Received-SPF: pass (google.com: domain of linux-kernel+bounces-81705-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:4601:e00::3 as permitted sender) client-ip=2604:1380:4601:e00::3;
From: Lai Jiangshan <jiangshanlai@gmail.com>
To: linux-kernel@vger.kernel.org
Cc: Hou Wenlong <houwenlong.hwl@antgroup.com>,
	Lai Jiangshan <jiangshan.ljs@antgroup.com>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Peter Zijlstra <peterz@infradead.org>,
	Sean Christopherson <seanjc@google.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Borislav Petkov <bp@alien8.de>,
	Ingo Molnar <mingo@redhat.com>,
	kvm@vger.kernel.org,
	Paolo Bonzini <pbonzini@redhat.com>,
	x86@kernel.org,
	Kees Cook <keescook@chromium.org>,
	Juergen Gross <jgross@suse.com>,
	Dave Hansen <dave.hansen@linux.intel.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Andy Lutomirski <luto@kernel.org>,
	Jonathan Corbet <corbet@lwn.net>,
	Josh Poimboeuf <jpoimboe@kernel.org>,
	Yuntao Wang <ytcoode@gmail.com>,
	Wang Jinchao <wangjinchao@xfusion.com>
Subject: [RFC PATCH 57/73] x86/pvm: Make cpu entry area and vmalloc area variable
Date: Mon, 26 Feb 2024 22:36:14 +0800
Message-Id: <20240226143630.33643-58-jiangshanlai@gmail.com>
In-Reply-To: <20240226143630.33643-1-jiangshanlai@gmail.com>
References: <20240226143630.33643-1-jiangshanlai@gmail.com>
Precedence: bulk
MIME-Version: 1.0
Content-Transfer-Encoding: 8bit

From: Hou Wenlong <houwenlong.hwl@antgroup.com>

For the PVM guest, the entire kernel layout should be within the allowed
virtual address range. Therefore, establish CPU_ENTRY_AREA_BASE and
VMEMORY_END as a variable for the PVM guest, allowing it to be
modified as necessary for the PVM guest.

Signed-off-by: Hou Wenlong <houwenlong.hwl@antgroup.com>
Signed-off-by: Lai Jiangshan <jiangshan.ljs@antgroup.com>
---
 arch/x86/include/asm/page_64.h          |  3 +++
 arch/x86/include/asm/pgtable_64_types.h | 14 ++++++++++++--
 arch/x86/kernel/head64.c                |  7 +++++++
 arch/x86/mm/dump_pagetables.c           |  3 ++-
 arch/x86/mm/kaslr.c                     |  4 ++--
 5 files changed, 26 insertions(+), 5 deletions(-)

diff --git a/arch/x86/include/asm/page_64.h b/arch/x86/include/asm/page_64.h
index b8692e6cc939..4f64f049f3d0 100644
--- a/arch/x86/include/asm/page_64.h
+++ b/arch/x86/include/asm/page_64.h
@@ -18,6 +18,9 @@ extern unsigned long page_offset_base;
 extern unsigned long vmalloc_base;
 extern unsigned long vmemmap_base;
 
+extern unsigned long cpu_entry_area_base;
+extern unsigned long vmemory_end;
+
 static __always_inline unsigned long __phys_addr_nodebug(unsigned long x)
 {
 	unsigned long y = x - KERNEL_MAP_BASE;
diff --git a/arch/x86/include/asm/pgtable_64_types.h b/arch/x86/include/asm/pgtable_64_types.h
index 6780f2e63717..66c8e7325d27 100644
--- a/arch/x86/include/asm/pgtable_64_types.h
+++ b/arch/x86/include/asm/pgtable_64_types.h
@@ -140,6 +140,7 @@ extern unsigned int ptrs_per_p4d;
 # define VMEMMAP_START		__VMEMMAP_BASE_L4
 #endif /* CONFIG_DYNAMIC_MEMORY_LAYOUT */
 
+#ifndef CONFIG_PVM_GUEST
 /*
  * End of the region for which vmalloc page tables are pre-allocated.
  * For non-KMSAN builds, this is the same as VMALLOC_END.
@@ -147,6 +148,10 @@ extern unsigned int ptrs_per_p4d;
  * VMALLOC_START..VMALLOC_END (see below).
  */
 #define VMEMORY_END		(VMALLOC_START + (VMALLOC_SIZE_TB << 40) - 1)
+#else
+#define RAW_VMEMORY_END		(__VMALLOC_BASE_L4 + (VMALLOC_SIZE_TB_L4 << 40) - 1)
+#define VMEMORY_END		vmemory_end
+#endif /* CONFIG_PVM_GUEST */
 
 #ifndef CONFIG_KMSAN
 #define VMALLOC_END		VMEMORY_END
@@ -166,7 +171,7 @@ extern unsigned int ptrs_per_p4d;
  *              KMSAN_MODULES_ORIGIN_START to
  *              KMSAN_MODULES_ORIGIN_START + MODULES_LEN - origins for modules.
  */
-#define VMALLOC_QUARTER_SIZE	((VMALLOC_SIZE_TB << 40) >> 2)
+#define VMALLOC_QUARTER_SIZE	((VMEMORY_END + 1 - VMALLOC_START) >> 2)
 #define VMALLOC_END		(VMALLOC_START + VMALLOC_QUARTER_SIZE - 1)
 
 /*
@@ -202,7 +207,12 @@ extern unsigned int ptrs_per_p4d;
 #define ESPFIX_BASE_ADDR	(ESPFIX_PGD_ENTRY << P4D_SHIFT)
 
 #define CPU_ENTRY_AREA_PGD	_AC(-4, UL)
-#define CPU_ENTRY_AREA_BASE	(CPU_ENTRY_AREA_PGD << P4D_SHIFT)
+#define RAW_CPU_ENTRY_AREA_BASE	(CPU_ENTRY_AREA_PGD << P4D_SHIFT)
+#ifdef CONFIG_PVM_GUEST
+#define CPU_ENTRY_AREA_BASE	cpu_entry_area_base
+#else
+#define CPU_ENTRY_AREA_BASE	RAW_CPU_ENTRY_AREA_BASE
+#endif
 
 #define EFI_VA_START		( -4 * (_AC(1, UL) << 30))
 #define EFI_VA_END		(-68 * (_AC(1, UL) << 30))
diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
index 0b0e460609e5..d0e8d648bd38 100644
--- a/arch/x86/kernel/head64.c
+++ b/arch/x86/kernel/head64.c
@@ -72,6 +72,13 @@ unsigned long kernel_map_base __ro_after_init = __START_KERNEL_map;
 EXPORT_SYMBOL(kernel_map_base);
 #endif
 
+#ifdef CONFIG_PVM_GUEST
+unsigned long cpu_entry_area_base __ro_after_init = RAW_CPU_ENTRY_AREA_BASE;
+EXPORT_SYMBOL(cpu_entry_area_base);
+unsigned long vmemory_end __ro_after_init = RAW_VMEMORY_END;
+EXPORT_SYMBOL(vmemory_end);
+#endif
+
 /* Wipe all early page tables except for the kernel symbol map */
 static void __init reset_early_page_tables(void)
 {
diff --git a/arch/x86/mm/dump_pagetables.c b/arch/x86/mm/dump_pagetables.c
index d5c6f61242aa..166c7d36d8ff 100644
--- a/arch/x86/mm/dump_pagetables.c
+++ b/arch/x86/mm/dump_pagetables.c
@@ -95,7 +95,7 @@ static struct addr_marker address_markers[] = {
 #ifdef CONFIG_MODIFY_LDT_SYSCALL
 	[LDT_NR]		= { 0UL,		"LDT remap" },
 #endif
-	[CPU_ENTRY_AREA_NR]	= { CPU_ENTRY_AREA_BASE,"CPU entry Area" },
+	[CPU_ENTRY_AREA_NR]	= { 0UL,		"CPU entry Area" },
 #ifdef CONFIG_X86_ESPFIX64
 	[ESPFIX_START_NR]	= { ESPFIX_BASE_ADDR,	"ESPfix Area", 16 },
 #endif
@@ -479,6 +479,7 @@ static int __init pt_dump_init(void)
 	address_markers[MODULES_VADDR_NR].start_address = MODULES_VADDR;
 	address_markers[MODULES_END_NR].start_address = MODULES_END;
 	address_markers[FIXADDR_START_NR].start_address = FIXADDR_START;
+	address_markers[CPU_ENTRY_AREA_NR].start_address = CPU_ENTRY_AREA_BASE;
 #endif
 #ifdef CONFIG_X86_32
 	address_markers[VMALLOC_START_NR].start_address = VMALLOC_START;
diff --git a/arch/x86/mm/kaslr.c b/arch/x86/mm/kaslr.c
index 37db264866b6..e3825c7542a3 100644
--- a/arch/x86/mm/kaslr.c
+++ b/arch/x86/mm/kaslr.c
@@ -38,7 +38,7 @@
  * highest amount of space for randomization available, but that's too hard
  * to keep straight and caused issues already.
  */
-static const unsigned long vaddr_end = CPU_ENTRY_AREA_BASE;
+static const unsigned long vaddr_end = RAW_CPU_ENTRY_AREA_BASE;
 
 /*
  * Memory regions randomized by KASLR (except modules that use a separate logic
@@ -79,7 +79,7 @@ void __init kernel_randomize_memory(void)
 	 * limited....
 	 */
 	BUILD_BUG_ON(vaddr_start >= vaddr_end);
-	BUILD_BUG_ON(vaddr_end != CPU_ENTRY_AREA_BASE);
+	BUILD_BUG_ON(vaddr_end != RAW_CPU_ENTRY_AREA_BASE);
 	BUILD_BUG_ON(vaddr_end > __START_KERNEL_map);
 
 	if (!kaslr_memory_enabled())
-- 
2.19.1.6.gb485710b