qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Cc: Chao Peng <chao.p.peng@linux.intel.com>,
	Xiaoyao Li <xiaoyao.li@intel.com>
Subject: [PULL 55/63] kvm: handle KVM_EXIT_MEMORY_FAULT
Date: Tue, 23 Apr 2024 17:09:43 +0200	[thread overview]
Message-ID: <20240423150951.41600-56-pbonzini@redhat.com> (raw)
In-Reply-To: <20240423150951.41600-1-pbonzini@redhat.com>

From: Chao Peng <chao.p.peng@linux.intel.com>

Upon an KVM_EXIT_MEMORY_FAULT exit, userspace needs to do the memory
conversion on the RAMBlock to turn the memory into desired attribute,
switching between private and shared.

Currently only KVM_MEMORY_EXIT_FLAG_PRIVATE in flags is valid when
KVM_EXIT_MEMORY_FAULT happens.

Note, KVM_EXIT_MEMORY_FAULT makes sense only when the RAMBlock has
guest_memfd memory backend.

Note, KVM_EXIT_MEMORY_FAULT returns with -EFAULT, so special handling is
added.

When page is converted from shared to private, the original shared
memory can be discarded via ram_block_discard_range(). Note, shared
memory can be discarded only when it's not back'ed by hugetlb because
hugetlb is supposed to be pre-allocated and no need for discarding.

Signed-off-by: Chao Peng <chao.p.peng@linux.intel.com>
Co-developed-by: Xiaoyao Li <xiaoyao.li@intel.com>
Signed-off-by: Xiaoyao Li <xiaoyao.li@intel.com>

Message-ID: <20240320083945.991426-13-michael.roth@amd.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 include/sysemu/kvm.h   |  2 +
 accel/kvm/kvm-all.c    | 98 +++++++++++++++++++++++++++++++++++++-----
 accel/kvm/trace-events |  2 +
 3 files changed, 92 insertions(+), 10 deletions(-)

diff --git a/include/sysemu/kvm.h b/include/sysemu/kvm.h
index 217f3fe17ba..47f9e8be1b3 100644
--- a/include/sysemu/kvm.h
+++ b/include/sysemu/kvm.h
@@ -542,4 +542,6 @@ int kvm_create_guest_memfd(uint64_t size, uint64_t flags, Error **errp);
 int kvm_set_memory_attributes_private(hwaddr start, uint64_t size);
 int kvm_set_memory_attributes_shared(hwaddr start, uint64_t size);
 
+int kvm_convert_memory(hwaddr start, hwaddr size, bool to_private);
+
 #endif
diff --git a/accel/kvm/kvm-all.c b/accel/kvm/kvm-all.c
index f49b2b95b54..9eef2c64003 100644
--- a/accel/kvm/kvm-all.c
+++ b/accel/kvm/kvm-all.c
@@ -2900,6 +2900,69 @@ static void kvm_eat_signals(CPUState *cpu)
     } while (sigismember(&chkset, SIG_IPI));
 }
 
+int kvm_convert_memory(hwaddr start, hwaddr size, bool to_private)
+{
+    MemoryRegionSection section;
+    ram_addr_t offset;
+    MemoryRegion *mr;
+    RAMBlock *rb;
+    void *addr;
+    int ret = -1;
+
+    trace_kvm_convert_memory(start, size, to_private ? "shared_to_private" : "private_to_shared");
+
+    if (!QEMU_PTR_IS_ALIGNED(start, qemu_real_host_page_size()) ||
+        !QEMU_PTR_IS_ALIGNED(size, qemu_real_host_page_size())) {
+        return -1;
+    }
+
+    if (!size) {
+        return -1;
+    }
+
+    section = memory_region_find(get_system_memory(), start, size);
+    mr = section.mr;
+    if (!mr) {
+        return -1;
+    }
+
+    if (!memory_region_has_guest_memfd(mr)) {
+        error_report("Converting non guest_memfd backed memory region "
+                     "(0x%"HWADDR_PRIx" ,+ 0x%"HWADDR_PRIx") to %s",
+                     start, size, to_private ? "private" : "shared");
+        goto out_unref;
+    }
+
+    if (to_private) {
+        ret = kvm_set_memory_attributes_private(start, size);
+    } else {
+        ret = kvm_set_memory_attributes_shared(start, size);
+    }
+    if (ret) {
+        goto out_unref;
+    }
+
+    addr = memory_region_get_ram_ptr(mr) + section.offset_within_region;
+    rb = qemu_ram_block_from_host(addr, false, &offset);
+
+    if (to_private) {
+        if (rb->page_size != qemu_real_host_page_size()) {
+            /*
+             * shared memory is backed by hugetlb, which is supposed to be
+             * pre-allocated and doesn't need to be discarded
+             */
+            goto out_unref;
+        }
+        ret = ram_block_discard_range(rb, offset, size);
+    } else {
+        ret = ram_block_discard_guest_memfd_range(rb, offset, size);
+    }
+
+out_unref:
+    memory_region_unref(mr);
+    return ret;
+}
+
 int kvm_cpu_exec(CPUState *cpu)
 {
     struct kvm_run *run = cpu->kvm_run;
@@ -2967,18 +3030,20 @@ int kvm_cpu_exec(CPUState *cpu)
                 ret = EXCP_INTERRUPT;
                 break;
             }
-            fprintf(stderr, "error: kvm run failed %s\n",
-                    strerror(-run_ret));
+            if (!(run_ret == -EFAULT && run->exit_reason == KVM_EXIT_MEMORY_FAULT)) {
+                fprintf(stderr, "error: kvm run failed %s\n",
+                        strerror(-run_ret));
 #ifdef TARGET_PPC
-            if (run_ret == -EBUSY) {
-                fprintf(stderr,
-                        "This is probably because your SMT is enabled.\n"
-                        "VCPU can only run on primary threads with all "
-                        "secondary threads offline.\n");
-            }
+                if (run_ret == -EBUSY) {
+                    fprintf(stderr,
+                            "This is probably because your SMT is enabled.\n"
+                            "VCPU can only run on primary threads with all "
+                            "secondary threads offline.\n");
+                }
 #endif
-            ret = -1;
-            break;
+                ret = -1;
+                break;
+            }
         }
 
         trace_kvm_run_exit(cpu->cpu_index, run->exit_reason);
@@ -3061,6 +3126,19 @@ int kvm_cpu_exec(CPUState *cpu)
                 break;
             }
             break;
+        case KVM_EXIT_MEMORY_FAULT:
+            trace_kvm_memory_fault(run->memory_fault.gpa,
+                                   run->memory_fault.size,
+                                   run->memory_fault.flags);
+            if (run->memory_fault.flags & ~KVM_MEMORY_EXIT_FLAG_PRIVATE) {
+                error_report("KVM_EXIT_MEMORY_FAULT: Unknown flag 0x%" PRIx64,
+                             (uint64_t)run->memory_fault.flags);
+                ret = -1;
+                break;
+            }
+            ret = kvm_convert_memory(run->memory_fault.gpa, run->memory_fault.size,
+                                     run->memory_fault.flags & KVM_MEMORY_EXIT_FLAG_PRIVATE);
+            break;
         default:
             ret = kvm_arch_handle_exit(cpu, run);
             break;
diff --git a/accel/kvm/trace-events b/accel/kvm/trace-events
index e8c52cb9e7a..681ccb667d6 100644
--- a/accel/kvm/trace-events
+++ b/accel/kvm/trace-events
@@ -31,3 +31,5 @@ kvm_cpu_exec(void) ""
 kvm_interrupt_exit_request(void) ""
 kvm_io_window_exit(void) ""
 kvm_run_exit_system_event(int cpu_index, uint32_t event_type) "cpu_index %d, system_even_type %"PRIu32
+kvm_convert_memory(uint64_t start, uint64_t size, const char *msg) "start 0x%" PRIx64 " size 0x%" PRIx64 " %s"
+kvm_memory_fault(uint64_t start, uint64_t size, uint64_t flags) "start 0x%" PRIx64 " size 0x%" PRIx64 " flags 0x%" PRIx64
-- 
2.44.0




  parent reply	other threads:[~2024-04-23 15:16 UTC|newest]

Thread overview: 83+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-04-23 15:08 [PULL 00/63] First batch of i386 and build system patch for QEMU 9.1 Paolo Bonzini
2024-04-23 15:08 ` [PULL 01/63] meson: do not link pixman automatically into all targets Paolo Bonzini
2024-04-23 15:08 ` [PULL 02/63] tests: only build plugins if TCG is enabled Paolo Bonzini
2024-04-23 15:08 ` [PULL 03/63] ebpf: Restrict to system emulation Paolo Bonzini
2024-04-23 15:08 ` [PULL 04/63] tests/unit: match some unit tests to corresponding feature switches Paolo Bonzini
2024-04-23 15:08 ` [PULL 05/63] yank: only build if needed Paolo Bonzini
2024-04-23 15:08 ` [PULL 06/63] util/qemu-config: Extract QMP commands to qemu-config-qmp.c Paolo Bonzini
2024-04-23 15:08 ` [PULL 07/63] hw/core: Move system emulation files to system_ss Paolo Bonzini
2024-04-23 15:08 ` [PULL 08/63] hw: Include minimal source set in user emulation build Paolo Bonzini
2024-04-23 15:08 ` [PULL 09/63] stubs: remove obsolete stubs Paolo Bonzini
2024-04-23 15:08 ` [PULL 10/63] hw/usb: move stubs out of stubs/ Paolo Bonzini
2024-04-23 15:08 ` [PULL 11/63] hw/virtio: " Paolo Bonzini
2024-08-03  2:29   ` Michael Tokarev
2024-09-05 16:27     ` Paolo Bonzini
2024-09-06  7:08       ` Michael Tokarev
2024-09-06  7:30         ` Paolo Bonzini
2024-04-23 15:09 ` [PULL 12/63] semihosting: " Paolo Bonzini
2024-04-23 15:09 ` [PULL 13/63] ramfb: " Paolo Bonzini
2024-04-23 15:09 ` [PULL 14/63] memory-device: " Paolo Bonzini
2024-04-23 15:09 ` [PULL 15/63] colo: " Paolo Bonzini
2024-04-23 15:09 ` [PULL 16/63] stubs: split record/replay stubs further Paolo Bonzini
2024-04-23 15:09 ` [PULL 17/63] stubs: include stubs only if needed Paolo Bonzini
2024-06-04 10:07   ` Daniel P. Berrangé
2024-06-05 14:46     ` Zhao Liu
2024-04-23 15:09 ` [PULL 18/63] stubs: move monitor_fdsets_cleanup with other fdset stubs Paolo Bonzini
2024-04-23 15:09 ` [PULL 19/63] vga: optimize computation of dirty memory region Paolo Bonzini
2024-04-23 15:09 ` [PULL 20/63] vga: move dirty memory region code together Paolo Bonzini
2024-04-23 15:09 ` [PULL 21/63] kvm: use configs/ definition to conditionalize debug support Paolo Bonzini
2024-04-23 15:09 ` [PULL 22/63] hw: Add compat machines for 9.1 Paolo Bonzini
2024-04-23 15:09 ` [PULL 23/63] target/i386: add guest-phys-bits cpu property Paolo Bonzini
2024-04-23 15:09 ` [PULL 24/63] kvm: add support for guest physical bits Paolo Bonzini
2024-04-23 15:09 ` [PULL 25/63] i386/kvm: Move architectural CPUID leaf generation to separate helper Paolo Bonzini
2024-04-23 15:29   ` Xiaoyao Li
2024-04-23 15:09 ` [PULL 26/63] target/i386: Introduce Icelake-Server-v7 to enable TSX Paolo Bonzini
2024-04-23 15:09 ` [PULL 27/63] target/i386: Add new CPU model SierraForest Paolo Bonzini
2024-04-23 15:09 ` [PULL 28/63] target/i386: Export RFDS bit to guests Paolo Bonzini
2024-04-23 15:09 ` [PULL 29/63] pci-host/q35: Move PAM initialization above SMRAM initialization Paolo Bonzini
2025-08-12 14:20   ` Michael Tokarev
2025-08-12 14:40     ` Michael Tokarev
2024-04-23 15:09 ` [PULL 30/63] q35: Introduce smm_ranges property for q35-pci-host Paolo Bonzini
2025-08-12 14:45   ` Philippe Mathieu-Daudé
2025-08-12 15:27   ` Michael Tokarev
2025-09-17 14:23     ` Michael Tokarev
2025-09-18  9:20       ` Gerd Hoffmann
2025-09-18  9:27         ` Michael Tokarev
2025-09-18 10:25           ` Gerd Hoffmann
2024-04-23 15:09 ` [PULL 31/63] hw/i386/acpi: Set PCAT_COMPAT bit only when pic is not disabled Paolo Bonzini
2024-04-23 15:09 ` [PULL 32/63] confidential guest support: Add kvm_init() and kvm_reset() in class Paolo Bonzini
2024-04-23 15:09 ` [PULL 33/63] i386/sev: Switch to use confidential_guest_kvm_init() Paolo Bonzini
2024-04-23 15:09 ` [PULL 34/63] ppc/pef: switch to use confidential_guest_kvm_init/reset() Paolo Bonzini
2024-04-23 15:09 ` [PULL 35/63] s390: Switch to use confidential_guest_kvm_init() Paolo Bonzini
2024-04-23 15:09 ` [PULL 36/63] scripts/update-linux-headers: Add setup_data.h to import list Paolo Bonzini
2024-04-23 15:09 ` [PULL 37/63] scripts/update-linux-headers: Add bits.h to file imports Paolo Bonzini
2024-04-23 15:09 ` [PULL 38/63] linux-headers: update to current kvm/next Paolo Bonzini
2024-04-23 15:09 ` [PULL 39/63] runstate: skip initial CPU reset if reset is not actually possible Paolo Bonzini
2024-04-23 15:09 ` [PULL 40/63] KVM: track whether guest state is encrypted Paolo Bonzini
2024-04-23 15:09 ` [PULL 41/63] KVM: remove kvm_arch_cpu_check_are_resettable Paolo Bonzini
2024-04-23 15:09 ` [PULL 42/63] target/i386: introduce x86-confidential-guest Paolo Bonzini
2024-04-23 15:09 ` [PULL 43/63] target/i386: Implement mc->kvm_type() to get VM type Paolo Bonzini
2024-04-24  8:36   ` Xiaoyao Li
2024-04-23 15:09 ` [PULL 44/63] target/i386: SEV: use KVM_SEV_INIT2 if possible Paolo Bonzini
2024-04-23 15:09 ` [PULL 45/63] i386/sev: Add 'legacy-vm-type' parameter for SEV guest objects Paolo Bonzini
2024-04-23 15:09 ` [PULL 46/63] hw/i386/sev: Use legacy SEV VM types for older machine types Paolo Bonzini
2024-04-23 15:09 ` [PULL 47/63] trace/kvm: Split address space and slot id in trace_kvm_set_user_memory() Paolo Bonzini
2024-04-23 15:09 ` [PULL 48/63] kvm: Introduce support for memory_attributes Paolo Bonzini
2024-04-23 15:09 ` [PULL 49/63] RAMBlock: Add support of KVM private guest memfd Paolo Bonzini
2024-04-23 15:09 ` [PULL 50/63] kvm: Enable KVM_SET_USER_MEMORY_REGION2 for memslot Paolo Bonzini
2024-04-23 15:09 ` [PULL 51/63] kvm/memory: Make memory type private by default if it has guest memfd backend Paolo Bonzini
2024-04-23 15:09 ` [PULL 52/63] HostMem: Add mechanism to opt in kvm guest memfd via MachineState Paolo Bonzini
2024-04-23 15:09 ` [PULL 53/63] RAMBlock: make guest_memfd require uncoordinated discard Paolo Bonzini
2024-04-23 15:09 ` [PULL 54/63] physmem: Introduce ram_block_discard_guest_memfd_range() Paolo Bonzini
2024-04-23 15:09 ` Paolo Bonzini [this message]
2024-04-26 13:40   ` [PULL 55/63] kvm: handle KVM_EXIT_MEMORY_FAULT Peter Maydell
2024-04-30 19:06     ` Paolo Bonzini
2024-04-23 15:09 ` [PULL 56/63] kvm/tdx: Don't complain when converting vMMIO region to shared Paolo Bonzini
2024-04-23 15:09 ` [PULL 57/63] kvm/tdx: Ignore memory conversion to shared of unassigned region Paolo Bonzini
2024-04-23 15:09 ` [PULL 58/63] target/i386/host-cpu: Consolidate the use of warn_report_once() Paolo Bonzini
2024-04-23 15:09 ` [PULL 59/63] target/i386/cpu: " Paolo Bonzini
2024-04-23 15:09 ` [PULL 60/63] target/i386/cpu: Merge the warning and error messages for AMD HT check Paolo Bonzini
2024-04-23 15:09 ` [PULL 61/63] accel/tcg/icount-common: Consolidate the use of warn_report_once() Paolo Bonzini
2024-04-23 15:09 ` [PULL 62/63] pythondeps.toml: warn about updates needed to docs/requirements.txt Paolo Bonzini
2024-04-23 15:09 ` [PULL 63/63] target/i386/translate.c: always write 32-bits for SGDT and SIDT Paolo Bonzini
2024-04-24  4:26 ` [PULL 00/63] First batch of i386 and build system patch for QEMU 9.1 Richard Henderson

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=20240423150951.41600-56-pbonzini@redhat.com \
    --to=pbonzini@redhat.com \
    --cc=chao.p.peng@linux.intel.com \
    --cc=qemu-devel@nongnu.org \
    --cc=xiaoyao.li@intel.com \
    /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).