kernel26-xen/patches-2.6.25-r1/1014-2.6.25-xen-auto-arch-x86_64.patch

Subject: xen3 arch-x86_64
From: http://xenbits.xensource.com/linux-2.6.18-xen.hg (tip 517:d71965a78c20)
Patch-mainline: obsolete
Acked-by: jbeulich@novell.com

Index: head-2008-04-15/arch/x86/kernel/asm-offsets_64.c
===================================================================
--- head-2008-04-15.orig/arch/x86/kernel/asm-offsets_64.c	2008-04-15 09:41:09.000000000 +0200
+++ head-2008-04-15/arch/x86/kernel/asm-offsets_64.c	2008-04-15 09:59:32.000000000 +0200
@@ -124,8 +124,10 @@ int main(void)
 	ENTRY(cr8);
 	BLANK();
 #undef ENTRY
+#ifndef CONFIG_X86_NO_TSS
 	DEFINE(TSS_ist, offsetof(struct tss_struct, x86_tss.ist));
 	BLANK();
+#endif
 	DEFINE(crypto_tfm_ctx_offset, offsetof(struct crypto_tfm, __crt_ctx));
 	BLANK();
 	DEFINE(__NR_syscall_max, sizeof(syscalls) - 1);
Index: head-2008-04-15/arch/x86/kernel/machine_kexec_64.c
===================================================================
--- head-2008-04-15.orig/arch/x86/kernel/machine_kexec_64.c	2008-04-15 09:41:09.000000000 +0200
+++ head-2008-04-15/arch/x86/kernel/machine_kexec_64.c	2008-04-15 09:59:32.000000000 +0200
@@ -25,6 +25,119 @@ static u64 kexec_pud1[512] PAGE_ALIGNED;
 static u64 kexec_pmd1[512] PAGE_ALIGNED;
 static u64 kexec_pte1[512] PAGE_ALIGNED;
 
+#ifdef CONFIG_XEN
+
+/* In the case of Xen, override hypervisor functions to be able to create
+ * a regular identity mapping page table...
+ */
+
+#include <xen/interface/kexec.h>
+#include <xen/interface/memory.h>
+
+#define x__pmd(x) ((pmd_t) { (x) } )
+#define x__pud(x) ((pud_t) { (x) } )
+#define x__pgd(x) ((pgd_t) { (x) } )
+
+#define x_pmd_val(x)   ((x).pmd)
+#define x_pud_val(x)   ((x).pud)
+#define x_pgd_val(x)   ((x).pgd)
+
+static inline void x_set_pmd(pmd_t *dst, pmd_t val)
+{
+	x_pmd_val(*dst) = x_pmd_val(val);
+}
+
+static inline void x_set_pud(pud_t *dst, pud_t val)
+{
+	x_pud_val(*dst) = phys_to_machine(x_pud_val(val));
+}
+
+static inline void x_pud_clear (pud_t *pud)
+{
+	x_pud_val(*pud) = 0;
+}
+
+static inline void x_set_pgd(pgd_t *dst, pgd_t val)
+{
+	x_pgd_val(*dst) = phys_to_machine(x_pgd_val(val));
+}
+
+static inline void x_pgd_clear (pgd_t * pgd)
+{
+	x_pgd_val(*pgd) = 0;
+}
+
+#define X__PAGE_KERNEL_LARGE_EXEC \
+         _PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_PSE
+#define X_KERNPG_TABLE _PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY
+
+#define __ma(x) (pfn_to_mfn(__pa((x)) >> PAGE_SHIFT) << PAGE_SHIFT)
+
+#if PAGES_NR > KEXEC_XEN_NO_PAGES
+#error PAGES_NR is greater than KEXEC_XEN_NO_PAGES - Xen support will break
+#endif
+
+#if PA_CONTROL_PAGE != 0
+#error PA_CONTROL_PAGE is non zero - Xen support will break
+#endif
+
+void machine_kexec_setup_load_arg(xen_kexec_image_t *xki, struct kimage *image)
+{
+	void *control_page;
+	void *table_page;
+
+	memset(xki->page_list, 0, sizeof(xki->page_list));
+
+	control_page = page_address(image->control_code_page) + PAGE_SIZE;
+	memcpy(control_page, relocate_kernel, PAGE_SIZE);
+
+	table_page = page_address(image->control_code_page);
+
+	xki->page_list[PA_CONTROL_PAGE] = __ma(control_page);
+	xki->page_list[PA_TABLE_PAGE] = __ma(table_page);
+
+	xki->page_list[PA_PGD] = __ma(kexec_pgd);
+	xki->page_list[PA_PUD_0] = __ma(kexec_pud0);
+	xki->page_list[PA_PUD_1] = __ma(kexec_pud1);
+	xki->page_list[PA_PMD_0] = __ma(kexec_pmd0);
+	xki->page_list[PA_PMD_1] = __ma(kexec_pmd1);
+	xki->page_list[PA_PTE_0] = __ma(kexec_pte0);
+	xki->page_list[PA_PTE_1] = __ma(kexec_pte1);
+}
+
+int __init machine_kexec_setup_resources(struct resource *hypervisor,
+					 struct resource *phys_cpus,
+					 int nr_phys_cpus)
+{
+	int k;
+
+	/* The per-cpu crash note resources belong to the hypervisor resource */
+	for (k = 0; k < nr_phys_cpus; k++)
+		request_resource(hypervisor, phys_cpus + k);
+
+	return 0;
+}
+
+void machine_kexec_register_resources(struct resource *res) { ; }
+
+#else /* CONFIG_XEN */
+
+#define x__pmd(x) __pmd(x)
+#define x__pud(x) __pud(x)
+#define x__pgd(x) __pgd(x)
+
+#define x_set_pmd(x, y) set_pmd(x, y)
+#define x_set_pud(x, y) set_pud(x, y)
+#define x_set_pgd(x, y) set_pgd(x, y)
+
+#define x_pud_clear(x) pud_clear(x)
+#define x_pgd_clear(x) pgd_clear(x)
+
+#define X__PAGE_KERNEL_LARGE_EXEC __PAGE_KERNEL_LARGE_EXEC
+#define X_KERNPG_TABLE _KERNPG_TABLE
+
+#endif /* CONFIG_XEN */
+
 static void init_level2_page(pmd_t *level2p, unsigned long addr)
 {
 	unsigned long end_addr;
@@ -32,7 +145,7 @@ static void init_level2_page(pmd_t *leve
 	addr &= PAGE_MASK;
 	end_addr = addr + PUD_SIZE;
 	while (addr < end_addr) {
-		set_pmd(level2p++, __pmd(addr | __PAGE_KERNEL_LARGE_EXEC));
+		x_set_pmd(level2p++, x__pmd(addr | X__PAGE_KERNEL_LARGE_EXEC));
 		addr += PMD_SIZE;
 	}
 }
@@ -57,12 +170,12 @@ static int init_level3_page(struct kimag
 		}
 		level2p = (pmd_t *)page_address(page);
 		init_level2_page(level2p, addr);
-		set_pud(level3p++, __pud(__pa(level2p) | _KERNPG_TABLE));
+		x_set_pud(level3p++, x__pud(__pa(level2p) | X_KERNPG_TABLE));
 		addr += PUD_SIZE;
 	}
 	/* clear the unused entries */
 	while (addr < end_addr) {
-		pud_clear(level3p++);
+		x_pud_clear(level3p++);
 		addr += PUD_SIZE;
 	}
 out:
@@ -93,12 +206,12 @@ static int init_level4_page(struct kimag
 		if (result) {
 			goto out;
 		}
-		set_pgd(level4p++, __pgd(__pa(level3p) | _KERNPG_TABLE));
+		x_set_pgd(level4p++, x__pgd(__pa(level3p) | X_KERNPG_TABLE));
 		addr += PGDIR_SIZE;
 	}
 	/* clear the unused entries */
 	while (addr < end_addr) {
-		pgd_clear(level4p++);
+		x_pgd_clear(level4p++);
 		addr += PGDIR_SIZE;
 	}
 out:
@@ -109,8 +222,14 @@ out:
 static int init_pgtable(struct kimage *image, unsigned long start_pgtable)
 {
 	pgd_t *level4p;
+	unsigned long x_end_pfn = end_pfn;
+
+#ifdef CONFIG_XEN
+	x_end_pfn = HYPERVISOR_memory_op(XENMEM_maximum_ram_page, NULL);
+#endif
+
 	level4p = (pgd_t *)__va(start_pgtable);
- 	return init_level4_page(image, level4p, 0, end_pfn << PAGE_SHIFT);
+ 	return init_level4_page(image, level4p, 0, x_end_pfn << PAGE_SHIFT);
 }
 
 int machine_kexec_prepare(struct kimage *image)
@@ -134,6 +253,7 @@ void machine_kexec_cleanup(struct kimage
 	return;
 }
 
+#ifndef CONFIG_XEN
 /*
  * Do not allocate memory (or fail in any way) in machine_kexec().
  * We are past the point of no return, committed to rebooting now.
@@ -172,6 +292,7 @@ NORET_TYPE void machine_kexec(struct kim
 	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
 			image->start);
 }
+#endif
 
 void arch_crash_save_vmcoreinfo(void)
 {
Index: head-2008-04-15/arch/x86/pci/Makefile_64
===================================================================
--- head-2008-04-15.orig/arch/x86/pci/Makefile_64	2008-04-15 09:41:09.000000000 +0200
+++ head-2008-04-15/arch/x86/pci/Makefile_64	2008-04-15 09:59:32.000000000 +0200
@@ -15,3 +15,7 @@ obj-$(CONFIG_PCI_MMCONFIG) += mmconfig_6
 
 obj-$(CONFIG_NUMA)	+= k8-bus_64.o
 
+# pcifront should be after mmconfig.o and direct.o as it should only
+# take over if direct access to the PCI bus is unavailable
+obj-$(CONFIG_XEN_PCIDEV_FRONTEND)	+= pcifront.o
+
Index: head-2008-04-15/arch/x86/power/cpu_64.c
===================================================================
--- head-2008-04-15.orig/arch/x86/power/cpu_64.c	2008-04-15 09:41:09.000000000 +0200
+++ head-2008-04-15/arch/x86/power/cpu_64.c	2008-04-15 09:59:32.000000000 +0200
@@ -135,6 +135,7 @@ void restore_processor_state(void)
 
 static void fix_processor_context(void)
 {
+#ifndef CONFIG_X86_NO_TSS
 	int cpu = smp_processor_id();
 	struct tss_struct *t = &per_cpu(init_tss, cpu);
 
@@ -146,6 +147,7 @@ static void fix_processor_context(void)
 	set_tss_desc(cpu, t);
 
 	get_cpu_gdt_table(cpu)[GDT_ENTRY_TSS].type = 9;
+#endif
 
 	syscall_init();                         /* This sets MSR_*STAR and related */
 	load_TR_desc();				/* This does ltr */
Index: head-2008-04-15/arch/x86/vdso/Makefile
===================================================================
--- head-2008-04-15.orig/arch/x86/vdso/Makefile	2008-04-15 09:41:09.000000000 +0200
+++ head-2008-04-15/arch/x86/vdso/Makefile	2008-04-15 09:59:32.000000000 +0200
@@ -64,6 +64,8 @@ obj-$(VDSO32-y)			+= vdso32-syms.lds
 vdso32.so-$(CONFIG_X86_32)	+= int80
 vdso32.so-$(CONFIG_COMPAT)	+= syscall
 vdso32.so-$(VDSO32-y)		+= sysenter
+xen-vdso32-$(subst 1,$(CONFIG_COMPAT),$(shell expr $(CONFIG_XEN_COMPAT)0 '<' 0x0302000)) += int80
+vdso32.so-$(CONFIG_XEN)		+= $(xen-vdso32-y)
 
 vdso32-images			= $(vdso32.so-y:%=vdso32-%.so)
 
1	Subject: xen3 arch-x86_64
2	From: http://xenbits.xensource.com/linux-2.6.18-xen.hg (tip 517:d71965a78c20)
3	Patch-mainline: obsolete
4	Acked-by: jbeulich@novell.com
5
6	Index: head-2008-04-15/arch/x86/kernel/asm-offsets_64.c
7	===================================================================
8	--- head-2008-04-15.orig/arch/x86/kernel/asm-offsets_64.c 2008-04-15 09:41:09.000000000 +0200
9	+++ head-2008-04-15/arch/x86/kernel/asm-offsets_64.c 2008-04-15 09:59:32.000000000 +0200
10	@@ -124,8 +124,10 @@ int main(void)
11	ENTRY(cr8);
12	BLANK();
13	#undef ENTRY
14	+#ifndef CONFIG_X86_NO_TSS
15	DEFINE(TSS_ist, offsetof(struct tss_struct, x86_tss.ist));
16	BLANK();
17	+#endif
18	DEFINE(crypto_tfm_ctx_offset, offsetof(struct crypto_tfm, __crt_ctx));
19	BLANK();
20	DEFINE(__NR_syscall_max, sizeof(syscalls) - 1);
21	Index: head-2008-04-15/arch/x86/kernel/machine_kexec_64.c
22	===================================================================
23	--- head-2008-04-15.orig/arch/x86/kernel/machine_kexec_64.c 2008-04-15 09:41:09.000000000 +0200
24	+++ head-2008-04-15/arch/x86/kernel/machine_kexec_64.c 2008-04-15 09:59:32.000000000 +0200
25	@@ -25,6 +25,119 @@ static u64 kexec_pud1[512] PAGE_ALIGNED;
26	static u64 kexec_pmd1[512] PAGE_ALIGNED;
27	static u64 kexec_pte1[512] PAGE_ALIGNED;
28
29	+#ifdef CONFIG_XEN
30	+
31	+/* In the case of Xen, override hypervisor functions to be able to create
32	+ * a regular identity mapping page table...
33	+ */
34	+
35	+#include <xen/interface/kexec.h>
36	+#include <xen/interface/memory.h>
37	+
38	+#define x__pmd(x) ((pmd_t) { (x) } )
39	+#define x__pud(x) ((pud_t) { (x) } )
40	+#define x__pgd(x) ((pgd_t) { (x) } )
41	+
42	+#define x_pmd_val(x) ((x).pmd)
43	+#define x_pud_val(x) ((x).pud)
44	+#define x_pgd_val(x) ((x).pgd)
45	+
46	+static inline void x_set_pmd(pmd_t *dst, pmd_t val)
47	+{
48	+ x_pmd_val(*dst) = x_pmd_val(val);
49	+}
50	+
51	+static inline void x_set_pud(pud_t *dst, pud_t val)
52	+{
53	+ x_pud_val(*dst) = phys_to_machine(x_pud_val(val));
54	+}
55	+
56	+static inline void x_pud_clear (pud_t *pud)
57	+{
58	+ x_pud_val(*pud) = 0;
59	+}
60	+
61	+static inline void x_set_pgd(pgd_t *dst, pgd_t val)
62	+{
63	+ x_pgd_val(*dst) = phys_to_machine(x_pgd_val(val));
64	+}
65	+
66	+static inline void x_pgd_clear (pgd_t * pgd)
67	+{
68	+ x_pgd_val(*pgd) = 0;
69	+}
70	+
71	+#define X__PAGE_KERNEL_LARGE_EXEC \
72	+ _PAGE_PRESENT \| _PAGE_RW \| _PAGE_DIRTY \| _PAGE_ACCESSED \| _PAGE_PSE
73	+#define X_KERNPG_TABLE _PAGE_PRESENT \| _PAGE_RW \| _PAGE_ACCESSED \| _PAGE_DIRTY
74	+
75	+#define __ma(x) (pfn_to_mfn(__pa((x)) >> PAGE_SHIFT) << PAGE_SHIFT)
76	+
77	+#if PAGES_NR > KEXEC_XEN_NO_PAGES
78	+#error PAGES_NR is greater than KEXEC_XEN_NO_PAGES - Xen support will break
79	+#endif
80	+
81	+#if PA_CONTROL_PAGE != 0
82	+#error PA_CONTROL_PAGE is non zero - Xen support will break
83	+#endif
84	+
85	+void machine_kexec_setup_load_arg(xen_kexec_image_t xki, struct kimage image)
86	+{
87	+ void *control_page;
88	+ void *table_page;
89	+
90	+ memset(xki->page_list, 0, sizeof(xki->page_list));
91	+
92	+ control_page = page_address(image->control_code_page) + PAGE_SIZE;
93	+ memcpy(control_page, relocate_kernel, PAGE_SIZE);
94	+
95	+ table_page = page_address(image->control_code_page);
96	+
97	+ xki->page_list[PA_CONTROL_PAGE] = __ma(control_page);
98	+ xki->page_list[PA_TABLE_PAGE] = __ma(table_page);
99	+
100	+ xki->page_list[PA_PGD] = __ma(kexec_pgd);
101	+ xki->page_list[PA_PUD_0] = __ma(kexec_pud0);
102	+ xki->page_list[PA_PUD_1] = __ma(kexec_pud1);
103	+ xki->page_list[PA_PMD_0] = __ma(kexec_pmd0);
104	+ xki->page_list[PA_PMD_1] = __ma(kexec_pmd1);
105	+ xki->page_list[PA_PTE_0] = __ma(kexec_pte0);
106	+ xki->page_list[PA_PTE_1] = __ma(kexec_pte1);
107	+}
108	+
109	+int __init machine_kexec_setup_resources(struct resource *hypervisor,
110	+ struct resource *phys_cpus,
111	+ int nr_phys_cpus)
112	+{
113	+ int k;
114	+
115	+ /* The per-cpu crash note resources belong to the hypervisor resource */
116	+ for (k = 0; k < nr_phys_cpus; k++)
117	+ request_resource(hypervisor, phys_cpus + k);
118	+
119	+ return 0;
120	+}
121	+
122	+void machine_kexec_register_resources(struct resource *res) { ; }
123	+
124	+#else /* CONFIG_XEN */
125	+
126	+#define x__pmd(x) __pmd(x)
127	+#define x__pud(x) __pud(x)
128	+#define x__pgd(x) __pgd(x)
129	+
130	+#define x_set_pmd(x, y) set_pmd(x, y)
131	+#define x_set_pud(x, y) set_pud(x, y)
132	+#define x_set_pgd(x, y) set_pgd(x, y)
133	+
134	+#define x_pud_clear(x) pud_clear(x)
135	+#define x_pgd_clear(x) pgd_clear(x)
136	+
137	+#define X__PAGE_KERNEL_LARGE_EXEC __PAGE_KERNEL_LARGE_EXEC
138	+#define X_KERNPG_TABLE _KERNPG_TABLE
139	+
140	+#endif /* CONFIG_XEN */
141	+
142	static void init_level2_page(pmd_t *level2p, unsigned long addr)
143	{
144	unsigned long end_addr;
145	@@ -32,7 +145,7 @@ static void init_level2_page(pmd_t *leve
146	addr &= PAGE_MASK;
147	end_addr = addr + PUD_SIZE;
148	while (addr < end_addr) {
149	- set_pmd(level2p++, __pmd(addr \| __PAGE_KERNEL_LARGE_EXEC));
150	+ x_set_pmd(level2p++, x__pmd(addr \| X__PAGE_KERNEL_LARGE_EXEC));
151	addr += PMD_SIZE;
152	}
153	}
154	@@ -57,12 +170,12 @@ static int init_level3_page(struct kimag
155	}
156	level2p = (pmd_t *)page_address(page);
157	init_level2_page(level2p, addr);
158	- set_pud(level3p++, __pud(__pa(level2p) \| _KERNPG_TABLE));
159	+ x_set_pud(level3p++, x__pud(__pa(level2p) \| X_KERNPG_TABLE));
160	addr += PUD_SIZE;
161	}
162	/* clear the unused entries */
163	while (addr < end_addr) {
164	- pud_clear(level3p++);
165	+ x_pud_clear(level3p++);
166	addr += PUD_SIZE;
167	}
168	out:
169	@@ -93,12 +206,12 @@ static int init_level4_page(struct kimag
170	if (result) {
171	goto out;
172	}
173	- set_pgd(level4p++, __pgd(__pa(level3p) \| _KERNPG_TABLE));
174	+ x_set_pgd(level4p++, x__pgd(__pa(level3p) \| X_KERNPG_TABLE));
175	addr += PGDIR_SIZE;
176	}
177	/* clear the unused entries */
178	while (addr < end_addr) {
179	- pgd_clear(level4p++);
180	+ x_pgd_clear(level4p++);
181	addr += PGDIR_SIZE;
182	}
183	out:
184	@@ -109,8 +222,14 @@ out:
185	static int init_pgtable(struct kimage *image, unsigned long start_pgtable)
186	{
187	pgd_t *level4p;
188	+ unsigned long x_end_pfn = end_pfn;
189	+
190	+#ifdef CONFIG_XEN
191	+ x_end_pfn = HYPERVISOR_memory_op(XENMEM_maximum_ram_page, NULL);
192	+#endif
193	+
194	level4p = (pgd_t *)__va(start_pgtable);
195	- return init_level4_page(image, level4p, 0, end_pfn << PAGE_SHIFT);
196	+ return init_level4_page(image, level4p, 0, x_end_pfn << PAGE_SHIFT);
197	}
198
199	int machine_kexec_prepare(struct kimage *image)
200	@@ -134,6 +253,7 @@ void machine_kexec_cleanup(struct kimage
201	return;
202	}
203
204	+#ifndef CONFIG_XEN
205	/*
206	* Do not allocate memory (or fail in any way) in machine_kexec().
207	* We are past the point of no return, committed to rebooting now.
208	@@ -172,6 +292,7 @@ NORET_TYPE void machine_kexec(struct kim
209	relocate_kernel((unsigned long)image->head, (unsigned long)page_list,
210	image->start);
211	}
212	+#endif
213
214	void arch_crash_save_vmcoreinfo(void)
215	{
216	Index: head-2008-04-15/arch/x86/pci/Makefile_64
217	===================================================================
218	--- head-2008-04-15.orig/arch/x86/pci/Makefile_64 2008-04-15 09:41:09.000000000 +0200
219	+++ head-2008-04-15/arch/x86/pci/Makefile_64 2008-04-15 09:59:32.000000000 +0200
220	@@ -15,3 +15,7 @@ obj-$(CONFIG_PCI_MMCONFIG) += mmconfig_6
221
222	obj-$(CONFIG_NUMA) += k8-bus_64.o
223
224	+# pcifront should be after mmconfig.o and direct.o as it should only
225	+# take over if direct access to the PCI bus is unavailable
226	+obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += pcifront.o
227	+
228	Index: head-2008-04-15/arch/x86/power/cpu_64.c
229	===================================================================
230	--- head-2008-04-15.orig/arch/x86/power/cpu_64.c 2008-04-15 09:41:09.000000000 +0200
231	+++ head-2008-04-15/arch/x86/power/cpu_64.c 2008-04-15 09:59:32.000000000 +0200
232	@@ -135,6 +135,7 @@ void restore_processor_state(void)
233
234	static void fix_processor_context(void)
235	{
236	+#ifndef CONFIG_X86_NO_TSS
237	int cpu = smp_processor_id();
238	struct tss_struct *t = &per_cpu(init_tss, cpu);
239
240	@@ -146,6 +147,7 @@ static void fix_processor_context(void)
241	set_tss_desc(cpu, t);
242
243	get_cpu_gdt_table(cpu)[GDT_ENTRY_TSS].type = 9;
244	+#endif
245
246	syscall_init(); /* This sets MSR_STAR and related /
247	load_TR_desc(); /* This does ltr */
248	Index: head-2008-04-15/arch/x86/vdso/Makefile
249	===================================================================
250	--- head-2008-04-15.orig/arch/x86/vdso/Makefile 2008-04-15 09:41:09.000000000 +0200
251	+++ head-2008-04-15/arch/x86/vdso/Makefile 2008-04-15 09:59:32.000000000 +0200
252	@@ -64,6 +64,8 @@ obj-$(VDSO32-y) += vdso32-syms.lds
253	vdso32.so-$(CONFIG_X86_32) += int80
254	vdso32.so-$(CONFIG_COMPAT) += syscall
255	vdso32.so-$(VDSO32-y) += sysenter
256	+xen-vdso32-$(subst 1,$(CONFIG_COMPAT),$(shell expr $(CONFIG_XEN_COMPAT)0 '<' 0x0302000)) += int80
257	+vdso32.so-$(CONFIG_XEN) += $(xen-vdso32-y)
258
259	vdso32-images = $(vdso32.so-y:%=vdso32-%.so)
260