qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: peter.maydell@linaro.org, "Emilio G. Cota" <cota@braap.org>
Subject: [Qemu-devel] [PULL 07/18] translate-all: make l1_map lockless
Date: Thu, 14 Jun 2018 09:31:36 -1000	[thread overview]
Message-ID: <20180614193147.29680-8-richard.henderson@linaro.org> (raw)
In-Reply-To: <20180614193147.29680-1-richard.henderson@linaro.org>

From: "Emilio G. Cota" <cota@braap.org>

Groundwork for supporting parallel TCG generation.

We never remove entries from the radix tree, so we can use cmpxchg
to implement lockless insertions.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
 accel/tcg/translate-all.c       | 24 ++++++++++++++----------
 docs/devel/multi-thread-tcg.txt |  4 ++--
 2 files changed, 16 insertions(+), 12 deletions(-)

diff --git a/accel/tcg/translate-all.c b/accel/tcg/translate-all.c
index 52e62125ed..52273e5846 100644
--- a/accel/tcg/translate-all.c
+++ b/accel/tcg/translate-all.c
@@ -469,20 +469,12 @@ static void page_init(void)
 #endif
 }
 
-/* If alloc=1:
- * Called with tb_lock held for system emulation.
- * Called with mmap_lock held for user-mode emulation.
- */
 static PageDesc *page_find_alloc(tb_page_addr_t index, int alloc)
 {
     PageDesc *pd;
     void **lp;
     int i;
 
-    if (alloc) {
-        assert_memory_lock();
-    }
-
     /* Level 1.  Always allocated.  */
     lp = l1_map + ((index >> v_l1_shift) & (v_l1_size - 1));
 
@@ -491,11 +483,17 @@ static PageDesc *page_find_alloc(tb_page_addr_t index, int alloc)
         void **p = atomic_rcu_read(lp);
 
         if (p == NULL) {
+            void *existing;
+
             if (!alloc) {
                 return NULL;
             }
             p = g_new0(void *, V_L2_SIZE);
-            atomic_rcu_set(lp, p);
+            existing = atomic_cmpxchg(lp, NULL, p);
+            if (unlikely(existing)) {
+                g_free(p);
+                p = existing;
+            }
         }
 
         lp = p + ((index >> (i * V_L2_BITS)) & (V_L2_SIZE - 1));
@@ -503,11 +501,17 @@ static PageDesc *page_find_alloc(tb_page_addr_t index, int alloc)
 
     pd = atomic_rcu_read(lp);
     if (pd == NULL) {
+        void *existing;
+
         if (!alloc) {
             return NULL;
         }
         pd = g_new0(PageDesc, V_L2_SIZE);
-        atomic_rcu_set(lp, pd);
+        existing = atomic_cmpxchg(lp, NULL, pd);
+        if (unlikely(existing)) {
+            g_free(pd);
+            pd = existing;
+        }
     }
 
     return pd + (index & (V_L2_SIZE - 1));
diff --git a/docs/devel/multi-thread-tcg.txt b/docs/devel/multi-thread-tcg.txt
index a99b4564c6..faf8918b23 100644
--- a/docs/devel/multi-thread-tcg.txt
+++ b/docs/devel/multi-thread-tcg.txt
@@ -134,8 +134,8 @@ tb_set_jmp_target() code. Modification to the linked lists that allow
 searching for linked pages are done under the protect of the
 tb_lock().
 
-The global page table is protected by the tb_lock() in system-mode and
-mmap_lock() in linux-user mode.
+The global page table is a lockless radix tree; cmpxchg is used
+to atomically insert new elements.
 
 The lookup caches are updated atomically and the lookup hash uses QHT
 which is designed for concurrent safe lookup.
-- 
2.17.1

  parent reply	other threads:[~2018-06-14 19:32 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-14 19:31 [Qemu-devel] [PULL 00/18] tcg queued patches Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 01/18] tcg/i386: Use byte form of xgetbv instruction Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 02/18] qht: require a default comparison function Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 03/18] qht: return existing entry when qht_insert fails Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 04/18] tcg: track TBs with per-region BST's Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 05/18] tcg: move tb_ctx.tb_phys_invalidate_count to tcg_ctx Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 06/18] translate-all: iterate over TBs in a page with PAGE_FOR_EACH_TB Richard Henderson
2018-06-14 19:31 ` Richard Henderson [this message]
2018-06-14 19:31 ` [Qemu-devel] [PULL 08/18] translate-all: remove hole in PageDesc Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 09/18] translate-all: work page-by-page in tb_invalidate_phys_range_1 Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 10/18] translate-all: move tb_invalidate_phys_page_range up in the file Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 11/18] translate-all: use per-page locking in !user-mode Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 12/18] translate-all: add page_locked assertions Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 13/18] translate-all: introduce assert_no_pages_locked Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 14/18] translate-all: discard TB when tb_link_page returns an existing matching TB Richard Henderson
2018-06-29  7:25   ` [Qemu-devel] [PULL, " Pavel Dovgalyuk
2018-06-29 18:48     ` Emilio G. Cota
2018-07-02  5:52       ` Pavel Dovgalyuk
2018-07-02 19:52         ` Emilio G. Cota
2018-07-03  5:38           ` Pavel Dovgalyuk
2018-07-04 19:38             ` Emilio G. Cota
2018-07-05  5:51               ` Pavel Dovgalyuk
2018-06-14 19:31 ` [Qemu-devel] [PULL 15/18] translate-all: protect TB jumps with a per-destination-TB lock Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 16/18] cputlb: remove tb_lock from tlb_flush functions Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 17/18] translate-all: remove tb_lock mention from cpu_restore_state_from_tb Richard Henderson
2018-06-14 19:31 ` [Qemu-devel] [PULL 18/18] tcg: remove tb_lock Richard Henderson
2018-06-15 10:41 ` [Qemu-devel] [PULL 00/18] tcg queued patches Peter Maydell
2018-06-15 14:01   ` Emilio G. Cota
2018-06-15 17:54     ` 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=20180614193147.29680-8-richard.henderson@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=cota@braap.org \
    --cc=peter.maydell@linaro.org \
    --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).