* [PATCH][QEMU] fix segmentation fault after hotplug pass-through device
@ 2010-04-09 9:01 Zhai, Edwin
0 siblings, 0 replies; only message in thread
From: Zhai, Edwin @ 2010-04-09 9:01 UTC (permalink / raw)
To: Ian Jackson; +Cc: Xen Developers, Zhai, Edwin
[-- Attachment #1: Type: text/plain, Size: 2708 bytes --]
This patch fixed the QEMU segmentation fault after hotplug
pass-through devices with MSI-X for many times.
There is a wrong boundary check in cpu_register_io_memory that uses
io_index rather than io_mem_nb. After many times of hotplug of MSI-X
pass-through device, io_mem_read[] got extended to overwrite mmio_cnt,
then cause QEMU segmentation fault.
This fix sync with upstream QEMU code in exec.c, and free unused
io_mem_XXX element after hot removal.
Signed-off-by: Zhai Edwin <edwin.zhai@intel.com>
Index: xen-dev/tools/ioemu-remote/hw/pt-msi.c
===================================================================
--- xen-dev.orig/tools/ioemu-remote/hw/pt-msi.c
+++ xen-dev/tools/ioemu-remote/hw/pt-msi.c
@@ -623,5 +623,11 @@ void pt_msix_delete(struct pt_dev *dev)
dev->msix->table_offset_adjust);
}
+ if (dev->msix->mmio_index > 0)
+ {
+ cpu_unregister_io_memory(dev->msix->mmio_index);
+ }
+
+
free(dev->msix);
}
Index: xen-dev/tools/ioemu-remote/i386-dm/exec-dm.c
===================================================================
--- xen-dev.orig/tools/ioemu-remote/i386-dm/exec-dm.c
+++ xen-dev/tools/ioemu-remote/i386-dm/exec-dm.c
@@ -125,7 +125,7 @@ unsigned long qemu_host_page_mask;
CPUWriteMemoryFunc *io_mem_write[IO_MEM_NB_ENTRIES][4];
CPUReadMemoryFunc *io_mem_read[IO_MEM_NB_ENTRIES][4];
void *io_mem_opaque[IO_MEM_NB_ENTRIES];
-static int io_mem_nb = 1;
+char io_mem_used[IO_MEM_NB_ENTRIES];
/* log support */
FILE *logfile;
@@ -310,6 +310,20 @@ void cpu_register_physical_memory(target
mmio[mmio_cnt++].size = size;
}
+static int get_free_io_mem_idx(void)
+{
+ int i;
+
+ /* Leave 1st element empty */
+ for (i = 1; i<IO_MEM_NB_ENTRIES; i++)
+ if (!io_mem_used[i]) {
+ io_mem_used[i] = 1;
+ return i;
+ }
+
+ return -1;
+}
+
/* mem_read and mem_write are arrays of functions containing the
function to access byte (index 0), word (index 1) and dword (index
2). All functions must be supplied. If io_index is non zero, the
@@ -324,9 +338,9 @@ int cpu_register_io_memory(int io_index,
int i;
if (io_index <= 0) {
- if (io_index >= IO_MEM_NB_ENTRIES)
- return -1;
- io_index = io_mem_nb++;
+ io_index = get_free_io_mem_idx();
+ if (io_index == -1)
+ return io_index;
} else {
if (io_index >= IO_MEM_NB_ENTRIES)
return -1;
@@ -357,6 +371,7 @@ void cpu_unregister_io_memory(int io_tab
io_mem_write[io_index][i] = NULL;
}
io_mem_opaque[io_index] = NULL;
+ io_mem_used[io_index] = 0;
}
void cpu_physical_memory_set_dirty(ram_addr_t addr)
--
best rgds,
edwin
[-- Attachment #2: qemu-segfault.patch --]
[-- Type: text/plain, Size: 2177 bytes --]
Index: xen-dev/tools/ioemu-remote/hw/pt-msi.c
===================================================================
--- xen-dev.orig/tools/ioemu-remote/hw/pt-msi.c
+++ xen-dev/tools/ioemu-remote/hw/pt-msi.c
@@ -623,5 +623,11 @@ void pt_msix_delete(struct pt_dev *dev)
dev->msix->table_offset_adjust);
}
+ if (dev->msix->mmio_index > 0)
+ {
+ cpu_unregister_io_memory(dev->msix->mmio_index);
+ }
+
+
free(dev->msix);
}
Index: xen-dev/tools/ioemu-remote/i386-dm/exec-dm.c
===================================================================
--- xen-dev.orig/tools/ioemu-remote/i386-dm/exec-dm.c
+++ xen-dev/tools/ioemu-remote/i386-dm/exec-dm.c
@@ -125,7 +125,7 @@ unsigned long qemu_host_page_mask;
CPUWriteMemoryFunc *io_mem_write[IO_MEM_NB_ENTRIES][4];
CPUReadMemoryFunc *io_mem_read[IO_MEM_NB_ENTRIES][4];
void *io_mem_opaque[IO_MEM_NB_ENTRIES];
-static int io_mem_nb = 1;
+char io_mem_used[IO_MEM_NB_ENTRIES];
/* log support */
FILE *logfile;
@@ -310,6 +310,20 @@ void cpu_register_physical_memory(target
mmio[mmio_cnt++].size = size;
}
+static int get_free_io_mem_idx(void)
+{
+ int i;
+
+ /* Leave 1st element empty */
+ for (i = 1; i<IO_MEM_NB_ENTRIES; i++)
+ if (!io_mem_used[i]) {
+ io_mem_used[i] = 1;
+ return i;
+ }
+
+ return -1;
+}
+
/* mem_read and mem_write are arrays of functions containing the
function to access byte (index 0), word (index 1) and dword (index
2). All functions must be supplied. If io_index is non zero, the
@@ -324,9 +338,9 @@ int cpu_register_io_memory(int io_index,
int i;
if (io_index <= 0) {
- if (io_index >= IO_MEM_NB_ENTRIES)
- return -1;
- io_index = io_mem_nb++;
+ io_index = get_free_io_mem_idx();
+ if (io_index == -1)
+ return io_index;
} else {
if (io_index >= IO_MEM_NB_ENTRIES)
return -1;
@@ -357,6 +371,7 @@ void cpu_unregister_io_memory(int io_tab
io_mem_write[io_index][i] = NULL;
}
io_mem_opaque[io_index] = NULL;
+ io_mem_used[io_index] = 0;
}
void cpu_physical_memory_set_dirty(ram_addr_t addr)
[-- Attachment #3: Type: text/plain, Size: 138 bytes --]
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xensource.com
http://lists.xensource.com/xen-devel
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2010-04-09 9:01 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2010-04-09 9:01 [PATCH][QEMU] fix segmentation fault after hotplug pass-through device Zhai, Edwin
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.