[PATCH] x86_64: Align and pad x86_64 GDT on page boundary

This patch is on the same lines as Zachary Amsden's i386 GDT page alignemnt
patch in -mm, but for x86_64.

Patch to align and pad x86_64 GDT on page boundries.

[AK: some minor cleanups and fixed incorrect TLS initialization
in CPU init.]
Signed-off-by: Nippun Goel <nippung@calsoftinc.com>
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Shai Fultheim <shai@scalex86.org>
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

arch/x86_64/kernel/head.S

@@ -379,7 +379,7 @@ gdt:
  * Also sysret mandates a special GDT layout 
  */
 		 		
-.align L1_CACHE_BYTES
+.align PAGE_SIZE
 
 /* The TLS descriptors are currently at a different place compared to i386.
    Hopefully nobody expects them at a fixed place (Wine?) */
@@ -401,10 +401,11 @@ ENTRY(cpu_gdt_table)
 gdt_end:	
 	/* asm/segment.h:GDT_ENTRIES must match this */	
 	/* This should be a multiple of the cache line size */
-	/* GDTs of other CPUs: */	
-	.fill (GDT_SIZE * NR_CPUS) - (gdt_end - cpu_gdt_table)
+	/* GDTs of other CPUs are now dynamically allocated */
+
+	/* zero the remaining page */
+	.fill PAGE_SIZE / 8 - GDT_ENTRIES,8,0
 
-	.align  L1_CACHE_BYTES
 ENTRY(idt_table)	
 	.rept   256
 	.quad   0

arch/x86_64/kernel/setup64.c

@@ -213,16 +213,14 @@ void __cpuinit cpu_init (void)
 	 * Initialize the per-CPU GDT with the boot GDT,
 	 * and set up the GDT descriptor:
 	 */
-	if (cpu) {
-		memcpy(cpu_gdt_table[cpu], cpu_gdt_table[0], GDT_SIZE);
-	}	
+	if (cpu)
+ 		memcpy(cpu_gdt(cpu), cpu_gdt_table, GDT_SIZE);
 
 	cpu_gdt_descr[cpu].size = GDT_SIZE;
-	cpu_gdt_descr[cpu].address = (unsigned long)cpu_gdt_table[cpu];
 	asm volatile("lgdt %0" :: "m" (cpu_gdt_descr[cpu]));
 	asm volatile("lidt %0" :: "m" (idt_descr));
 
-	memcpy(me->thread.tls_array, cpu_gdt_table[cpu], GDT_ENTRY_TLS_ENTRIES * 8);
+	memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
 	syscall_init();
 
 	wrmsrl(MSR_FS_BASE, 0);

arch/x86_64/kernel/smpboot.c

@@ -744,6 +744,13 @@ static int __cpuinit do_boot_cpu(int cpu, int apicid)
 	};
 	DECLARE_WORK(work, do_fork_idle, &c_idle);
 
+	/* allocate memory for gdts of secondary cpus. Hotplug is considered */
+	if (!cpu_gdt_descr[cpu].address &&
+		!(cpu_gdt_descr[cpu].address = get_zeroed_page(GFP_KERNEL))) {
+		printk(KERN_ERR "Failed to allocate GDT for CPU %d\n", cpu);
+		return -1;
+	}
+
 	c_idle.idle = get_idle_for_cpu(cpu);
 
 	if (c_idle.idle) {

arch/x86_64/kernel/suspend.c

@@ -120,7 +120,7 @@ void fix_processor_context(void)
 
 	set_tss_desc(cpu,t);	/* This just modifies memory; should not be neccessary. But... This is neccessary, because 386 hardware has concept of busy TSS or some similar stupidity. */
 
-	cpu_gdt_table[cpu][GDT_ENTRY_TSS].type = 9;
+	cpu_gdt(cpu)[GDT_ENTRY_TSS].type = 9;
 
 	syscall_init();                         /* This sets MSR_*STAR and related */
 	load_TR_desc();				/* This does ltr */

include/asm-x86_64/desc.h

@@ -25,7 +25,7 @@ struct n_desc_struct {
 	unsigned int a,b;
 }; 	
 
-extern struct desc_struct cpu_gdt_table[NR_CPUS][GDT_ENTRIES];
+extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
 
 enum { 
 	GATE_INTERRUPT = 0xE, 
@@ -79,6 +79,9 @@ extern struct desc_struct default_ldt[];
 extern struct gate_struct idt_table[]; 
 extern struct desc_ptr cpu_gdt_descr[];
 
+/* the cpu gdt accessor */
+#define cpu_gdt(_cpu) ((struct desc_struct *)cpu_gdt_descr[_cpu].address)
+
 static inline void _set_gate(void *adr, unsigned type, unsigned long func, unsigned dpl, unsigned ist)  
 {
 	struct gate_struct s; 	
@@ -144,20 +147,20 @@ static inline void set_tss_desc(unsigned cpu, void *addr)
 	 * -1? seg base+limit should be pointing to the address of the
 	 * last valid byte
 	 */
-	set_tssldt_descriptor(&cpu_gdt_table[cpu][GDT_ENTRY_TSS],
+	set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_TSS],
 		(unsigned long)addr, DESC_TSS,
 		IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
 } 
 
 static inline void set_ldt_desc(unsigned cpu, void *addr, int size)
 { 
-	set_tssldt_descriptor(&cpu_gdt_table[cpu][GDT_ENTRY_LDT], (unsigned long)addr, 
+	set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_LDT], (unsigned long)addr,
 			      DESC_LDT, size * 8 - 1);
 }
 
 static inline void set_seg_base(unsigned cpu, int entry, void *base)
 { 
-	struct desc_struct *d = &cpu_gdt_table[cpu][entry];
+	struct desc_struct *d = &cpu_gdt(cpu)[entry];
 	u32 addr = (u32)(u64)base;
 	BUG_ON((u64)base >> 32); 
 	d->base0 = addr & 0xffff;
@@ -199,7 +202,7 @@ static inline void set_seg_base(unsigned cpu, int entry, void *base)
 
 static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
 {
-	u64 *gdt = (u64 *)(cpu_gdt_table[cpu] + GDT_ENTRY_TLS_MIN);
+	u64 *gdt = (u64 *)(cpu_gdt(cpu) + GDT_ENTRY_TLS_MIN);
 	gdt[0] = t->tls_array[0];
 	gdt[1] = t->tls_array[1];
 	gdt[2] = t->tls_array[2];