Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1501116 > unrolled thread

[PATCH 0/8] PVH v2 support

Back to article view | Back to linux.kernel

Contents

  [PATCH 0/8] PVH v2 support Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 20:20 +0200
    [PATCH 7/8] xen/pvh: PVH guests always have PV devices Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 20:20 +0200
      Re: [Xen-devel] [PATCH 7/8] xen/pvh: PVH guests always have PV  devices Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> - 2016-10-14 21:30 +0200
      Re: [PATCH 7/8] xen/pvh: PVH guests always have PV devices Juergen Gross <jgross@suse.com> - 2016-10-18 18:00 +0200
    [PATCH 5/8] xen/pvh: Prevent PVH guests from using PIC, RTC and IOAPIC Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 20:20 +0200
      Re: [Xen-devel] [PATCH 5/8] xen/pvh: Prevent PVH guests from using  PIC, RTC and IOAPIC Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> - 2016-10-14 21:20 +0200
        Re: [Xen-devel] [PATCH 5/8] xen/pvh: Prevent PVH guests from using  PIC, RTC and IOAPIC Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 21:40 +0200
      Re: [PATCH 5/8] xen/pvh: Prevent PVH guests from using PIC, RTC and  IOAPIC Roger Pau Monné <roger.pau@citrix.com> - 2016-10-26 12:50 +0200
        Re: [PATCH 5/8] xen/pvh: Prevent PVH guests from using PIC, RTC and  IOAPIC Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-26 16:50 +0200
          Re: [PATCH 5/8] xen/pvh: Prevent PVH guests from using PIC, RTC and  IOAPIC Roger Pau Monné <roger.pau@citrix.com> - 2016-10-26 17:20 +0200
            Re: [PATCH 5/8] xen/pvh: Prevent PVH guests from using PIC, RTC and  IOAPIC Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-26 18:10 +0200
    [PATCH 8/8] xen/pvh: Enable CPU hotplug Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 20:20 +0200
      Re: [Xen-devel] [PATCH 8/8] xen/pvh: Enable CPU hotplug Andrew Cooper <andrew.cooper3@citrix.com> - 2016-10-14 20:50 +0200
        Re: [Xen-devel] [PATCH 8/8] xen/pvh: Enable CPU hotplug Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 21:10 +0200
    [PATCH 4/8] xen/pvh: Bootstrap PVH guest Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 20:20 +0200
      Re: [Xen-devel] [PATCH 4/8] xen/pvh: Bootstrap PVH guest Andrew Cooper <andrew.cooper3@citrix.com> - 2016-10-14 20:50 +0200
        Re: [Xen-devel] [PATCH 4/8] xen/pvh: Bootstrap PVH guest Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 21:00 +0200
          Re: [Xen-devel] [PATCH 4/8] xen/pvh: Bootstrap PVH guest Andrew Cooper <andrew.cooper3@citrix.com> - 2016-10-14 21:20 +0200
      Re: [Xen-devel] [PATCH 4/8] xen/pvh: Bootstrap PVH guest Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> - 2016-10-14 21:20 +0200
        Re: [Xen-devel] [PATCH 4/8] xen/pvh: Bootstrap PVH guest Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 21:40 +0200
    [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 20:20 +0200
      Re: [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup hpa@zytor.com - 2016-10-14 20:40 +0200
        Re: [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 20:50 +0200
          Re: [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup hpa@zytor.com - 2016-10-14 21:10 +0200
            Re: [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 21:30 +0200
    [PATCH 3/8] xen/pvh: Import PVH-related Xen public interfaces Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 20:20 +0200
      Re: [Xen-devel] [PATCH 3/8] xen/pvh: Import PVH-related Xen public  interfaces Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> - 2016-10-14 20:40 +0200
      Re: [PATCH 3/8] xen/pvh: Import PVH-related Xen public interfaces Juergen Gross <jgross@suse.com> - 2016-10-21 13:00 +0200
    [PATCH 1/8] xen/x86: Remove PVH support Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-14 20:20 +0200
      Re: [Xen-devel] [PATCH 1/8] xen/x86: Remove PVH support Konrad Rzeszutek Wilk <konrad.wilk@oracle.com> - 2016-10-14 20:40 +0200
      Re: [PATCH 1/8] xen/x86: Remove PVH support Juergen Gross <jgross@suse.com> - 2016-10-18 15:50 +0200
        Re: [PATCH 1/8] xen/x86: Remove PVH support Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-18 16:50 +0200
          Re: [PATCH 1/8] xen/x86: Remove PVH support Juergen Gross <jgross@suse.com> - 2016-10-18 17:40 +0200
            Re: [PATCH 1/8] xen/x86: Remove PVH support Boris Ostrovsky <boris.ostrovsky@oracle.com> - 2016-10-18 17:50 +0200

Page 2 of 2 — ← Prev page 1 [2]

#1501121 — [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup

From: Matt Fleming <matt@codeblueprint.co.uk>

The new Xen PVH entry point requires page tables to be setup by the
kernel since it is entered with paging disabled.

Pull the common code out of head_32.S and into pgtable_32.S so that
setup_pgtable_32 can be invoked from both the new Xen entry point and
the existing startup_32 code.

Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: x86@kernel.org
Signed-off-by: Matt Fleming <matt@codeblueprint.co.uk>
---
 arch/x86/Makefile            |   2 +
 arch/x86/kernel/Makefile     |   2 +
 arch/x86/kernel/head_32.S    | 168 +------------------------------------
 arch/x86/kernel/pgtable_32.S | 196 +++++++++++++++++++++++++++++++++++++++++++
 4 files changed, 201 insertions(+), 167 deletions(-)
 create mode 100644 arch/x86/kernel/pgtable_32.S

diff --git a/arch/x86/Makefile b/arch/x86/Makefile
index 2d44933..67cc771 100644
--- a/arch/x86/Makefile
+++ b/arch/x86/Makefile
@@ -204,6 +204,8 @@ head-y += arch/x86/kernel/head$(BITS).o
 head-y += arch/x86/kernel/ebda.o
 head-y += arch/x86/kernel/platform-quirks.o
 
+head-$(CONFIG_X86_32) += arch/x86/kernel/pgtable_32.o
+
 libs-y  += arch/x86/lib/
 
 # See arch/x86/Kbuild for content of core part of the kernel
diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
index 4dd5d50..eae85a5 100644
--- a/arch/x86/kernel/Makefile
+++ b/arch/x86/kernel/Makefile
@@ -8,6 +8,8 @@ extra-y	+= ebda.o
 extra-y	+= platform-quirks.o
 extra-y	+= vmlinux.lds
 
+extra-$(CONFIG_X86_32) += pgtable_32.o
+
 CPPFLAGS_vmlinux.lds += -U$(UTS_MACHINE)
 
 ifdef CONFIG_FUNCTION_TRACER
diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
index 5f40126..0db066e 100644
--- a/arch/x86/kernel/head_32.S
+++ b/arch/x86/kernel/head_32.S
@@ -41,51 +41,6 @@
 #define X86_VENDOR_ID	new_cpu_data+CPUINFO_x86_vendor_id
 
 /*
- * This is how much memory in addition to the memory covered up to
- * and including _end we need mapped initially.
- * We need:
- *     (KERNEL_IMAGE_SIZE/4096) / 1024 pages (worst case, non PAE)
- *     (KERNEL_IMAGE_SIZE/4096) / 512 + 4 pages (worst case for PAE)
- *
- * Modulo rounding, each megabyte assigned here requires a kilobyte of
- * memory, which is currently unreclaimed.
- *
- * This should be a multiple of a page.
- *
- * KERNEL_IMAGE_SIZE should be greater than pa(_end)
- * and small than max_low_pfn, otherwise will waste some page table entries
- */
-
-#if PTRS_PER_PMD > 1
-#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) + PTRS_PER_PGD)
-#else
-#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD)
-#endif
-
-/*
- * Number of possible pages in the lowmem region.
- *
- * We shift 2 by 31 instead of 1 by 32 to the left in order to avoid a
- * gas warning about overflowing shift count when gas has been compiled
- * with only a host target support using a 32-bit type for internal
- * representation.
- */
-LOWMEM_PAGES = (((2<<31) - __PAGE_OFFSET) >> PAGE_SHIFT)
-
-/* Enough space to fit pagetables for the low memory linear map */
-MAPPING_BEYOND_END = PAGE_TABLE_SIZE(LOWMEM_PAGES) << PAGE_SHIFT
-
-/*
- * Worst-case size of the kernel mapping we need to make:
- * a relocatable kernel can live anywhere in lowmem, so we need to be able
- * to map all of lowmem.
- */
-KERNEL_PAGES = LOWMEM_PAGES
-
-INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_PAGES) * PAGE_SIZE
-RESERVE_BRK(pagetables, INIT_MAP_SIZE)
-
-/*
  * 32-bit kernel entrypoint; only used by the boot CPU.  On entry,
  * %esi points to the real-mode code as a 32-bit pointer.
  * CS and DS must be 4 GB flat segments, but we don't depend on
@@ -157,92 +112,7 @@ ENTRY(startup_32)
 	call load_ucode_bsp
 #endif
 
-/*
- * Initialize page tables.  This creates a PDE and a set of page
- * tables, which are located immediately beyond __brk_base.  The variable
- * _brk_end is set up to point to the first "safe" location.
- * Mappings are created both at virtual address 0 (identity mapping)
- * and PAGE_OFFSET for up to _end.
- */
-#ifdef CONFIG_X86_PAE
-
-	/*
-	 * In PAE mode initial_page_table is statically defined to contain
-	 * enough entries to cover the VMSPLIT option (that is the top 1, 2 or 3
-	 * entries). The identity mapping is handled by pointing two PGD entries
-	 * to the first kernel PMD.
-	 *
-	 * Note the upper half of each PMD or PTE are always zero at this stage.
-	 */
-
-#define KPMDS (((-__PAGE_OFFSET) >> 30) & 3) /* Number of kernel PMDs */
-
-	xorl %ebx,%ebx				/* %ebx is kept at zero */
-
-	movl $pa(__brk_base), %edi
-	movl $pa(initial_pg_pmd), %edx
-	movl $PTE_IDENT_ATTR, %eax
-10:
-	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PMD entry */
-	movl %ecx,(%edx)			/* Store PMD entry */
-						/* Upper half already zero */
-	addl $8,%edx
-	movl $512,%ecx
-11:
-	stosl
-	xchgl %eax,%ebx
-	stosl
-	xchgl %eax,%ebx
-	addl $0x1000,%eax
-	loop 11b
-
-	/*
-	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
-	 */
-	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
-	cmpl %ebp,%eax
-	jb 10b
-1:
-	addl $__PAGE_OFFSET, %edi
-	movl %edi, pa(_brk_end)
-	shrl $12, %eax
-	movl %eax, pa(max_pfn_mapped)
-
-	/* Do early initialization of the fixmap area */
-	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
-	movl %eax,pa(initial_pg_pmd+0x1000*KPMDS-8)
-#else	/* Not PAE */
-
-page_pde_offset = (__PAGE_OFFSET >> 20);
-
-	movl $pa(__brk_base), %edi
-	movl $pa(initial_page_table), %edx
-	movl $PTE_IDENT_ATTR, %eax
-10:
-	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PDE entry */
-	movl %ecx,(%edx)			/* Store identity PDE entry */
-	movl %ecx,page_pde_offset(%edx)		/* Store kernel PDE entry */
-	addl $4,%edx
-	movl $1024, %ecx
-11:
-	stosl
-	addl $0x1000,%eax
-	loop 11b
-	/*
-	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
-	 */
-	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
-	cmpl %ebp,%eax
-	jb 10b
-	addl $__PAGE_OFFSET, %edi
-	movl %edi, pa(_brk_end)
-	shrl $12, %eax
-	movl %eax, pa(max_pfn_mapped)
-
-	/* Do early initialization of the fixmap area */
-	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
-	movl %eax,pa(initial_page_table+0xffc)
-#endif
+	call setup_pgtable_32
 
 #ifdef CONFIG_PARAVIRT
 	/* This is can only trip for a broken bootloader... */
@@ -660,47 +530,11 @@ ENTRY(setup_once_ref)
  */
 __PAGE_ALIGNED_BSS
 	.align PAGE_SIZE
-#ifdef CONFIG_X86_PAE
-initial_pg_pmd:
-	.fill 1024*KPMDS,4,0
-#else
-ENTRY(initial_page_table)
-	.fill 1024,4,0
-#endif
-initial_pg_fixmap:
-	.fill 1024,4,0
 ENTRY(empty_zero_page)
 	.fill 4096,1,0
 ENTRY(swapper_pg_dir)
 	.fill 1024,4,0
 
-/*
- * This starts the data section.
- */
-#ifdef CONFIG_X86_PAE
-__PAGE_ALIGNED_DATA
-	/* Page-aligned for the benefit of paravirt? */
-	.align PAGE_SIZE
-ENTRY(initial_page_table)
-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0	/* low identity map */
-# if KPMDS == 3
-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0
-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x2000),0
-# elif KPMDS == 2
-	.long	0,0
-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0
-# elif KPMDS == 1
-	.long	0,0
-	.long	0,0
-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
-# else
-#  error "Kernel PMDs should be 1, 2 or 3"
-# endif
-	.align PAGE_SIZE		/* needs to be page-sized too */
-#endif
-
 .data
 .balign 4
 ENTRY(initial_stack)
diff --git a/arch/x86/kernel/pgtable_32.S b/arch/x86/kernel/pgtable_32.S
new file mode 100644
index 0000000..aded718
--- /dev/null
+++ b/arch/x86/kernel/pgtable_32.S
@@ -0,0 +1,196 @@
+#include <linux/threads.h>
+#include <linux/init.h>
+#include <linux/linkage.h>
+#include <asm/segment.h>
+#include <asm/page_types.h>
+#include <asm/pgtable_types.h>
+#include <asm/cache.h>
+#include <asm/thread_info.h>
+#include <asm/asm-offsets.h>
+#include <asm/setup.h>
+#include <asm/processor-flags.h>
+#include <asm/msr-index.h>
+#include <asm/cpufeatures.h>
+#include <asm/percpu.h>
+#include <asm/nops.h>
+#include <asm/bootparam.h>
+
+/* Physical address */
+#define pa(X) ((X) - __PAGE_OFFSET)
+
+/*
+ * This is how much memory in addition to the memory covered up to
+ * and including _end we need mapped initially.
+ * We need:
+ *     (KERNEL_IMAGE_SIZE/4096) / 1024 pages (worst case, non PAE)
+ *     (KERNEL_IMAGE_SIZE/4096) / 512 + 4 pages (worst case for PAE)
+ *
+ * Modulo rounding, each megabyte assigned here requires a kilobyte of
+ * memory, which is currently unreclaimed.
+ *
+ * This should be a multiple of a page.
+ *
+ * KERNEL_IMAGE_SIZE should be greater than pa(_end)
+ * and small than max_low_pfn, otherwise will waste some page table entries
+ */
+
+#if PTRS_PER_PMD > 1
+#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) + PTRS_PER_PGD)
+#else
+#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD)
+#endif
+
+/*
+ * Number of possible pages in the lowmem region.
+ *
+ * We shift 2 by 31 instead of 1 by 32 to the left in order to avoid a
+ * gas warning about overflowing shift count when gas has been compiled
+ * with only a host target support using a 32-bit type for internal
+ * representation.
+ */
+LOWMEM_PAGES = (((2<<31) - __PAGE_OFFSET) >> PAGE_SHIFT)
+
+/* Enough space to fit pagetables for the low memory linear map */
+MAPPING_BEYOND_END = PAGE_TABLE_SIZE(LOWMEM_PAGES) << PAGE_SHIFT
+
+/*
+ * Worst-case size of the kernel mapping we need to make:
+ * a relocatable kernel can live anywhere in lowmem, so we need to be able
+ * to map all of lowmem.
+ */
+KERNEL_PAGES = LOWMEM_PAGES
+
+INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_PAGES) * PAGE_SIZE
+RESERVE_BRK(pagetables, INIT_MAP_SIZE)
+
+/*
+ * Initialize page tables.  This creates a PDE and a set of page
+ * tables, which are located immediately beyond __brk_base.  The variable
+ * _brk_end is set up to point to the first "safe" location.
+ * Mappings are created both at virtual address 0 (identity mapping)
+ * and PAGE_OFFSET for up to _end.
+ */
+	.text
+ENTRY(setup_pgtable_32)
+#ifdef CONFIG_X86_PAE
+	/*
+	 * In PAE mode initial_page_table is statically defined to contain
+	 * enough entries to cover the VMSPLIT option (that is the top 1, 2 or 3
+	 * entries). The identity mapping is handled by pointing two PGD entries
+	 * to the first kernel PMD.
+	 *
+	 * Note the upper half of each PMD or PTE are always zero at this stage.
+	 */
+
+#define KPMDS (((-__PAGE_OFFSET) >> 30) & 3) /* Number of kernel PMDs */
+
+	xorl %ebx,%ebx				/* %ebx is kept at zero */
+
+	movl $pa(__brk_base), %edi
+	movl $pa(initial_pg_pmd), %edx
+	movl $PTE_IDENT_ATTR, %eax
+10:
+	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PMD entry */
+	movl %ecx,(%edx)			/* Store PMD entry */
+						/* Upper half already zero */
+	addl $8,%edx
+	movl $512,%ecx
+11:
+	stosl
+	xchgl %eax,%ebx
+	stosl
+	xchgl %eax,%ebx
+	addl $0x1000,%eax
+	loop 11b
+
+	/*
+	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
+	 */
+	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
+	cmpl %ebp,%eax
+	jb 10b
+1:
+	addl $__PAGE_OFFSET, %edi
+	movl %edi, pa(_brk_end)
+	shrl $12, %eax
+	movl %eax, pa(max_pfn_mapped)
+
+	/* Do early initialization of the fixmap area */
+	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
+	movl %eax,pa(initial_pg_pmd+0x1000*KPMDS-8)
+#else	/* Not PAE */
+
+page_pde_offset = (__PAGE_OFFSET >> 20);
+
+	movl $pa(__brk_base), %edi
+	movl $pa(initial_page_table), %edx
+	movl $PTE_IDENT_ATTR, %eax
+10:
+	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PDE entry */
+	movl %ecx,(%edx)			/* Store identity PDE entry */
+	movl %ecx,page_pde_offset(%edx)		/* Store kernel PDE entry */
+	addl $4,%edx
+	movl $1024, %ecx
+11:
+	stosl
+	addl $0x1000,%eax
+	loop 11b
+	/*
+	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
+	 */
+	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
+	cmpl %ebp,%eax
+	jb 10b
+	addl $__PAGE_OFFSET, %edi
+	movl %edi, pa(_brk_end)
+	shrl $12, %eax
+	movl %eax, pa(max_pfn_mapped)
+
+	/* Do early initialization of the fixmap area */
+	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
+	movl %eax,pa(initial_page_table+0xffc)
+#endif
+	ret
+ENDPROC(setup_pgtable_32)
+
+/*
+ * BSS section
+ */
+__PAGE_ALIGNED_BSS
+	.align PAGE_SIZE
+#ifdef CONFIG_X86_PAE
+initial_pg_pmd:
+	.fill 1024*KPMDS,4,0
+#else
+ENTRY(initial_page_table)
+	.fill 1024,4,0
+#endif
+initial_pg_fixmap:
+	.fill 1024,4,0
+
+/*
+ * This starts the data section.
+ */
+#ifdef CONFIG_X86_PAE
+__PAGE_ALIGNED_DATA
+	/* Page-aligned for the benefit of paravirt? */
+	.align PAGE_SIZE
+ENTRY(initial_page_table)
+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0	/* low identity map */
+# if KPMDS == 3
+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0
+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x2000),0
+# elif KPMDS == 2
+	.long	0,0
+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0
+# elif KPMDS == 1
+	.long	0,0
+	.long	0,0
+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
+# else
+#  error "Kernel PMDs should be 1, 2 or 3"
+# endif
+	.align PAGE_SIZE		/* needs to be page-sized too */
+#endif
-- 
1.8.3.1

[toc] | [prev] | [next] | [standalone]

#1501138 — Re: [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup

On October 14, 2016 11:05:12 AM PDT, Boris Ostrovsky <boris.ostrovsky@oracle.com> wrote:
>From: Matt Fleming <matt@codeblueprint.co.uk>
>
>The new Xen PVH entry point requires page tables to be setup by the
>kernel since it is entered with paging disabled.
>
>Pull the common code out of head_32.S and into pgtable_32.S so that
>setup_pgtable_32 can be invoked from both the new Xen entry point and
>the existing startup_32 code.
>
>Cc: Boris Ostrovsky <boris.ostrovsky@oracle.com>
>Cc: Thomas Gleixner <tglx@linutronix.de>
>Cc: Ingo Molnar <mingo@redhat.com>
>Cc: "H. Peter Anvin" <hpa@zytor.com>
>Cc: x86@kernel.org
>Signed-off-by: Matt Fleming <matt@codeblueprint.co.uk>
>---
> arch/x86/Makefile            |   2 +
> arch/x86/kernel/Makefile     |   2 +
>arch/x86/kernel/head_32.S    | 168
>+------------------------------------
>arch/x86/kernel/pgtable_32.S | 196
>+++++++++++++++++++++++++++++++++++++++++++
> 4 files changed, 201 insertions(+), 167 deletions(-)
> create mode 100644 arch/x86/kernel/pgtable_32.S
>
>diff --git a/arch/x86/Makefile b/arch/x86/Makefile
>index 2d44933..67cc771 100644
>--- a/arch/x86/Makefile
>+++ b/arch/x86/Makefile
>@@ -204,6 +204,8 @@ head-y += arch/x86/kernel/head$(BITS).o
> head-y += arch/x86/kernel/ebda.o
> head-y += arch/x86/kernel/platform-quirks.o
> 
>+head-$(CONFIG_X86_32) += arch/x86/kernel/pgtable_32.o
>+
> libs-y  += arch/x86/lib/
> 
> # See arch/x86/Kbuild for content of core part of the kernel
>diff --git a/arch/x86/kernel/Makefile b/arch/x86/kernel/Makefile
>index 4dd5d50..eae85a5 100644
>--- a/arch/x86/kernel/Makefile
>+++ b/arch/x86/kernel/Makefile
>@@ -8,6 +8,8 @@ extra-y	+= ebda.o
> extra-y	+= platform-quirks.o
> extra-y	+= vmlinux.lds
> 
>+extra-$(CONFIG_X86_32) += pgtable_32.o
>+
> CPPFLAGS_vmlinux.lds += -U$(UTS_MACHINE)
> 
> ifdef CONFIG_FUNCTION_TRACER
>diff --git a/arch/x86/kernel/head_32.S b/arch/x86/kernel/head_32.S
>index 5f40126..0db066e 100644
>--- a/arch/x86/kernel/head_32.S
>+++ b/arch/x86/kernel/head_32.S
>@@ -41,51 +41,6 @@
> #define X86_VENDOR_ID	new_cpu_data+CPUINFO_x86_vendor_id
> 
> /*
>- * This is how much memory in addition to the memory covered up to
>- * and including _end we need mapped initially.
>- * We need:
>- *     (KERNEL_IMAGE_SIZE/4096) / 1024 pages (worst case, non PAE)
>- *     (KERNEL_IMAGE_SIZE/4096) / 512 + 4 pages (worst case for PAE)
>- *
>- * Modulo rounding, each megabyte assigned here requires a kilobyte of
>- * memory, which is currently unreclaimed.
>- *
>- * This should be a multiple of a page.
>- *
>- * KERNEL_IMAGE_SIZE should be greater than pa(_end)
>- * and small than max_low_pfn, otherwise will waste some page table
>entries
>- */
>-
>-#if PTRS_PER_PMD > 1
>-#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) +
>PTRS_PER_PGD)
>-#else
>-#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD)
>-#endif
>-
>-/*
>- * Number of possible pages in the lowmem region.
>- *
>- * We shift 2 by 31 instead of 1 by 32 to the left in order to avoid a
>- * gas warning about overflowing shift count when gas has been
>compiled
>- * with only a host target support using a 32-bit type for internal
>- * representation.
>- */
>-LOWMEM_PAGES = (((2<<31) - __PAGE_OFFSET) >> PAGE_SHIFT)
>-
>-/* Enough space to fit pagetables for the low memory linear map */
>-MAPPING_BEYOND_END = PAGE_TABLE_SIZE(LOWMEM_PAGES) << PAGE_SHIFT
>-
>-/*
>- * Worst-case size of the kernel mapping we need to make:
>- * a relocatable kernel can live anywhere in lowmem, so we need to be
>able
>- * to map all of lowmem.
>- */
>-KERNEL_PAGES = LOWMEM_PAGES
>-
>-INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_PAGES) * PAGE_SIZE
>-RESERVE_BRK(pagetables, INIT_MAP_SIZE)
>-
>-/*
>  * 32-bit kernel entrypoint; only used by the boot CPU.  On entry,
>  * %esi points to the real-mode code as a 32-bit pointer.
>  * CS and DS must be 4 GB flat segments, but we don't depend on
>@@ -157,92 +112,7 @@ ENTRY(startup_32)
> 	call load_ucode_bsp
> #endif
> 
>-/*
>- * Initialize page tables.  This creates a PDE and a set of page
>- * tables, which are located immediately beyond __brk_base.  The
>variable
>- * _brk_end is set up to point to the first "safe" location.
>- * Mappings are created both at virtual address 0 (identity mapping)
>- * and PAGE_OFFSET for up to _end.
>- */
>-#ifdef CONFIG_X86_PAE
>-
>-	/*
>-	 * In PAE mode initial_page_table is statically defined to contain
>-	 * enough entries to cover the VMSPLIT option (that is the top 1, 2
>or 3
>-	 * entries). The identity mapping is handled by pointing two PGD
>entries
>-	 * to the first kernel PMD.
>-	 *
>-	 * Note the upper half of each PMD or PTE are always zero at this
>stage.
>-	 */
>-
>-#define KPMDS (((-__PAGE_OFFSET) >> 30) & 3) /* Number of kernel PMDs
>*/
>-
>-	xorl %ebx,%ebx				/* %ebx is kept at zero */
>-
>-	movl $pa(__brk_base), %edi
>-	movl $pa(initial_pg_pmd), %edx
>-	movl $PTE_IDENT_ATTR, %eax
>-10:
>-	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PMD entry */
>-	movl %ecx,(%edx)			/* Store PMD entry */
>-						/* Upper half already zero */
>-	addl $8,%edx
>-	movl $512,%ecx
>-11:
>-	stosl
>-	xchgl %eax,%ebx
>-	stosl
>-	xchgl %eax,%ebx
>-	addl $0x1000,%eax
>-	loop 11b
>-
>-	/*
>-	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
>-	 */
>-	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
>-	cmpl %ebp,%eax
>-	jb 10b
>-1:
>-	addl $__PAGE_OFFSET, %edi
>-	movl %edi, pa(_brk_end)
>-	shrl $12, %eax
>-	movl %eax, pa(max_pfn_mapped)
>-
>-	/* Do early initialization of the fixmap area */
>-	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
>-	movl %eax,pa(initial_pg_pmd+0x1000*KPMDS-8)
>-#else	/* Not PAE */
>-
>-page_pde_offset = (__PAGE_OFFSET >> 20);
>-
>-	movl $pa(__brk_base), %edi
>-	movl $pa(initial_page_table), %edx
>-	movl $PTE_IDENT_ATTR, %eax
>-10:
>-	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PDE entry */
>-	movl %ecx,(%edx)			/* Store identity PDE entry */
>-	movl %ecx,page_pde_offset(%edx)		/* Store kernel PDE entry */
>-	addl $4,%edx
>-	movl $1024, %ecx
>-11:
>-	stosl
>-	addl $0x1000,%eax
>-	loop 11b
>-	/*
>-	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
>-	 */
>-	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
>-	cmpl %ebp,%eax
>-	jb 10b
>-	addl $__PAGE_OFFSET, %edi
>-	movl %edi, pa(_brk_end)
>-	shrl $12, %eax
>-	movl %eax, pa(max_pfn_mapped)
>-
>-	/* Do early initialization of the fixmap area */
>-	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
>-	movl %eax,pa(initial_page_table+0xffc)
>-#endif
>+	call setup_pgtable_32
> 
> #ifdef CONFIG_PARAVIRT
> 	/* This is can only trip for a broken bootloader... */
>@@ -660,47 +530,11 @@ ENTRY(setup_once_ref)
>  */
> __PAGE_ALIGNED_BSS
> 	.align PAGE_SIZE
>-#ifdef CONFIG_X86_PAE
>-initial_pg_pmd:
>-	.fill 1024*KPMDS,4,0
>-#else
>-ENTRY(initial_page_table)
>-	.fill 1024,4,0
>-#endif
>-initial_pg_fixmap:
>-	.fill 1024,4,0
> ENTRY(empty_zero_page)
> 	.fill 4096,1,0
> ENTRY(swapper_pg_dir)
> 	.fill 1024,4,0
> 
>-/*
>- * This starts the data section.
>- */
>-#ifdef CONFIG_X86_PAE
>-__PAGE_ALIGNED_DATA
>-	/* Page-aligned for the benefit of paravirt? */
>-	.align PAGE_SIZE
>-ENTRY(initial_page_table)
>-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0	/* low identity map */
>-# if KPMDS == 3
>-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
>-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0
>-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x2000),0
>-# elif KPMDS == 2
>-	.long	0,0
>-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
>-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0
>-# elif KPMDS == 1
>-	.long	0,0
>-	.long	0,0
>-	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
>-# else
>-#  error "Kernel PMDs should be 1, 2 or 3"
>-# endif
>-	.align PAGE_SIZE		/* needs to be page-sized too */
>-#endif
>-
> .data
> .balign 4
> ENTRY(initial_stack)
>diff --git a/arch/x86/kernel/pgtable_32.S
>b/arch/x86/kernel/pgtable_32.S
>new file mode 100644
>index 0000000..aded718
>--- /dev/null
>+++ b/arch/x86/kernel/pgtable_32.S
>@@ -0,0 +1,196 @@
>+#include <linux/threads.h>
>+#include <linux/init.h>
>+#include <linux/linkage.h>
>+#include <asm/segment.h>
>+#include <asm/page_types.h>
>+#include <asm/pgtable_types.h>
>+#include <asm/cache.h>
>+#include <asm/thread_info.h>
>+#include <asm/asm-offsets.h>
>+#include <asm/setup.h>
>+#include <asm/processor-flags.h>
>+#include <asm/msr-index.h>
>+#include <asm/cpufeatures.h>
>+#include <asm/percpu.h>
>+#include <asm/nops.h>
>+#include <asm/bootparam.h>
>+
>+/* Physical address */
>+#define pa(X) ((X) - __PAGE_OFFSET)
>+
>+/*
>+ * This is how much memory in addition to the memory covered up to
>+ * and including _end we need mapped initially.
>+ * We need:
>+ *     (KERNEL_IMAGE_SIZE/4096) / 1024 pages (worst case, non PAE)
>+ *     (KERNEL_IMAGE_SIZE/4096) / 512 + 4 pages (worst case for PAE)
>+ *
>+ * Modulo rounding, each megabyte assigned here requires a kilobyte of
>+ * memory, which is currently unreclaimed.
>+ *
>+ * This should be a multiple of a page.
>+ *
>+ * KERNEL_IMAGE_SIZE should be greater than pa(_end)
>+ * and small than max_low_pfn, otherwise will waste some page table
>entries
>+ */
>+
>+#if PTRS_PER_PMD > 1
>+#define PAGE_TABLE_SIZE(pages) (((pages) / PTRS_PER_PMD) +
>PTRS_PER_PGD)
>+#else
>+#define PAGE_TABLE_SIZE(pages) ((pages) / PTRS_PER_PGD)
>+#endif
>+
>+/*
>+ * Number of possible pages in the lowmem region.
>+ *
>+ * We shift 2 by 31 instead of 1 by 32 to the left in order to avoid a
>+ * gas warning about overflowing shift count when gas has been
>compiled
>+ * with only a host target support using a 32-bit type for internal
>+ * representation.
>+ */
>+LOWMEM_PAGES = (((2<<31) - __PAGE_OFFSET) >> PAGE_SHIFT)
>+
>+/* Enough space to fit pagetables for the low memory linear map */
>+MAPPING_BEYOND_END = PAGE_TABLE_SIZE(LOWMEM_PAGES) << PAGE_SHIFT
>+
>+/*
>+ * Worst-case size of the kernel mapping we need to make:
>+ * a relocatable kernel can live anywhere in lowmem, so we need to be
>able
>+ * to map all of lowmem.
>+ */
>+KERNEL_PAGES = LOWMEM_PAGES
>+
>+INIT_MAP_SIZE = PAGE_TABLE_SIZE(KERNEL_PAGES) * PAGE_SIZE
>+RESERVE_BRK(pagetables, INIT_MAP_SIZE)
>+
>+/*
>+ * Initialize page tables.  This creates a PDE and a set of page
>+ * tables, which are located immediately beyond __brk_base.  The
>variable
>+ * _brk_end is set up to point to the first "safe" location.
>+ * Mappings are created both at virtual address 0 (identity mapping)
>+ * and PAGE_OFFSET for up to _end.
>+ */
>+	.text
>+ENTRY(setup_pgtable_32)
>+#ifdef CONFIG_X86_PAE
>+	/*
>+	 * In PAE mode initial_page_table is statically defined to contain
>+	 * enough entries to cover the VMSPLIT option (that is the top 1, 2
>or 3
>+	 * entries). The identity mapping is handled by pointing two PGD
>entries
>+	 * to the first kernel PMD.
>+	 *
>+	 * Note the upper half of each PMD or PTE are always zero at this
>stage.
>+	 */
>+
>+#define KPMDS (((-__PAGE_OFFSET) >> 30) & 3) /* Number of kernel PMDs
>*/
>+
>+	xorl %ebx,%ebx				/* %ebx is kept at zero */
>+
>+	movl $pa(__brk_base), %edi
>+	movl $pa(initial_pg_pmd), %edx
>+	movl $PTE_IDENT_ATTR, %eax
>+10:
>+	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PMD entry */
>+	movl %ecx,(%edx)			/* Store PMD entry */
>+						/* Upper half already zero */
>+	addl $8,%edx
>+	movl $512,%ecx
>+11:
>+	stosl
>+	xchgl %eax,%ebx
>+	stosl
>+	xchgl %eax,%ebx
>+	addl $0x1000,%eax
>+	loop 11b
>+
>+	/*
>+	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
>+	 */
>+	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
>+	cmpl %ebp,%eax
>+	jb 10b
>+1:
>+	addl $__PAGE_OFFSET, %edi
>+	movl %edi, pa(_brk_end)
>+	shrl $12, %eax
>+	movl %eax, pa(max_pfn_mapped)
>+
>+	/* Do early initialization of the fixmap area */
>+	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
>+	movl %eax,pa(initial_pg_pmd+0x1000*KPMDS-8)
>+#else	/* Not PAE */
>+
>+page_pde_offset = (__PAGE_OFFSET >> 20);
>+
>+	movl $pa(__brk_base), %edi
>+	movl $pa(initial_page_table), %edx
>+	movl $PTE_IDENT_ATTR, %eax
>+10:
>+	leal PDE_IDENT_ATTR(%edi),%ecx		/* Create PDE entry */
>+	movl %ecx,(%edx)			/* Store identity PDE entry */
>+	movl %ecx,page_pde_offset(%edx)		/* Store kernel PDE entry */
>+	addl $4,%edx
>+	movl $1024, %ecx
>+11:
>+	stosl
>+	addl $0x1000,%eax
>+	loop 11b
>+	/*
>+	 * End condition: we must map up to the end + MAPPING_BEYOND_END.
>+	 */
>+	movl $pa(_end) + MAPPING_BEYOND_END + PTE_IDENT_ATTR, %ebp
>+	cmpl %ebp,%eax
>+	jb 10b
>+	addl $__PAGE_OFFSET, %edi
>+	movl %edi, pa(_brk_end)
>+	shrl $12, %eax
>+	movl %eax, pa(max_pfn_mapped)
>+
>+	/* Do early initialization of the fixmap area */
>+	movl $pa(initial_pg_fixmap)+PDE_IDENT_ATTR,%eax
>+	movl %eax,pa(initial_page_table+0xffc)
>+#endif
>+	ret
>+ENDPROC(setup_pgtable_32)
>+
>+/*
>+ * BSS section
>+ */
>+__PAGE_ALIGNED_BSS
>+	.align PAGE_SIZE
>+#ifdef CONFIG_X86_PAE
>+initial_pg_pmd:
>+	.fill 1024*KPMDS,4,0
>+#else
>+ENTRY(initial_page_table)
>+	.fill 1024,4,0
>+#endif
>+initial_pg_fixmap:
>+	.fill 1024,4,0
>+
>+/*
>+ * This starts the data section.
>+ */
>+#ifdef CONFIG_X86_PAE
>+__PAGE_ALIGNED_DATA
>+	/* Page-aligned for the benefit of paravirt? */
>+	.align PAGE_SIZE
>+ENTRY(initial_page_table)
>+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0	/* low identity map */
>+# if KPMDS == 3
>+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
>+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0
>+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x2000),0
>+# elif KPMDS == 2
>+	.long	0,0
>+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
>+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR+0x1000),0
>+# elif KPMDS == 1
>+	.long	0,0
>+	.long	0,0
>+	.long	pa(initial_pg_pmd+PGD_IDENT_ATTR),0
>+# else
>+#  error "Kernel PMDs should be 1, 2 or 3"
>+# endif
>+	.align PAGE_SIZE		/* needs to be page-sized too */
>+#endif

And why does it need a separate entry point as opposed to the plain one?
-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.

[toc] | [prev] | [next] | [standalone]

#1501145 — Re: [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup

On 10/14/2016 02:31 PM, hpa@zytor.com wrote:
> On October 14, 2016 11:05:12 AM PDT, Boris Ostrovsky <boris.ostrovsky@oracle.com> wrote:
>> From: Matt Fleming <matt@codeblueprint.co.uk>
>>
>> The new Xen PVH entry point requires page tables to be setup by the
>> kernel since it is entered with paging disabled.
>>
>> Pull the common code out of head_32.S and into pgtable_32.S so that
>> setup_pgtable_32 can be invoked from both the new Xen entry point and
>> the existing startup_32 code.
>>
> And why does it need a separate entry point as opposed to the plain one?

One reason is that we need to prepare boot_params before jumping to
startup_{32|64}.

When the guest is loaded (always in 32-bit mode) the only thing we have
is a pointer to Xen-specific datastructure. The early PVH code will
prepare zeropage based on that structure and then jump to regular
startup_*() code.

-boris

[toc] | [prev] | [next] | [standalone]

#1501154 — Re: [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup

On October 14, 2016 11:44:18 AM PDT, Boris Ostrovsky <boris.ostrovsky@oracle.com> wrote:
>On 10/14/2016 02:31 PM, hpa@zytor.com wrote:
>> On October 14, 2016 11:05:12 AM PDT, Boris Ostrovsky
><boris.ostrovsky@oracle.com> wrote:
>>> From: Matt Fleming <matt@codeblueprint.co.uk>
>>>
>>> The new Xen PVH entry point requires page tables to be setup by the
>>> kernel since it is entered with paging disabled.
>>>
>>> Pull the common code out of head_32.S and into pgtable_32.S so that
>>> setup_pgtable_32 can be invoked from both the new Xen entry point
>and
>>> the existing startup_32 code.
>>>
>> And why does it need a separate entry point as opposed to the plain
>one?
>
>One reason is that we need to prepare boot_params before jumping to
>startup_{32|64}.
>
>When the guest is loaded (always in 32-bit mode) the only thing we have
>is a pointer to Xen-specific datastructure. The early PVH code will
>prepare zeropage based on that structure and then jump to regular
>startup_*() code.
>
>-boris

And why not just resume execution at start_32 then?
-- 
Sent from my Android device with K-9 Mail. Please excuse my brevity.

[toc] | [prev] | [next] | [standalone]

#1501163 — Re: [PATCH 2/8] x86/head: Refactor 32-bit pgtable setup

On 10/14/2016 03:04 PM, hpa@zytor.com wrote:
> On October 14, 2016 11:44:18 AM PDT, Boris Ostrovsky <boris.ostrovsky@oracle.com> wrote:
>> On 10/14/2016 02:31 PM, hpa@zytor.com wrote:
>>> On October 14, 2016 11:05:12 AM PDT, Boris Ostrovsky
>> <boris.ostrovsky@oracle.com> wrote:
>>>> From: Matt Fleming <matt@codeblueprint.co.uk>
>>>>
>>>> The new Xen PVH entry point requires page tables to be setup by the
>>>> kernel since it is entered with paging disabled.
>>>>
>>>> Pull the common code out of head_32.S and into pgtable_32.S so that
>>>> setup_pgtable_32 can be invoked from both the new Xen entry point
>> and
>>>> the existing startup_32 code.
>>>>
>>> And why does it need a separate entry point as opposed to the plain
>> one?
>>
>> One reason is that we need to prepare boot_params before jumping to
>> startup_{32|64}.
>>
>> When the guest is loaded (always in 32-bit mode) the only thing we have
>> is a pointer to Xen-specific datastructure. The early PVH code will
>> prepare zeropage based on that structure and then jump to regular
>> startup_*() code.
>>
>> -boris
> And why not just resume execution at start_32 then?

I am not sure what start_32 is.

If you meant startup_32 then that's exactly what we do (for 32-bit
guests) once zeropage is set up.

-boris


-boris

[toc] | [prev] | [next] | [standalone]

#1501122 — [PATCH 3/8] xen/pvh: Import PVH-related Xen public interfaces

Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
---
 include/xen/interface/elfnote.h        |  12 ++-
 include/xen/interface/hvm/hvm_vcpu.h   | 143 +++++++++++++++++++++++++++++++++
 include/xen/interface/hvm/start_info.h |  98 ++++++++++++++++++++++
 3 files changed, 252 insertions(+), 1 deletion(-)
 create mode 100644 include/xen/interface/hvm/hvm_vcpu.h
 create mode 100644 include/xen/interface/hvm/start_info.h

diff --git a/include/xen/interface/elfnote.h b/include/xen/interface/elfnote.h
index f90b034..9e9f9bf 100644
--- a/include/xen/interface/elfnote.h
+++ b/include/xen/interface/elfnote.h
@@ -193,9 +193,19 @@
 #define XEN_ELFNOTE_SUPPORTED_FEATURES 17
 
 /*
+ * Physical entry point into the kernel.
+ *
+ * 32bit entry point into the kernel. When requested to launch the
+ * guest kernel in a HVM container, Xen will use this entry point to
+ * launch the guest in 32bit protected mode with paging disabled.
+ * Ignored otherwise.
+ */
+#define XEN_ELFNOTE_PHYS32_ENTRY 18
+
+/*
  * The number of the highest elfnote defined.
  */
-#define XEN_ELFNOTE_MAX XEN_ELFNOTE_SUPPORTED_FEATURES
+#define XEN_ELFNOTE_MAX XEN_ELFNOTE_PHYS32_ENTRY
 
 #endif /* __XEN_PUBLIC_ELFNOTE_H__ */
 
diff --git a/include/xen/interface/hvm/hvm_vcpu.h b/include/xen/interface/hvm/hvm_vcpu.h
new file mode 100644
index 0000000..32ca83e
--- /dev/null
+++ b/include/xen/interface/hvm/hvm_vcpu.h
@@ -0,0 +1,143 @@
+/*
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Copyright (c) 2015, Roger Pau Monne <roger.pau@citrix.com>
+ */
+
+#ifndef __XEN_PUBLIC_HVM_HVM_VCPU_H__
+#define __XEN_PUBLIC_HVM_HVM_VCPU_H__
+
+#include "../xen.h"
+
+struct vcpu_hvm_x86_32 {
+    uint32_t eax;
+    uint32_t ecx;
+    uint32_t edx;
+    uint32_t ebx;
+    uint32_t esp;
+    uint32_t ebp;
+    uint32_t esi;
+    uint32_t edi;
+    uint32_t eip;
+    uint32_t eflags;
+
+    uint32_t cr0;
+    uint32_t cr3;
+    uint32_t cr4;
+
+    uint32_t pad1;
+
+    /*
+     * EFER should only be used to set the NXE bit (if required)
+     * when starting a vCPU in 32bit mode with paging enabled or
+     * to set the LME/LMA bits in order to start the vCPU in
+     * compatibility mode.
+     */
+    uint64_t efer;
+
+    uint32_t cs_base;
+    uint32_t ds_base;
+    uint32_t ss_base;
+    uint32_t es_base;
+    uint32_t tr_base;
+    uint32_t cs_limit;
+    uint32_t ds_limit;
+    uint32_t ss_limit;
+    uint32_t es_limit;
+    uint32_t tr_limit;
+    uint16_t cs_ar;
+    uint16_t ds_ar;
+    uint16_t ss_ar;
+    uint16_t es_ar;
+    uint16_t tr_ar;
+
+    uint16_t pad2[3];
+};
+
+/*
+ * The layout of the _ar fields of the segment registers is the
+ * following:
+ *
+ * Bits   [0,3]: type (bits 40-43).
+ * Bit        4: s    (descriptor type, bit 44).
+ * Bit    [5,6]: dpl  (descriptor privilege level, bits 45-46).
+ * Bit        7: p    (segment-present, bit 47).
+ * Bit        8: avl  (available for system software, bit 52).
+ * Bit        9: l    (64-bit code segment, bit 53).
+ * Bit       10: db   (meaning depends on the segment, bit 54).
+ * Bit       11: g    (granularity, bit 55)
+ * Bits [12,15]: unused, must be blank.
+ *
+ * A more complete description of the meaning of this fields can be
+ * obtained from the Intel SDM, Volume 3, section 3.4.5.
+ */
+
+struct vcpu_hvm_x86_64 {
+    uint64_t rax;
+    uint64_t rcx;
+    uint64_t rdx;
+    uint64_t rbx;
+    uint64_t rsp;
+    uint64_t rbp;
+    uint64_t rsi;
+    uint64_t rdi;
+    uint64_t rip;
+    uint64_t rflags;
+
+    uint64_t cr0;
+    uint64_t cr3;
+    uint64_t cr4;
+    uint64_t efer;
+
+    /*
+     * Using VCPU_HVM_MODE_64B implies that the vCPU is launched
+     * directly in long mode, so the cached parts of the segment
+     * registers get set to match that environment.
+     *
+     * If the user wants to launch the vCPU in compatibility mode
+     * the 32-bit structure should be used instead.
+     */
+};
+
+struct vcpu_hvm_context {
+#define VCPU_HVM_MODE_32B 0  /* 32bit fields of the structure will be used. */
+#define VCPU_HVM_MODE_64B 1  /* 64bit fields of the structure will be used. */
+    uint32_t mode;
+
+    uint32_t pad;
+
+    /* CPU registers. */
+    union {
+        struct vcpu_hvm_x86_32 x86_32;
+        struct vcpu_hvm_x86_64 x86_64;
+    } cpu_regs;
+};
+typedef struct vcpu_hvm_context vcpu_hvm_context_t;
+
+#endif /* __XEN_PUBLIC_HVM_HVM_VCPU_H__ */
+
+/*
+ * Local variables:
+ * mode: C
+ * c-file-style: "BSD"
+ * c-basic-offset: 4
+ * tab-width: 4
+ * indent-tabs-mode: nil
+ * End:
+ */
diff --git a/include/xen/interface/hvm/start_info.h b/include/xen/interface/hvm/start_info.h
new file mode 100644
index 0000000..6484159
--- /dev/null
+++ b/include/xen/interface/hvm/start_info.h
@@ -0,0 +1,98 @@
+/*
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to
+ * deal in the Software without restriction, including without limitation the
+ * rights to use, copy, modify, merge, publish, distribute, sublicense, and/or
+ * sell copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
+ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER
+ * DEALINGS IN THE SOFTWARE.
+ *
+ * Copyright (c) 2016, Citrix Systems, Inc.
+ */
+
+#ifndef __XEN_PUBLIC_ARCH_X86_HVM_START_INFO_H__
+#define __XEN_PUBLIC_ARCH_X86_HVM_START_INFO_H__
+
+/*
+ * Start of day structure passed to PVH guests and to HVM guests in %ebx.
+ *
+ * NOTE: nothing will be loaded at physical address 0, so a 0 value in any
+ * of the address fields should be treated as not present.
+ *
+ *  0 +----------------+
+ *    | magic          | Contains the magic value XEN_HVM_START_MAGIC_VALUE
+ *    |                | ("xEn3" with the 0x80 bit of the "E" set).
+ *  4 +----------------+
+ *    | version        | Version of this structure. Current version is 0. New
+ *    |                | versions are guaranteed to be backwards-compatible.
+ *  8 +----------------+
+ *    | flags          | SIF_xxx flags.
+ * 12 +----------------+
+ *    | nr_modules     | Number of modules passed to the kernel.
+ * 16 +----------------+
+ *    | modlist_paddr  | Physical address of an array of modules
+ *    |                | (layout of the structure below).
+ * 24 +----------------+
+ *    | cmdline_paddr  | Physical address of the command line,
+ *    |                | a zero-terminated ASCII string.
+ * 32 +----------------+
+ *    | rsdp_paddr     | Physical address of the RSDP ACPI data structure.
+ * 40 +----------------+
+ *
+ * The layout of each entry in the module structure is the following:
+ *
+ *  0 +----------------+
+ *    | paddr          | Physical address of the module.
+ *  8 +----------------+
+ *    | size           | Size of the module in bytes.
+ * 16 +----------------+
+ *    | cmdline_paddr  | Physical address of the command line,
+ *    |                | a zero-terminated ASCII string.
+ * 24 +----------------+
+ *    | reserved       |
+ * 32 +----------------+
+ *
+ * The address and sizes are always a 64bit little endian unsigned integer.
+ *
+ * NB: Xen on x86 will always try to place all the data below the 4GiB
+ * boundary.
+ */
+#define XEN_HVM_START_MAGIC_VALUE 0x336ec578
+
+/*
+ * C representation of the x86/HVM start info layout.
+ *
+ * The canonical definition of this layout is above, this is just a way to
+ * represent the layout described there using C types.
+ */
+struct hvm_start_info {
+    uint32_t magic;             /* Contains the magic value 0x336ec578       */
+                                /* ("xEn3" with the 0x80 bit of the "E" set).*/
+    uint32_t version;           /* Version of this structure.                */
+    uint32_t flags;             /* SIF_xxx flags.                            */
+    uint32_t nr_modules;        /* Number of modules passed to the kernel.   */
+    uint64_t modlist_paddr;     /* Physical address of an array of           */
+                                /* hvm_modlist_entry.                        */
+    uint64_t cmdline_paddr;     /* Physical address of the command line.     */
+    uint64_t rsdp_paddr;        /* Physical address of the RSDP ACPI data    */
+                                /* structure.                                */
+};
+
+struct hvm_modlist_entry {
+    uint64_t paddr;             /* Physical address of the module.           */
+    uint64_t size;              /* Size of the module in bytes.              */
+    uint64_t cmdline_paddr;     /* Physical address of the command line.     */
+    uint64_t reserved;
+};
+
+#endif /* __XEN_PUBLIC_ARCH_X86_HVM_START_INFO_H__ */
-- 
1.8.3.1

[toc] | [prev] | [next] | [standalone]

#1501137 — Re: [Xen-devel] [PATCH 3/8] xen/pvh: Import PVH-related Xen public interfaces

On Fri, Oct 14, 2016 at 02:05:13PM -0400, Boris Ostrovsky wrote:
> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>

Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>

[toc] | [prev] | [next] | [standalone]

#1505774 — Re: [PATCH 3/8] xen/pvh: Import PVH-related Xen public interfaces

On 14/10/16 20:05, Boris Ostrovsky wrote:
> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>

Reviewed-by: Juergen Gross <jgross@suse.com>


Juergen

[toc] | [prev] | [next] | [standalone]

#1501123 — [PATCH 1/8] xen/x86: Remove PVH support

We are replacing existing PVH guests with new implementation.

Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
---
 arch/x86/xen/enlighten.c         | 140 ++++++---------------------------------
 arch/x86/xen/mmu.c               |  21 +-----
 arch/x86/xen/setup.c             |  37 +----------
 arch/x86/xen/smp.c               |  78 ++++++++--------------
 arch/x86/xen/smp.h               |   8 ---
 arch/x86/xen/xen-head.S          |  62 ++---------------
 arch/x86/xen/xen-ops.h           |   1 -
 drivers/xen/events/events_base.c |   1 -
 include/xen/xen.h                |  13 +---
 9 files changed, 54 insertions(+), 307 deletions(-)

diff --git a/arch/x86/xen/enlighten.c b/arch/x86/xen/enlighten.c
index c0fdd57..dc4ed0c 100644
--- a/arch/x86/xen/enlighten.c
+++ b/arch/x86/xen/enlighten.c
@@ -1149,10 +1149,11 @@ void xen_setup_vcpu_info_placement(void)
 		xen_vcpu_setup(cpu);
 	}
 
-	/* xen_vcpu_setup managed to place the vcpu_info within the
-	 * percpu area for all cpus, so make use of it. Note that for
-	 * PVH we want to use native IRQ mechanism. */
-	if (have_vcpu_info_placement && !xen_pvh_domain()) {
+	/*
+	 * xen_vcpu_setup managed to place the vcpu_info within the
+	 * percpu area for all cpus, so make use of it.
+	 */
+	if (have_vcpu_info_placement) {
 		pv_irq_ops.save_fl = __PV_IS_CALLEE_SAVE(xen_save_fl_direct);
 		pv_irq_ops.restore_fl = __PV_IS_CALLEE_SAVE(xen_restore_fl_direct);
 		pv_irq_ops.irq_disable = __PV_IS_CALLEE_SAVE(xen_irq_disable_direct);
@@ -1426,49 +1427,9 @@ static void __init xen_boot_params_init_edd(void)
  * Set up the GDT and segment registers for -fstack-protector.  Until
  * we do this, we have to be careful not to call any stack-protected
  * function, which is most of the kernel.
- *
- * Note, that it is __ref because the only caller of this after init
- * is PVH which is not going to use xen_load_gdt_boot or other
- * __init functions.
  */
-static void __ref xen_setup_gdt(int cpu)
+static void xen_setup_gdt(int cpu)
 {
-	if (xen_feature(XENFEAT_auto_translated_physmap)) {
-#ifdef CONFIG_X86_64
-		unsigned long dummy;
-
-		load_percpu_segment(cpu); /* We need to access per-cpu area */
-		switch_to_new_gdt(cpu); /* GDT and GS set */
-
-		/* We are switching of the Xen provided GDT to our HVM mode
-		 * GDT. The new GDT has  __KERNEL_CS with CS.L = 1
-		 * and we are jumping to reload it.
-		 */
-		asm volatile ("pushq %0\n"
-			      "leaq 1f(%%rip),%0\n"
-			      "pushq %0\n"
-			      "lretq\n"
-			      "1:\n"
-			      : "=&r" (dummy) : "0" (__KERNEL_CS));
-
-		/*
-		 * While not needed, we also set the %es, %ds, and %fs
-		 * to zero. We don't care about %ss as it is NULL.
-		 * Strictly speaking this is not needed as Xen zeros those
-		 * out (and also MSR_FS_BASE, MSR_GS_BASE, MSR_KERNEL_GS_BASE)
-		 *
-		 * Linux zeros them in cpu_init() and in secondary_startup_64
-		 * (for BSP).
-		 */
-		loadsegment(es, 0);
-		loadsegment(ds, 0);
-		loadsegment(fs, 0);
-#else
-		/* PVH: TODO Implement. */
-		BUG();
-#endif
-		return; /* PVH does not need any PV GDT ops. */
-	}
 	pv_cpu_ops.write_gdt_entry = xen_write_gdt_entry_boot;
 	pv_cpu_ops.load_gdt = xen_load_gdt_boot;
 
@@ -1479,59 +1440,6 @@ static void __ref xen_setup_gdt(int cpu)
 	pv_cpu_ops.load_gdt = xen_load_gdt;
 }
 
-#ifdef CONFIG_XEN_PVH
-/*
- * A PV guest starts with default flags that are not set for PVH, set them
- * here asap.
- */
-static void xen_pvh_set_cr_flags(int cpu)
-{
-
-	/* Some of these are setup in 'secondary_startup_64'. The others:
-	 * X86_CR0_TS, X86_CR0_PE, X86_CR0_ET are set by Xen for HVM guests
-	 * (which PVH shared codepaths), while X86_CR0_PG is for PVH. */
-	write_cr0(read_cr0() | X86_CR0_MP | X86_CR0_NE | X86_CR0_WP | X86_CR0_AM);
-
-	if (!cpu)
-		return;
-	/*
-	 * For BSP, PSE PGE are set in probe_page_size_mask(), for APs
-	 * set them here. For all, OSFXSR OSXMMEXCPT are set in fpu__init_cpu().
-	*/
-	if (boot_cpu_has(X86_FEATURE_PSE))
-		cr4_set_bits_and_update_boot(X86_CR4_PSE);
-
-	if (boot_cpu_has(X86_FEATURE_PGE))
-		cr4_set_bits_and_update_boot(X86_CR4_PGE);
-}
-
-/*
- * Note, that it is ref - because the only caller of this after init
- * is PVH which is not going to use xen_load_gdt_boot or other
- * __init functions.
- */
-void __ref xen_pvh_secondary_vcpu_init(int cpu)
-{
-	xen_setup_gdt(cpu);
-	xen_pvh_set_cr_flags(cpu);
-}
-
-static void __init xen_pvh_early_guest_init(void)
-{
-	if (!xen_feature(XENFEAT_auto_translated_physmap))
-		return;
-
-	BUG_ON(!xen_feature(XENFEAT_hvm_callback_vector));
-
-	xen_pvh_early_cpu_init(0, false);
-	xen_pvh_set_cr_flags(0);
-
-#ifdef CONFIG_X86_32
-	BUG(); /* PVH: Implement proper support. */
-#endif
-}
-#endif    /* CONFIG_XEN_PVH */
-
 static void __init xen_dom0_set_legacy_features(void)
 {
 	x86_platform.legacy.rtc = 1;
@@ -1568,24 +1476,17 @@ asmlinkage __visible void __init xen_start_kernel(void)
 	xen_domain_type = XEN_PV_DOMAIN;
 
 	xen_setup_features();
-#ifdef CONFIG_XEN_PVH
-	xen_pvh_early_guest_init();
-#endif
+
 	xen_setup_machphys_mapping();
 
 	/* Install Xen paravirt ops */
 	pv_info = xen_info;
 	pv_init_ops = xen_init_ops;
-	if (!xen_pvh_domain()) {
-		pv_cpu_ops = xen_cpu_ops;
+	pv_cpu_ops = xen_cpu_ops;
 
-		x86_platform.get_nmi_reason = xen_get_nmi_reason;
-	}
+	x86_platform.get_nmi_reason = xen_get_nmi_reason;
 
-	if (xen_feature(XENFEAT_auto_translated_physmap))
-		x86_init.resources.memory_setup = xen_auto_xlated_memory_setup;
-	else
-		x86_init.resources.memory_setup = xen_memory_setup;
+	x86_init.resources.memory_setup = xen_memory_setup;
 	x86_init.oem.arch_setup = xen_arch_setup;
 	x86_init.oem.banner = xen_banner;
 
@@ -1678,18 +1579,15 @@ asmlinkage __visible void __init xen_start_kernel(void)
 	/* set the limit of our address space */
 	xen_reserve_top();
 
-	/* PVH: runs at default kernel iopl of 0 */
-	if (!xen_pvh_domain()) {
-		/*
-		 * We used to do this in xen_arch_setup, but that is too late
-		 * on AMD were early_cpu_init (run before ->arch_setup()) calls
-		 * early_amd_init which pokes 0xcf8 port.
-		 */
-		set_iopl.iopl = 1;
-		rc = HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
-		if (rc != 0)
-			xen_raw_printk("physdev_op failed %d\n", rc);
-	}
+	/*
+	 * We used to do this in xen_arch_setup, but that is too late
+	 * on AMD were early_cpu_init (run before ->arch_setup()) calls
+	 * early_amd_init which pokes 0xcf8 port.
+	 */
+	set_iopl.iopl = 1;
+	rc = HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
+	if (rc != 0)
+		xen_raw_printk("physdev_op failed %d\n", rc);
 
 #ifdef CONFIG_X86_32
 	/* set up basic CPUID stuff */
diff --git a/arch/x86/xen/mmu.c b/arch/x86/xen/mmu.c
index 7d5afdb..f6740b5 100644
--- a/arch/x86/xen/mmu.c
+++ b/arch/x86/xen/mmu.c
@@ -1792,10 +1792,6 @@ static void __init set_page_prot_flags(void *addr, pgprot_t prot,
 	unsigned long pfn = __pa(addr) >> PAGE_SHIFT;
 	pte_t pte = pfn_pte(pfn, prot);
 
-	/* For PVH no need to set R/O or R/W to pin them or unpin them. */
-	if (xen_feature(XENFEAT_auto_translated_physmap))
-		return;
-
 	if (HYPERVISOR_update_va_mapping((unsigned long)addr, pte, flags))
 		BUG();
 }
@@ -1902,8 +1898,7 @@ static void __init check_pt_base(unsigned long *pt_base, unsigned long *pt_end,
  * level2_ident_pgt, and level2_kernel_pgt.  This means that only the
  * kernel has a physical mapping to start with - but that's enough to
  * get __va working.  We need to fill in the rest of the physical
- * mapping once some sort of allocator has been set up.  NOTE: for
- * PVH, the page tables are native.
+ * mapping once some sort of allocator has been set up.
  */
 void __init xen_setup_kernel_pagetable(pgd_t *pgd, unsigned long max_pfn)
 {
@@ -2812,16 +2807,6 @@ static int do_remap_gfn(struct vm_area_struct *vma,
 
 	BUG_ON(!((vma->vm_flags & (VM_PFNMAP | VM_IO)) == (VM_PFNMAP | VM_IO)));
 
-	if (xen_feature(XENFEAT_auto_translated_physmap)) {
-#ifdef CONFIG_XEN_PVH
-		/* We need to update the local page tables and the xen HAP */
-		return xen_xlate_remap_gfn_array(vma, addr, gfn, nr, err_ptr,
-						 prot, domid, pages);
-#else
-		return -EINVAL;
-#endif
-        }
-
 	rmd.mfn = gfn;
 	rmd.prot = prot;
 	/* We use the err_ptr to indicate if there we are doing a contiguous
@@ -2915,10 +2900,6 @@ int xen_unmap_domain_gfn_range(struct vm_area_struct *vma,
 	if (!pages || !xen_feature(XENFEAT_auto_translated_physmap))
 		return 0;
 
-#ifdef CONFIG_XEN_PVH
-	return xen_xlate_unmap_gfn_range(vma, numpgs, pages);
-#else
 	return -EINVAL;
-#endif
 }
 EXPORT_SYMBOL_GPL(xen_unmap_domain_gfn_range);
diff --git a/arch/x86/xen/setup.c b/arch/x86/xen/setup.c
index f8960fc..6999016 100644
--- a/arch/x86/xen/setup.c
+++ b/arch/x86/xen/setup.c
@@ -915,39 +915,6 @@ char * __init xen_memory_setup(void)
 }
 
 /*
- * Machine specific memory setup for auto-translated guests.
- */
-char * __init xen_auto_xlated_memory_setup(void)
-{
-	struct xen_memory_map memmap;
-	int i;
-	int rc;
-
-	memmap.nr_entries = E820MAX;
-	set_xen_guest_handle(memmap.buffer, xen_e820_map);
-
-	rc = HYPERVISOR_memory_op(XENMEM_memory_map, &memmap);
-	if (rc < 0)
-		panic("No memory map (%d)\n", rc);
-
-	xen_e820_map_entries = memmap.nr_entries;
-
-	sanitize_e820_map(xen_e820_map, ARRAY_SIZE(xen_e820_map),
-			  &xen_e820_map_entries);
-
-	for (i = 0; i < xen_e820_map_entries; i++)
-		e820_add_region(xen_e820_map[i].addr, xen_e820_map[i].size,
-				xen_e820_map[i].type);
-
-	/* Remove p2m info, it is not needed. */
-	xen_start_info->mfn_list = 0;
-	xen_start_info->first_p2m_pfn = 0;
-	xen_start_info->nr_p2m_frames = 0;
-
-	return "Xen";
-}
-
-/*
  * Set the bit indicating "nosegneg" library variants should be used.
  * We only need to bother in pure 32-bit mode; compat 32-bit processes
  * can have un-truncated segments, so wrapping around is allowed.
@@ -1032,8 +999,8 @@ void __init xen_pvmmu_arch_setup(void)
 void __init xen_arch_setup(void)
 {
 	xen_panic_handler_init();
-	if (!xen_feature(XENFEAT_auto_translated_physmap))
-		xen_pvmmu_arch_setup();
+
+	xen_pvmmu_arch_setup();
 
 #ifdef CONFIG_ACPI
 	if (!(xen_start_info->flags & SIF_INITDOMAIN)) {
diff --git a/arch/x86/xen/smp.c b/arch/x86/xen/smp.c
index 9fa27ce..914e320 100644
--- a/arch/x86/xen/smp.c
+++ b/arch/x86/xen/smp.c
@@ -105,18 +105,8 @@ static void cpu_bringup(void)
 	local_irq_enable();
 }
 
-/*
- * Note: cpu parameter is only relevant for PVH. The reason for passing it
- * is we can't do smp_processor_id until the percpu segments are loaded, for
- * which we need the cpu number! So we pass it in rdi as first parameter.
- */
-asmlinkage __visible void cpu_bringup_and_idle(int cpu)
+asmlinkage __visible void cpu_bringup_and_idle(void)
 {
-#ifdef CONFIG_XEN_PVH
-	if (xen_feature(XENFEAT_auto_translated_physmap) &&
-	    xen_feature(XENFEAT_supervisor_mode_kernel))
-		xen_pvh_secondary_vcpu_init(cpu);
-#endif
 	cpu_bringup();
 	cpu_startup_entry(CPUHP_AP_ONLINE_IDLE);
 }
@@ -410,61 +400,47 @@ static void __init xen_smp_prepare_cpus(unsigned int max_cpus)
 	gdt = get_cpu_gdt_table(cpu);
 
 #ifdef CONFIG_X86_32
-	/* Note: PVH is not yet supported on x86_32. */
 	ctxt->user_regs.fs = __KERNEL_PERCPU;
 	ctxt->user_regs.gs = __KERNEL_STACK_CANARY;
 #endif
 	memset(&ctxt->fpu_ctxt, 0, sizeof(ctxt->fpu_ctxt));
 
-	if (!xen_feature(XENFEAT_auto_translated_physmap)) {
-		ctxt->user_regs.eip = (unsigned long)cpu_bringup_and_idle;
-		ctxt->flags = VGCF_IN_KERNEL;
-		ctxt->user_regs.eflags = 0x1000; /* IOPL_RING1 */
-		ctxt->user_regs.ds = __USER_DS;
-		ctxt->user_regs.es = __USER_DS;
-		ctxt->user_regs.ss = __KERNEL_DS;
+	ctxt->user_regs.eip = (unsigned long)cpu_bringup_and_idle;
+	ctxt->flags = VGCF_IN_KERNEL;
+	ctxt->user_regs.eflags = 0x1000; /* IOPL_RING1 */
+	ctxt->user_regs.ds = __USER_DS;
+	ctxt->user_regs.es = __USER_DS;
+	ctxt->user_regs.ss = __KERNEL_DS;
 
-		xen_copy_trap_info(ctxt->trap_ctxt);
+	xen_copy_trap_info(ctxt->trap_ctxt);
 
-		ctxt->ldt_ents = 0;
+	ctxt->ldt_ents = 0;
 
-		BUG_ON((unsigned long)gdt & ~PAGE_MASK);
+	BUG_ON((unsigned long)gdt & ~PAGE_MASK);
 
-		gdt_mfn = arbitrary_virt_to_mfn(gdt);
-		make_lowmem_page_readonly(gdt);
-		make_lowmem_page_readonly(mfn_to_virt(gdt_mfn));
+	gdt_mfn = arbitrary_virt_to_mfn(gdt);
+	make_lowmem_page_readonly(gdt);
+	make_lowmem_page_readonly(mfn_to_virt(gdt_mfn));
 
-		ctxt->gdt_frames[0] = gdt_mfn;
-		ctxt->gdt_ents      = GDT_ENTRIES;
+	ctxt->gdt_frames[0] = gdt_mfn;
+	ctxt->gdt_ents      = GDT_ENTRIES;
 
-		ctxt->kernel_ss = __KERNEL_DS;
-		ctxt->kernel_sp = idle->thread.sp0;
+	ctxt->kernel_ss = __KERNEL_DS;
+	ctxt->kernel_sp = idle->thread.sp0;
 
 #ifdef CONFIG_X86_32
-		ctxt->event_callback_cs     = __KERNEL_CS;
-		ctxt->failsafe_callback_cs  = __KERNEL_CS;
+	ctxt->event_callback_cs     = __KERNEL_CS;
+	ctxt->failsafe_callback_cs  = __KERNEL_CS;
 #else
-		ctxt->gs_base_kernel = per_cpu_offset(cpu);
-#endif
-		ctxt->event_callback_eip    =
-					(unsigned long)xen_hypervisor_callback;
-		ctxt->failsafe_callback_eip =
-					(unsigned long)xen_failsafe_callback;
-		ctxt->user_regs.cs = __KERNEL_CS;
-		per_cpu(xen_cr3, cpu) = __pa(swapper_pg_dir);
-	}
-#ifdef CONFIG_XEN_PVH
-	else {
-		/*
-		 * The vcpu comes on kernel page tables which have the NX pte
-		 * bit set. This means before DS/SS is touched, NX in
-		 * EFER must be set. Hence the following assembly glue code.
-		 */
-		ctxt->user_regs.eip = (unsigned long)xen_pvh_early_cpu_init;
-		ctxt->user_regs.rdi = cpu;
-		ctxt->user_regs.rsi = true;  /* entry == true */
-	}
+	ctxt->gs_base_kernel = per_cpu_offset(cpu);
 #endif
+	ctxt->event_callback_eip    =
+		(unsigned long)xen_hypervisor_callback;
+	ctxt->failsafe_callback_eip =
+		(unsigned long)xen_failsafe_callback;
+	ctxt->user_regs.cs = __KERNEL_CS;
+	per_cpu(xen_cr3, cpu) = __pa(swapper_pg_dir);
+
 	ctxt->user_regs.esp = idle->thread.sp0 - sizeof(struct pt_regs);
 	ctxt->ctrlreg[3] = xen_pfn_to_cr3(virt_to_gfn(swapper_pg_dir));
 	if (HYPERVISOR_vcpu_op(VCPUOP_initialise, xen_vcpu_nr(cpu), ctxt))
diff --git a/arch/x86/xen/smp.h b/arch/x86/xen/smp.h
index c5c16dc..9beef33 100644
--- a/arch/x86/xen/smp.h
+++ b/arch/x86/xen/smp.h
@@ -21,12 +21,4 @@ static inline int xen_smp_intr_init(unsigned int cpu)
 static inline void xen_smp_intr_free(unsigned int cpu) {}
 #endif /* CONFIG_SMP */
 
-#ifdef CONFIG_XEN_PVH
-extern void xen_pvh_early_cpu_init(int cpu, bool entry);
-#else
-static inline void xen_pvh_early_cpu_init(int cpu, bool entry)
-{
-}
-#endif
-
 #endif
diff --git a/arch/x86/xen/xen-head.S b/arch/x86/xen/xen-head.S
index 7f8d8ab..37794e4 100644
--- a/arch/x86/xen/xen-head.S
+++ b/arch/x86/xen/xen-head.S
@@ -16,25 +16,6 @@
 #include <xen/interface/xen-mca.h>
 #include <asm/xen/interface.h>
 
-#ifdef CONFIG_XEN_PVH
-#define PVH_FEATURES_STR  "|writable_descriptor_tables|auto_translated_physmap|supervisor_mode_kernel"
-/* Note the lack of 'hvm_callback_vector'. Older hypervisor will
- * balk at this being part of XEN_ELFNOTE_FEATURES, so we put it in
- * XEN_ELFNOTE_SUPPORTED_FEATURES which older hypervisors will ignore.
- */
-#define PVH_FEATURES ((1 << XENFEAT_writable_page_tables) | \
-		      (1 << XENFEAT_auto_translated_physmap) | \
-		      (1 << XENFEAT_supervisor_mode_kernel) | \
-		      (1 << XENFEAT_hvm_callback_vector))
-/* The XENFEAT_writable_page_tables is not stricly necessary as we set that
- * up regardless whether this CONFIG option is enabled or not, but it
- * clarifies what the right flags need to be.
- */
-#else
-#define PVH_FEATURES_STR  ""
-#define PVH_FEATURES (0)
-#endif
-
 	__INIT
 ENTRY(startup_xen)
 	cld
@@ -54,41 +35,6 @@ ENTRY(startup_xen)
 
 	__FINIT
 
-#ifdef CONFIG_XEN_PVH
-/*
- * xen_pvh_early_cpu_init() - early PVH VCPU initialization
- * @cpu:   this cpu number (%rdi)
- * @entry: true if this is a secondary vcpu coming up on this entry
- *         point, false if this is the boot CPU being initialized for
- *         the first time (%rsi)
- *
- * Note: This is called as a function on the boot CPU, and is the entry point
- *       on the secondary CPU.
- */
-ENTRY(xen_pvh_early_cpu_init)
-	mov     %rsi, %r11
-
-	/* Gather features to see if NX implemented. */
-	mov     $0x80000001, %eax
-	cpuid
-	mov     %edx, %esi
-
-	mov     $MSR_EFER, %ecx
-	rdmsr
-	bts     $_EFER_SCE, %eax
-
-	bt      $20, %esi
-	jnc     1f      	/* No NX, skip setting it */
-	bts     $_EFER_NX, %eax
-1:	wrmsr
-#ifdef CONFIG_SMP
-	cmp     $0, %r11b
-	jne     cpu_bringup_and_idle
-#endif
-	ret
-
-#endif /* CONFIG_XEN_PVH */
-
 .pushsection .text
 	.balign PAGE_SIZE
 ENTRY(hypercall_page)
@@ -114,10 +60,10 @@ ENTRY(hypercall_page)
 #endif
 	ELFNOTE(Xen, XEN_ELFNOTE_ENTRY,          _ASM_PTR startup_xen)
 	ELFNOTE(Xen, XEN_ELFNOTE_HYPERCALL_PAGE, _ASM_PTR hypercall_page)
-	ELFNOTE(Xen, XEN_ELFNOTE_FEATURES,       .ascii "!writable_page_tables|pae_pgdir_above_4gb"; .asciz PVH_FEATURES_STR)
-	ELFNOTE(Xen, XEN_ELFNOTE_SUPPORTED_FEATURES, .long (PVH_FEATURES) |
-						(1 << XENFEAT_writable_page_tables) |
-						(1 << XENFEAT_dom0))
+	ELFNOTE(Xen, XEN_ELFNOTE_FEATURES,
+		.ascii "!writable_page_tables|pae_pgdir_above_4gb")
+	ELFNOTE(Xen, XEN_ELFNOTE_SUPPORTED_FEATURES,
+		.long (1 << XENFEAT_writable_page_tables) | (1 << XENFEAT_dom0))
 	ELFNOTE(Xen, XEN_ELFNOTE_PAE_MODE,       .asciz "yes")
 	ELFNOTE(Xen, XEN_ELFNOTE_LOADER,         .asciz "generic")
 	ELFNOTE(Xen, XEN_ELFNOTE_L1_MFN_VALID,
diff --git a/arch/x86/xen/xen-ops.h b/arch/x86/xen/xen-ops.h
index 3cbce3b..3688ecc 100644
--- a/arch/x86/xen/xen-ops.h
+++ b/arch/x86/xen/xen-ops.h
@@ -146,5 +146,4 @@ static inline void __init xen_efi_init(void)
 
 extern int xen_panic_handler_init(void);
 
-void xen_pvh_secondary_vcpu_init(int cpu);
 #endif /* XEN_OPS_H */
diff --git a/drivers/xen/events/events_base.c b/drivers/xen/events/events_base.c
index 9ecfcdc..6a0aa5c 100644
--- a/drivers/xen/events/events_base.c
+++ b/drivers/xen/events/events_base.c
@@ -1706,7 +1706,6 @@ void __init xen_init_IRQ(void)
 		pirq_eoi_map = (void *)__get_free_page(GFP_KERNEL|__GFP_ZERO);
 		eoi_gmfn.gmfn = virt_to_gfn(pirq_eoi_map);
 		rc = HYPERVISOR_physdev_op(PHYSDEVOP_pirq_eoi_gmfn_v2, &eoi_gmfn);
-		/* TODO: No PVH support for PIRQ EOI */
 		if (rc != 0) {
 			free_page((unsigned long) pirq_eoi_map);
 			pirq_eoi_map = NULL;
diff --git a/include/xen/xen.h b/include/xen/xen.h
index f0f0252..d0f9684 100644
--- a/include/xen/xen.h
+++ b/include/xen/xen.h
@@ -29,17 +29,6 @@ enum xen_domain_type {
 #define xen_initial_domain()	(0)
 #endif	/* CONFIG_XEN_DOM0 */
 
-#ifdef CONFIG_XEN_PVH
-/* This functionality exists only for x86. The XEN_PVHVM support exists
- * only in x86 world - hence on ARM it will be always disabled.
- * N.B. ARM guests are neither PV nor HVM nor PVHVM.
- * It's a bit like PVH but is different also (it's further towards the H
- * end of the spectrum than even PVH).
- */
-#include <xen/features.h>
-#define xen_pvh_domain() (xen_pv_domain() && \
-			  xen_feature(XENFEAT_auto_translated_physmap))
-#else
 #define xen_pvh_domain()	(0)
-#endif
+
 #endif	/* _XEN_XEN_H */
-- 
1.8.3.1

[toc] | [prev] | [next] | [standalone]

#1501136 — Re: [Xen-devel] [PATCH 1/8] xen/x86: Remove PVH support

On Fri, Oct 14, 2016 at 02:05:11PM -0400, Boris Ostrovsky wrote:
> We are replacing existing PVH guests with new implementation.
> 
> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>

Reviewed-by: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>

[toc] | [prev] | [next] | [standalone]

#1502998 — Re: [PATCH 1/8] xen/x86: Remove PVH support

On 14/10/16 20:05, Boris Ostrovsky wrote:
> We are replacing existing PVH guests with new implementation.
> 
> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>

Reviewed-by: Juergen Gross <jgross@suse.com>

with the following addressed:

> diff --git a/include/xen/xen.h b/include/xen/xen.h
> index f0f0252..d0f9684 100644
> --- a/include/xen/xen.h
> +++ b/include/xen/xen.h
> @@ -29,17 +29,6 @@ enum xen_domain_type {
>  #define xen_initial_domain()	(0)
>  #endif	/* CONFIG_XEN_DOM0 */
>  
> -#ifdef CONFIG_XEN_PVH
> -/* This functionality exists only for x86. The XEN_PVHVM support exists
> - * only in x86 world - hence on ARM it will be always disabled.
> - * N.B. ARM guests are neither PV nor HVM nor PVHVM.
> - * It's a bit like PVH but is different also (it's further towards the H
> - * end of the spectrum than even PVH).
> - */
> -#include <xen/features.h>
> -#define xen_pvh_domain() (xen_pv_domain() && \
> -			  xen_feature(XENFEAT_auto_translated_physmap))
> -#else
>  #define xen_pvh_domain()	(0)

Any reason you don't remove this, too (together with its last user in
arch/x86/xen/grant-table.c) ?


Juergen

[toc] | [prev] | [next] | [standalone]

#1503065 — Re: [PATCH 1/8] xen/x86: Remove PVH support

On 10/18/2016 09:46 AM, Juergen Gross wrote:
> On 14/10/16 20:05, Boris Ostrovsky wrote:
>> We are replacing existing PVH guests with new implementation.
>>
>> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
> Reviewed-by: Juergen Gross <jgross@suse.com>
>
> with the following addressed:
>
>> diff --git a/include/xen/xen.h b/include/xen/xen.h
>> index f0f0252..d0f9684 100644
>> --- a/include/xen/xen.h
>> +++ b/include/xen/xen.h
>> @@ -29,17 +29,6 @@ enum xen_domain_type {
>>  #define xen_initial_domain()	(0)
>>  #endif	/* CONFIG_XEN_DOM0 */
>>  
>> -#ifdef CONFIG_XEN_PVH
>> -/* This functionality exists only for x86. The XEN_PVHVM support exists
>> - * only in x86 world - hence on ARM it will be always disabled.
>> - * N.B. ARM guests are neither PV nor HVM nor PVHVM.
>> - * It's a bit like PVH but is different also (it's further towards the H
>> - * end of the spectrum than even PVH).
>> - */
>> -#include <xen/features.h>
>> -#define xen_pvh_domain() (xen_pv_domain() && \
>> -			  xen_feature(XENFEAT_auto_translated_physmap))
>> -#else
>>  #define xen_pvh_domain()	(0)
> Any reason you don't remove this, too (together with its last user in
> arch/x86/xen/grant-table.c) ?

grant-table.c is in fact one of the reasons: we will be using that code
for PVHv2 again so I kept it to avoid unnecessary code churn.

Also, we want to have a nop definition of xen_pvh_domain() for
!CONFIG_XEN_PVH.

-boris

[toc] | [prev] | [next] | [standalone]

#1503111 — Re: [PATCH 1/8] xen/x86: Remove PVH support

On 18/10/16 16:45, Boris Ostrovsky wrote:
> On 10/18/2016 09:46 AM, Juergen Gross wrote:
>> On 14/10/16 20:05, Boris Ostrovsky wrote:
>>> We are replacing existing PVH guests with new implementation.
>>>
>>> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
>> Reviewed-by: Juergen Gross <jgross@suse.com>
>>
>> with the following addressed:
>>
>>> diff --git a/include/xen/xen.h b/include/xen/xen.h
>>> index f0f0252..d0f9684 100644
>>> --- a/include/xen/xen.h
>>> +++ b/include/xen/xen.h
>>> @@ -29,17 +29,6 @@ enum xen_domain_type {
>>>  #define xen_initial_domain()	(0)
>>>  #endif	/* CONFIG_XEN_DOM0 */
>>>  
>>> -#ifdef CONFIG_XEN_PVH
>>> -/* This functionality exists only for x86. The XEN_PVHVM support exists
>>> - * only in x86 world - hence on ARM it will be always disabled.
>>> - * N.B. ARM guests are neither PV nor HVM nor PVHVM.
>>> - * It's a bit like PVH but is different also (it's further towards the H
>>> - * end of the spectrum than even PVH).
>>> - */
>>> -#include <xen/features.h>
>>> -#define xen_pvh_domain() (xen_pv_domain() && \
>>> -			  xen_feature(XENFEAT_auto_translated_physmap))
>>> -#else
>>>  #define xen_pvh_domain()	(0)
>> Any reason you don't remove this, too (together with its last user in
>> arch/x86/xen/grant-table.c) ?
> 
> grant-table.c is in fact one of the reasons: we will be using that code
> for PVHv2 again so I kept it to avoid unnecessary code churn.
> 
> Also, we want to have a nop definition of xen_pvh_domain() for
> !CONFIG_XEN_PVH.

Okay, could you mention this in the commit message, please?


Juergen

[toc] | [prev] | [next] | [standalone]

#1503135 — Re: [PATCH 1/8] xen/x86: Remove PVH support

On 10/18/2016 11:33 AM, Juergen Gross wrote:
> On 18/10/16 16:45, Boris Ostrovsky wrote:
>> On 10/18/2016 09:46 AM, Juergen Gross wrote:
>>> On 14/10/16 20:05, Boris Ostrovsky wrote:
>>>> We are replacing existing PVH guests with new implementation.
>>>>
>>>> Signed-off-by: Boris Ostrovsky <boris.ostrovsky@oracle.com>
>>> Reviewed-by: Juergen Gross <jgross@suse.com>
>>>
>>> with the following addressed:
>>>
>>>> diff --git a/include/xen/xen.h b/include/xen/xen.h
>>>> index f0f0252..d0f9684 100644
>>>> --- a/include/xen/xen.h
>>>> +++ b/include/xen/xen.h
>>>> @@ -29,17 +29,6 @@ enum xen_domain_type {
>>>>  #define xen_initial_domain()	(0)
>>>>  #endif	/* CONFIG_XEN_DOM0 */
>>>>  
>>>> -#ifdef CONFIG_XEN_PVH
>>>> -/* This functionality exists only for x86. The XEN_PVHVM support exists
>>>> - * only in x86 world - hence on ARM it will be always disabled.
>>>> - * N.B. ARM guests are neither PV nor HVM nor PVHVM.
>>>> - * It's a bit like PVH but is different also (it's further towards the H
>>>> - * end of the spectrum than even PVH).
>>>> - */
>>>> -#include <xen/features.h>
>>>> -#define xen_pvh_domain() (xen_pv_domain() && \
>>>> -			  xen_feature(XENFEAT_auto_translated_physmap))
>>>> -#else
>>>>  #define xen_pvh_domain()	(0)
>>> Any reason you don't remove this, too (together with its last user in
>>> arch/x86/xen/grant-table.c) ?
>> grant-table.c is in fact one of the reasons: we will be using that code
>> for PVHv2 again so I kept it to avoid unnecessary code churn.
>>
>> Also, we want to have a nop definition of xen_pvh_domain() for
>> !CONFIG_XEN_PVH.
> Okay, could you mention this in the commit message, please?


Will do.

-boris

[toc] | [prev] | [standalone]

Page 2 of 2 — ← Prev page 1 [2]

Back to top | Article view | linux.kernel

csiph-web