From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: alistair.francis@wdc.com
Subject: [PATCH 9/9] cputlb: Hoist timestamp outside of loops over tlbs
Date: Thu, 9 Jan 2020 13:49:07 +1100 [thread overview]
Message-ID: <20200109024907.2730-10-richard.henderson@linaro.org> (raw)
In-Reply-To: <20200109024907.2730-1-richard.henderson@linaro.org>
Do not call get_clock_realtime() in tlb_mmu_resize_locked,
but hoist outside of any loop over a set of tlbs. This is
only two (indirect) callers, tlb_flush_by_mmuidx_async_work
and tlb_flush_page_locked, so not onerous.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
accel/tcg/cputlb.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
diff --git a/accel/tcg/cputlb.c b/accel/tcg/cputlb.c
index 761e9d44d7..9f6cb36921 100644
--- a/accel/tcg/cputlb.c
+++ b/accel/tcg/cputlb.c
@@ -137,12 +137,12 @@ static void tlb_window_reset(CPUTLBDesc *desc, int64_t ns,
* high), since otherwise we are likely to have a significant amount of
* conflict misses.
*/
-static void tlb_mmu_resize_locked(CPUTLBDesc *desc, CPUTLBDescFast *fast)
+static void tlb_mmu_resize_locked(CPUTLBDesc *desc, CPUTLBDescFast *fast,
+ int64_t now)
{
size_t old_size = tlb_n_entries(fast);
size_t rate;
size_t new_size = old_size;
- int64_t now = get_clock_realtime();
int64_t window_len_ms = 100;
int64_t window_len_ns = window_len_ms * 1000 * 1000;
bool window_expired = now > desc->window_begin_ns + window_len_ns;
@@ -222,12 +222,13 @@ static void tlb_mmu_flush_locked(CPUTLBDesc *desc, CPUTLBDescFast *fast)
memset(desc->vtable, -1, sizeof(desc->vtable));
}
-static void tlb_flush_one_mmuidx_locked(CPUArchState *env, int mmu_idx)
+static void tlb_flush_one_mmuidx_locked(CPUArchState *env, int mmu_idx,
+ int64_t now)
{
CPUTLBDesc *desc = &env_tlb(env)->d[mmu_idx];
CPUTLBDescFast *fast = &env_tlb(env)->f[mmu_idx];
- tlb_mmu_resize_locked(desc, fast);
+ tlb_mmu_resize_locked(desc, fast, now);
tlb_mmu_flush_locked(desc, fast);
}
@@ -310,6 +311,7 @@ static void tlb_flush_by_mmuidx_async_work(CPUState *cpu, run_on_cpu_data data)
CPUArchState *env = cpu->env_ptr;
uint16_t asked = data.host_int;
uint16_t all_dirty, work, to_clean;
+ int64_t now = get_clock_realtime();
assert_cpu_is_self(cpu);
@@ -324,7 +326,7 @@ static void tlb_flush_by_mmuidx_async_work(CPUState *cpu, run_on_cpu_data data)
for (work = to_clean; work != 0; work &= work - 1) {
int mmu_idx = ctz32(work);
- tlb_flush_one_mmuidx_locked(env, mmu_idx);
+ tlb_flush_one_mmuidx_locked(env, mmu_idx, now);
}
qemu_spin_unlock(&env_tlb(env)->c.lock);
@@ -446,7 +448,7 @@ static void tlb_flush_page_locked(CPUArchState *env, int midx,
tlb_debug("forcing full flush midx %d ("
TARGET_FMT_lx "/" TARGET_FMT_lx ")\n",
midx, lp_addr, lp_mask);
- tlb_flush_one_mmuidx_locked(env, midx);
+ tlb_flush_one_mmuidx_locked(env, midx, get_clock_realtime());
} else {
if (tlb_flush_entry_locked(tlb_entry(env, midx, page), page)) {
tlb_n_used_entries_dec(env, midx);
--
2.20.1
next prev parent reply other threads:[~2020-01-09 2:56 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-09 2:48 [PATCH 0/9] cputlb: Various cleanups Richard Henderson
2020-01-09 2:48 ` [PATCH 1/9] cputlb: Merge tlb_table_flush_by_mmuidx into tlb_flush_one_mmuidx_locked Richard Henderson
2020-01-20 0:34 ` Alistair Francis
2020-01-20 13:34 ` Alex Bennée
2020-01-09 2:49 ` [PATCH 2/9] cputlb: Make tlb_n_entries private to cputlb.c Richard Henderson
2020-01-20 0:36 ` Alistair Francis
2020-01-20 8:53 ` Philippe Mathieu-Daudé
2020-01-20 13:34 ` Alex Bennée
2020-01-09 2:49 ` [PATCH 3/9] cputlb: Pass CPUTLBDescFast to tlb_n_entries and sizeof_tlb Richard Henderson
2020-01-20 0:37 ` Alistair Francis
2020-01-20 8:54 ` Philippe Mathieu-Daudé
2020-01-20 13:36 ` Alex Bennée
2020-01-20 13:39 ` Alex Bennée
2020-01-09 2:49 ` [PATCH 4/9] cputlb: Hoist tlb portions in tlb_mmu_resize_locked Richard Henderson
2020-01-20 8:58 ` Philippe Mathieu-Daudé
2020-01-20 12:05 ` Alistair Francis
2020-01-20 13:40 ` Alex Bennée
2020-01-09 2:49 ` [PATCH 5/9] cputlb: Hoist tlb portions in tlb_flush_one_mmuidx_locked Richard Henderson
2020-01-20 8:58 ` Philippe Mathieu-Daudé
2020-01-20 13:41 ` Alex Bennée
2020-01-20 22:56 ` Alistair Francis
2020-01-09 2:49 ` [PATCH 6/9] cputlb: Split out tlb_mmu_flush_locked Richard Henderson
2020-01-20 8:59 ` Philippe Mathieu-Daudé
2020-01-20 13:41 ` Alex Bennée
2020-01-20 22:58 ` Alistair Francis
2020-01-09 2:49 ` [PATCH 7/9] cputlb: Partially merge tlb_dyn_init into tlb_init Richard Henderson
2020-01-20 9:01 ` Philippe Mathieu-Daudé
2020-01-20 14:33 ` Alex Bennée
2020-01-20 23:00 ` Alistair Francis
2020-01-09 2:49 ` [PATCH 8/9] cputlb: Initialize tlbs as flushed Richard Henderson
2020-01-20 14:33 ` Alex Bennée
2020-01-20 23:01 ` Alistair Francis
2020-01-09 2:49 ` Richard Henderson [this message]
2020-01-20 9:02 ` [PATCH 9/9] cputlb: Hoist timestamp outside of loops over tlbs Philippe Mathieu-Daudé
2020-01-20 14:36 ` Alex Bennée
2020-01-20 23:02 ` Alistair Francis
2020-01-20 23:03 ` [PATCH 0/9] cputlb: Various cleanups Alistair Francis
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=20200109024907.2730-10-richard.henderson@linaro.org \
--to=richard.henderson@linaro.org \
--cc=alistair.francis@wdc.com \
--cc=qemu-devel@nongnu.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;
as well as URLs for NNTP newsgroup(s).