From: Avi Kivity <avi@qumranet.com>
To: linux-kernel@vger.kernel.org, kvm-devel@lists.sourceforge.net
Cc: Marcelo Tosatti <marcelo@kvack.org>
Subject: [PATCH 02/33] KVM: MMU: emulated cmpxchg8b should be atomic on i386
Date: Sun, 6 Jan 2008 16:39:08 +0200 [thread overview]
Message-ID: <1199630379-28638-3-git-send-email-avi@qumranet.com> (raw)
In-Reply-To: <1199630379-28638-1-git-send-email-avi@qumranet.com>
From: Marcelo Tosatti <marcelo@kvack.org>
Emulate cmpxchg8b atomically on i386. This is required to avoid a guest
pte walker from seeing a splitted write.
[avi: make it compile]
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Avi Kivity <avi@qumranet.com>
---
drivers/kvm/x86.c | 26 ++++++++++++++++++++++++++
1 files changed, 26 insertions(+), 0 deletions(-)
diff --git a/drivers/kvm/x86.c b/drivers/kvm/x86.c
index 9db4e32..3b79684 100644
--- a/drivers/kvm/x86.c
+++ b/drivers/kvm/x86.c
@@ -25,6 +25,7 @@
#include <linux/vmalloc.h>
#include <linux/module.h>
#include <linux/mman.h>
+#include <linux/highmem.h>
#include <asm/uaccess.h>
#include <asm/msr.h>
@@ -1674,6 +1675,31 @@ static int emulator_cmpxchg_emulated(unsigned long addr,
reported = 1;
printk(KERN_WARNING "kvm: emulating exchange as write\n");
}
+#ifndef CONFIG_X86_64
+ /* guests cmpxchg8b have to be emulated atomically */
+ if (bytes == 8) {
+ gpa_t gpa = vcpu->mmu.gva_to_gpa(vcpu, addr);
+ struct page *page;
+ char *addr;
+ u64 val;
+
+ if (gpa == UNMAPPED_GVA ||
+ (gpa & PAGE_MASK) == APIC_DEFAULT_PHYS_BASE)
+ goto emul_write;
+
+ if (((gpa + bytes - 1) & PAGE_MASK) != (gpa & PAGE_MASK))
+ goto emul_write;
+
+ val = *(u64 *)new;
+ page = gfn_to_page(vcpu->kvm, gpa >> PAGE_SHIFT);
+ addr = kmap_atomic(page, KM_USER0);
+ set_64bit((u64 *)(addr + offset_in_page(gpa)), val);
+ kunmap_atomic(addr, KM_USER0);
+ kvm_release_page_dirty(page);
+ }
+emul_write:
+#endif
+
return emulator_write_emulated(addr, new, bytes, vcpu);
}
--
1.5.3.7
next prev parent reply other threads:[~2008-01-06 14:40 UTC|newest]
Thread overview: 36+ messages / expand[flat|nested] mbox.gz Atom feed top
2008-01-06 14:39 [PATCH 00/33] KVM patch queue review for 2.6.25 merge window (part V) Avi Kivity
2008-01-06 14:39 ` [PATCH 01/33] KVM: SVM: support writing 0 to K8 performance counter control registers Avi Kivity
2008-01-06 14:39 ` Avi Kivity [this message]
2008-01-06 18:57 ` [kvm-devel] [PATCH 02/33] KVM: MMU: emulated cmpxchg8b should be atomic on i386 Carlo Marcelo Arenas Belon
2008-01-07 9:24 ` Avi Kivity
2008-01-06 14:39 ` [PATCH 03/33] KVM: Portability: Move kvm{pic,ioapic} accesors to x86 specific code Avi Kivity
2008-01-06 14:39 ` [PATCH 04/33] KVM: Portability: Introduce kvm_vcpu_arch Avi Kivity
2008-01-06 14:39 ` [PATCH 05/33] KVM: Portability: Split mmu-related static inline functions to mmu.h Avi Kivity
2008-01-06 14:39 ` [PATCH 06/33] KVM: Portability: Move kvm_vcpu definition back to kvm.h Avi Kivity
2008-01-06 14:39 ` [PATCH 07/33] KVM: Portability: Expand the KVM_VCPU_COMM in kvm_vcpu structure Avi Kivity
2008-01-06 14:39 ` [PATCH 08/33] KVM: Portability: Move kvm_vcpu_stat to x86.h Avi Kivity
2008-01-06 14:39 ` [PATCH 09/33] KVM: Portability: Move memslot aliases to new struct kvm_arch Avi Kivity
2008-01-06 14:39 ` [PATCH 10/33] KVM: Portability: Move mmu-related fields to kvm_arch Avi Kivity
2008-01-06 14:39 ` [PATCH 11/33] KVM: Portability: move vpic and vioapic " Avi Kivity
2008-01-06 14:39 ` [PATCH 12/33] KVM: Portability: Move round_robin_prev_vcpu and tss_addr " Avi Kivity
2008-01-06 14:39 ` [PATCH 13/33] KVM: Portability: Move kvm_vm_stat to x86.h Avi Kivity
2008-01-06 14:39 ` [PATCH 14/33] KVM: VMX: Add printk_ratelimit in vmx_intr_assist Avi Kivity
2008-01-06 14:39 ` [PATCH 15/33] KVM: Move arch dependent files to new directory arch/x86/kvm/ Avi Kivity
2008-01-06 14:39 ` [PATCH 16/33] KVM: Move drivers/kvm/* to virt/kvm/ Avi Kivity
2008-01-06 14:39 ` [PATCH 17/33] KVM: Move irqchip declarations into new ioapic.h and lapic.h Avi Kivity
2008-01-06 14:39 ` [PATCH 18/33] KVM: Move ioapic code to common directory Avi Kivity
2008-01-06 14:39 ` [PATCH 19/33] KVM: Move kvm_vcpu_kick() to x86.c Avi Kivity
2008-01-06 14:39 ` [PATCH 20/33] KVM: Expose ioapic to ia64 save/restore APIs Avi Kivity
2008-01-06 14:39 ` [PATCH 21/33] KVM: MMU: Coalesce remote tlb flushes Avi Kivity
2008-01-06 14:39 ` [PATCH 22/33] KVM: MMU: Add cache miss statistic Avi Kivity
2008-01-06 14:39 ` [PATCH 23/33] KVM: Print data for unimplemented wrmsr Avi Kivity
2008-01-06 14:39 ` [PATCH 24/33] KVM: local APIC TPR access reporting facility Avi Kivity
2008-01-06 14:39 ` [PATCH 25/33] KVM: Accelerated apic support Avi Kivity
2008-01-06 14:39 ` [PATCH 26/33] KVM: Disable vapic support on Intel machines with FlexPriority Avi Kivity
2008-01-06 14:39 ` [PATCH 27/33] KVM: MMU: Concurrent guest walkers Avi Kivity
2008-01-06 14:39 ` [PATCH 28/33] KVM: Add kvm_read_guest_atomic() Avi Kivity
2008-01-06 14:39 ` [PATCH 29/33] KVM: MMU: Avoid calling gfn_to_page() in mmu_set_spte() Avi Kivity
2008-01-06 14:39 ` [PATCH 30/33] KVM: MMU: Switch to mmu spinlock Avi Kivity
2008-01-06 14:39 ` [PATCH 31/33] KVM: MMU: Move kvm_free_some_pages() into critical section Avi Kivity
2008-01-06 14:39 ` [PATCH 32/33] KVM: MMU: Merge shadow level check in FNAME(fetch) Avi Kivity
2008-01-06 14:39 ` [PATCH 33/33] KVM: x86 emulator: Only allow VMCALL/VMMCALL trapped by #UD Avi Kivity
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1199630379-28638-3-git-send-email-avi@qumranet.com \
--to=avi@qumranet.com \
--cc=kvm-devel@lists.sourceforge.net \
--cc=linux-kernel@vger.kernel.org \
--cc=marcelo@kvack.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox
all inboxes | Powered by JetHome®