From 069f9911a35c6191ea0cace0b5b5c8061e9b7720 Mon Sep 17 00:00:00 2001
From: Tom Lendacky <thomas.lendacky@amd.com>
Date: Fri, 8 Mar 2024 07:31:11 -0800
Subject: [PATCH] OvmfPkg/BaseMemEncryptSevLib: Maximize Page State Change
 efficiency

BZ: https://bugzilla.tianocore.org/show_bug.cgi?id=4654

When building the Page State Change entries for a range of memory, it can
happen that multiple calls to BuildPageStateBuffer() need to be made. If
the size of the input work area passed to BuildPageStateBuffer() exceeds
the number of entries that can be passed to the hypervisor using the GHCB
shared buffer, the Page State Change VMGEXIT support will issue multiple
VMGEXITs to process all entries in the buffer.

However, it could be that the final VMGEXIT for each round of Page State
Changes is only for a small number of entries and subsequent VMGEXITs may
still be issued to handle the full range of memory requested. To maximize
the number of entries processed during the Page State Change VMGEXIT,
limit BuildPageStateBuffer() to not build entries that exceed the maximum
number of entries that can be handled in a single Page State Change
VMGEXIT.

Cc: Ard Biesheuvel <ardb+tianocore@kernel.org>
Cc: Erdem Aktas <erdemaktas@google.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Jiewen Yao <jiewen.yao@intel.com>
Cc: Laszlo Ersek <lersek@redhat.com>
Cc: Michael Roth <michael.roth@amd.com>
Cc: Min Xu <min.m.xu@intel.com>
Reviewed-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
---
 .../X64/SnpPageStateChangeInternal.c                  | 11 +++++++++++
 1 file changed, 11 insertions(+)

diff --git a/OvmfPkg/Library/BaseMemEncryptSevLib/X64/SnpPageStateChangeInternal.c b/OvmfPkg/Library/BaseMemEncryptSevLib/X64/SnpPageStateChangeInternal.c
index bcc0798d6b..f1883239a6 100644
--- a/OvmfPkg/Library/BaseMemEncryptSevLib/X64/SnpPageStateChangeInternal.c
+++ b/OvmfPkg/Library/BaseMemEncryptSevLib/X64/SnpPageStateChangeInternal.c
@@ -145,6 +145,7 @@ BuildPageStateBuffer (
   UINTN                 RmpPageSize;
   UINTN                 Index;
   UINTN                 IndexMax;
+  UINTN                 PscIndexMax;
 
   // Clear the page state structure
   SetMem (Info, InfoSize, 0);
@@ -153,6 +154,16 @@ BuildPageStateBuffer (
   IndexMax    = (InfoSize - sizeof (Info->Header)) / sizeof (Info->Entry[0]);
   NextAddress = EndAddress;
 
+  //
+  // Make the use of the work area as efficient as possible relative to
+  // exiting from the guest to the hypervisor. Maximize the number of entries
+  // that can be processed per exit.
+  //
+  PscIndexMax = (IndexMax / SNP_PAGE_STATE_MAX_ENTRY) * SNP_PAGE_STATE_MAX_ENTRY;
+  if (PscIndexMax > 0) {
+    IndexMax = MIN (IndexMax, PscIndexMax);
+  }
+
   //
   // Populate the page state entry structure
   //