[xen staging] xen/livepatch: register livepatch regions when loaded

[patchbot@xxxxxxx]

commit a57b4074ab39bee78b6c116277f0a9963bd8e687
Author:     Roger Pau Monné <roger.pau@xxxxxxxxxx>
AuthorDate: Tue Feb 27 12:25:24 2024 +0100
Commit:     Andrew Cooper <andrew.cooper3@xxxxxxxxxx>
CommitDate: Wed Feb 28 16:57:25 2024 +0000

    xen/livepatch: register livepatch regions when loaded
    
    Currently livepatch regions are registered as virtual regions only after the
    livepatch has been applied.
    
    This can lead to issues when using the pre-apply or post-revert hooks, as at
    that point the livepatch is not in the virtual regions list.  If a livepatch
    pre-apply hook contains a WARN() it would trigger an hypervisor crash, as 
the
    code to handle the bug frame won't be able to find the instruction pointer 
that
    triggered the #UD in any of the registered virtual regions, and hence crash.
    
    Fix this by adding the livepatch payloads as virtual regions as soon as 
loaded,
    and only remove them once the payload is unloaded.  This requires some 
changes
    to the virtual regions code, as the removal of the virtual regions is no 
longer
    done in stop machine context, and hence an RCU barrier is added in order to
    make sure there are no users of the virtual region after it's been removed 
from
    the list.
    
    Fixes: 8313c864fa95 ('livepatch: Implement pre-|post- apply|revert hooks')
    Signed-off-by: Roger Pau Monné <roger.pau@xxxxxxxxxx>
    Reviewed-by: Ross Lagerwall <ross.lagerwall@xxxxxxxxxx>
---
 xen/common/livepatch.c      |  4 ++--
 xen/common/virtual_region.c | 44 +++++++++++++++++---------------------------
 2 files changed, 19 insertions(+), 29 deletions(-)

diff --git a/xen/common/livepatch.c b/xen/common/livepatch.c
index 2c4b843827..d7f50e1018 100644
--- a/xen/common/livepatch.c
+++ b/xen/common/livepatch.c
@@ -1071,6 +1071,7 @@ static int build_symbol_table(struct payload *payload,
 static void free_payload(struct payload *data)
 {
     ASSERT(spin_is_locked(&payload_lock));
+    unregister_virtual_region(&data->region);
     list_del(&data->list);
     payload_cnt--;
     payload_version++;
@@ -1170,6 +1171,7 @@ static int livepatch_upload(struct 
xen_sysctl_livepatch_upload *upload)
         INIT_LIST_HEAD(&data->list);
         INIT_LIST_HEAD(&data->applied_list);
 
+        register_virtual_region(&data->region);
         list_add_tail(&data->list, &payload_list);
         payload_cnt++;
         payload_version++;
@@ -1386,7 +1388,6 @@ static inline void apply_payload_tail(struct payload 
*data)
      * The applied_list is iterated by the trap code.
      */
     list_add_tail_rcu(&data->applied_list, &applied_list);
-    register_virtual_region(&data->region);
 
     data->state = LIVEPATCH_STATE_APPLIED;
 }
@@ -1432,7 +1433,6 @@ static inline void revert_payload_tail(struct payload 
*data)
      * The applied_list is iterated by the trap code.
      */
     list_del_rcu(&data->applied_list);
-    unregister_virtual_region(&data->region);
 
     data->reverted = true;
     data->state = LIVEPATCH_STATE_CHECKED;
diff --git a/xen/common/virtual_region.c b/xen/common/virtual_region.c
index ddac5c9147..cefef3e47e 100644
--- a/xen/common/virtual_region.c
+++ b/xen/common/virtual_region.c
@@ -23,14 +23,8 @@ static struct virtual_region core_init __initdata = {
 };
 
 /*
- * RCU locking. Additions are done either at startup (when there is only
- * one CPU) or when all CPUs are running without IRQs.
- *
- * Deletions are bit tricky. We do it when Live Patch (all CPUs running
- * without IRQs) or during bootup (when clearing the init).
- *
- * Hence we use list_del_rcu (which sports an memory fence) and a spinlock
- * on deletion.
+ * RCU locking. Modifications to the list must be done in exclusive mode, and
+ * hence need to hold the spinlock.
  *
  * All readers of virtual_region_list MUST use list_for_each_entry_rcu.
  */
@@ -58,41 +52,36 @@ const struct virtual_region *find_text_region(unsigned long 
addr)
 
 void register_virtual_region(struct virtual_region *r)
 {
-    ASSERT(!local_irq_is_enabled());
+    unsigned long flags;
 
+    spin_lock_irqsave(&virtual_region_lock, flags);
     list_add_tail_rcu(&r->list, &virtual_region_list);
+    spin_unlock_irqrestore(&virtual_region_lock, flags);
 }
 
-static void remove_virtual_region(struct virtual_region *r)
+/*
+ * Suggest inline so when !CONFIG_LIVEPATCH the function is not left
+ * unreachable after init code is removed.
+ */
+static void inline remove_virtual_region(struct virtual_region *r)
 {
     unsigned long flags;
 
     spin_lock_irqsave(&virtual_region_lock, flags);
     list_del_rcu(&r->list);
     spin_unlock_irqrestore(&virtual_region_lock, flags);
-    /*
-     * We do not need to invoke call_rcu.
-     *
-     * This is due to the fact that on the deletion we have made sure
-     * to use spinlocks (to guard against somebody else calling
-     * unregister_virtual_region) and list_deletion spiced with
-     * memory barrier.
-     *
-     * That protects us from corrupting the list as the readers all
-     * use list_for_each_entry_rcu which is safe against concurrent
-     * deletions.
-     */
 }
 
+#ifdef CONFIG_LIVEPATCH
 void unregister_virtual_region(struct virtual_region *r)
 {
-    /* Expected to be called from Live Patch - which has IRQs disabled. */
-    ASSERT(!local_irq_is_enabled());
-
     remove_virtual_region(r);
+
+    /* Assert that no CPU might be using the removed region. */
+    rcu_barrier();
 }
 
-#if defined(CONFIG_LIVEPATCH) && defined(CONFIG_X86)
+#ifdef CONFIG_X86
 void relax_virtual_region_perms(void)
 {
     const struct virtual_region *region;
@@ -116,7 +105,8 @@ void tighten_virtual_region_perms(void)
                                  PAGE_HYPERVISOR_RX);
     rcu_read_unlock(&rcu_virtual_region_lock);
 }
-#endif
+#endif /* CONFIG_X86 */
+#endif /* CONFIG_LIVEPATCH */
 
 void __init unregister_init_virtual_region(void)
 {
--
generated by git-patchbot for /home/xen/git/xen.git#staging