Build on Jeff Roberson's linker-set based dynamic per-CPU allocator

(DPCPU), as suggested by Peter Wemm, and implement a new per-virtual
network stack memory allocator.  Modify vnet to use the allocator
instead of monolithic global container structures (vinet, ...).  This
change solves many binary compatibility problems associated with
VIMAGE, and restores ELF symbols for virtualized global variables.

Each virtualized global variable exists as a "reference copy", and also
once per virtual network stack.  Virtualized global variables are
tagged at compile-time, placing the in a special linker set, which is
loaded into a contiguous region of kernel memory.  Virtualized global
variables in the base kernel are linked as normal, but those in modules
are copied and relocated to a reserved portion of the kernel's vnet
region with the help of a the kernel linker.

Virtualized global variables exist in per-vnet memory set up when the
network stack instance is created, and are initialized statically from
the reference copy.  Run-time access occurs via an accessor macro, which
converts from the current vnet and requested symbol to a per-vnet
address.  When "options VIMAGE" is not compiled into the kernel, normal
global ELF symbols will be used instead and indirection is avoided.

This change restores static initialization for network stack global
variables, restores support for non-global symbols and types, eliminates
the need for many subsystem constructors, eliminates large per-subsystem
structures that caused many binary compatibility issues both for
monitoring applications (netstat) and kernel modules, removes the
per-function INIT_VNET_*() macros throughout the stack, eliminates the
need for vnet_symmap ksym(2) munging, and eliminates duplicate
definitions of virtualized globals under VIMAGE_GLOBALS.

Bump __FreeBSD_version and update UPDATING.

Portions submitted by:  bz
Reviewed by:            bz, zec
Discussed with:         gnn, jamie, jeff, jhb, julian, sam
Suggested by:           peter
Approved by:            re (kensmith)

1 files changed, 55 insertions, 0 deletions

diff --git a/sys/kern/link_elf.c b/sys/kern/link_elf.c
index cd0f3e9..b389ace 100644
--- a/sys/kern/link_elf.c
+++ b/sys/kern/link_elf.c
@@ -49,6 +49,8 @@ __FBSDID("$FreeBSD$");
 
 #include <machine/elf.h>
 
+#include <net/vnet.h>
+
 #include <security/mac/mac_framework.h>
 
 #include <vm/vm.h>
@@ -111,6 +113,11 @@ typedef struct elf_file {
     Elf_Addr		pcpu_start;	/* Pre-relocation pcpu set start. */
     Elf_Addr		pcpu_stop;	/* Pre-relocation pcpu set stop. */
     Elf_Addr		pcpu_base;	/* Relocated pcpu set address. */
+#ifdef VIMAGE
+    Elf_Addr		vnet_start;	/* Pre-relocation vnet set start. */
+    Elf_Addr		vnet_stop;	/* Pre-relocation vnet set stop. */
+    Elf_Addr		vnet_base;	/* Relocated vnet set address. */
+#endif
 #ifdef GDB
     struct link_map	gdb;		/* hooks for gdb */
 #endif
@@ -506,6 +513,36 @@ parse_dpcpu(elf_file_t ef)
     return (0);
 }
 
+#ifdef VIMAGE
+static int
+parse_vnet(elf_file_t ef)
+{ 
+    int count;
+    int error;
+
+    ef->vnet_start = 0;
+    ef->vnet_stop = 0;
+    error = link_elf_lookup_set(&ef->lf, "vnet", (void ***)&ef->vnet_start,
+                               (void ***)&ef->vnet_stop, &count);
+    /* Error just means there is no vnet data set to relocate. */
+    if (error)
+        return (0);
+    count *= sizeof(void *);
+    /*
+     * Allocate space in the primary vnet area.  Copy in our initialization
+     * from the data section and then initialize all per-vnet storage from
+     * that.
+     */
+    ef->vnet_base = (Elf_Addr)(uintptr_t)vnet_data_alloc(count);
+    if (ef->vnet_base == (Elf_Addr)NULL)
+        return (ENOSPC);
+    memcpy((void *)ef->vnet_base, (void *)ef->vnet_start, count);
+    vnet_data_copy((void *)ef->vnet_base, count);
+
+    return (0);
+}
+#endif
+
 static int
 link_elf_link_preload(linker_class_t cls,
 		      const char* filename, linker_file_t *result)
@@ -553,6 +590,10 @@ link_elf_link_preload(linker_class_t cls,
     error = parse_dynamic(ef);
     if (error == 0)
         error = parse_dpcpu(ef);
+#ifdef VIMAGE
+    if (error == 0)
+	error = parse_vnet(ef);
+#endif
     if (error) {
 	linker_file_unload(lf, LINKER_UNLOAD_FORCE);
 	return error;
@@ -838,6 +879,11 @@ link_elf_load_file(linker_class_t cls, const char* filename,
     error = parse_dpcpu(ef);
     if (error)
         goto out;
+#ifdef VIMAGE
+    error = parse_vnet(ef);
+    if (error)
+        goto out;
+#endif
     link_elf_reloc_local(lf);
 
     VOP_UNLOCK(nd.ni_vp, 0);
@@ -942,6 +988,10 @@ elf_relocaddr(linker_file_t lf, Elf_Addr x)
     ef = (elf_file_t)lf;
     if (x >= ef->pcpu_start && x < ef->pcpu_stop)
 	return ((x - ef->pcpu_start) + ef->pcpu_base);
+#ifdef VIMAGE
+    if (x >= ef->vnet_start && x < ef->vnet_stop)
+	return ((x - ef->vnet_start) + ef->vnet_base);
+#endif
     return (x);
 }
 
@@ -954,6 +1004,11 @@ link_elf_unload_file(linker_file_t file)
     if (ef->pcpu_base) {
         dpcpu_free((void *)ef->pcpu_base, ef->pcpu_stop - ef->pcpu_start);
     }
+#ifdef VIMAGE
+    if (ef->vnet_base) {
+        vnet_data_free((void *)ef->vnet_base, ef->vnet_stop - ef->vnet_start);
+    }
+#endif
 #ifdef GDB
     if (ef->gdb.l_ld) {
 	GDB_STATE(RT_DELETE);