kvm.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Claudio Imbrenda <imbrenda@linux.ibm.com>
To: kvm@vger.kernel.org
Cc: cohuck@redhat.com, borntraeger@de.ibm.com, frankja@linux.ibm.com,
	thuth@redhat.com, pasic@linux.ibm.com, david@redhat.com,
	linux-s390@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH v6 13/17] KVM: s390: pv: cleanup leftover protected VMs if needed
Date: Fri,  3 Dec 2021 17:58:10 +0100	[thread overview]
Message-ID: <20211203165814.73016-14-imbrenda@linux.ibm.com> (raw)
In-Reply-To: <20211203165814.73016-1-imbrenda@linux.ibm.com>

In upcoming patches it will be possible to start tearing down a
protected VM, and finish the teardown concurrently in a different
thread.

Protected VMs that are pending for tear down ("leftover") need to be
cleaned properly when the userspace process (e.g. qemu) terminates.

This patch makes sure that all "leftover" protected VMs are always
properly torn down.

Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
---
 arch/s390/include/asm/kvm_host.h |  2 +
 arch/s390/kvm/kvm-s390.c         |  1 +
 arch/s390/kvm/pv.c               | 69 ++++++++++++++++++++++++++++++++
 3 files changed, 72 insertions(+)

diff --git a/arch/s390/include/asm/kvm_host.h b/arch/s390/include/asm/kvm_host.h
index 8b288cff2a3a..948c9d4b0b51 100644
--- a/arch/s390/include/asm/kvm_host.h
+++ b/arch/s390/include/asm/kvm_host.h
@@ -921,6 +921,8 @@ struct kvm_s390_pv {
 	u64 guest_len;
 	unsigned long stor_base;
 	void *stor_var;
+	void *async_deinit;
+	struct list_head need_cleanup;
 	struct mmu_notifier mmu_notifier;
 };
 
diff --git a/arch/s390/kvm/kvm-s390.c b/arch/s390/kvm/kvm-s390.c
index 850d86995083..1b83fd3b1322 100644
--- a/arch/s390/kvm/kvm-s390.c
+++ b/arch/s390/kvm/kvm-s390.c
@@ -2817,6 +2817,7 @@ int kvm_arch_init_vm(struct kvm *kvm, unsigned long type)
 	kvm_s390_vsie_init(kvm);
 	if (use_gisa)
 		kvm_s390_gisa_init(kvm);
+	INIT_LIST_HEAD(&kvm->arch.pv.need_cleanup);
 	KVM_EVENT(3, "vm 0x%pK created by pid %u", kvm, current->pid);
 
 	return 0;
diff --git a/arch/s390/kvm/pv.c b/arch/s390/kvm/pv.c
index a54f6e98fa6f..825d69601fe5 100644
--- a/arch/s390/kvm/pv.c
+++ b/arch/s390/kvm/pv.c
@@ -16,6 +16,19 @@
 #include <linux/sched/mm.h>
 #include "kvm-s390.h"
 
+/**
+ * @struct deferred_priv
+ * Represents a "leftover" protected VM that does not correspond to any
+ * active KVM VM.
+ */
+struct deferred_priv {
+	struct list_head list;
+	unsigned long old_table;
+	u64 handle;
+	void *stor_var;
+	unsigned long stor_base;
+};
+
 static void kvm_s390_clear_pv_state(struct kvm *kvm)
 {
 	kvm->arch.pv.handle = 0;
@@ -165,6 +178,60 @@ static int kvm_s390_pv_alloc_vm(struct kvm *kvm)
 	return -ENOMEM;
 }
 
+/**
+ * kvm_s390_pv_cleanup_deferred - Clean up one leftover protected VM.
+ * @kvm the KVM that was associated with this leftover protected VM
+ * @deferred details about the leftover protected VM that needs a clean up
+ * Return: 0 in case of success, otherwise 1
+ */
+static int kvm_s390_pv_cleanup_deferred(struct kvm *kvm, struct deferred_priv *deferred)
+{
+	u16 rc, rrc;
+	int cc;
+
+	cc = uv_cmd_nodata(deferred->handle, UVC_CMD_DESTROY_SEC_CONF, &rc, &rrc);
+	KVM_UV_EVENT(kvm, 3, "PROTVIRT DESTROY VM: rc %x rrc %x", rc, rrc);
+	WARN_ONCE(cc, "protvirt destroy vm failed rc %x rrc %x", rc, rrc);
+	if (cc)
+		return cc;
+	/*
+	 * Intentionally leak unusable memory. If the UVC fails, the memory
+	 * used for the VM and its metadata is permanently unusable.
+	 * This can only happen in case of a serious KVM or hardware bug; it
+	 * is not expected to happen in normal operation.
+	 */
+	free_pages(deferred->stor_base, get_order(uv_info.guest_base_stor_len));
+	free_pages(deferred->old_table, CRST_ALLOC_ORDER);
+	vfree(deferred->stor_var);
+	return 0;
+}
+
+/**
+ * kvm_s390_pv_cleanup_leftovers - Clean up all leftover protected VMs.
+ * @kvm the KVM whose leftover protected VMs are to be cleaned up
+ * Return: 0 in case of success, otherwise 1
+ */
+static int kvm_s390_pv_cleanup_leftovers(struct kvm *kvm)
+{
+	struct deferred_priv *deferred;
+	int cc = 0;
+
+	if (kvm->arch.pv.async_deinit)
+		list_add(kvm->arch.pv.async_deinit, &kvm->arch.pv.need_cleanup);
+
+	while (!list_empty(&kvm->arch.pv.need_cleanup)) {
+		deferred = list_first_entry(&kvm->arch.pv.need_cleanup, typeof(*deferred), list);
+		if (kvm_s390_pv_cleanup_deferred(kvm, deferred))
+			cc = 1;
+		else
+			atomic_dec(&kvm->mm->context.protected_count);
+		list_del(&deferred->list);
+		kfree(deferred);
+	}
+	kvm->arch.pv.async_deinit = NULL;
+	return cc;
+}
+
 /* this should not fail, but if it does, we must not free the donated memory */
 int kvm_s390_pv_deinit_vm(struct kvm *kvm, u16 *rc, u16 *rrc)
 {
@@ -192,6 +259,8 @@ int kvm_s390_pv_deinit_vm(struct kvm *kvm, u16 *rc, u16 *rrc)
 	KVM_UV_EVENT(kvm, 3, "PROTVIRT DESTROY VM: rc %x rrc %x", *rc, *rrc);
 	WARN_ONCE(cc, "protvirt destroy vm failed rc %x rrc %x", *rc, *rrc);
 
+	cc |= kvm_s390_pv_cleanup_leftovers(kvm);
+
 	return cc ? -EIO : 0;
 }
 
-- 
2.31.1


  parent reply	other threads:[~2021-12-03 16:59 UTC|newest]

Thread overview: 23+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-12-03 16:57 [PATCH v6 00/17] KVM: s390: pv: implement lazy destroy for reboot Claudio Imbrenda
2021-12-03 16:57 ` [PATCH v6 01/17] KVM: s390: pv: leak the topmost page table when destroy fails Claudio Imbrenda
2021-12-03 16:57 ` [PATCH v6 02/17] KVM: s390: pv: handle secure storage violations for protected guests Claudio Imbrenda
2022-01-13  9:54   ` Janosch Frank
2021-12-03 16:58 ` [PATCH v6 03/17] KVM: s390: pv: handle secure storage exceptions for normal guests Claudio Imbrenda
2022-01-13  9:58   ` Janosch Frank
2021-12-03 16:58 ` [PATCH v6 04/17] KVM: s390: pv: refactor s390_reset_acc Claudio Imbrenda
2021-12-03 16:58 ` [PATCH v6 05/17] KVM: s390: pv: usage counter instead of flag Claudio Imbrenda
2021-12-03 16:58 ` [PATCH v6 06/17] KVM: s390: pv: add export before import Claudio Imbrenda
2021-12-03 16:58 ` [PATCH v6 07/17] KVM: s390: pv: module parameter to fence lazy destroy Claudio Imbrenda
2021-12-03 16:58 ` [PATCH v6 08/17] KVM: s390: pv: make kvm_s390_cpus_from_pv global Claudio Imbrenda
2021-12-03 16:58 ` [PATCH v6 09/17] KVM: s390: pv: clear the state without memset Claudio Imbrenda
2022-01-13 10:30   ` Janosch Frank
2021-12-03 16:58 ` [PATCH v6 10/17] KVM: s390: pv: add mmu_notifier Claudio Imbrenda
2021-12-04  2:32   ` kernel test robot
2021-12-03 16:58 ` [PATCH v6 11/17] s390/mm: KVM: pv: when tearing down, try to destroy protected pages Claudio Imbrenda
2022-01-13 10:38   ` Janosch Frank
2021-12-03 16:58 ` [PATCH v6 12/17] KVM: s390: pv: refactoring of kvm_s390_pv_deinit_vm Claudio Imbrenda
2021-12-03 16:58 ` Claudio Imbrenda [this message]
2021-12-03 16:58 ` [PATCH v6 14/17] KVM: s390: pv: asynchronous destroy for reboot Claudio Imbrenda
2021-12-03 16:58 ` [PATCH v6 15/17] KVM: s390: pv: api documentation for asynchronous destroy Claudio Imbrenda
2021-12-03 16:58 ` [PATCH v6 16/17] KVM: s390: pv: add KVM_CAP_S390_PROT_REBOOT_ASYNC Claudio Imbrenda
2021-12-03 16:58 ` [PATCH v6 17/17] KVM: s390: pv: avoid export before import if possible Claudio Imbrenda

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=20211203165814.73016-14-imbrenda@linux.ibm.com \
    --to=imbrenda@linux.ibm.com \
    --cc=borntraeger@de.ibm.com \
    --cc=cohuck@redhat.com \
    --cc=david@redhat.com \
    --cc=frankja@linux.ibm.com \
    --cc=kvm@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-s390@vger.kernel.org \
    --cc=pasic@linux.ibm.com \
    --cc=thuth@redhat.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).