[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Qemu-devel] [PULL 39/51] translate-all: report correct avg host TB size
From: |
Richard Henderson |
Subject: |
[Qemu-devel] [PULL 39/51] translate-all: report correct avg host TB size |
Date: |
Wed, 25 Oct 2017 11:35:23 +0200 |
From: "Emilio G. Cota" <address@hidden>
Since commit 6e3b2bfd6 ("tcg: allocate TB structs before the
corresponding translated code") we are not fully utilizing
code_gen_buffer for translated code, and therefore are
incorrectly reporting the amount of translated code as well as
the average host TB size. Address this by:
- Making the conscious choice of misreporting the total translated code;
doing otherwise would mislead users into thinking "-tb-size" is not
honoured.
- Expanding tb_tree_stats to accurately count the bytes of translated code on
the host, and using this for reporting the average tb host size,
as well as the expansion ratio.
In the future we might want to consider reporting the accurate numbers for
the total translated code, together with a "bookkeeping/overhead" field to
account for the TB structs.
Reviewed-by: Richard Henderson <address@hidden>
Signed-off-by: Emilio G. Cota <address@hidden>
Signed-off-by: Richard Henderson <address@hidden>
---
accel/tcg/translate-all.c | 32 +++++++++++++++++++++++---------
1 file changed, 23 insertions(+), 9 deletions(-)
diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
index e929ccb30b..7e2c0cdb98 100644
--- a/accel/tcg/translate-all.c
+++ b/accel/tcg/translate-all.c
@@ -921,6 +921,15 @@ static void page_flush_tb(void)
}
}
+static gboolean tb_host_size_iter(gpointer key, gpointer value, gpointer data)
+{
+ const TranslationBlock *tb = value;
+ size_t *size = data;
+
+ *size += tb->tc.size;
+ return false;
+}
+
/* flush all the translation blocks */
static void do_tb_flush(CPUState *cpu, run_on_cpu_data tb_flush_count)
{
@@ -935,11 +944,12 @@ static void do_tb_flush(CPUState *cpu, run_on_cpu_data
tb_flush_count)
if (DEBUG_TB_FLUSH_GATE) {
size_t nb_tbs = g_tree_nnodes(tcg_ctx.tb_ctx.tb_tree);
+ size_t host_size = 0;
- printf("qemu: flush code_size=%td nb_tbs=%zu avg_tb_size=%td\n",
+ g_tree_foreach(tcg_ctx.tb_ctx.tb_tree, tb_host_size_iter, &host_size);
+ printf("qemu: flush code_size=%td nb_tbs=%zu avg_tb_size=%zu\n",
tcg_ctx.code_gen_ptr - tcg_ctx.code_gen_buffer, nb_tbs,
- nb_tbs > 0 ?
- (tcg_ctx.code_gen_ptr - tcg_ctx.code_gen_buffer) / nb_tbs : 0);
+ nb_tbs > 0 ? host_size / nb_tbs : 0);
}
if ((unsigned long)(tcg_ctx.code_gen_ptr - tcg_ctx.code_gen_buffer)
> tcg_ctx.code_gen_buffer_size) {
@@ -1882,6 +1892,7 @@ static void print_qht_statistics(FILE *f,
fprintf_function cpu_fprintf,
}
struct tb_tree_stats {
+ size_t host_size;
size_t target_size;
size_t max_target_size;
size_t direct_jmp_count;
@@ -1894,6 +1905,7 @@ static gboolean tb_tree_stats_iter(gpointer key, gpointer
value, gpointer data)
const TranslationBlock *tb = value;
struct tb_tree_stats *tst = data;
+ tst->host_size += tb->tc.size;
tst->target_size += tb->size;
if (tb->size > tst->max_target_size) {
tst->max_target_size = tb->size;
@@ -1922,6 +1934,11 @@ void dump_exec_info(FILE *f, fprintf_function
cpu_fprintf)
g_tree_foreach(tcg_ctx.tb_ctx.tb_tree, tb_tree_stats_iter, &tst);
/* XXX: avoid using doubles ? */
cpu_fprintf(f, "Translation buffer state:\n");
+ /*
+ * Report total code size including the padding and TB structs;
+ * otherwise users might think "-tb-size" is not honoured.
+ * For avg host size we use the precise numbers from tb_tree_stats though.
+ */
cpu_fprintf(f, "gen code size %td/%zd\n",
tcg_ctx.code_gen_ptr - tcg_ctx.code_gen_buffer,
tcg_ctx.code_gen_highwater - tcg_ctx.code_gen_buffer);
@@ -1929,12 +1946,9 @@ void dump_exec_info(FILE *f, fprintf_function
cpu_fprintf)
cpu_fprintf(f, "TB avg target size %zu max=%zu bytes\n",
nb_tbs ? tst.target_size / nb_tbs : 0,
tst.max_target_size);
- cpu_fprintf(f, "TB avg host size %td bytes (expansion ratio: %0.1f)\n",
- nb_tbs ? (tcg_ctx.code_gen_ptr -
- tcg_ctx.code_gen_buffer) / nb_tbs : 0,
- tst.target_size ? (double) (tcg_ctx.code_gen_ptr -
- tcg_ctx.code_gen_buffer) /
- tst.target_size : 0);
+ cpu_fprintf(f, "TB avg host size %zu bytes (expansion ratio: %0.1f)\n",
+ nb_tbs ? tst.host_size / nb_tbs : 0,
+ tst.target_size ? (double)tst.host_size / tst.target_size : 0);
cpu_fprintf(f, "cross page TB count %zu (%zu%%)\n", tst.cross_page,
nb_tbs ? (tst.cross_page * 100) / nb_tbs : 0);
cpu_fprintf(f, "direct jump count %zu (%zu%%) (2 jumps=%zu %zu%%)\n",
--
2.13.6
- [Qemu-devel] [PULL 27/51] target/hppa: check CF_PARALLEL instead of parallel_cpus, (continued)
- [Qemu-devel] [PULL 27/51] target/hppa: check CF_PARALLEL instead of parallel_cpus, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 28/51] target/i386: check CF_PARALLEL instead of parallel_cpus, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 31/51] target/sh4: check CF_PARALLEL instead of parallel_cpus, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 25/51] tcg: convert tb->cflags reads to tb_cflags(tb), Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 29/51] target/m68k: check CF_PARALLEL instead of parallel_cpus, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 33/51] tcg: check CF_PARALLEL instead of parallel_cpus, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 30/51] target/s390x: check CF_PARALLEL instead of parallel_cpus, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 34/51] cpu-exec: lookup/generate TB outside exclusive region during step_atomic, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 32/51] target/sparc: check CF_PARALLEL instead of parallel_cpus, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 35/51] tcg: Add CF_LAST_IO + CF_USE_ICOUNT to CF_HASH_MASK, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 39/51] translate-all: report correct avg host TB size,
Richard Henderson <=
- [Qemu-devel] [PULL 37/51] translate-all: use a binary search tree to track TBs in TBContext, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 36/51] tcg: Remove CF_IGNORE_ICOUNT, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 38/51] exec-all: rename tb_free to tb_remove, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 43/51] tcg: introduce **tcg_ctxs to keep track of all TCGContext's, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 42/51] gen-icount: fold exitreq_label into TCGContext, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 40/51] tcg: take tb_ctx out of TCGContext, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 44/51] tcg: distribute profiling counters across TCGContext's, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 45/51] tcg: allocate optimizer temps with tcg_malloc, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 46/51] osdep: introduce qemu_mprotect_rwx/none, Richard Henderson, 2017/10/25
- [Qemu-devel] [PULL 41/51] tcg: define tcg_init_ctx and make tcg_ctx a pointer, Richard Henderson, 2017/10/25