summaryrefslogtreecommitdiff
path: root/arch/x86
diff options
context:
space:
mode:
authorMatias Zabaljauregui <zabaljauregui@gmail.com>2009-06-12 22:27:07 -0600
committerRusty Russell <rusty@rustcorp.com.au>2009-06-12 22:27:08 +0930
commitacdd0b6292b282c4511897ac2691a47befbf1c6a (patch)
tree1bfcfc32b11d35e99fec5bbf52b19d6ee038f25e /arch/x86
parentcefcad1773197523e11e18b669f245e6a8d32058 (diff)
lguest: PAE support
This version requires that host and guest have the same PAE status. NX cap is not offered to the guest, yet. Signed-off-by: Matias Zabaljauregui <zabaljauregui@gmail.com> Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Diffstat (limited to 'arch/x86')
-rw-r--r--arch/x86/include/asm/lguest.h7
-rw-r--r--arch/x86/include/asm/lguest_hcall.h3
-rw-r--r--arch/x86/lguest/Kconfig1
-rw-r--r--arch/x86/lguest/boot.c71
4 files changed, 74 insertions, 8 deletions
diff --git a/arch/x86/include/asm/lguest.h b/arch/x86/include/asm/lguest.h
index 1caf57628b9c..313389cd50d2 100644
--- a/arch/x86/include/asm/lguest.h
+++ b/arch/x86/include/asm/lguest.h
@@ -17,8 +17,13 @@
/* Pages for switcher itself, then two pages per cpu */
#define TOTAL_SWITCHER_PAGES (SHARED_SWITCHER_PAGES + 2 * nr_cpu_ids)
-/* We map at -4M for ease of mapping into the guest (one PTE page). */
+/* We map at -4M (-2M when PAE is activated) for ease of mapping
+ * into the guest (one PTE page). */
+#ifdef CONFIG_X86_PAE
+#define SWITCHER_ADDR 0xFFE00000
+#else
#define SWITCHER_ADDR 0xFFC00000
+#endif
/* Found in switcher.S */
extern unsigned long default_idt_entries[];
diff --git a/arch/x86/include/asm/lguest_hcall.h b/arch/x86/include/asm/lguest_hcall.h
index b14b3552a4db..d31c4a684078 100644
--- a/arch/x86/include/asm/lguest_hcall.h
+++ b/arch/x86/include/asm/lguest_hcall.h
@@ -12,6 +12,7 @@
#define LHCALL_TS 8
#define LHCALL_SET_CLOCKEVENT 9
#define LHCALL_HALT 10
+#define LHCALL_SET_PMD 13
#define LHCALL_SET_PTE 14
#define LHCALL_SET_PGD 15
#define LHCALL_LOAD_TLS 16
@@ -33,7 +34,7 @@
* operations? There are two ways: the direct way is to make a "hypercall",
* to make requests of the Host Itself.
*
- * We use the KVM hypercall mechanism. Eighteen hypercalls are
+ * We use the KVM hypercall mechanism. Seventeen hypercalls are
* available: the hypercall number is put in the %eax register, and the
* arguments (when required) are placed in %ebx, %ecx, %edx and %esi.
* If a return value makes sense, it's returned in %eax.
diff --git a/arch/x86/lguest/Kconfig b/arch/x86/lguest/Kconfig
index 8dab8f7844d3..38718041efc3 100644
--- a/arch/x86/lguest/Kconfig
+++ b/arch/x86/lguest/Kconfig
@@ -2,7 +2,6 @@ config LGUEST_GUEST
bool "Lguest guest support"
select PARAVIRT
depends on X86_32
- depends on !X86_PAE
select VIRTIO
select VIRTIO_RING
select VIRTIO_CONSOLE
diff --git a/arch/x86/lguest/boot.c b/arch/x86/lguest/boot.c
index d12f554e5f6a..7bc65f0f62c4 100644
--- a/arch/x86/lguest/boot.c
+++ b/arch/x86/lguest/boot.c
@@ -167,6 +167,7 @@ static void lazy_hcall3(unsigned long call,
async_hcall(call, arg1, arg2, arg3, 0);
}
+#ifdef CONFIG_X86_PAE
static void lazy_hcall4(unsigned long call,
unsigned long arg1,
unsigned long arg2,
@@ -178,6 +179,7 @@ static void lazy_hcall4(unsigned long call,
else
async_hcall(call, arg1, arg2, arg3, arg4);
}
+#endif
/* When lazy mode is turned off reset the per-cpu lazy mode variable and then
* issue the do-nothing hypercall to flush any stored calls. */
@@ -380,8 +382,8 @@ static void lguest_cpuid(unsigned int *ax, unsigned int *bx,
case 1: /* Basic feature request. */
/* We only allow kernel to see SSE3, CMPXCHG16B and SSSE3 */
*cx &= 0x00002201;
- /* SSE, SSE2, FXSR, MMX, CMOV, CMPXCHG8B, TSC, FPU. */
- *dx &= 0x07808111;
+ /* SSE, SSE2, FXSR, MMX, CMOV, CMPXCHG8B, TSC, FPU, PAE. */
+ *dx &= 0x07808151;
/* The Host can do a nice optimization if it knows that the
* kernel mappings (addresses above 0xC0000000 or whatever
* PAGE_OFFSET is set to) haven't changed. But Linux calls
@@ -400,6 +402,11 @@ static void lguest_cpuid(unsigned int *ax, unsigned int *bx,
if (*ax > 0x80000008)
*ax = 0x80000008;
break;
+ case 0x80000001:
+ /* Here we should fix nx cap depending on host. */
+ /* For this version of PAE, we just clear NX bit. */
+ *dx &= ~(1 << 20);
+ break;
}
}
@@ -533,7 +540,12 @@ static void lguest_write_cr4(unsigned long val)
static void lguest_pte_update(struct mm_struct *mm, unsigned long addr,
pte_t *ptep)
{
+#ifdef CONFIG_X86_PAE
+ lazy_hcall4(LHCALL_SET_PTE, __pa(mm->pgd), addr,
+ ptep->pte_low, ptep->pte_high);
+#else
lazy_hcall3(LHCALL_SET_PTE, __pa(mm->pgd), addr, ptep->pte_low);
+#endif
}
static void lguest_set_pte_at(struct mm_struct *mm, unsigned long addr,
@@ -543,15 +555,37 @@ static void lguest_set_pte_at(struct mm_struct *mm, unsigned long addr,
lguest_pte_update(mm, addr, ptep);
}
-/* The Guest calls this to set a top-level entry. Again, we set the entry then
- * tell the Host which top-level page we changed, and the index of the entry we
- * changed. */
+/* The Guest calls lguest_set_pud to set a top-level entry and lguest_set_pmd
+ * to set a middle-level entry when PAE is activated.
+ * Again, we set the entry then tell the Host which page we changed,
+ * and the index of the entry we changed. */
+#ifdef CONFIG_X86_PAE
+static void lguest_set_pud(pud_t *pudp, pud_t pudval)
+{
+ native_set_pud(pudp, pudval);
+
+ /* 32 bytes aligned pdpt address and the index. */
+ lazy_hcall2(LHCALL_SET_PGD, __pa(pudp) & 0xFFFFFFE0,
+ (__pa(pudp) & 0x1F) / sizeof(pud_t));
+}
+
+static void lguest_set_pmd(pmd_t *pmdp, pmd_t pmdval)
+{
+ native_set_pmd(pmdp, pmdval);
+ lazy_hcall2(LHCALL_SET_PMD, __pa(pmdp) & PAGE_MASK,
+ (__pa(pmdp) & (PAGE_SIZE - 1)) / sizeof(pmd_t));
+}
+#else
+
+/* The Guest calls lguest_set_pmd to set a top-level entry when PAE is not
+ * activated. */
static void lguest_set_pmd(pmd_t *pmdp, pmd_t pmdval)
{
native_set_pmd(pmdp, pmdval);
lazy_hcall2(LHCALL_SET_PGD, __pa(pmdp) & PAGE_MASK,
(__pa(pmdp) & (PAGE_SIZE - 1)) / sizeof(pmd_t));
}
+#endif
/* There are a couple of legacy places where the kernel sets a PTE, but we
* don't know the top level any more. This is useless for us, since we don't
@@ -569,6 +603,26 @@ static void lguest_set_pte(pte_t *ptep, pte_t pteval)
lazy_hcall1(LHCALL_FLUSH_TLB, 1);
}
+#ifdef CONFIG_X86_PAE
+static void lguest_set_pte_atomic(pte_t *ptep, pte_t pte)
+{
+ native_set_pte_atomic(ptep, pte);
+ if (cr3_changed)
+ lazy_hcall1(LHCALL_FLUSH_TLB, 1);
+}
+
+void lguest_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
+{
+ native_pte_clear(mm, addr, ptep);
+ lguest_pte_update(mm, addr, ptep);
+}
+
+void lguest_pmd_clear(pmd_t *pmdp)
+{
+ lguest_set_pmd(pmdp, __pmd(0));
+}
+#endif
+
/* Unfortunately for Lguest, the pv_mmu_ops for page tables were based on
* native page table operations. On native hardware you can set a new page
* table entry whenever you want, but if you want to remove one you have to do
@@ -1035,6 +1089,7 @@ __init void lguest_init(void)
pv_info.name = "lguest";
pv_info.paravirt_enabled = 1;
pv_info.kernel_rpl = 1;
+ pv_info.shared_kernel_pmd = 1;
/* We set up all the lguest overrides for sensitive operations. These
* are detailed with the operations themselves. */
@@ -1080,6 +1135,12 @@ __init void lguest_init(void)
pv_mmu_ops.set_pte = lguest_set_pte;
pv_mmu_ops.set_pte_at = lguest_set_pte_at;
pv_mmu_ops.set_pmd = lguest_set_pmd;
+#ifdef CONFIG_X86_PAE
+ pv_mmu_ops.set_pte_atomic = lguest_set_pte_atomic;
+ pv_mmu_ops.pte_clear = lguest_pte_clear;
+ pv_mmu_ops.pmd_clear = lguest_pmd_clear;
+ pv_mmu_ops.set_pud = lguest_set_pud;
+#endif
pv_mmu_ops.read_cr2 = lguest_read_cr2;
pv_mmu_ops.read_cr3 = lguest_read_cr3;
pv_mmu_ops.lazy_mode.enter = paravirt_enter_lazy_mmu;