dump: rebase from host-private RAMBlock offsets to guest-physical addresses

RAMBlock.offset                   --> GuestPhysBlock.target_start
RAMBlock.offset + RAMBlock.length --> GuestPhysBlock.target_end
RAMBlock.length                   --> GuestPhysBlock.target_end -
                                      GuestPhysBlock.target_start

"GuestPhysBlock.host_addr" is only used when writing the dump contents.

This patch enables "crash" to work with the vmcore by rebasing the vmcore
from the left side of the following diagram to the right side:

host-private
offset
relative
to ram_addr   RAMBlock                  guest-visible paddrs
            0 +-------------------+.....+-------------------+ 0
              |         ^         |     |        ^          |
              |       640 KB      |     |      640 KB       |
              |         v         |     |        v          |
  0x0000a0000 +-------------------+.....+-------------------+ 0x0000a0000
              |         ^         |     |XXXXXXXXXXXXXXXXXXX|
              |       384 KB      |     |XXXXXXXXXXXXXXXXXXX|
              |         v         |     |XXXXXXXXXXXXXXXXXXX|
  0x000100000 +-------------------+.....+-------------------+ 0x000100000
              |         ^         |     |        ^          |
              |       3583 MB     |     |      3583 MB      |
              |         v         |     |        v          |
  0x0e0000000 +-------------------+.....+-------------------+ 0x0e0000000
              |         ^         |.    |XXXXXXXXXXXXXXXXXXX|
              | above_4g_mem_size | .   |XXXX PCI hole XXXXX|
              |         v         |  .  |XXXX          XXXXX|
     ram_size +-------------------+   . |XXXX  512 MB  XXXXX|
                                   .   .|XXXXXXXXXXXXXXXXXXX|
                                    .   +-------------------+ 0x100000000
                                     .  |         ^         |
                                      . | above_4g_mem_size |
                                       .|         v         |
                                        +-------------------+ ram_size
                                                              + 512 MB

Related RHBZ: https://bugzilla.redhat.com/show_bug.cgi?id=981582

Signed-off-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>

dump.c

@@ -70,7 +70,7 @@ typedef struct DumpState {
     hwaddr memory_offset;
     int fd;
 
-    RAMBlock *block;
+    GuestPhysBlock *next_block;
     ram_addr_t start;
     bool has_filter;
     int64_t begin;
@@ -391,14 +391,14 @@ static int write_data(DumpState *s, void *buf, int length)
 }
 
 /* write the memroy to vmcore. 1 page per I/O. */
-static int write_memory(DumpState *s, RAMBlock *block, ram_addr_t start,
+static int write_memory(DumpState *s, GuestPhysBlock *block, ram_addr_t start,
                         int64_t size)
 {
     int64_t i;
     int ret;
 
     for (i = 0; i < size / TARGET_PAGE_SIZE; i++) {
-        ret = write_data(s, block->host + start + i * TARGET_PAGE_SIZE,
+        ret = write_data(s, block->host_addr + start + i * TARGET_PAGE_SIZE,
                          TARGET_PAGE_SIZE);
         if (ret < 0) {
             return ret;
@@ -406,7 +406,7 @@ static int write_memory(DumpState *s, RAMBlock *block, ram_addr_t start,
     }
 
     if ((size % TARGET_PAGE_SIZE) != 0) {
-        ret = write_data(s, block->host + start + i * TARGET_PAGE_SIZE,
+        ret = write_data(s, block->host_addr + start + i * TARGET_PAGE_SIZE,
                          size % TARGET_PAGE_SIZE);
         if (ret < 0) {
             return ret;
@@ -423,7 +423,7 @@ static void get_offset_range(hwaddr phys_addr,
                              hwaddr *p_offset,
                              hwaddr *p_filesz)
 {
-    RAMBlock *block;
+    GuestPhysBlock *block;
     hwaddr offset = s->memory_offset;
     int64_t size_in_block, start;
 
@@ -437,35 +437,34 @@ static void get_offset_range(hwaddr phys_addr,
         }
     }
 
-    QTAILQ_FOREACH(block, &ram_list.blocks, next) {
+    QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
         if (s->has_filter) {
-            if (block->offset >= s->begin + s->length ||
-                block->offset + block->length <= s->begin) {
+            if (block->target_start >= s->begin + s->length ||
+                block->target_end <= s->begin) {
                 /* This block is out of the range */
                 continue;
             }
 
-            if (s->begin <= block->offset) {
-                start = block->offset;
+            if (s->begin <= block->target_start) {
+                start = block->target_start;
             } else {
                 start = s->begin;
             }
 
-            size_in_block = block->length - (start - block->offset);
-            if (s->begin + s->length < block->offset + block->length) {
-                size_in_block -= block->offset + block->length -
-                                 (s->begin + s->length);
+            size_in_block = block->target_end - start;
+            if (s->begin + s->length < block->target_end) {
+                size_in_block -= block->target_end - (s->begin + s->length);
             }
         } else {
-            start = block->offset;
-            size_in_block = block->length;
+            start = block->target_start;
+            size_in_block = block->target_end - block->target_start;
         }
 
         if (phys_addr >= start && phys_addr < start + size_in_block) {
             *p_offset = phys_addr - start + offset;
 
             /* The offset range mapped from the vmcore file must not spill over
-             * the RAMBlock, clamp it. The rest of the mapping will be
+             * the GuestPhysBlock, clamp it. The rest of the mapping will be
              * zero-filled in memory at load time; see
              * <http://refspecs.linuxbase.org/elf/gabi4+/ch5.pheader.html>.
              */
@@ -613,7 +612,7 @@ static int dump_completed(DumpState *s)
     return 0;
 }
 
-static int get_next_block(DumpState *s, RAMBlock *block)
+static int get_next_block(DumpState *s, GuestPhysBlock *block)
 {
     while (1) {
         block = QTAILQ_NEXT(block, next);
@@ -623,16 +622,16 @@ static int get_next_block(DumpState *s, RAMBlock *block)
         }
 
         s->start = 0;
-        s->block = block;
+        s->next_block = block;
         if (s->has_filter) {
-            if (block->offset >= s->begin + s->length ||
-                block->offset + block->length <= s->begin) {
+            if (block->target_start >= s->begin + s->length ||
+                block->target_end <= s->begin) {
                 /* This block is out of the range */
                 continue;
             }
 
-            if (s->begin > block->offset) {
-                s->start = s->begin - block->offset;
+            if (s->begin > block->target_start) {
+                s->start = s->begin - block->target_start;
             }
         }
 
@@ -643,18 +642,18 @@ static int get_next_block(DumpState *s, RAMBlock *block)
 /* write all memory to vmcore */
 static int dump_iterate(DumpState *s)
 {
-    RAMBlock *block;
+    GuestPhysBlock *block;
     int64_t size;
     int ret;
 
     while (1) {
-        block = s->block;
+        block = s->next_block;
 
-        size = block->length;
+        size = block->target_end - block->target_start;
         if (s->has_filter) {
             size -= s->start;
-            if (s->begin + s->length < block->offset + block->length) {
-                size -= block->offset + block->length - (s->begin + s->length);
+            if (s->begin + s->length < block->target_end) {
+                size -= block->target_end - (s->begin + s->length);
             }
         }
         ret = write_memory(s, block, s->start, size);
@@ -689,23 +688,23 @@ static int create_vmcore(DumpState *s)
 
 static ram_addr_t get_start_block(DumpState *s)
 {
-    RAMBlock *block;
+    GuestPhysBlock *block;
 
     if (!s->has_filter) {
-        s->block = QTAILQ_FIRST(&ram_list.blocks);
+        s->next_block = QTAILQ_FIRST(&s->guest_phys_blocks.head);
         return 0;
     }
 
-    QTAILQ_FOREACH(block, &ram_list.blocks, next) {
-        if (block->offset >= s->begin + s->length ||
-            block->offset + block->length <= s->begin) {
+    QTAILQ_FOREACH(block, &s->guest_phys_blocks.head, next) {
+        if (block->target_start >= s->begin + s->length ||
+            block->target_end <= s->begin) {
             /* This block is out of the range */
             continue;
         }
 
-        s->block = block;
-        if (s->begin > block->offset) {
-            s->start = s->begin - block->offset;
+        s->next_block = block;
+        if (s->begin > block->target_start) {
+            s->start = s->begin - block->target_start;
         } else {
             s->start = 0;
         }
@@ -758,7 +757,7 @@ static int dump_init(DumpState *s, int fd, bool paging, bool has_filter,
      * If the target architecture is not supported, cpu_get_dump_info() will
      * return -1.
      */
-    ret = cpu_get_dump_info(&s->dump_info);
+    ret = cpu_get_dump_info(&s->dump_info, &s->guest_phys_blocks);
     if (ret < 0) {
         error_set(errp, QERR_UNSUPPORTED);
         goto cleanup;
@@ -774,13 +773,13 @@ static int dump_init(DumpState *s, int fd, bool paging, bool has_filter,
     /* get memory mapping */
     memory_mapping_list_init(&s->list);
     if (paging) {
-        qemu_get_guest_memory_mapping(&s->list, &err);
+        qemu_get_guest_memory_mapping(&s->list, &s->guest_phys_blocks, &err);
         if (err != NULL) {
             error_propagate(errp, err);
             goto cleanup;
         }
     } else {
-        qemu_get_guest_simple_memory_mapping(&s->list);
+        qemu_get_guest_simple_memory_mapping(&s->list, &s->guest_phys_blocks);
     }
 
     if (s->has_filter) {

include/sysemu/dump.h

@@ -20,7 +20,9 @@ typedef struct ArchDumpInfo {
     int d_class;    /* ELFCLASS32 or ELFCLASS64 */
 } ArchDumpInfo;
 
-int cpu_get_dump_info(ArchDumpInfo *info);
+struct GuestPhysBlockList; /* memory_mapping.h */
+int cpu_get_dump_info(ArchDumpInfo *info,
+                      const struct GuestPhysBlockList *guest_phys_blocks);
 ssize_t cpu_get_note_size(int class, int machine, int nr_cpus);
 
 #endif

include/sysemu/memory_mapping.h

@@ -68,10 +68,13 @@ void guest_phys_blocks_free(GuestPhysBlockList *list);
 void guest_phys_blocks_init(GuestPhysBlockList *list);
 void guest_phys_blocks_append(GuestPhysBlockList *list);
 
-void qemu_get_guest_memory_mapping(MemoryMappingList *list, Error **errp);
+void qemu_get_guest_memory_mapping(MemoryMappingList *list,
+                                   const GuestPhysBlockList *guest_phys_blocks,
+                                   Error **errp);
 
 /* get guest's memory mapping without do paging(virtual address is 0). */
-void qemu_get_guest_simple_memory_mapping(MemoryMappingList *list);
+void qemu_get_guest_simple_memory_mapping(MemoryMappingList *list,
+                                  const GuestPhysBlockList *guest_phys_blocks);
 
 void memory_mapping_filter(MemoryMappingList *list, int64_t begin,
                            int64_t length);

memory_mapping.c

@@ -279,10 +279,12 @@ static CPUState *find_paging_enabled_cpu(CPUState *start_cpu)
     return NULL;
 }
 
-void qemu_get_guest_memory_mapping(MemoryMappingList *list, Error **errp)
+void qemu_get_guest_memory_mapping(MemoryMappingList *list,
+                                   const GuestPhysBlockList *guest_phys_blocks,
+                                   Error **errp)
 {
     CPUState *cpu, *first_paging_enabled_cpu;
-    RAMBlock *block;
+    GuestPhysBlock *block;
     ram_addr_t offset, length;
 
     first_paging_enabled_cpu = find_paging_enabled_cpu(first_cpu);
@@ -302,19 +304,21 @@ void qemu_get_guest_memory_mapping(MemoryMappingList *list, Error **errp)
      * If the guest doesn't use paging, the virtual address is equal to physical
      * address.
      */
-    QTAILQ_FOREACH(block, &ram_list.blocks, next) {
-        offset = block->offset;
-        length = block->length;
+    QTAILQ_FOREACH(block, &guest_phys_blocks->head, next) {
+        offset = block->target_start;
+        length = block->target_end - block->target_start;
         create_new_memory_mapping(list, offset, offset, length);
     }
 }
 
-void qemu_get_guest_simple_memory_mapping(MemoryMappingList *list)
+void qemu_get_guest_simple_memory_mapping(MemoryMappingList *list,
+                                   const GuestPhysBlockList *guest_phys_blocks)
 {
-    RAMBlock *block;
+    GuestPhysBlock *block;
 
-    QTAILQ_FOREACH(block, &ram_list.blocks, next) {
-        create_new_memory_mapping(list, block->offset, 0, block->length);
+    QTAILQ_FOREACH(block, &guest_phys_blocks->head, next) {
+        create_new_memory_mapping(list, block->target_start, 0,
+                                  block->target_end - block->target_start);
     }
 }
 

stubs/dump.c

@@ -16,7 +16,8 @@
 #include "qapi/qmp/qerror.h"
 #include "qmp-commands.h"
 
-int cpu_get_dump_info(ArchDumpInfo *info)
+int cpu_get_dump_info(ArchDumpInfo *info,
+                      const struct GuestPhysBlockList *guest_phys_blocks)
 {
     return -1;
 }

target-i386/arch_dump.c

@@ -15,6 +15,7 @@
 #include "exec/cpu-all.h"
 #include "sysemu/dump.h"
 #include "elf.h"
+#include "sysemu/memory_mapping.h"
 
 #ifdef TARGET_X86_64
 typedef struct {
@@ -389,10 +390,11 @@ int x86_cpu_write_elf32_qemunote(WriteCoreDumpFunction f, CPUState *cs,
     return cpu_write_qemu_note(f, &cpu->env, opaque, 0);
 }
 
-int cpu_get_dump_info(ArchDumpInfo *info)
+int cpu_get_dump_info(ArchDumpInfo *info,
+                      const GuestPhysBlockList *guest_phys_blocks)
 {
     bool lma = false;
-    RAMBlock *block;
+    GuestPhysBlock *block;
 
 #ifdef TARGET_X86_64
     X86CPU *first_x86_cpu = X86_CPU(first_cpu);
@@ -412,8 +414,8 @@ int cpu_get_dump_info(ArchDumpInfo *info)
     } else {
         info->d_class = ELFCLASS32;
 
-        QTAILQ_FOREACH(block, &ram_list.blocks, next) {
-            if (block->offset + block->length > UINT_MAX) {
+        QTAILQ_FOREACH(block, &guest_phys_blocks->head, next) {
+            if (block->target_end > UINT_MAX) {
                 /* The memory size is greater than 4G */
                 info->d_class = ELFCLASS64;
                 break;

target-s390x/arch_dump.c

@@ -176,7 +176,8 @@ int s390_cpu_write_elf64_note(WriteCoreDumpFunction f, CPUState *cs,
     return s390x_write_all_elf64_notes("CORE", f, cpu, cpuid, opaque);
 }
 
-int cpu_get_dump_info(ArchDumpInfo *info)
+int cpu_get_dump_info(ArchDumpInfo *info,
+                      const struct GuestPhysBlockList *guest_phys_blocks)
 {
     info->d_machine = EM_S390;
     info->d_endian = ELFDATA2MSB;