct __user_cap_header_struct hdr = {
+		.version = _LINUX_CAPABILITY_VERSION_3,
+	};
+	int rc;
+
+	rc = capget(&hdr, data);
+	TEST_ASSERT_EQ(0, rc);
+	TEST_REQUIRE((data->effective & SYS_ADMIN_CAP) > 0);
+
+	TEST_REQUIRE(kvm_has_cap(KVM_CAP_S390_UCONTROL));
+}
+
+/**
+ * Assert HPAGE CAP cannot be enabled on UCONTROL VM
+ */
+TEST(uc_cap_hpage)
+{
+	int rc, kvm_fd, vm_fd, vcpu_fd;
+	struct kvm_enable_cap cap = {
+		.cap = KVM_CAP_S390_HPAGE_1M,
+	};
+
+	require_ucontrol_admin();
+
+	kvm_fd = open_kvm_dev_path_or_exit();
+	vm_fd = ioctl(kvm_fd, KVM_CREATE_VM, KVM_VM_S390_UCONTROL);
+	ASSERT_GE(vm_fd, 0);
+
+	/* assert hpages are not supported on ucontrol vm */
+	rc = ioctl(vm_fd, KVM_CHECK_EXTENSION, KVM_CAP_S390_HPAGE_1M);
+	EXPECT_EQ(0, rc);
+
+	/* Test that KVM_CAP_S390_HPAGE_1M can't be enabled for a ucontrol vm */
+	rc = ioctl(vm_fd, KVM_ENABLE_CAP, cap);
+	EXPECT_EQ(-1, rc);
+	EXPECT_EQ(EINVAL, errno);
+
+	/* assert HPAGE CAP is rejected after vCPU creation */
+	vcpu_fd = ioctl(vm_fd, KVM_CREATE_VCPU, 0);
+	ASSERT_GE(vcpu_fd, 0);
+	rc = ioctl(vm_fd, KVM_ENABLE_CAP, cap);
+	EXPECT_EQ(-1, rc);
+	EXPECT_EQ(EBUSY, errno);
+
+	close(vcpu_fd);
+	close(vm_fd);
+	close(kvm_fd);
+}
+
+TEST_HARNESS_MAIN
-- 
2.45.2

[PATCH v2 04/10] selftests: kvm: s390: Add test fixture and simple VM setup tests

[Christoph Schlameuss]

Add a uc_kvm fixture to create and destroy a ucontrol VM.

* uc_sie_assertions asserts basic settings in the SIE as setup by the
  kernel.
* uc_attr_mem_limit asserts the memory limit is max value and cannot be
  set (not supported).
* uc_no_dirty_log asserts dirty log is not supported.

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
---
 .../selftests/kvm/s390x/ucontrol_test.c       | 129 ++++++++++++++++++
 1 file changed, 129 insertions(+)

diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
index a706a9f2b9ea..d8a81b407467 100644
--- a/tools/testing/selftests/kvm/s390x/ucontrol_test.c
+++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
@@ -9,12 +9,16 @@
  */
 #include "kselftest_harness.h"
 #include "kvm_util.h"
+#include "processor.h"
+#include "sie.h"
 
 #include <linux/capability.h>
 #include <linux/sizes.h>
 
 #define SYS_ADMIN_CAP 0x200000
 
+#define VM_MEM_SIZE (4 * SZ_1M)
+
 /* so directly declare capget to check caps without libcap */
 int capget(cap_user_header_t header, cap_user_data_t data);
 
@@ -38,6 +42,131 @@ void require_ucontrol_admin(void)
 	TEST_REQUIRE(kvm_has_cap(KVM_CAP_S390_UCONTROL));
 }
 
+FIXTURE(uc_kvm)
+{
+	struct kvm_s390_sie_block *sie_block;
+	struct kvm_run *run;
+	uintptr_t base_gpa;
+	uintptr_t code_gpa;
+	uintptr_t base_hva;
+	uintptr_t code_hva;
+	int kvm_run_size;
+	void *vm_mem;
+	int vcpu_fd;
+	int kvm_fd;
+	int vm_fd;
+};
+
+/**
+ * create VM with single vcpu, map kvm_run and SIE control block for easy access
+ */
+FIXTURE_SETUP(uc_kvm)
+{
+	struct kvm_s390_vm_cpu_processor info;
+	int rc;
+
+	require_ucontrol_admin();
+
+	self->kvm_fd = open_kvm_dev_path_or_exit();
+	self->vm_fd = ioctl(self->kvm_fd, KVM_CREATE_VM, KVM_VM_S390_UCONTROL);
+	ASSERT_GE(self->vm_fd, 0);
+
+	kvm_device_attr_get(self->vm_fd, KVM_S390_VM_CPU_MODEL,
+			    KVM_S390_VM_CPU_PROCESSOR, &info);
+	TH_LOG("create VM 0x%llx", info.cpuid);
+
+	self->vcpu_fd = ioctl(self->vm_fd, KVM_CREATE_VCPU, 0);
+	ASSERT_GE(self->vcpu_fd, 0);
+
+	self->kvm_run_size = ioctl(self->kvm_fd, KVM_GET_VCPU_MMAP_SIZE, NULL);
+	ASSERT_GE(self->kvm_run_size, sizeof(struct kvm_run))
+		  TH_LOG(KVM_IOCTL_ERROR(KVM_GET_VCPU_MMAP_SIZE, self->kvm_run_size));
+	self->run = (struct kvm_run *)mmap(NULL, self->kvm_run_size,
+		    PROT_READ | PROT_WRITE, MAP_SHARED, self->vcpu_fd, 0);
+	ASSERT_NE(self->run, MAP_FAILED);
+	/**
+	 * For virtual cpus that have been created with S390 user controlled
+	 * virtual machines, the resulting vcpu fd can be memory mapped at page
+	 * offset KVM_S390_SIE_PAGE_OFFSET in order to obtain a memory map of
+	 * the virtual cpu's hardware control block.
+	 */
+	self->sie_block = (struct kvm_s390_sie_block *)mmap(NULL, PAGE_SIZE,
+			  PROT_READ | PROT_WRITE, MAP_SHARED,
+			  self->vcpu_fd, KVM_S390_SIE_PAGE_OFFSET << PAGE_SHIFT);
+	ASSERT_NE(self->sie_block, MAP_FAILED);
+
+	TH_LOG("VM created %p %p", self->run, self->sie_block);
+
+	self->base_gpa = 0;
+	self->code_gpa = self->base_gpa + (3 * SZ_1M);
+
+	self->vm_mem = aligned_alloc(SZ_1M, VM_MEM_SIZE);
+	ASSERT_NE(NULL, self->vm_mem) TH_LOG("malloc failed %u", errno);
+	self->base_hva = (uintptr_t)self->vm_mem + SZ_1M;
+	self->code_hva = self->base_hva - self->base_gpa + self->code_gpa;
+	struct kvm_s390_ucas_mapping map = {
+		.user_addr = self->base_hva,
+		.vcpu_addr = self->base_gpa,
+		.length = VM_MEM_SIZE,
+	};
+	TH_LOG("ucas map %p %p 0x%llx",
+	       (void *)map.user_addr, (void *)map.vcpu_addr, map.length);
+	rc = ioctl(self->vcpu_fd, KVM_S390_UCAS_MAP, &map);
+	ASSERT_EQ(0, rc) TH_LOG("ucas map result %d not expected, %s",
+				rc, strerror(errno));
+
+	TH_LOG("page in %p", (void *)self->base_gpa);
+	rc = ioctl(self->vcpu_fd, KVM_S390_VCPU_FAULT, self->base_gpa);
+	ASSERT_EQ(0, rc) TH_LOG("vcpu fault (%p) result %d not expected, %s",
+				(void *)self->base_hva, rc, strerror(errno));
+}
+
+FIXTURE_TEARDOWN(uc_kvm)
+{
+	munmap(self->sie_block, PAGE_SIZE);
+	munmap(self->run, self->kvm_run_size);
+	close(self->vcpu_fd);
+	close(self->vm_fd);
+	close(self->kvm_fd);
+	free(self->vm_mem);
+}
+
+TEST_F(uc_kvm, uc_sie_assertions)
+{
+	/* assert interception of Code 08 (Program Interruption) is set */
+	EXPECT_EQ(0, self->sie_block->ecb & ECB_SPECI);
+}
+
+TEST_F(uc_kvm, uc_attr_mem_limit)
+{
+	u64 limit;
+	struct kvm_device_attr attr = {
+		.group = KVM_S390_VM_MEM_CTRL,
+		.attr = KVM_S390_VM_MEM_LIMIT_SIZE,
+		.addr = (unsigned long)&limit,
+	};
+	int rc;
+
+	rc = ioctl(self->vm_fd, KVM_GET_DEVICE_ATTR, &attr);
+	EXPECT_EQ(0, rc);
+	EXPECT_EQ(~0UL, limit);
+
+	/* assert set not supported */
+	rc = ioctl(self->vm_fd, KVM_SET_DEVICE_ATTR, &attr);
+	EXPECT_EQ(-1, rc);
+	EXPECT_EQ(EINVAL, errno);
+}
+
+TEST_F(uc_kvm, uc_no_dirty_log)
+{
+	struct kvm_dirty_log dlog;
+	int rc;
+
+	rc = ioctl(self->vm_fd, KVM_GET_DIRTY_LOG, &dlog);
+	EXPECT_EQ(-1, rc);
+	EXPECT_EQ(EINVAL, errno);
+}
+
 /**
  * Assert HPAGE CAP cannot be enabled on UCONTROL VM
  */
-- 
2.45.2

[PATCH v2 05/10] selftests: kvm: s390: Add debug print functions

[Christoph Schlameuss]

Add functions to simply print some basic state information in selftests.

The output can be enabled by setting:

    #define TH_LOG_ENABLED 1
    #define DEBUG 1

* print_psw: current SIE state description and VM run state
* print_hex_bytes: print memory with some counting markers
* print_hex: PRINT_HEX with 512 bytes
* print_run: use print_psw and print_hex to print contents of VM run
  state and SIE state description
* print_regs: print content of general and control registers

All prints use pr_debug for the output and can be configured using
DEBUG.

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
---
 .../selftests/kvm/include/s390x/debug_print.h | 69 +++++++++++++++++++
 .../selftests/kvm/s390x/ucontrol_test.c       |  1 +
 2 files changed, 70 insertions(+)
 create mode 100644 tools/testing/selftests/kvm/include/s390x/debug_print.h

diff --git a/tools/testing/selftests/kvm/include/s390x/debug_print.h b/tools/testing/selftests/kvm/include/s390x/debug_print.h
new file mode 100644
index 000000000000..1bf275631cc6
--- /dev/null
+++ b/tools/testing/selftests/kvm/include/s390x/debug_print.h
@@ -0,0 +1,69 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/*
+ * Definition for kernel virtual machines on s390x
+ *
+ * Copyright IBM Corp. 2024
+ *
+ * Authors:
+ *  Christoph Schlameuss <schlameuss@linux.ibm.com>
+ */
+
+#ifndef SELFTEST_KVM_DEBUG_PRINT_H
+#define SELFTEST_KVM_DEBUG_PRINT_H
+
+#include "asm/ptrace.h"
+#include "kvm_util.h"
+#include "sie.h"
+
+static inline void print_hex_bytes(const char *name, u64 addr, size_t len)
+{
+	u64 pos;
+
+	pr_debug("%s (%p)\n", name, (void *)addr);
+	pr_debug("            0/0x00---------|");
+	if (len > 8)
+		pr_debug(" 8/0x08---------|");
+	if (len > 16)
+		pr_debug(" 16/0x10--------|");
+	if (len > 24)
+		pr_debug(" 24/0x18--------|");
+	for (pos = 0; pos < len; pos += 8) {
+		if ((pos % 32) == 0)
+			pr_debug("\n %3lu 0x%.3lx ", pos, pos);
+		pr_debug(" %16lx", *((u64 *)(addr + pos)));
+	}
+	pr_debug("\n");
+}
+
+static inline void print_hex(const char *name, u64 addr)
+{
+	print_hex_bytes(name, addr, 512);
+}
+
+static inline void print_psw(struct kvm_run *run, struct kvm_s390_sie_block *sie_block)
+{
+	pr_debug("flags:0x%x psw:0x%.16llx:0x%.16llx exit:%u %s\n",
+		 run->flags,
+		 run->psw_mask, run->psw_addr,
+		 run->exit_reason, exit_reason_str(run->exit_reason));
+	pr_debug("sie_block psw:0x%.16llx:0x%.16llx\n",
+		 sie_block->psw_mask, sie_block->psw_addr);
+}
+
+static inline void print_run(struct kvm_run *run, struct kvm_s390_sie_block *sie_block)
+{
+	print_hex_bytes("run", (u64)run, 0x150);
+	print_hex("sie_block", (u64)sie_block);
+	print_psw(run, sie_block);
+}
+
+static inline void print_regs(struct kvm_run *run)
+{
+	struct kvm_sync_regs *sync_regs = &run->s.regs;
+
+	print_hex_bytes("GPRS", (u64)sync_regs->gprs, 8 * NUM_GPRS);
+	print_hex_bytes("ACRS", (u64)sync_regs->acrs, 4 * NUM_ACRS);
+	print_hex_bytes("CRS", (u64)sync_regs->crs, 8 * NUM_CRS);
+}
+
+#endif /* SELFTEST_KVM_DEBUG_PRINT_H */
diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
index d8a81b407467..527c431a9758 100644
--- a/tools/testing/selftests/kvm/s390x/ucontrol_test.c
+++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
@@ -7,6 +7,7 @@
  * Authors:
  *  Christoph Schlameuss <schlameuss@linux.ibm.com>
  */
+#include "debug_print.h"
 #include "kselftest_harness.h"
 #include "kvm_util.h"
 #include "processor.h"
-- 
2.45.2

[PATCH v2 06/10] selftests: kvm: s390: Add VM run test case

[Christoph Schlameuss]

Add test case running code interacting with registers within a
ucontrol VM.

* Add uc_gprs test case

The test uses the same VM setup using the fixture and debug macros
introduced in earlier patches in this series.

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
---
 .../selftests/kvm/s390x/ucontrol_test.c       | 132 ++++++++++++++++++
 1 file changed, 132 insertions(+)

diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
index 527c431a9758..c98d5a3a315b 100644
--- a/tools/testing/selftests/kvm/s390x/ucontrol_test.c
+++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
@@ -43,6 +43,23 @@ void require_ucontrol_admin(void)
 	TEST_REQUIRE(kvm_has_cap(KVM_CAP_S390_UCONTROL));
 }
 
+/* Test program setting some registers and looping */
+extern char test_gprs_pgm[];
+asm("test_gprs_pgm:\n"
+	"xgr	%r0, %r0\n"
+	"lgfi	%r1,1\n"
+	"lgfi	%r2,2\n"
+	"lgfi	%r3,3\n"
+	"lgfi	%r4,4\n"
+	"lgfi	%r5,5\n"
+	"lgfi	%r6,6\n"
+	"lgfi	%r7,7\n"
+	"0:\n"
+	"	diag	0,0,0x44\n"
+	"	ahi	%r0,1\n"
+	"	j	0b\n"
+);
+
 FIXTURE(uc_kvm)
 {
 	struct kvm_s390_sie_block *sie_block;
@@ -205,4 +222,119 @@ TEST(uc_cap_hpage)
 	close(kvm_fd);
 }
 
+/* verify SIEIC exit
+ * * reset stop requests
+ * * fail on codes not expected in the test cases
+ */
+static bool uc_handle_sieic(FIXTURE_DATA(uc_kvm) * self)
+{
+	struct kvm_s390_sie_block *sie_block = self->sie_block;
+	struct kvm_run *run = self->run;
+
+	/* check SIE interception code */
+	pr_info("sieic: 0x%2x 0x%4x 0x%4x\n",
+		run->s390_sieic.icptcode,
+		run->s390_sieic.ipa,
+		run->s390_sieic.ipb);
+	switch (run->s390_sieic.icptcode) {
+	case ICPT_STOP:
+		/* stopped via sie V P --> ignore */
+		/* reset stop request */
+		sie_block->cpuflags = sie_block->cpuflags & ~CPUSTAT_STOP_INT;
+		pr_info("sie V P - cleared %.4x\n", sie_block->cpuflags);
+		break;
+	case ICPT_INST:
+		/* end execution in caller on intercepted instruction */
+		return false;
+	default:
+		TEST_FAIL("UNEXPECTED SIEIC CODE %d", run->s390_sieic.icptcode);
+	}
+	return true;
+}
+
+/* verify VM state on exit */
+static bool uc_handle_exit(FIXTURE_DATA(uc_kvm) * self)
+{
+	struct kvm_run *run = self->run;
+
+	switch (run->exit_reason) {
+	case KVM_EXIT_S390_SIEIC:
+		return uc_handle_sieic(self);
+	default:
+		pr_info("exit_reason %2d not handled\n", run->exit_reason);
+	}
+	return true;
+}
+
+/* run the VM until interrupted */
+static int uc_run_once(FIXTURE_DATA(uc_kvm) * self)
+{
+	int rc;
+
+	rc = ioctl(self->vcpu_fd, KVM_RUN, NULL);
+	print_run(self->run, self->sie_block);
+	print_regs(self->run);
+	pr_debug("run %d / %d %s\n", rc, errno, strerror(errno));
+	return rc;
+}
+
+static void uc_assert_diag44(FIXTURE_DATA(uc_kvm) * self)
+{
+	struct kvm_s390_sie_block *sie_block = self->sie_block;
+
+	/* assert vm was interrupted by diag 0x0044 */
+	TEST_ASSERT_EQ(KVM_EXIT_S390_SIEIC, self->run->exit_reason);
+	TEST_ASSERT_EQ(ICPT_INST, sie_block->icptcode);
+	TEST_ASSERT_EQ(0x8300, sie_block->ipa);
+	TEST_ASSERT_EQ(0x440000, sie_block->ipb);
+}
+
+TEST_F(uc_kvm, uc_gprs)
+{
+	struct kvm_s390_sie_block *sie_block = self->sie_block;
+	struct kvm_sync_regs *sync_regs = &self->run->s.regs;
+	struct kvm_run *run = self->run;
+	struct kvm_regs regs = {};
+
+	/* Set registers to values that are different from the ones that we expect below */
+	for (int i = 0; i < 8; i++)
+		sync_regs->gprs[i] = 8;
+	run->kvm_dirty_regs |= KVM_SYNC_GPRS;
+
+	/* copy test_gprs_pgm to code_hva / code_gpa */
+	TH_LOG("copy code %p to vm mapped memory %p / %p",
+	       &test_gprs_pgm, (void *)self->code_hva, (void *)self->code_gpa);
+	memcpy((void *)self->code_hva, &test_gprs_pgm, PAGE_SIZE);
+
+	run->psw_mask = 0x0000000180000000ULL;  /* DAT disabled + 64 bit mode */
+	run->psw_addr = self->code_gpa;
+
+	/* run VM and expect immediate stop */
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(true, uc_handle_exit(self));
+	ASSERT_EQ(self->code_gpa, sie_block->psw_addr);
+
+	/* run and expect interception of diag 44 */
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(false, uc_handle_exit(self));
+	uc_assert_diag44(self);
+
+	/* Retrieve and check guest register values */
+	ASSERT_EQ(0, ioctl(self->vcpu_fd, KVM_GET_REGS, &regs));
+	for (int i = 0; i < 8; i++) {
+		ASSERT_EQ(i, regs.gprs[i]);
+		ASSERT_EQ(i, sync_regs->gprs[i]);
+	}
+
+	/* run and expect interception of diag 44 again */
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(false, uc_handle_exit(self));
+	uc_assert_diag44(self);
+
+	/* check continued increment of register 0 value */
+	ASSERT_EQ(0, ioctl(self->vcpu_fd, KVM_GET_REGS, &regs));
+	ASSERT_EQ(1, regs.gprs[0]);
+	ASSERT_EQ(1, sync_regs->gprs[0]);
+}
+
 TEST_HARNESS_MAIN
-- 
2.45.2

[PATCH v2 07/10] selftests: kvm: s390: Add uc_map_unmap VM test case

[Christoph Schlameuss]

Add a test case verifying basic running and interaction of ucontrol VMs.
Fill the segment and page tables for allocated memory and map memory on
first access.

* uc_map_unmap
  Store and load data to mapped and unmapped memory and use pic segment
  translation handling to map memory on access.

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
---
 .../selftests/kvm/s390x/ucontrol_test.c       | 171 +++++++++++++++++-
 1 file changed, 170 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
index c98d5a3a315b..a39c560ba42e 100644
--- a/tools/testing/selftests/kvm/s390x/ucontrol_test.c
+++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
@@ -18,7 +18,13 @@
 
 #define SYS_ADMIN_CAP 0x200000
 
+#define UC_PIC_SEGMENT_TRANSLATION 0x10
+
 #define VM_MEM_SIZE (4 * SZ_1M)
+#define VM_MEM_EXT_SIZE (2 * SZ_1M)
+#define VM_MEM_MAX (VM_MEM_SIZE + VM_MEM_EXT_SIZE)
+
+#define PAGES_PER_SEGMENT 4
 
 /* so directly declare capget to check caps without libcap */
 int capget(cap_user_header_t header, cap_user_data_t data);
@@ -60,6 +66,23 @@ asm("test_gprs_pgm:\n"
 	"	j	0b\n"
 );
 
+/* Test program manipulating memory */
+extern char test_mem_pgm[];
+asm("test_mem_pgm:\n"
+	"xgr	%r0, %r0\n"
+
+	"0:\n"
+	"	ahi	%r0,1\n"
+	"	st	%r1,0(%r5,%r6)\n"
+
+	"	xgr	%r1, %r1\n"
+	"	l	%r1,0(%r5,%r6)\n"
+	"	ahi	%r0,1\n"
+	"	diag	0,0,0x44\n"
+
+	"	j	0b\n"
+);
+
 FIXTURE(uc_kvm)
 {
 	struct kvm_s390_sie_block *sie_block;
@@ -69,6 +92,7 @@ FIXTURE(uc_kvm)
 	uintptr_t base_hva;
 	uintptr_t code_hva;
 	int kvm_run_size;
+	vm_paddr_t pgd;
 	void *vm_mem;
 	int vcpu_fd;
 	int kvm_fd;
@@ -118,7 +142,7 @@ FIXTURE_SETUP(uc_kvm)
 	self->base_gpa = 0;
 	self->code_gpa = self->base_gpa + (3 * SZ_1M);
 
-	self->vm_mem = aligned_alloc(SZ_1M, VM_MEM_SIZE);
+	self->vm_mem = aligned_alloc(SZ_1M, VM_MEM_SIZE + VM_MEM_EXT_SIZE);
 	ASSERT_NE(NULL, self->vm_mem) TH_LOG("malloc failed %u", errno);
 	self->base_hva = (uintptr_t)self->vm_mem + SZ_1M;
 	self->code_hva = self->base_hva - self->base_gpa + self->code_gpa;
@@ -222,6 +246,75 @@ TEST(uc_cap_hpage)
 	close(kvm_fd);
 }
 
+/* calculate host virtual addr from guest physical addr */
+static void *gpa2hva(FIXTURE_DATA(uc_kvm) * self, u64 gpa)
+{
+	return (void *)(self->base_hva - self->base_gpa + gpa);
+}
+
+/* initialize segment and page tables for uc_kvm tests */
+static void init_st_pt(FIXTURE_DATA(uc_kvm) * self)
+{
+	struct kvm_sync_regs *sync_regs = &self->run->s.regs;
+	struct kvm_run *run = self->run;
+	void *se_addr;
+	int si, pi;
+	u64 *phd;
+
+	self->pgd = self->base_gpa + SZ_1M; /* set PASCE addr */
+	phd = gpa2hva(self, self->pgd);
+	memset(phd, 0xff, PAGES_PER_SEGMENT * PAGE_SIZE);
+
+	for (si = 0; si < ((VM_MEM_SIZE + VM_MEM_EXT_SIZE) / SZ_1M); si++) {
+		/* create ste */
+		phd[si] = (self->pgd
+			+ (PAGES_PER_SEGMENT * PAGE_SIZE
+				* ((VM_MEM_SIZE + VM_MEM_EXT_SIZE) / SZ_1M))
+			+ (PAGES_PER_SEGMENT * PAGE_SIZE * si)) & ~0x7fful;
+		se_addr = gpa2hva(self, phd[si]);
+		memset(se_addr, 0xff, PAGES_PER_SEGMENT * PAGE_SIZE);
+		for (pi = 0; pi < (SZ_1M / PAGE_SIZE); pi++) {
+			/* create pte */
+			((u64 *)se_addr)[pi] = (self->base_gpa
+				+ (si * SZ_1M) + (pi * PAGE_SIZE)) & ~0xffful;
+		}
+	}
+	pr_debug("segment table entry %p (0x%lx) --> %p\n",
+		 phd, phd[0], gpa2hva(self, (phd[0] & ~0x7fful)));
+	print_hex_bytes("st", (u64)phd, 64);
+	print_hex_bytes("pt", (u64)gpa2hva(self, phd[0]), 128);
+	print_hex_bytes("pt+", (u64)
+			gpa2hva(self, phd[0] + (PAGES_PER_SEGMENT * PAGE_SIZE
+			* ((VM_MEM_SIZE + VM_MEM_EXT_SIZE) / SZ_1M)) - 0x64), 128);
+
+	sync_regs->crs[1] = self->pgd | 0x3;	/* PASCE TT=00 for segment table */
+	run->kvm_dirty_regs |= KVM_SYNC_CRS;
+}
+
+static void uc_handle_exit_ucontrol(FIXTURE_DATA(uc_kvm) * self)
+{
+	struct kvm_run *run = self->run;
+
+	TEST_ASSERT_EQ(KVM_EXIT_S390_UCONTROL, run->exit_reason);
+	switch (run->s390_ucontrol.pgm_code) {
+	case UC_PIC_SEGMENT_TRANSLATION:
+		pr_info("ucontrol pic segment translation 0x%llx\n",
+			run->s390_ucontrol.trans_exc_code);
+		/* map / make additional memory available */
+		struct kvm_s390_ucas_mapping map2 = {
+			.user_addr = (u64)gpa2hva(self, run->s390_ucontrol.trans_exc_code),
+			.vcpu_addr = run->s390_ucontrol.trans_exc_code,
+			.length = VM_MEM_EXT_SIZE,
+		};
+		pr_info("ucas map %p %p 0x%llx\n",
+			(void *)map2.user_addr, (void *)map2.vcpu_addr, map2.length);
+		TEST_ASSERT_EQ(0, ioctl(self->vcpu_fd, KVM_S390_UCAS_MAP, &map2));
+		break;
+	default:
+		TEST_FAIL("UNEXPECTED PGM CODE %d", run->s390_ucontrol.pgm_code);
+	}
+}
+
 /* verify SIEIC exit
  * * reset stop requests
  * * fail on codes not expected in the test cases
@@ -245,7 +338,11 @@ static bool uc_handle_sieic(FIXTURE_DATA(uc_kvm) * self)
 		break;
 	case ICPT_INST:
 		/* end execution in caller on intercepted instruction */
+		pr_info("sie instruction interception\n");
 		return false;
+	case ICPT_OPEREXC:
+		/* operation exception */
+		TEST_FAIL("sie exception on %.4x%.8x", sie_block->ipa, sie_block->ipb);
 	default:
 		TEST_FAIL("UNEXPECTED SIEIC CODE %d", run->s390_sieic.icptcode);
 	}
@@ -258,6 +355,11 @@ static bool uc_handle_exit(FIXTURE_DATA(uc_kvm) * self)
 	struct kvm_run *run = self->run;
 
 	switch (run->exit_reason) {
+	case KVM_EXIT_S390_UCONTROL:
+		/* check program interruption code */
+		/* handle page fault --> ucas map */
+		uc_handle_exit_ucontrol(self);
+		break;
 	case KVM_EXIT_S390_SIEIC:
 		return uc_handle_sieic(self);
 	default:
@@ -289,6 +391,73 @@ static void uc_assert_diag44(FIXTURE_DATA(uc_kvm) * self)
 	TEST_ASSERT_EQ(0x440000, sie_block->ipb);
 }
 
+TEST_F(uc_kvm, uc_map_unmap)
+{
+	struct kvm_sync_regs *sync_regs = &self->run->s.regs;
+	struct kvm_s390_sie_block *sie_block = self->sie_block;
+	struct kvm_run *run = self->run;
+	int rc;
+
+	init_st_pt(self);
+
+	/* copy test_mem_pgm to code_hva / code_gpa */
+	TH_LOG("copy code %p to vm mapped memory %p / %p",
+	       &test_mem_pgm, (void *)self->code_hva, (void *)self->code_gpa);
+	memcpy((void *)self->code_hva, &test_mem_pgm, PAGE_SIZE);
+
+	run->psw_mask = 0x0400000180000000ULL;  /* DAT enabled + 64 bit mode */
+	run->psw_addr = self->code_gpa;
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(0, sync_regs->gprs[0]);
+	ASSERT_EQ(KVM_EXIT_S390_SIEIC, run->exit_reason);
+	ASSERT_EQ(ICPT_STOP, sie_block->icptcode);
+	ASSERT_EQ(true, uc_handle_exit(self));
+
+	/* set register content for test_mem_pgm to access not mapped memory*/
+	sync_regs->gprs[1] = 0x55;
+	sync_regs->gprs[5] = self->base_gpa;
+	sync_regs->gprs[6] = VM_MEM_SIZE;
+	run->kvm_dirty_regs |= KVM_SYNC_GPRS;
+
+	/* run and expect to fail witch ucontrol pic segment translation */
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(1, sync_regs->gprs[0]);
+	ASSERT_EQ(KVM_EXIT_S390_UCONTROL, run->exit_reason);
+
+	ASSERT_EQ(UC_PIC_SEGMENT_TRANSLATION, run->s390_ucontrol.pgm_code);
+	ASSERT_EQ(self->base_gpa + VM_MEM_SIZE, run->s390_ucontrol.trans_exc_code);
+	/* map / make additional memory available */
+	struct kvm_s390_ucas_mapping map2 = {
+		.user_addr = (u64)gpa2hva(self, self->base_gpa + VM_MEM_SIZE),
+		.vcpu_addr = self->base_gpa + VM_MEM_SIZE,
+		.length = VM_MEM_EXT_SIZE,
+	};
+	TH_LOG("ucas map %p %p 0x%llx",
+	       (void *)map2.user_addr, (void *)map2.vcpu_addr, map2.length);
+	rc = ioctl(self->vcpu_fd, KVM_S390_UCAS_MAP, &map2);
+	ASSERT_EQ(0, rc)
+		TH_LOG("ucas map result %d not expected, %s", rc, strerror(errno));
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(false, uc_handle_exit(self));
+	uc_assert_diag44(self);
+
+	/* assert registers and memory are in expected state */
+	ASSERT_EQ(2, sync_regs->gprs[0]);
+	ASSERT_EQ(0x55, sync_regs->gprs[1]);
+	ASSERT_EQ(0x55, *(u32 *)gpa2hva(self, self->base_gpa + VM_MEM_SIZE));
+
+	/* unmap and run loop again */
+	TH_LOG("ucas unmap %p %p 0x%llx",
+	       (void *)map2.user_addr, (void *)map2.vcpu_addr, map2.length);
+	rc = ioctl(self->vcpu_fd, KVM_S390_UCAS_UNMAP, &map2);
+	ASSERT_EQ(0, rc)
+		TH_LOG("ucas map result %d not expected, %s", rc, strerror(errno));
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(3, sync_regs->gprs[0]);
+	ASSERT_EQ(KVM_EXIT_S390_UCONTROL, run->exit_reason);
+	ASSERT_EQ(true, uc_handle_exit(self));
+}
+
 TEST_F(uc_kvm, uc_gprs)
 {
 	struct kvm_s390_sie_block *sie_block = self->sie_block;
-- 
2.45.2

[PATCH v2 08/10] selftests: kvm: s390: Add uc_skey VM test case

[Christoph Schlameuss]

Add a test case manipulating s390 storage keys from within the ucontrol
VM.

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
---
 .../selftests/kvm/s390x/ucontrol_test.c       | 85 ++++++++++++++++++-
 1 file changed, 84 insertions(+), 1 deletion(-)

diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
index a39c560ba42e..79992621c0b5 100644
--- a/tools/testing/selftests/kvm/s390x/ucontrol_test.c
+++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
@@ -83,6 +83,32 @@ asm("test_mem_pgm:\n"
 	"	j	0b\n"
 );
 
+/* Test program manipulating storage keys */
+extern char test_skey_pgm[];
+asm("test_skey_pgm:\n"
+	"xgr	%r0, %r0\n"
+
+	"0:\n"
+	"	ahi	%r0,1\n"
+	"	st	%r1,0(%r5,%r6)\n"
+
+	"	iske	%r1,%r6\n"
+	"	ahi	%r0,1\n"
+	"	diag	0,0,0x44\n"
+
+	"	sske	%r1,%r6\n"
+	"	iske	%r1,%r6\n"
+	"	ahi	%r0,1\n"
+	"	diag	0,0,0x44\n"
+
+	"	rrbe	%r1,%r6\n"
+	"	iske	%r1,%r6\n"
+	"	ahi	%r0,1\n"
+	"	diag	0,0,0x44\n"
+
+	"	j	0b\n"
+);
+
 FIXTURE(uc_kvm)
 {
 	struct kvm_s390_sie_block *sie_block;
@@ -391,10 +417,67 @@ static void uc_assert_diag44(FIXTURE_DATA(uc_kvm) * self)
 	TEST_ASSERT_EQ(0x440000, sie_block->ipb);
 }
 
-TEST_F(uc_kvm, uc_map_unmap)
+TEST_F(uc_kvm, uc_skey)
 {
+	u64 test_vaddr = self->base_gpa + VM_MEM_SIZE - (SZ_1M / 2);
+	struct kvm_s390_sie_block *sie_block = self->sie_block;
 	struct kvm_sync_regs *sync_regs = &self->run->s.regs;
+	struct kvm_run *run = self->run;
+	u8 skeyvalue = 0x34;
+
+	init_st_pt(self);
+
+	/* copy test_skey_pgm to code_hva / code_gpa */
+	TH_LOG("copy code %p to vm mapped memory %p / %p",
+	       &test_skey_pgm, (void *)self->code_hva, (void *)self->code_gpa);
+	memcpy((void *)self->code_hva, &test_skey_pgm, PAGE_SIZE);
+
+	/* set register content for test_skey_pgm to access not mapped memory */
+	sync_regs->gprs[1] = skeyvalue;
+	sync_regs->gprs[5] = self->base_gpa;
+	sync_regs->gprs[6] = test_vaddr;
+	run->kvm_dirty_regs |= KVM_SYNC_GPRS;
+
+	run->kvm_dirty_regs |= KVM_SYNC_CRS;
+	TH_LOG("set CR0 to 0x%llx", sync_regs->crs[0]);
+
+	self->sie_block->ictl |= ICTL_OPEREXC | ICTL_PINT;
+	self->sie_block->cpuflags &= ~CPUSTAT_KSS;
+	run->psw_mask = 0x0400000180000000ULL;  /* DAT enabled + 64 bit mode */
+	run->psw_addr = self->code_gpa;
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(true, uc_handle_exit(self));
+	ASSERT_EQ(0, sync_regs->gprs[0]);
+	ASSERT_EQ(KVM_EXIT_S390_SIEIC, run->exit_reason);
+	ASSERT_EQ(ICPT_STOP, sie_block->icptcode);
+
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(false, uc_handle_exit(self));
+	ASSERT_EQ(2, sync_regs->gprs[0]);
+	ASSERT_EQ(0x06, sync_regs->gprs[1]);
+	uc_assert_diag44(self);
+
+	sync_regs->gprs[1] = skeyvalue;
+	run->kvm_dirty_regs |= KVM_SYNC_GPRS;
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(false, uc_handle_exit(self));
+	ASSERT_EQ(3, sync_regs->gprs[0]);
+	ASSERT_EQ(skeyvalue, sync_regs->gprs[1]);
+	uc_assert_diag44(self);
+
+	sync_regs->gprs[1] = skeyvalue;
+	run->kvm_dirty_regs |= KVM_SYNC_GPRS;
+	ASSERT_EQ(0, uc_run_once(self));
+	ASSERT_EQ(false, uc_handle_exit(self));
+	ASSERT_EQ(4, sync_regs->gprs[0]);
+	ASSERT_EQ(skeyvalue & 0xfb, sync_regs->gprs[1]);
+	uc_assert_diag44(self);
+}
+
+TEST_F(uc_kvm, uc_map_unmap)
+{
 	struct kvm_s390_sie_block *sie_block = self->sie_block;
+	struct kvm_sync_regs *sync_regs = &self->run->s.regs;
 	struct kvm_run *run = self->run;
 	int rc;
 
-- 
2.45.2

[PATCH v2 09/10] selftests: kvm: s390: Verify reject memory region operations for ucontrol VMs

[Christoph Schlameuss]

Add a test case verifying KVM_SET_USER_MEMORY_REGION and
KVM_SET_USER_MEMORY_REGION2 cannot be executed on ucontrol VMs.

Executing this test case on not patched kernels will cause a null
pointer dereference in the host kernel.
This is fixed with commit:
commit 7816e58967d0 ("kvm: s390: Reject memory region operations for ucontrol VMs")

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
---
 .../selftests/kvm/s390x/ucontrol_test.c       | 22 +++++++++++++++++++
 1 file changed, 22 insertions(+)

diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
index 79992621c0b5..853d8c6cea5c 100644
--- a/tools/testing/selftests/kvm/s390x/ucontrol_test.c
+++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
@@ -589,4 +589,26 @@ TEST_F(uc_kvm, uc_gprs)
 	ASSERT_EQ(1, sync_regs->gprs[0]);
 }
 
+TEST_F(uc_kvm, uc_no_user_region)
+{
+	struct kvm_userspace_memory_region region = {
+		.slot = 1,
+		.guest_phys_addr = self->code_gpa,
+		.memory_size = VM_MEM_EXT_SIZE,
+		.userspace_addr = (uintptr_t)self->code_hva,
+	};
+	struct kvm_userspace_memory_region2 region2 = {
+		.slot = 1,
+		.guest_phys_addr = self->code_gpa,
+		.memory_size = VM_MEM_EXT_SIZE,
+		.userspace_addr = (uintptr_t)self->code_hva,
+	};
+
+	ASSERT_EQ(-1, ioctl(self->vm_fd, KVM_SET_USER_MEMORY_REGION, &region));
+	ASSERT_EQ(EINVAL, errno);
+
+	ASSERT_EQ(-1, ioctl(self->vm_fd, KVM_SET_USER_MEMORY_REGION2, &region2));
+	ASSERT_EQ(EINVAL, errno);
+}
+
 TEST_HARNESS_MAIN
-- 
2.45.2

[PATCH v2 10/10] s390: Enable KVM_S390_UCONTROL config in debug_defconfig

[Christoph Schlameuss]

To simplify testing enable UCONTROL KVM by default in debug kernels.

Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
---
 arch/s390/Kconfig.debug | 3 +++
 1 file changed, 3 insertions(+)

diff --git a/arch/s390/Kconfig.debug b/arch/s390/Kconfig.debug
index c4300ea4abf8..4798dded6d76 100644
--- a/arch/s390/Kconfig.debug
+++ b/arch/s390/Kconfig.debug
@@ -20,3 +20,6 @@ config CIO_INJECT
 	  This option provides a debugging facility to inject certain artificial events
 	  and instruction responses to the CIO layer of Linux kernel. The newly created
 	  debugfs user-interfaces will be at /sys/kernel/debug/s390/cio/*
+
+config KVM_S390_UCONTROL
+	def_bool y
-- 
2.45.2

Re: [PATCH v2 10/10] s390: Enable KVM_S390_UCONTROL config in debug_defconfig

[Christian Borntraeger]

Am 23.07.24 um 11:31 schrieb Christoph Schlameuss:
> To simplify testing enable UCONTROL KVM by default in debug kernels.
> 
> Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
> ---
>   arch/s390/Kconfig.debug | 3 +++
>   1 file changed, 3 insertions(+)
> 
> diff --git a/arch/s390/Kconfig.debug b/arch/s390/Kconfig.debug
> index c4300ea4abf8..4798dded6d76 100644
> --- a/arch/s390/Kconfig.debug
> +++ b/arch/s390/Kconfig.debug
> @@ -20,3 +20,6 @@ config CIO_INJECT
>   	  This option provides a debugging facility to inject certain artificial events
>   	  and instruction responses to the CIO layer of Linux kernel. The newly created
>   	  debugfs user-interfaces will be at /sys/kernel/debug/s390/cio/*
> +
> +config KVM_S390_UCONTROL
> +	def_bool y

Isnt that enabling ucontrol for all configs? YOu probably wanted to patch
arch/s390/configs/debug_defconfig

Re: [PATCH v2 10/10] s390: Enable KVM_S390_UCONTROL config in debug_defconfig

[Christoph Schlameuss]

On Tue, 23 Jul 2024 11:40:35 +0200
Christian Borntraeger <borntraeger@linux.ibm.com> wrote:

> Am 23.07.24 um 11:31 schrieb Christoph Schlameuss:
> > To simplify testing enable UCONTROL KVM by default in debug kernels.
> > 
> > Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
> > ---
> >   arch/s390/Kconfig.debug | 3 +++
> >   1 file changed, 3 insertions(+)
> > 
> > diff --git a/arch/s390/Kconfig.debug b/arch/s390/Kconfig.debug
> > index c4300ea4abf8..4798dded6d76 100644
> > --- a/arch/s390/Kconfig.debug
> > +++ b/arch/s390/Kconfig.debug
> > @@ -20,3 +20,6 @@ config CIO_INJECT
> >   	  This option provides a debugging facility to inject certain artificial events
> >   	  and instruction responses to the CIO layer of Linux kernel. The newly created
> >   	  debugfs user-interfaces will be at /sys/kernel/debug/s390/cio/*
> > +
> > +config KVM_S390_UCONTROL
> > +	def_bool y  
> 
> Isnt that enabling ucontrol for all configs? YOu probably wanted to patch
> arch/s390/configs/debug_defconfig

Yes, I double checked. You are of cause right.
I did verify that is generated into the ".config" with "make
debug_defconfig". But then I missed to check that it does not get added
with "make defconfig".

I will correct that in the next version.

Christoph

Re: [PATCH v2 01/10] selftests: kvm: s390: Define page sizes in shared header

[Janosch Frank]

On 7/23/24 11:31, Christoph Schlameuss wrote:
> Multiple test cases need page size and shift definitions.
> By moving the definitions to a single architecture specific header we
> limit the repetition.
> 
> Make use of PAGE_SIZE, PAGE_SHIFT and PAGE_MASK defines in existing
> code.
> 
> Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
> Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>

Reviewed-by: Janosch Frank <frankja@linux.ibm.com>

Re: [PATCH v2 02/10] selftests: kvm: s390: Add kvm_s390_sie_block definition for userspace tests

[Janosch Frank]

On 7/23/24 11:31, Christoph Schlameuss wrote:
> Subsequent tests do require direct manipulation of the SIE instruction.
> This commit introduces the SIE definition for the selftests.

AFAIK ucontrol doesn't allow to manipulate the instruction, it allows 
manipulation of the *SIE control block* that's handed over as an operand 
to SIE.


> 
> There are already definitions of this within the kernel.
> This differs in two ways.
> * This is the first definition of this in userspace.
> * This does not require atomicity for the flags.
> 
> With the userspace definition of the SIE block layout now being present
> we can reuse the values in other tests where applicable.
> 
> Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>

Code looks fine though.

Re: [PATCH v2 03/10] selftests: kvm: s390: Add s390x ucontrol test suite with hpage test

[Janosch Frank]

On 7/23/24 11:31, Christoph Schlameuss wrote:
> Add test suite to validate the s390x architecture specific ucontrol KVM
> interface.
> 
> Make use of the selftest test harness.
> 
> * uc_cap_hpage testcase verifies that a ucontrol VM cannot be run with
>    hugepages.
> 
> To allow testing of the ucontrol interface the kernel needs a
> non-default config containing CONFIG_KVM_S390_UCONTROL.
> This config needs to be set to built-in (y) as this cannot be built as
> module.
> 
> Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
> Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
> ---

[...]

> +#include "kselftest_harness.h"
> +#include "kvm_util.h"
> +
> +#include <linux/capability.h>
> +#include <linux/sizes.h>
> +
> +#define SYS_ADMIN_CAP 0x200000

This looked suspicious to me.
Surely this would be available in some form in capability.h since 
CAP_SYS_ADMIN is something that's regularly checked.

[...]

> +
> +/* so directly declare capget to check caps without libcap */
> +int capget(cap_user_header_t header, cap_user_data_t data);
> +
> +/**
> + * In order to create user controlled virtual machines on S390,
> + * check KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL
> + * as privileged user (SYS_ADMIN).
> + */
> +void require_ucontrol_admin(void)
> +{
> +	struct __user_cap_data_struct data[_LINUX_CAPABILITY_U32S_3];
> +	struct __user_cap_header_struct hdr = {
> +		.version = _LINUX_CAPABILITY_VERSION_3,
> +	};
> +	int rc;
> +
> +	rc = capget(&hdr, data);
> +	TEST_ASSERT_EQ(0, rc);
> +	TEST_REQUIRE((data->effective & SYS_ADMIN_CAP) > 0);

And in fact capability.h does have defines which hide the magic constant:
data->effective & CAP_TO_MASK(CAP_SYS_ADMIN)


> +
> +	TEST_REQUIRE(kvm_has_cap(KVM_CAP_S390_UCONTROL));
> +}
> +

Re: [PATCH v2 02/10] selftests: kvm: s390: Add kvm_s390_sie_block definition for userspace tests

[Christoph Schlameuss]

On Wed, 24 Jul 2024 16:39:28 +0200
Janosch Frank <frankja@linux.ibm.com> wrote:

> On 7/23/24 11:31, Christoph Schlameuss wrote:
> > Subsequent tests do require direct manipulation of the SIE instruction.
> > This commit introduces the SIE definition for the selftests.  
> 
> AFAIK ucontrol doesn't allow to manipulate the instruction, it allows 
> manipulation of the *SIE control block* that's handed over as an operand 
> to SIE.
> 

I will clarify that.

Christoph

> > 
> > There are already definitions of this within the kernel.
> > This differs in two ways.
> > * This is the first definition of this in userspace.
> > * This does not require atomicity for the flags.
> > 
> > With the userspace definition of the SIE block layout now being present
> > we can reuse the values in other tests where applicable.
> > 
> > Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>  
> 
> Code looks fine though.

Re: [PATCH v2 03/10] selftests: kvm: s390: Add s390x ucontrol test suite with hpage test

[Christoph Schlameuss]

On Wed, 24 Jul 2024 16:52:00 +0200
Janosch Frank <frankja@linux.ibm.com> wrote:

> On 7/23/24 11:31, Christoph Schlameuss wrote:
> > Add test suite to validate the s390x architecture specific ucontrol KVM
> > interface.
> > 
> > Make use of the selftest test harness.
> > 
> > * uc_cap_hpage testcase verifies that a ucontrol VM cannot be run with
> >    hugepages.
> > 
> > To allow testing of the ucontrol interface the kernel needs a
> > non-default config containing CONFIG_KVM_S390_UCONTROL.
> > This config needs to be set to built-in (y) as this cannot be built as
> > module.
> > 
> > Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
> > Reviewed-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
> > ---  
> 
> [...]
> 
> > +#include "kselftest_harness.h"
> > +#include "kvm_util.h"
> > +
> > +#include <linux/capability.h>
> > +#include <linux/sizes.h>
> > +
> > +#define SYS_ADMIN_CAP 0x200000  
> 
> This looked suspicious to me.
> Surely this would be available in some form in capability.h since 
> CAP_SYS_ADMIN is something that's regularly checked.
> 
> [...]
> 
> > +
> > +/* so directly declare capget to check caps without libcap */
> > +int capget(cap_user_header_t header, cap_user_data_t data);
> > +
> > +/**
> > + * In order to create user controlled virtual machines on S390,
> > + * check KVM_CAP_S390_UCONTROL and use the flag KVM_VM_S390_UCONTROL
> > + * as privileged user (SYS_ADMIN).
> > + */
> > +void require_ucontrol_admin(void)
> > +{
> > +	struct __user_cap_data_struct data[_LINUX_CAPABILITY_U32S_3];
> > +	struct __user_cap_header_struct hdr = {
> > +		.version = _LINUX_CAPABILITY_VERSION_3,
> > +	};
> > +	int rc;
> > +
> > +	rc = capget(&hdr, data);
> > +	TEST_ASSERT_EQ(0, rc);
> > +	TEST_REQUIRE((data->effective & SYS_ADMIN_CAP) > 0);  
> 
> And in fact capability.h does have defines which hide the magic constant:
> data->effective & CAP_TO_MASK(CAP_SYS_ADMIN)
> 

Yes, thank you, that is much better. Will use that instead and remove
the custom SYS_ADMIN_CAP define above.

> > +
> > +	TEST_REQUIRE(kvm_has_cap(KVM_CAP_S390_UCONTROL));
> > +}
> > +  
> 

Re: [PATCH v2 06/10] selftests: kvm: s390: Add VM run test case

[Janosch Frank]

On 7/23/24 11:31, Christoph Schlameuss wrote:
> Add test case running code interacting with registers within a
> ucontrol VM.
> 
> * Add uc_gprs test case
> 
> The test uses the same VM setup using the fixture and debug macros
> introduced in earlier patches in this series.
> 
> Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
> ---
>   .../selftests/kvm/s390x/ucontrol_test.c       | 132 ++++++++++++++++++
>   1 file changed, 132 insertions(+)
> 
> diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
> index 527c431a9758..c98d5a3a315b 100644
> --- a/tools/testing/selftests/kvm/s390x/ucontrol_test.c
> +++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
> @@ -43,6 +43,23 @@ void require_ucontrol_admin(void)
>   	TEST_REQUIRE(kvm_has_cap(KVM_CAP_S390_UCONTROL));
>   }
>   
> +/* Test program setting some registers and looping */
> +extern char test_gprs_pgm[];
> +asm("test_gprs_pgm:\n"
> +	"xgr	%r0, %r0\n"
> +	"lgfi	%r1,1\n"
[...]

Naming something PGM for handling anything else than Program Exceptions 
is not recommendable. PGM has been a stable name on s390 for code 
related to PGM exceptions

When first reading this I expected to find some kind of exception code.

Re: [PATCH v2 06/10] selftests: kvm: s390: Add VM run test case

[Christoph Schlameuss]

On Thu, 25 Jul 2024 14:03:12 +0200
Janosch Frank <frankja@linux.ibm.com> wrote:

> On 7/23/24 11:31, Christoph Schlameuss wrote:
> > Add test case running code interacting with registers within a
> > ucontrol VM.
> > 
> > * Add uc_gprs test case
> > 
> > The test uses the same VM setup using the fixture and debug macros
> > introduced in earlier patches in this series.
> > 
> > Signed-off-by: Christoph Schlameuss <schlameuss@linux.ibm.com>
> > ---
> >   .../selftests/kvm/s390x/ucontrol_test.c       | 132 ++++++++++++++++++
> >   1 file changed, 132 insertions(+)
> > 
> > diff --git a/tools/testing/selftests/kvm/s390x/ucontrol_test.c b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
> > index 527c431a9758..c98d5a3a315b 100644
> > --- a/tools/testing/selftests/kvm/s390x/ucontrol_test.c
> > +++ b/tools/testing/selftests/kvm/s390x/ucontrol_test.c
> > @@ -43,6 +43,23 @@ void require_ucontrol_admin(void)
> >   	TEST_REQUIRE(kvm_has_cap(KVM_CAP_S390_UCONTROL));
> >   }
> >   
> > +/* Test program setting some registers and looping */
> > +extern char test_gprs_pgm[];
> > +asm("test_gprs_pgm:\n"
> > +	"xgr	%r0, %r0\n"
> > +	"lgfi	%r1,1\n"  
> [...]
> 
> Naming something PGM for handling anything else than Program Exceptions 
> is not recommendable. PGM has been a stable name on s390 for code 
> related to PGM exceptions
> 
> When first reading this I expected to find some kind of exception code.
> 

Ok, I get it now. Previously I was thinking only "test_pgm" was the
hangup. But I do understand now.
I will rename these here and in the other patches to "test_xxx_asm" to
hopefully reduce the confusion.

Christoph