[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Xen-devel] [RFC][PATCH 4/5] tools:firmware:hvmloader: reserve RMRR mappings in e820
On 2014/8/7 20:03, Andrew Cooper wrote:
On 07/08/14 12:02, Tiejun Chen wrote:
We need to reserve all RMRR mappings in e820 to avoid any
potential guest memory conflict.
Signed-off-by: Tiejun Chen <tiejun.chen@xxxxxxxxx>
---
tools/firmware/hvmloader/e820.c | 14 ++++++++++++++
tools/firmware/hvmloader/e820.h | 6 ++++++
tools/firmware/hvmloader/util.c | 13 +++++++++++++
tools/firmware/hvmloader/util.h | 1 +
4 files changed, 34 insertions(+)
diff --git a/tools/firmware/hvmloader/e820.c b/tools/firmware/hvmloader/e820.c
index 2e05e93..8cf8c75 100644
--- a/tools/firmware/hvmloader/e820.c
+++ b/tools/firmware/hvmloader/e820.c
@@ -74,6 +74,8 @@ int build_e820_table(struct e820entry *e820,
unsigned int bios_image_base)
{
unsigned int nr = 0;
+ struct e820map *e820_rmrr_map;
+ unsigned int i = 0;
if ( !lowmem_reserved_base )
lowmem_reserved_base = 0xA0000;
@@ -124,6 +126,18 @@ int build_e820_table(struct e820entry *e820,
e820[nr].type = E820_RAM;
nr++;
+ /* We'd better reserve RMRR mapping for each VM to avoid potential
+ * memory conflict.
+ */
+ e820_rmrr_map = get_rmrr_map_info();
+ for ( i = 0; i <= e820_rmrr_map->nr_map; i++ )
+ {
+ e820[nr].addr = e820_rmrr_map->map[i].addr;
+ e820[nr].size = e820_rmrr_map->map[i].size + 1;
+ e820[nr].type = E820_RESERVED;
+ nr++;
+ }
+
/*
* Explicitly reserve space for special pages.
* This space starts at RESERVED_MEMBASE an extends to cover various
diff --git a/tools/firmware/hvmloader/e820.h b/tools/firmware/hvmloader/e820.h
index b2ead7f..ae4cc55 100644
--- a/tools/firmware/hvmloader/e820.h
+++ b/tools/firmware/hvmloader/e820.h
@@ -15,6 +15,12 @@ struct e820entry {
uint32_t type;
} __attribute__((packed));
+#define E820MAX 128
+
+struct e820map {
+ int nr_map;
This value should be unsigned.
I'm not sure if we need to change this since here I just copy this from
the xen/include/asm-x86/e820.h file,
struct e820map {
int nr_map;
struct e820entry map[E820MAX];
};
+ struct e820entry map[E820MAX];
+};
#endif /* __HVMLOADER_E820_H__ */
/*
diff --git a/tools/firmware/hvmloader/util.c b/tools/firmware/hvmloader/util.c
index 80d822f..270700f 100644
--- a/tools/firmware/hvmloader/util.c
+++ b/tools/firmware/hvmloader/util.c
@@ -766,6 +766,19 @@ struct shared_info *get_shared_info(void)
return shared_info;
}
+struct e820map *get_rmrr_map_info(void)
+{
+ static struct e820map *e820_map = NULL;
+
+ if ( e820_map != NULL )
+ return e820_map;
+
+ if ( hypercall_memory_op(XENMEM_RMRR_memory_map, e820_map) != 0 )
+ BUG();
This instructs Xen to clobber the memory starting at 0, and works
because HVMLoader is in protected, non-paging mode at this point. I
don't think this is what you meant to do, and will repeatedly make the
Sorry I can't understand this explicitly. Here I just want a way to get
RMRR mapping info under hvmloader circumstance.
Could you elaborate what you mean? Or show me a proper way I should do
as you expect.
Thanks
Tiejun
hypercall rather than caching the result.
+
+ return e820_map;
+}
+
uint16_t get_cpu_mhz(void)
{
struct shared_info *shared_info = get_shared_info();
diff --git a/tools/firmware/hvmloader/util.h b/tools/firmware/hvmloader/util.h
index a70e4aa..26bfb8c 100644
--- a/tools/firmware/hvmloader/util.h
+++ b/tools/firmware/hvmloader/util.h
@@ -236,6 +236,7 @@ unsigned long create_pir_tables(void);
void smp_initialise(void);
#include "e820.h"
+struct e820map *get_rmrr_map_info(void);
int build_e820_table(struct e820entry *e820,
unsigned int lowmem_reserved_base,
unsigned int bios_image_base);
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel
|
