All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Emilio G. Cota" <cota@braap.org>
To: QEMU Developers <qemu-devel@nongnu.org>,
	MTTCG Devel <mttcg@listserver.greensocs.com>
Cc: "Alex Bennée" <alex.bennee@linaro.org>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	"Peter Crosthwaite" <crosthwaite.peter@gmail.com>,
	"Richard Henderson" <rth@twiddle.net>,
	"Peter Maydell" <peter.maydell@linaro.org>,
	"Sergey Fedorov" <serge.fdrv@gmail.com>
Subject: [Qemu-devel] [PATCH v3 07/11] tb hash: hash phys_pc, pc, and flags with xxhash
Date: Tue, 19 Apr 2016 19:07:46 -0400	[thread overview]
Message-ID: <1461107270-19234-8-git-send-email-cota@braap.org> (raw)
In-Reply-To: <1461107270-19234-1-git-send-email-cota@braap.org>

For some workloads such as arm bootup, tb_phys_hash is performance-critical.
The is due to the high frequency of accesses to the hash table, originated
by (frequent) TLB flushes that wipe out the cpu-private tb_jmp_cache's.
More info:
  https://lists.nongnu.org/archive/html/qemu-devel/2016-03/msg05098.html

To dig further into this I modified an arm image booting debian jessie to
immediately shut down after boot. Analysis revealed that quite a bit of time
is unnecessarily spent in tb_phys_hash: the cause is poor hashing that
results in very uneven loading of chains in the hash table's buckets;
the longest observed chain had ~550 elements.

The appended addresses this with two changes:

1) Use xxhash as the hash table's hash function. xxhash is a fast,
   high-quality hashing function.

2) Feed the hashing function with not just tb_phys, but also pc and flags.

This improves performance over using just tb_phys for hashing, since that
resulted in some hash buckets having many TB's, while others getting very few;
with these changes, the longest observed chain on a single hash bucket is
brought down from ~550 to ~40.

Tests show that the other element checked for in tb_find_physical,
cs_base, is always a match when tb_phys+pc+flags are a match,
so hashing cs_base is wasteful. It could be that this is an ARM-only
thing, though.

BTW, after this change the hash table should not be called "tb_hash_phys"
anymore; this is addressed later in this series.

This change gives consistent bootup time improvements. I tested two
host machines:
- Intel Xeon E5-2690: 11.6% less time
- Intel i7-4790K: 19.2% less time

Increasing the number of hash buckets yields further improvements. However,
using a larger, fixed number of buckets can degrade performance for other
workloads that do not translate as many blocks (600K+ for debian-jessie arm
bootup). This is dealt with later in this series.

Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
---
 cpu-exec.c             | 2 +-
 include/exec/tb-hash.h | 8 ++++++--
 translate-all.c        | 6 +++---
 3 files changed, 10 insertions(+), 6 deletions(-)

diff --git a/cpu-exec.c b/cpu-exec.c
index debc65c..a889cf1 100644
--- a/cpu-exec.c
+++ b/cpu-exec.c
@@ -233,7 +233,7 @@ static TranslationBlock *tb_find_physical(CPUState *cpu,
     /* find translated block using physical mappings */
     phys_pc = get_page_addr_code(env, pc);
     phys_page1 = phys_pc & TARGET_PAGE_MASK;
-    h = tb_phys_hash_func(phys_pc);
+    h = tb_hash_func(phys_pc, pc, flags);
     ptb1 = &tcg_ctx.tb_ctx.tb_phys_hash[h];
     for(;;) {
         tb = *ptb1;
diff --git a/include/exec/tb-hash.h b/include/exec/tb-hash.h
index 0f4e8a0..4b9635a 100644
--- a/include/exec/tb-hash.h
+++ b/include/exec/tb-hash.h
@@ -20,6 +20,9 @@
 #ifndef EXEC_TB_HASH
 #define EXEC_TB_HASH
 
+#include "exec/exec-all.h"
+#include "exec/tb-hash-xx.h"
+
 /* Only the bottom TB_JMP_PAGE_BITS of the jump cache hash bits vary for
    addresses on the same page.  The top bits are the same.  This allows
    TLB invalidation to quickly clear a subset of the hash table.  */
@@ -43,9 +46,10 @@ static inline unsigned int tb_jmp_cache_hash_func(target_ulong pc)
            | (tmp & TB_JMP_ADDR_MASK));
 }
 
-static inline unsigned int tb_phys_hash_func(tb_page_addr_t pc)
+static inline
+uint32_t tb_hash_func(tb_page_addr_t phys_pc, target_ulong pc, int flags)
 {
-    return (pc >> 2) & (CODE_GEN_PHYS_HASH_SIZE - 1);
+    return tb_hash_func5(phys_pc, pc, flags) & (CODE_GEN_PHYS_HASH_SIZE - 1);
 }
 
 #endif
diff --git a/translate-all.c b/translate-all.c
index 1a8f68b..eca2f16 100644
--- a/translate-all.c
+++ b/translate-all.c
@@ -972,7 +972,7 @@ void tb_phys_invalidate(TranslationBlock *tb, tb_page_addr_t page_addr)
 
     /* remove the TB from the hash list */
     phys_pc = tb->page_addr[0] + (tb->pc & ~TARGET_PAGE_MASK);
-    h = tb_phys_hash_func(phys_pc);
+    h = tb_hash_func(phys_pc, tb->pc, tb->flags);
     tb_hash_remove(&tcg_ctx.tb_ctx.tb_phys_hash[h], tb);
 
     /* remove the TB from the page list */
@@ -1474,8 +1474,8 @@ static void tb_link_page(TranslationBlock *tb, tb_page_addr_t phys_pc,
     unsigned int h;
     TranslationBlock **ptb;
 
-    /* add in the physical hash table */
-    h = tb_phys_hash_func(phys_pc);
+    /* add in the hash table */
+    h = tb_hash_func(phys_pc, tb->pc, tb->flags);
     ptb = &tcg_ctx.tb_ctx.tb_phys_hash[h];
     tb->phys_hash_next = *ptb;
     *ptb = tb;
-- 
2.5.0

  parent reply	other threads:[~2016-04-19 23:07 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-19 23:07 [Qemu-devel] [PATCH v3 00/11] tb hash improvements Emilio G. Cota
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 01/11] compiler.h: add QEMU_ALIGNED() to enforce struct alignment Emilio G. Cota
2016-04-22  9:32   ` Alex Bennée
2016-04-22  9:35   ` Peter Maydell
2016-04-22 15:50     ` Emilio G. Cota
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 02/11] seqlock: remove optional mutex Emilio G. Cota
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 03/11] seqlock: rename write_lock/unlock to write_begin/end Emilio G. Cota
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 04/11] include/processor.h: define cpu_relax() Emilio G. Cota
2016-04-20 12:15   ` KONRAD Frederic
2016-04-20 17:16     ` Emilio G. Cota
2016-04-20 17:18       ` Peter Maydell
2016-04-20 17:32   ` [Qemu-devel] [UPDATED " Emilio G. Cota
2016-04-22  9:35   ` [Qemu-devel] [PATCH " Alex Bennée
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 05/11] qemu-thread: add simple test-and-set spinlock Emilio G. Cota
2016-04-20 15:18   ` Richard Henderson
2016-04-20 17:17     ` Emilio G. Cota
2016-04-20 17:55       ` Richard Henderson
2016-04-20 18:11         ` Emilio G. Cota
2016-04-20 19:39           ` Richard Henderson
2016-04-21 16:24             ` Emilio G. Cota
2016-04-20 17:35   ` [Qemu-devel] [UPDATED " Emilio G. Cota
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 06/11] exec: add tb_hash_func5, derived from xxhash Emilio G. Cota
2016-04-20 15:19   ` Richard Henderson
2016-04-22 12:58   ` Alex Bennée
2016-04-19 23:07 ` Emilio G. Cota [this message]
2016-04-22 12:58   ` [Qemu-devel] [PATCH v3 07/11] tb hash: hash phys_pc, pc, and flags with xxhash Alex Bennée
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 08/11] qht: QEMU's fast, resizable and scalable Hash Table Emilio G. Cota
2016-04-22 14:04   ` Alex Bennée
2016-04-24 20:01   ` Richard Henderson
2016-04-24 21:58     ` Emilio G. Cota
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 09/11] qht: add test program Emilio G. Cota
2016-04-22 14:35   ` Alex Bennée
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 10/11] tb hash: track translated blocks with qht Emilio G. Cota
2016-04-28 13:27   ` Alex Bennée
2016-04-19 23:07 ` [Qemu-devel] [PATCH v3 11/11] translate-all: add tb hash bucket info to 'info jit' dump Emilio G. Cota
2016-04-20 15:21   ` Richard Henderson
2016-04-22 14:38   ` Alex Bennée
2016-04-22 17:41   ` Richard Henderson
2016-04-22 19:23     ` Emilio G. Cota
2016-04-22 19:59     ` Richard Henderson
2016-04-22 23:57       ` Emilio G. Cota
2016-04-24 19:46         ` Richard Henderson
2016-04-24 22:06           ` Emilio G. Cota
2016-04-27  2:43             ` Emilio G. Cota
2016-04-28 16:37               ` Richard Henderson

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=1461107270-19234-8-git-send-email-cota@braap.org \
    --to=cota@braap.org \
    --cc=alex.bennee@linaro.org \
    --cc=crosthwaite.peter@gmail.com \
    --cc=mttcg@listserver.greensocs.com \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    --cc=serge.fdrv@gmail.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.