All of lore.kernel.org
 help / color / mirror / Atom feed
From: Liu Ping Fan <pingfank@linux.vnet.ibm.com>
To: qemu-devel@nongnu.org
Cc: Stefan Hajnoczi <stefanha@gmail.com>,
	Marcelo Tosatti <mtosatti@redhat.com>,
	Avi Kivity <avi@redhat.com>,
	Anthony Liguori <anthony@codemonkey.ws>,
	Jan Kiszka <jan.kiszka@siemens.com>,
	Paolo Bonzini <pbonzini@redhat.com>
Subject: [Qemu-devel] [patch v4 07/16] memory: make mmio dispatch able to be out of biglock
Date: Mon, 22 Oct 2012 17:23:50 +0800	[thread overview]
Message-ID: <1350897839-29593-8-git-send-email-pingfank@linux.vnet.ibm.com> (raw)
In-Reply-To: <1350897839-29593-1-git-send-email-pingfank@linux.vnet.ibm.com>

Without biglock, we try to protect the mr by increase refcnt.
If we can inc refcnt, go backward and resort to biglock.

Another point is memory radix-tree can be flushed by another
thread, so we should get the copy of terminal mr to survive
from such issue.

Signed-off-by: Liu Ping Fan <pingfank@linux.vnet.ibm.com>
---
 exec.c |  125 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++----
 1 files changed, 117 insertions(+), 8 deletions(-)

diff --git a/exec.c b/exec.c
index 5834766..91b859b 100644
--- a/exec.c
+++ b/exec.c
@@ -200,6 +200,8 @@ struct PhysPageEntry {
     uint16_t ptr : 15;
 };
 
+static QemuMutex mem_map_lock;
+
 /* Simple allocator for PhysPageEntry nodes */
 static PhysPageEntry (*phys_map_nodes)[L2_SIZE];
 static unsigned phys_map_nodes_nb, phys_map_nodes_nb_alloc;
@@ -212,6 +214,8 @@ static PhysPageEntry phys_map = { .ptr = PHYS_MAP_NODE_NIL, .is_leaf = 0 };
 
 static void io_mem_init(void);
 static void memory_map_init(void);
+static int phys_page_lookup(target_phys_addr_t addr, MemoryRegionSection *mrs);
+
 
 static MemoryRegion io_mem_watch;
 #endif
@@ -2245,6 +2249,7 @@ static void register_subpage(MemoryRegionSection *section)
     subpage_t *subpage;
     target_phys_addr_t base = section->offset_within_address_space
         & TARGET_PAGE_MASK;
+    /* Already under the protection of mem_map_lock */
     MemoryRegionSection *existing = phys_page_find(base >> TARGET_PAGE_BITS);
     MemoryRegionSection subsection = {
         .offset_within_address_space = base,
@@ -3165,6 +3170,8 @@ static void io_mem_init(void)
 
 static void core_begin(MemoryListener *listener)
 {
+    /* protect the updating process of mrs in memory core agaist readers */
+    qemu_mutex_lock(&mem_map_lock);
     destroy_all_mappings();
     phys_sections_clear();
     phys_map.ptr = PHYS_MAP_NODE_NIL;
@@ -3184,17 +3191,32 @@ static void core_commit(MemoryListener *listener)
     for(env = first_cpu; env != NULL; env = env->next_cpu) {
         tlb_flush(env, 1);
     }
+    qemu_mutex_unlock(&mem_map_lock);
 }
 
 static void core_region_add(MemoryListener *listener,
                             MemoryRegionSection *section)
 {
+    MemoryRegion *mr = section->mr;
+
+    if (mr->ops) {
+        if (mr->ops->ref) {
+            mr->ops->ref(mr);
+        }
+    }
     cpu_register_physical_memory_log(section, section->readonly);
 }
 
 static void core_region_del(MemoryListener *listener,
                             MemoryRegionSection *section)
 {
+    MemoryRegion *mr = section->mr;
+
+    if (mr->ops) {
+        if (mr->ops->unref) {
+            mr->ops->unref(mr);
+        }
+    }
 }
 
 static void core_region_nop(MemoryListener *listener,
@@ -3348,6 +3370,8 @@ static void memory_map_init(void)
     memory_region_init(system_io, "io", 65536);
     set_system_io_map(system_io);
 
+    qemu_mutex_init(&mem_map_lock);
+
     memory_listener_register(&core_memory_listener, system_memory);
     memory_listener_register(&io_memory_listener, system_io);
 }
@@ -3406,6 +3430,58 @@ int cpu_memory_rw_debug(CPUArchState *env, target_ulong addr,
 }
 
 #else
+
+static MemoryRegionSection *subpage_get_terminal(subpage_t *mmio,
+    target_phys_addr_t addr)
+{
+    MemoryRegionSection *section;
+    unsigned int idx = SUBPAGE_IDX(addr);
+
+    section = &phys_sections[mmio->sub_section[idx]];
+    return section;
+}
+
+static int memory_region_section_ref(MemoryRegionSection *mrs)
+{
+    MemoryRegion *mr;
+    int ret = 0;
+
+    mr = mrs->mr;
+    if (mr->ops) {
+        if (mr->ops->ref) {
+            ret = mr->ops->ref(mr);
+        }
+    }
+    return ret;
+}
+
+static void memory_region_section_unref(MemoryRegionSection *mrs)
+{
+    MemoryRegion *mr;
+
+    mr = mrs->mr;
+    if (mr->ops) {
+        if (mr->ops->unref) {
+            mr->ops->unref(mr);
+        }
+    }
+}
+
+static int phys_page_lookup(target_phys_addr_t addr, MemoryRegionSection *mrs)
+{
+    MemoryRegionSection *section;
+    int ret;
+
+    section = phys_page_find(addr >> TARGET_PAGE_BITS);
+    if (section->mr->subpage) {
+        section = subpage_get_terminal(section->mr->opaque, addr);
+    }
+    *mrs = *section;
+    ret = memory_region_section_ref(mrs);
+
+    return ret;
+}
+
 void cpu_physical_memory_rw(target_phys_addr_t addr, uint8_t *buf,
                             int len, int is_write)
 {
@@ -3413,14 +3489,28 @@ void cpu_physical_memory_rw(target_phys_addr_t addr, uint8_t *buf,
     uint8_t *ptr;
     uint32_t val;
     target_phys_addr_t page;
-    MemoryRegionSection *section;
+    MemoryRegionSection *section, obj_mrs;
+    int safe_ref;
 
     while (len > 0) {
         page = addr & TARGET_PAGE_MASK;
         l = (page + TARGET_PAGE_SIZE) - addr;
         if (l > len)
             l = len;
-        section = phys_page_find(page >> TARGET_PAGE_BITS);
+        qemu_mutex_lock(&mem_map_lock);
+        safe_ref = phys_page_lookup(page, &obj_mrs);
+        qemu_mutex_unlock(&mem_map_lock);
+        if (safe_ref == 0) {
+            qemu_mutex_lock_iothread();
+            qemu_mutex_lock(&mem_map_lock);
+            /* At the 2nd try, mem map can change, so need to judge it again */
+            safe_ref = phys_page_lookup(page, &obj_mrs);
+            qemu_mutex_unlock(&mem_map_lock);
+            if (safe_ref > 0) {
+                qemu_mutex_unlock_iothread();
+            }
+        }
+        section = &obj_mrs;
 
         if (is_write) {
             if (!memory_region_is_ram(section->mr)) {
@@ -3491,10 +3581,16 @@ void cpu_physical_memory_rw(target_phys_addr_t addr, uint8_t *buf,
                 qemu_put_ram_ptr(ptr);
             }
         }
+
+        memory_region_section_unref(&obj_mrs);
         len -= l;
         buf += l;
         addr += l;
+        if (safe_ref == 0) {
+            qemu_mutex_unlock_iothread();
+        }
     }
+
 }
 
 /* used for ROM loading : can write in RAM and ROM */
@@ -3504,14 +3600,18 @@ void cpu_physical_memory_write_rom(target_phys_addr_t addr,
     int l;
     uint8_t *ptr;
     target_phys_addr_t page;
-    MemoryRegionSection *section;
+    MemoryRegionSection *section, mr_obj;
 
     while (len > 0) {
         page = addr & TARGET_PAGE_MASK;
         l = (page + TARGET_PAGE_SIZE) - addr;
         if (l > len)
             l = len;
-        section = phys_page_find(page >> TARGET_PAGE_BITS);
+
+        qemu_mutex_lock(&mem_map_lock);
+        phys_page_lookup(page, &mr_obj);
+        qemu_mutex_unlock(&mem_map_lock);
+        section = &mr_obj;
 
         if (!(memory_region_is_ram(section->mr) ||
               memory_region_is_romd(section->mr))) {
@@ -3528,6 +3628,7 @@ void cpu_physical_memory_write_rom(target_phys_addr_t addr,
         len -= l;
         buf += l;
         addr += l;
+        memory_region_section_unref(&mr_obj);
     }
 }
 
@@ -3592,7 +3693,7 @@ void *cpu_physical_memory_map(target_phys_addr_t addr,
     target_phys_addr_t todo = 0;
     int l;
     target_phys_addr_t page;
-    MemoryRegionSection *section;
+    MemoryRegionSection *section, mr_obj;
     ram_addr_t raddr = RAM_ADDR_MAX;
     ram_addr_t rlen;
     void *ret;
@@ -3602,7 +3703,10 @@ void *cpu_physical_memory_map(target_phys_addr_t addr,
         l = (page + TARGET_PAGE_SIZE) - addr;
         if (l > len)
             l = len;
-        section = phys_page_find(page >> TARGET_PAGE_BITS);
+        qemu_mutex_lock(&mem_map_lock);
+        phys_page_lookup(page, &mr_obj);
+        qemu_mutex_unlock(&mem_map_lock);
+        section = &mr_obj;
 
         if (!(memory_region_is_ram(section->mr) && !section->readonly)) {
             if (todo || bounce.buffer) {
@@ -3616,6 +3720,7 @@ void *cpu_physical_memory_map(target_phys_addr_t addr,
             }
 
             *plen = l;
+            memory_region_section_unref(&mr_obj);
             return bounce.buffer;
         }
         if (!todo) {
@@ -3630,6 +3735,7 @@ void *cpu_physical_memory_map(target_phys_addr_t addr,
     rlen = todo;
     ret = qemu_ram_ptr_length(raddr, &rlen);
     *plen = rlen;
+    memory_region_section_unref(&mr_obj);
     return ret;
 }
 
@@ -4239,9 +4345,12 @@ bool virtio_is_big_endian(void)
 #ifndef CONFIG_USER_ONLY
 bool cpu_physical_memory_is_io(target_phys_addr_t phys_addr)
 {
-    MemoryRegionSection *section;
+    MemoryRegionSection *section, mr_obj;
 
-    section = phys_page_find(phys_addr >> TARGET_PAGE_BITS);
+    qemu_mutex_lock(&mem_map_lock);
+    phys_page_lookup(phys_addr, &mr_obj);
+    qemu_mutex_unlock(&mem_map_lock);
+    section = &mr_obj;
 
     return !(memory_region_is_ram(section->mr) ||
              memory_region_is_romd(section->mr));
-- 
1.7.4.4

  parent reply	other threads:[~2012-10-22  9:24 UTC|newest]

Thread overview: 102+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-10-22  9:23 [Qemu-devel] [patch v4 00/16] push mmio dispatch out of big lock Liu Ping Fan
2012-10-22  9:23 ` [Qemu-devel] [patch v4 01/16] atomic: introduce atomic operations Liu Ping Fan
2012-10-22  9:23 ` [Qemu-devel] [patch v4 02/16] qom: apply atomic on object's refcount Liu Ping Fan
2012-10-22  9:23 ` [Qemu-devel] [patch v4 03/16] hotplug: introduce qdev_unplug_complete() to remove device from views Liu Ping Fan
2012-10-22  9:23 ` [Qemu-devel] [patch v4 04/16] pci: remove pci device from mem view when unplug Liu Ping Fan
2012-10-22  9:23 ` [Qemu-devel] [patch v4 05/16] memory: introduce ref, unref interface for MemoryRegionOps Liu Ping Fan
2012-10-22  9:38   ` Avi Kivity
2012-10-23 11:51     ` Paolo Bonzini
2012-10-23 11:55       ` Avi Kivity
2012-10-23 11:57         ` Paolo Bonzini
2012-10-23 12:02           ` Avi Kivity
2012-10-23 12:06             ` Paolo Bonzini
2012-10-23 12:15               ` Avi Kivity
2012-10-23 12:32                 ` Paolo Bonzini
2012-10-23 14:49                   ` Avi Kivity
2012-10-23 15:26                     ` Paolo Bonzini
2012-10-23 16:09                       ` Avi Kivity
2012-10-24  7:29                         ` Paolo Bonzini
2012-10-25 16:28                           ` Avi Kivity
2012-10-26 15:05                             ` Paolo Bonzini
2012-10-23 12:04         ` Jan Kiszka
2012-10-23 12:12           ` Paolo Bonzini
2012-10-23 12:16             ` Jan Kiszka
2012-10-23 12:28               ` Avi Kivity
2012-10-23 12:40                 ` Jan Kiszka
2012-10-23 14:37                   ` Avi Kivity
2012-10-22  9:23 ` [Qemu-devel] [patch v4 06/16] memory: document ref, unref interface Liu Ping Fan
2012-10-22  9:23 ` Liu Ping Fan [this message]
2012-10-23 12:12   ` [Qemu-devel] [patch v4 07/16] memory: make mmio dispatch able to be out of biglock Jan Kiszka
2012-10-23 12:36     ` Avi Kivity
2012-10-24  6:31       ` liu ping fan
2012-10-24  6:56         ` liu ping fan
2012-10-25  8:57           ` Avi Kivity
2012-10-22  9:23 ` [Qemu-devel] [patch v4 08/16] QemuThread: make QemuThread as tls to store extra info Liu Ping Fan
2012-10-22  9:30   ` Jan Kiszka
2012-10-22 17:13     ` Peter Maydell
2012-10-23  5:58       ` liu ping fan
2012-10-23 11:48       ` Paolo Bonzini
2012-10-23 11:50         ` Peter Maydell
2012-10-23 11:51           ` Jan Kiszka
2012-10-23 12:00           ` Paolo Bonzini
2012-10-23 12:27             ` Peter Maydell
2012-11-18 10:02             ` Brad Smith
2012-11-18 16:14               ` Paolo Bonzini
2012-11-18 16:15                 ` Paolo Bonzini
2012-10-22  9:23 ` [Qemu-devel] [patch v4 09/16] memory: introduce mmio request pending to anti nested DMA Liu Ping Fan
2012-10-22 10:28   ` Avi Kivity
2012-10-23 12:38   ` Gleb Natapov
2012-10-24  6:31     ` liu ping fan
2012-10-22  9:23 ` [Qemu-devel] [patch v4 10/16] memory: introduce lock ops for MemoryRegionOps Liu Ping Fan
2012-10-22 10:30   ` Avi Kivity
2012-10-23  5:53     ` liu ping fan
2012-10-23  8:53       ` Jan Kiszka
2012-10-22  9:23 ` [Qemu-devel] [patch v4 11/16] vcpu: push mmio dispatcher out of big lock Liu Ping Fan
2012-10-22 10:31   ` Avi Kivity
2012-10-22 10:36     ` Jan Kiszka
2012-10-22  9:23 ` [Qemu-devel] [patch v4 12/16] e1000: apply fine lock on e1000 Liu Ping Fan
2012-10-22 10:37   ` Avi Kivity
2012-10-23  9:04   ` Jan Kiszka
2012-10-24  6:31     ` liu ping fan
2012-10-24  7:17       ` Jan Kiszka
2012-10-25  9:01         ` Avi Kivity
2012-10-25  9:31           ` Jan Kiszka
2012-10-25 16:21             ` Avi Kivity
2012-10-25 16:39               ` Jan Kiszka
2012-10-25 17:02                 ` Avi Kivity
2012-10-25 18:48                   ` Jan Kiszka
2012-10-29  5:24                     ` liu ping fan
2012-10-24  7:29     ` liu ping fan
2012-10-25 13:34       ` Jan Kiszka
2012-10-25 16:23         ` Avi Kivity
2012-10-25 16:41           ` Jan Kiszka
2012-10-25 17:03             ` Avi Kivity
2012-10-29  5:24         ` liu ping fan
2012-10-31  7:03           ` Jan Kiszka
2012-10-22  9:23 ` [Qemu-devel] [patch v4 13/16] e1000: add busy flag to anti broken device state Liu Ping Fan
2012-10-22 10:40   ` Avi Kivity
2012-10-23  5:52     ` liu ping fan
2012-10-23  9:06       ` Avi Kivity
2012-10-23  9:07       ` Jan Kiszka
2012-10-23  9:32         ` liu ping fan
2012-10-23  9:37           ` Avi Kivity
2012-10-24  6:36             ` liu ping fan
2012-10-25  8:55               ` Avi Kivity
2012-10-25  9:00             ` Peter Maydell
2012-10-25  9:04               ` Avi Kivity
2012-10-26  3:05                 ` liu ping fan
2012-10-26  3:08                   ` liu ping fan
2012-10-26 10:25                     ` Jan Kiszka
2012-10-29  5:24                       ` liu ping fan
2012-10-29  7:50                         ` Peter Maydell
2012-10-22  9:23 ` [Qemu-devel] [patch v4 14/16] qdev: introduce stopping state Liu Ping Fan
2012-10-22  9:23 ` [Qemu-devel] [patch v4 15/16] e1000: introduce unmap() to fix unplug issue Liu Ping Fan
2012-10-22  9:23 ` [Qemu-devel] [patch v4 16/16] e1000: implement MemoryRegionOps's ref&lock interface Liu Ping Fan
2012-10-25 14:04 ` [Qemu-devel] [patch v4 00/16] push mmio dispatch out of big lock Peter Maydell
2012-10-25 16:44   ` Jan Kiszka
2012-10-25 17:07   ` Avi Kivity
2012-10-25 17:13     ` Peter Maydell
2012-10-25 18:13       ` Marcelo Tosatti
2012-10-25 19:00         ` Jan Kiszka
2012-10-25 19:06           ` Peter Maydell
2012-10-29 15:24       ` Avi Kivity

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=1350897839-29593-8-git-send-email-pingfank@linux.vnet.ibm.com \
    --to=pingfank@linux.vnet.ibm.com \
    --cc=anthony@codemonkey.ws \
    --cc=avi@redhat.com \
    --cc=jan.kiszka@siemens.com \
    --cc=mtosatti@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@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.