The goal is to sleep qemu whenever the guest clock
is in advance compared to the host clock (we use
the monotonic clocks). The amount of time to sleep
is calculated in the execution loop in cpu_exec.
Basically, using QEMU_CLOCK_REALTIME, we calculate
the real time duration of the execution (meaning
generating TBs and executing them) and using the
the fields icount_decr.u16.low and icount_extra
(from the CPUState structure) shifted by icount_time_shift
we calculate the theoretical virtual time elapsed.
Having these 2 values, we can determine if the
guest is in advance and sleep.
Signed-off-by: Sebastian Tanase <sebastian.tan...@openwide.fr>
Tested-by: Camille Bégué <camille.be...@openwide.fr>
---
At first, we tried to approximate at each for loop the real time elapsed
while searching for a TB (generating or retrieving from cache) and
executing it. We would then approximate the virtual time corresponding
to the number of virtual instructions executed. The difference between
these 2 values would allow us to know if the guest is in advance or delayed.
However, the function used for measuring the real time
(qemu_clock_get_ns(QEMU_CLOCK_REALTIME)) proved to be very expensive.
We had an added overhead of 13% of the total run time.
Therefore, we modified the algorithm and only take into account the
difference between the 2 clocks at the begining of the cpu_exec function.
During the for loop we try to reduce the advance of the guest only by
computing the virtual time elapsed and sleeping if necessary. The overhead
is thus reduced to 3%. Even though this method still has a noticeable
overhead, it no longer is a bottleneck in trying to achieve a better
guest frequency for which the guest clock is faster than the host one.
As for the the alignement of the 2 clocks, with the first algorithm
the guest clock was oscillating between -1 and 1ms compared to the host clock.
Using the second algorithm we notice that the guest is 5ms behind the host,
which
is still acceptable for our use case.
The tests where conducted using fio and stress. The host machine in an i5 CPU at
3.10GHz running Debian Jessie (kernel 3.12). The guest machine is an arm
versatile-pb built
with buildroot.
Currently, on our test machine, the lowest icount we can achieve that is
suitable for
aligning the 2 clocks is 6. However, we observe that the IO tests (using fio)
are
slower than the cpu tests (using stress).
---
cpu-exec.c | 73 ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
1 file changed, 73 insertions(+)
diff --git a/cpu-exec.c b/cpu-exec.c
index 38e5f02..eef3cb5 100644
--- a/cpu-exec.c
+++ b/cpu-exec.c
@@ -22,6 +22,7 @@
#include "tcg.h"
#include "qemu/atomic.h"
#include "sysemu/qtest.h"
+#include "qemu/timer.h"
void cpu_loop_exit(CPUState *cpu)
{
@@ -209,6 +210,26 @@ static void cpu_handle_debug_exception(CPUArchState *env)
}
}
+/* Allow the guest to have a max 300us advance.
+ * The difference between the 2 clocks could therefore
+ * oscillate around 0.
+ */
+#define VM_CLOCK_ADVANCE 300000
+
+static int64_t diff_clk;
+static void delay_host(void)
+{
+ static struct timespec sleep_delay, rem_delay;
+ sleep_delay.tv_sec = diff_clk / 1000000000LL;
+ sleep_delay.tv_nsec = diff_clk % 1000000000LL;
+ if (nanosleep(&sleep_delay, &rem_delay) < 0) {
+ diff_clk -= (sleep_delay.tv_sec - rem_delay.tv_sec) * 1000000000LL;
+ diff_clk -= sleep_delay.tv_nsec - rem_delay.tv_nsec;
+ } else {
+ diff_clk = 0;
+ }
+}
+
/* main execution loop */
volatile sig_atomic_t exit_request;
@@ -227,6 +248,8 @@ int cpu_exec(CPUArchState *env)
TranslationBlock *tb;
uint8_t *tc_ptr;
uintptr_t next_tb;
+ /* Delay algorithm */
+ int64_t instr_exec_time, original_extra, original_low;
/* This must be volatile so it is not trashed by longjmp() */
volatile bool have_tb_lock = false;
@@ -283,6 +306,16 @@ int cpu_exec(CPUArchState *env)
#endif
cpu->exception_index = -1;
+ if (icount_align_option) {
+ /* Calculate difference between guest clock and host clock.
+ This delay includes the delay of the last cycle, so
+ what we have to do is sleep until it is 0. As for the
+ advance/delay we gain here, we try to fix it next time. */
+ diff_clk = qemu_clock_get_ns(QEMU_CLOCK_VIRTUAL) -
+ qemu_clock_get_ns(QEMU_CLOCK_REALTIME);
+ original_extra = cpu->icount_extra;
+ original_low = cpu->icount_decr.u16.low;
+ }
/* prepare setjmp context for exception handling */
for(;;) {
if (sigsetjmp(cpu->jmp_env, 0) == 0) {
@@ -600,6 +633,11 @@ int cpu_exec(CPUArchState *env)
}
}
if (unlikely(cpu->exit_request)) {
+ if (icount_align_option) {
+ if (diff_clk > VM_CLOCK_ADVANCE) {
+ delay_host();
+ }
+ }
cpu->exit_request = 0;
cpu->exception_index = EXCP_INTERRUPT;
cpu_loop_exit(cpu);
@@ -668,10 +706,26 @@ int cpu_exec(CPUArchState *env)
}
cpu->icount_extra -= insns_left;
cpu->icount_decr.u16.low = insns_left;
+ if (icount_align_option) {
+ /* Instruction counters have been
+ updated so we update ours */
+ original_extra = cpu->icount_extra;
+ original_low = cpu->icount_decr.u16.low;
+ }
} else {
if (insns_left > 0) {
/* Execute remaining instructions. */
cpu_exec_nocache(env, insns_left, tb);
+ if (icount_align_option) {
+ instr_exec_time = original_low -
+ cpu->icount_decr.u16.low;
+ instr_exec_time = instr_exec_time <<
+ icount_time_shift;
+ diff_clk += instr_exec_time;
+ if (diff_clk > VM_CLOCK_ADVANCE) {
+ delay_host();
+ }
+ }
}
cpu->exception_index = EXCP_INTERRUPT;
next_tb = 0;
@@ -679,11 +733,30 @@ int cpu_exec(CPUArchState *env)
}
break;
}
+ case 1:
+ case 0:
+ if (icount_align_option) {
+ instr_exec_time = original_extra -
+ cpu->icount_extra +
+ original_low -
+ cpu->icount_decr.u16.low;
+ instr_exec_time = instr_exec_time <<
+ icount_time_shift;
+ diff_clk += instr_exec_time;
+ original_extra = cpu->icount_extra;
+ original_low = cpu->icount_decr.u16.low;
+ }
+ break;
default:
break;
}
}
cpu->current_tb = NULL;
+ if (icount_align_option) {
+ if (diff_clk > VM_CLOCK_ADVANCE) {
+ delay_host();
+ }
+ }
/* reset soft MMU for next block (it can currently
only be set by a memory fault) */
} /* for(;;) */
--
2.0.0.rc2
