All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Alex Bennée" <alex.bennee@linaro.org>
To: mttcg@listserver.greensocs.com, qemu-devel@nongnu.org,
	fred.konrad@greensocs.com, a.rigo@virtualopensystems.com,
	cota@braap.org, bobby.prani@gmail.com, nikunj@linux.vnet.ibm.com
Cc: mark.burton@greensocs.com, pbonzini@redhat.com,
	jan.kiszka@siemens.com, serge.fdrv@gmail.com, rth@twiddle.net,
	peter.maydell@linaro.org, claudio.fontana@huawei.com,
	"Alex Bennée" <alex.bennee@linaro.org>,
	"Peter Crosthwaite" <crosthwaite.peter@gmail.com>
Subject: [Qemu-devel] [RFC v4 27/28] cputlb: make tlb_reset_dirty safe for MTTCG
Date: Thu, 11 Aug 2016 16:24:23 +0100	[thread overview]
Message-ID: <1470929064-4092-28-git-send-email-alex.bennee@linaro.org> (raw)
In-Reply-To: <1470929064-4092-1-git-send-email-alex.bennee@linaro.org>

The main use case for tlb_reset_dirty is to set the TLB_NOTDIRTY flags
in TLB entries to force the slow-path on writes. This is used to mark
page ranges containing code which has been translated so it can be
invalidated if written to. To do this safely we need to ensure the TLB
entries in question for all vCPUs are updated before we attempt to run
the code otherwise a race could be introduced.

To achieve this we atomically set the flag in tlb_reset_dirty_range and
take care when setting it when the TLB entry is filled.

The helper function is made static as it isn't used outside of cputlb.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
---
 cputlb.c              | 55 +++++++++++++++++++++++++++++++++++----------------
 include/exec/cputlb.h |  2 --
 2 files changed, 38 insertions(+), 19 deletions(-)

diff --git a/cputlb.c b/cputlb.c
index 945ea02..faeb195 100644
--- a/cputlb.c
+++ b/cputlb.c
@@ -275,32 +275,50 @@ void tlb_unprotect_code(ram_addr_t ram_addr)
     cpu_physical_memory_set_dirty_flag(ram_addr, DIRTY_MEMORY_CODE);
 }
 
-static bool tlb_is_dirty_ram(CPUTLBEntry *tlbe)
-{
-    return (tlbe->addr_write & (TLB_INVALID_MASK|TLB_MMIO|TLB_NOTDIRTY)) == 0;
-}
 
-void tlb_reset_dirty_range(CPUTLBEntry *tlb_entry, uintptr_t start,
+/*
+ * Dirty write flag handling
+ *
+ * When the TCG code writes to a location it looks up the address in
+ * the TLB and uses that data to compute the final address. If any of
+ * the lower bits of the address are set then the slow path is forced.
+ * There are a number of reasons to do this but for normal RAM the
+ * most usual is detecting writes to code regions which may invalidate
+ * generated code.
+ *
+ * Because we want other vCPUs to respond to changes straight away we
+ * update the te->addr_write field atomically. If the TLB entry has
+ * been changed by the vCPU in the mean time we skip the update.
+ */
+
+static void tlb_reset_dirty_range(CPUTLBEntry *tlb_entry, uintptr_t start,
                            uintptr_t length)
 {
-    uintptr_t addr;
+    /* paired with atomic_mb_set in tlb_set_page_with_attrs */
+    uintptr_t orig_addr = atomic_mb_read(&tlb_entry->addr_write);
+    uintptr_t addr = orig_addr;
 
-    if (tlb_is_dirty_ram(tlb_entry)) {
-        addr = (tlb_entry->addr_write & TARGET_PAGE_MASK) + tlb_entry->addend;
+    if ((addr & (TLB_INVALID_MASK | TLB_MMIO | TLB_NOTDIRTY)) == 0) {
+        addr &= TARGET_PAGE_MASK;
+        addr += atomic_read(&tlb_entry->addend);
         if ((addr - start) < length) {
-            tlb_entry->addr_write |= TLB_NOTDIRTY;
+            uintptr_t notdirty_addr = orig_addr | TLB_NOTDIRTY;
+            atomic_cmpxchg(&tlb_entry->addr_write, orig_addr, notdirty_addr);
         }
     }
 }
 
+/* This is a cross vCPU call (i.e. another vCPU resetting the flags of
+ * the target vCPU). As such care needs to be taken that we don't
+ * dangerously race with another vCPU update. The only thing actually
+ * updated is the target TLB entry ->addr_write flags.
+ */
 void tlb_reset_dirty(CPUState *cpu, ram_addr_t start1, ram_addr_t length)
 {
     CPUArchState *env;
 
     int mmu_idx;
 
-    assert_cpu_is_self(cpu);
-
     env = cpu->env_ptr;
     for (mmu_idx = 0; mmu_idx < NB_MMU_MODES; mmu_idx++) {
         unsigned int i;
@@ -386,7 +404,7 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
     MemoryRegionSection *section;
     unsigned int index;
     target_ulong address;
-    target_ulong code_address;
+    target_ulong code_address, write_address;
     uintptr_t addend;
     CPUTLBEntry *te;
     hwaddr iotlb, xlat, sz;
@@ -443,21 +461,24 @@ void tlb_set_page_with_attrs(CPUState *cpu, target_ulong vaddr,
     } else {
         te->addr_code = -1;
     }
+
+    write_address = -1;
     if (prot & PAGE_WRITE) {
         if ((memory_region_is_ram(section->mr) && section->readonly)
             || memory_region_is_romd(section->mr)) {
             /* Write access calls the I/O callback.  */
-            te->addr_write = address | TLB_MMIO;
+            write_address = address | TLB_MMIO;
         } else if (memory_region_is_ram(section->mr)
                    && cpu_physical_memory_is_clean(
                         memory_region_get_ram_addr(section->mr) + xlat)) {
-            te->addr_write = address | TLB_NOTDIRTY;
+            write_address = address | TLB_NOTDIRTY;
         } else {
-            te->addr_write = address;
+            write_address = address;
         }
-    } else {
-        te->addr_write = -1;
     }
+
+    /* Pairs with flag setting in tlb_reset_dirty_range */
+    atomic_mb_set(&te->addr_write, write_address);
 }
 
 /* Add a new TLB entry, but without specifying the memory
diff --git a/include/exec/cputlb.h b/include/exec/cputlb.h
index d454c00..3f94178 100644
--- a/include/exec/cputlb.h
+++ b/include/exec/cputlb.h
@@ -23,8 +23,6 @@
 /* cputlb.c */
 void tlb_protect_code(ram_addr_t ram_addr);
 void tlb_unprotect_code(ram_addr_t ram_addr);
-void tlb_reset_dirty_range(CPUTLBEntry *tlb_entry, uintptr_t start,
-                           uintptr_t length);
 extern int tlb_flush_count;
 
 #endif
-- 
2.7.4

  parent reply	other threads:[~2016-08-11 15:30 UTC|newest]

Thread overview: 64+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-08-11 15:23 [Qemu-devel] [RFC v4 00/28] Base enabling patches for MTTCG Alex Bennée
2016-08-11 15:23 ` [Qemu-devel] [RFC v4 01/28] cpus: make all_vcpus_paused() return bool Alex Bennée
2016-08-11 15:23 ` [Qemu-devel] [RFC v4 02/28] translate_all: DEBUG_FLUSH -> DEBUG_TB_FLUSH Alex Bennée
2016-08-11 15:23 ` [Qemu-devel] [RFC v4 03/28] translate-all: add DEBUG_LOCKING asserts Alex Bennée
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 04/28] cpu-exec: include cpu_index in CPU_LOG_EXEC messages Alex Bennée
2016-09-07  2:21   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 05/28] docs: new design document multi-thread-tcg.txt (DRAFTING) Alex Bennée
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 06/28] tcg: comment on which functions have to be called with tb_lock held Alex Bennée
2016-09-07  2:30   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 07/28] linux-user/elfload: ensure mmap_lock() held while setting up Alex Bennée
2016-09-07  2:34   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 08/28] translate-all: Add assert_(memory|tb)_lock annotations Alex Bennée
2016-09-07  2:41   ` Richard Henderson
2016-09-07  7:08     ` Alex Bennée
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 09/28] tcg: protect TBContext with tb_lock Alex Bennée
2016-09-07  2:48   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 10/28] target-arm/arm-powerctl: wake up sleeping CPUs Alex Bennée
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 11/28] tcg: move tcg_exec_all and helpers above thread fn Alex Bennée
2016-09-07  2:53   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 12/28] tcg: cpus rm tcg_exec_all() Alex Bennée
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 13/28] tcg: add options for enabling MTTCG Alex Bennée
2016-09-07  3:06   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 14/28] tcg: add kick timer for single-threaded vCPU emulation Alex Bennée
2016-09-07  3:25   ` Richard Henderson
2016-09-07  5:40     ` Paolo Bonzini
2016-09-07 10:15       ` Alex Bennée
2016-09-07 10:19     ` Alex Bennée
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 15/28] tcg: rename tcg_current_cpu to tcg_current_rr_cpu Alex Bennée
2016-09-07  3:34   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 16/28] tcg: drop global lock during TCG code execution Alex Bennée
2016-09-07  4:03   ` Richard Henderson
2016-09-07  5:43     ` Paolo Bonzini
2016-09-07  6:43       ` Richard Henderson
2016-09-07 15:15         ` Paolo Bonzini
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 17/28] cpus: re-factor out handle_icount_deadline Alex Bennée
2016-09-07  4:06   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 18/28] tcg: remove global exit_request Alex Bennée
2016-09-07  4:11   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 19/28] tcg: move locking for tb_invalidate_phys_page_range up Alex Bennée
2016-09-27 15:56   ` Paolo Bonzini
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 20/28] cpus: tweak sleeping and safe_work rules for MTTCG Alex Bennée
2016-09-07  4:22   ` Richard Henderson
2016-09-07 10:05   ` Paolo Bonzini
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 21/28] tcg: enable tb_lock() for SoftMMU Alex Bennée
2016-09-07  4:26   ` Richard Henderson
2016-09-27 16:16   ` Paolo Bonzini
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 22/28] tcg: enable thread-per-vCPU Alex Bennée
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 23/28] atomic: introduce cmpxchg_bool Alex Bennée
2016-09-08  0:12   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 24/28] cputlb: add assert_cpu_is_self checks Alex Bennée
2016-09-08 17:19   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 25/28] cputlb: introduce tlb_flush_* async work Alex Bennée
2016-09-07 10:08   ` Paolo Bonzini
2016-09-08 17:23   ` Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 26/28] cputlb: tweak qemu_ram_addr_from_host_nofail reporting Alex Bennée
2016-09-08 17:24   ` Richard Henderson
2016-08-11 15:24 ` Alex Bennée [this message]
2016-09-08 17:34   ` [Qemu-devel] [RFC v4 27/28] cputlb: make tlb_reset_dirty safe for MTTCG Richard Henderson
2016-08-11 15:24 ` [Qemu-devel] [RFC v4 28/28] cputlb: make tlb_flush_by_mmuidx " Alex Bennée
2016-09-07 10:09   ` Paolo Bonzini
2016-09-08 17:54   ` Richard Henderson
2016-08-11 17:22 ` [Qemu-devel] [RFC v4 00/28] Base enabling patches " Alex Bennée
2016-08-12  8:02   ` Alex Bennée
2016-09-06  9:24 ` Alex Bennée

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=1470929064-4092-28-git-send-email-alex.bennee@linaro.org \
    --to=alex.bennee@linaro.org \
    --cc=a.rigo@virtualopensystems.com \
    --cc=bobby.prani@gmail.com \
    --cc=claudio.fontana@huawei.com \
    --cc=cota@braap.org \
    --cc=crosthwaite.peter@gmail.com \
    --cc=fred.konrad@greensocs.com \
    --cc=jan.kiszka@siemens.com \
    --cc=mark.burton@greensocs.com \
    --cc=mttcg@listserver.greensocs.com \
    --cc=nikunj@linux.vnet.ibm.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.