* [Qemu-devel] [PATCH] KVM: flush icache after writing to RAM
@ 2011-04-08 18:00 Scott Wood
0 siblings, 0 replies; only message in thread
From: Scott Wood @ 2011-04-08 18:00 UTC (permalink / raw)
To: qemu-devel
This is required so that the guest does not execute any stale
instructions.
qemu-kvm does this in cpu_physical_memory_rw, but not in
cpu_physical_memory_write_rom.
Signed-off-by: Scott Wood <scottwood@freescale.com>
---
exec.c | 14 ++++++++++++++
1 files changed, 14 insertions(+), 0 deletions(-)
diff --git a/exec.c b/exec.c
index 983c0db..055d304 100644
--- a/exec.c
+++ b/exec.c
@@ -33,6 +33,7 @@
#include "osdep.h"
#include "kvm.h"
#include "qemu-timer.h"
+#include "cache-utils.h"
#if defined(CONFIG_USER_ONLY)
#include <qemu.h>
#include <signal.h>
@@ -3768,6 +3769,12 @@ void cpu_physical_memory_rw(target_phys_addr_t addr, uint8_t *buf,
cpu_physical_memory_set_dirty_flags(
addr1, (0xff & ~CODE_DIRTY_FLAG));
}
+ /* qemu doesn't execute guest code directly, but kvm does
+ therefore flush instruction caches */
+ if (kvm_enabled()) {
+ flush_icache_range((unsigned long)ptr,
+ (unsigned long)ptr + l);
+ }
}
} else {
if ((pd & ~TARGET_PAGE_MASK) > IO_MEM_ROM &&
@@ -3838,6 +3845,13 @@ void cpu_physical_memory_write_rom(target_phys_addr_t addr,
/* ROM/RAM case */
ptr = qemu_get_ram_ptr(addr1);
memcpy(ptr, buf, l);
+
+ /* qemu doesn't execute guest code directly, but kvm does
+ therefore flush instruction caches */
+ if (kvm_enabled()) {
+ flush_icache_range((unsigned long)ptr,
+ (unsigned long)ptr + l);
+ }
}
len -= l;
buf += l;
--
1.7.1
^ permalink raw reply related [flat|nested] only message in thread
only message in thread, other threads:[~2011-04-08 18:00 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-04-08 18:00 [Qemu-devel] [PATCH] KVM: flush icache after writing to RAM Scott Wood
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).