KVM: selftests: Add ucall pool based implementation

To play nice with guests whose stack memory is encrypted, e.g. AMD SEV,
introduce a new "ucall pool" implementation that passes the ucall struct
via dedicated memory (which can be mapped shared, a.k.a. as plain text).

Because not all architectures have access to the vCPU index in the guest,
use a bitmap with atomic accesses to track which entries in the pool are
free/used.  A list+lock could also work in theory, but synchronizing the
individual pointers to the guest would be a mess.

Note, there's no need to rewalk the bitmap to ensure success.  If all
vCPUs are simply allocating, success is guaranteed because there are
enough entries for all vCPUs.  If one or more vCPUs are freeing and then
reallocating, success is guaranteed because vCPUs _always_ walk the
bitmap from 0=>N; if vCPU frees an entry and then wins a race to
re-allocate, then either it will consume the entry it just freed (bit is
the first free bit), or the losing vCPU is guaranteed to see the freed
bit (winner consumes an earlier bit, which the loser hasn't yet visited).

Reviewed-by: Andrew Jones <[email protected]>
Signed-off-by: Peter Gonda <[email protected]>
Co-developed-by: Sean Christopherson <[email protected]>
Signed-off-by: Sean Christopherson <[email protected]>
Link: https://lore.kernel.org/r/[email protected]

1 files changed, 68 insertions, 4 deletions

diff --git a/tools/testing/selftests/kvm/lib/ucall_common.c b/tools/testing/selftests/kvm/lib/ucall_common.c
index ced480860746..fcae96461e46 100644
--- a/tools/testing/selftests/kvm/lib/ucall_common.c
+++ b/tools/testing/selftests/kvm/lib/ucall_common.c
@@ -1,22 +1,86 @@
 // SPDX-License-Identifier: GPL-2.0-only
 #include "kvm_util.h"
+#include "linux/types.h"
+#include "linux/bitmap.h"
+#include "linux/atomic.h"
+
+struct ucall_header {
+	DECLARE_BITMAP(in_use, KVM_MAX_VCPUS);
+	struct ucall ucalls[KVM_MAX_VCPUS];
+};
+
+/*
+ * ucall_pool holds per-VM values (global data is duplicated by each VM), it
+ * must not be accessed from host code.
+ */
+static struct ucall_header *ucall_pool;
+
+void ucall_init(struct kvm_vm *vm, vm_paddr_t mmio_gpa)
+{
+	struct ucall_header *hdr;
+	struct ucall *uc;
+	vm_vaddr_t vaddr;
+	int i;
+
+	vaddr = vm_vaddr_alloc(vm, sizeof(*hdr), KVM_UTIL_MIN_VADDR);
+	hdr = (struct ucall_header *)addr_gva2hva(vm, vaddr);
+	memset(hdr, 0, sizeof(*hdr));
+
+	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+		uc = &hdr->ucalls[i];
+		uc->hva = uc;
+	}
+
+	write_guest_global(vm, ucall_pool, (struct ucall_header *)vaddr);
+
+	ucall_arch_init(vm, mmio_gpa);
+}
+
+static struct ucall *ucall_alloc(void)
+{
+	struct ucall *uc;
+	int i;
+
+	GUEST_ASSERT(ucall_pool);
+
+	for (i = 0; i < KVM_MAX_VCPUS; ++i) {
+		if (!atomic_test_and_set_bit(i, ucall_pool->in_use)) {
+			uc = &ucall_pool->ucalls[i];
+			memset(uc->args, 0, sizeof(uc->args));
+			return uc;
+		}
+	}
+
+	GUEST_ASSERT(0);
+	return NULL;
+}
+
+static void ucall_free(struct ucall *uc)
+{
+	/* Beware, here be pointer arithmetic.  */
+	clear_bit(uc - ucall_pool->ucalls, ucall_pool->in_use);
+}
 
 void ucall(uint64_t cmd, int nargs, ...)
 {
-	struct ucall uc = {};
+	struct ucall *uc;
 	va_list va;
 	int i;
 
-	WRITE_ONCE(uc.cmd, cmd);
+	uc = ucall_alloc();
+
+	WRITE_ONCE(uc->cmd, cmd);
 
 	nargs = min(nargs, UCALL_MAX_ARGS);
 
 	va_start(va, nargs);
 	for (i = 0; i < nargs; ++i)
-		WRITE_ONCE(uc.args[i], va_arg(va, uint64_t));
+		WRITE_ONCE(uc->args[i], va_arg(va, uint64_t));
 	va_end(va);
 
-	ucall_arch_do_ucall((vm_vaddr_t)&uc);
+	ucall_arch_do_ucall((vm_vaddr_t)uc->hva);
+
+	ucall_free(uc);
 }
 
 uint64_t get_ucall(struct kvm_vcpu *vcpu, struct ucall *uc)