[RFC PATCH v5 21/45] x86/tdx: Add APIs to support get/put of DPAMT entries from KVM, under spinlock

[Sean Christopherson <seanjc@google.com>]

From: Rick Edgecombe <rick.p.edgecombe@intel.com>

Implement a PAMT "caching" scheme, similar to KVM's pre-allocated cache of
MMU assets, along with APIs to allow KVM to pre-allocate PAMT pages before
acquiring its mmu_lock spinlock, but wait until S-EPT entries are created
to actually update the Dynamic PAMT.

Signed-off-by: Rick Edgecombe <rick.p.edgecombe@intel.com>
Co-developed-by: Sean Christopherson <seanjc@google.com>
Signed-off-by: Sean Christopherson <seanjc@google.com>
---
 arch/x86/include/asm/tdx.h  | 17 ++++++++++
 arch/x86/virt/vmx/tdx/tdx.c | 65 +++++++++++++++++++++++++++++++++----
 2 files changed, 76 insertions(+), 6 deletions(-)

diff --git a/arch/x86/include/asm/tdx.h b/arch/x86/include/asm/tdx.h
index fa29be18498c..c39e2920d0c3 100644
--- a/arch/x86/include/asm/tdx.h
+++ b/arch/x86/include/asm/tdx.h
@@ -136,6 +136,23 @@ static inline bool tdx_supports_dynamic_pamt(const struct tdx_sys_info *sysinfo)
 	return false; /* To be enabled when kernel is ready */
 }
 
+/* Simple structure for pre-allocating Dynamic PAMT pages outside of locks. */
+struct tdx_pamt_cache {
+	struct list_head page_list;
+	int cnt;
+};
+
+static inline void tdx_init_pamt_cache(struct tdx_pamt_cache *cache)
+{
+	INIT_LIST_HEAD(&cache->page_list);
+	cache->cnt = 0;
+}
+
+void tdx_free_pamt_cache(struct tdx_pamt_cache *cache);
+int tdx_topup_pamt_cache(struct tdx_pamt_cache *cache, unsigned long npages);
+int tdx_pamt_get(struct page *page, struct tdx_pamt_cache *cache);
+void tdx_pamt_put(struct page *page);
+
 void tdx_quirk_reset_page(struct page *page);
 
 int tdx_guest_keyid_alloc(void);
diff --git a/arch/x86/virt/vmx/tdx/tdx.c b/arch/x86/virt/vmx/tdx/tdx.c
index d333d2790913..53b29c827520 100644
--- a/arch/x86/virt/vmx/tdx/tdx.c
+++ b/arch/x86/virt/vmx/tdx/tdx.c
@@ -2064,13 +2064,34 @@ u64 tdh_phymem_page_wbinvd_hkid(u64 hkid, struct page *page)
 }
 EXPORT_SYMBOL_FOR_KVM(tdh_phymem_page_wbinvd_hkid);
 
-static int alloc_pamt_array(u64 *pa_array)
+static struct page *tdx_alloc_page_pamt_cache(struct tdx_pamt_cache *cache)
+{
+	struct page *page;
+
+	page = list_first_entry_or_null(&cache->page_list, struct page, lru);
+	if (page) {
+		list_del(&page->lru);
+		cache->cnt--;
+	}
+
+	return page;
+}
+
+static struct page *alloc_dpamt_page(struct tdx_pamt_cache *cache)
+{
+	if (cache)
+		return tdx_alloc_page_pamt_cache(cache);
+
+	return alloc_page(GFP_KERNEL_ACCOUNT);
+}
+
+static int alloc_pamt_array(u64 *pa_array, struct tdx_pamt_cache *cache)
 {
 	struct page *page;
 	int i;
 
 	for (i = 0; i < tdx_dpamt_entry_pages(); i++) {
-		page = alloc_page(GFP_KERNEL_ACCOUNT);
+		page = alloc_dpamt_page(cache);
 		if (!page)
 			goto err;
 		pa_array[i] = page_to_phys(page);
@@ -2151,7 +2172,7 @@ static u64 tdh_phymem_pamt_remove(struct page *page, u64 *pamt_pa_array)
 static DEFINE_SPINLOCK(pamt_lock);
 
 /* Bump PAMT refcount for the given page and allocate PAMT memory if needed */
-static int tdx_pamt_get(struct page *page)
+int tdx_pamt_get(struct page *page, struct tdx_pamt_cache *cache)
 {
 	u64 pamt_pa_array[MAX_NR_DPAMT_ARGS];
 	atomic_t *pamt_refcount;
@@ -2170,7 +2191,7 @@ static int tdx_pamt_get(struct page *page)
 	if (atomic_inc_not_zero(pamt_refcount))
 		return 0;
 
-	ret = alloc_pamt_array(pamt_pa_array);
+	ret = alloc_pamt_array(pamt_pa_array, cache);
 	if (ret)
 		goto out_free;
 
@@ -2222,12 +2243,13 @@ static int tdx_pamt_get(struct page *page)
 	free_pamt_array(pamt_pa_array);
 	return ret;
 }
+EXPORT_SYMBOL_FOR_KVM(tdx_pamt_get);
 
 /*
  * Drop PAMT refcount for the given page and free PAMT memory if it is no
  * longer needed.
  */
-static void tdx_pamt_put(struct page *page)
+void tdx_pamt_put(struct page *page)
 {
 	u64 pamt_pa_array[MAX_NR_DPAMT_ARGS];
 	atomic_t *pamt_refcount;
@@ -2281,6 +2303,37 @@ static void tdx_pamt_put(struct page *page)
 	 */
 	free_pamt_array(pamt_pa_array);
 }
+EXPORT_SYMBOL_FOR_KVM(tdx_pamt_put);
+
+void tdx_free_pamt_cache(struct tdx_pamt_cache *cache)
+{
+	struct page *page;
+
+	while ((page = tdx_alloc_page_pamt_cache(cache)))
+		__free_page(page);
+}
+EXPORT_SYMBOL_FOR_KVM(tdx_free_pamt_cache);
+
+int tdx_topup_pamt_cache(struct tdx_pamt_cache *cache, unsigned long npages)
+{
+	if (WARN_ON_ONCE(!tdx_supports_dynamic_pamt(&tdx_sysinfo)))
+		return 0;
+
+	npages *= tdx_dpamt_entry_pages();
+
+	while (cache->cnt < npages) {
+		struct page *page = alloc_page(GFP_KERNEL_ACCOUNT);
+
+		if (!page)
+			return -ENOMEM;
+
+		list_add(&page->lru, &cache->page_list);
+		cache->cnt++;
+	}
+
+	return 0;
+}
+EXPORT_SYMBOL_FOR_KVM(tdx_topup_pamt_cache);
 
 /*
  * Return a page that can be gifted to the TDX-Module for use as a "control"
@@ -2296,7 +2349,7 @@ struct page *__tdx_alloc_control_page(gfp_t gfp)
 	if (!page)
 		return NULL;
 
-	if (tdx_pamt_get(page)) {
+	if (tdx_pamt_get(page, NULL)) {
 		__free_page(page);
 		return NULL;
 	}
-- 
2.53.0.rc1.217.geba53bf80e-goog