-/* Simple program to layout "physical" memory for new lguest guest.
- * Linked high to avoid likely physical memory. */
+/*P:100 This is the Launcher code, a simple program which lays out the
+ * "physical" memory for the new Guest by mapping the kernel image and the
+ * virtual devices, then reads repeatedly from /dev/lguest to run the Guest.
+:*/
#define _LARGEFILE64_SOURCE
#define _GNU_SOURCE
#include <stdio.h>
#include <stdlib.h>
#include <elf.h>
#include <sys/mman.h>
+#include <sys/param.h>
#include <sys/types.h>
#include <sys/stat.h>
#include <sys/wait.h>
#include <termios.h>
#include <getopt.h>
#include <zlib.h>
+#include <assert.h>
+#include <sched.h>
+/*L:110 We can ignore the 30 include files we need for this program, but I do
+ * want to draw attention to the use of kernel-style types.
+ *
+ * As Linus said, "C is a Spartan language, and so should your naming be." I
+ * like these abbreviations and the header we need uses them, so we define them
+ * here.
+ */
typedef unsigned long long u64;
typedef uint32_t u32;
typedef uint16_t u16;
typedef uint8_t u8;
-#include "../../include/linux/lguest_launcher.h"
-#include "../../include/asm-i386/e820.h"
+#include "linux/lguest_launcher.h"
+#include "linux/pci_ids.h"
+#include "linux/virtio_config.h"
+#include "linux/virtio_net.h"
+#include "linux/virtio_blk.h"
+#include "linux/virtio_console.h"
+#include "linux/virtio_ring.h"
+#include "asm-x86/e820.h"
+/*:*/
#define PAGE_PRESENT 0x7 /* Present, RW, Execute */
#define NET_PEERNUM 1
#ifndef SIOCBRADDIF
#define SIOCBRADDIF 0x89a2 /* add interface to bridge */
#endif
+/* We can have up to 256 pages for devices. */
+#define DEVICE_PAGES 256
+/* This fits nicely in a single 4096-byte page. */
+#define VIRTQUEUE_NUM 127
+/*L:120 verbose is both a global flag and a macro. The C preprocessor allows
+ * this, and although I wouldn't recommend it, it works quite nicely here. */
static bool verbose;
#define verbose(args...) \
do { if (verbose) printf(args); } while(0)
+/*:*/
+
+/* The pipe to send commands to the waker process */
static int waker_fd;
+/* The pointer to the start of guest memory. */
+static void *guest_base;
+/* The maximum guest physical address allowed, and maximum possible. */
+static unsigned long guest_limit, guest_max;
+/* This is our list of devices. */
struct device_list
{
+ /* Summary information about the devices in our list: ready to pass to
+ * select() to ask which need servicing.*/
fd_set infds;
int max_infd;
+ /* Counter to assign interrupt numbers. */
+ unsigned int next_irq;
+
+ /* Counter to print out convenient device numbers. */
+ unsigned int device_num;
+
+ /* The descriptor page for the devices. */
+ u8 *descpage;
+
+ /* The tail of the last descriptor. */
+ unsigned int desc_used;
+
+ /* A single linked list of devices. */
struct device *dev;
+ /* ... And an end pointer so we can easily append new devices */
struct device **lastdev;
};
+/* The list of Guest devices, based on command line arguments. */
+static struct device_list devices;
+
+/* The device structure describes a single device. */
struct device
{
+ /* The linked-list pointer. */
struct device *next;
+
+ /* The this device's descriptor, as mapped into the Guest. */
struct lguest_device_desc *desc;
- void *mem;
- /* Watch this fd if handle_input non-NULL. */
+ /* The name of this device, for --verbose. */
+ const char *name;
+
+ /* If handle_input is set, it wants to be called when this file
+ * descriptor is ready. */
int fd;
bool (*handle_input)(int fd, struct device *me);
- /* Watch DMA to this key if handle_input non-NULL. */
- unsigned long watch_key;
- u32 (*handle_output)(int fd, const struct iovec *iov,
- unsigned int num, struct device *me);
+ /* Any queues attached to this device */
+ struct virtqueue *vq;
/* Device-specific data. */
void *priv;
};
+/* The virtqueue structure describes a queue attached to a device. */
+struct virtqueue
+{
+ struct virtqueue *next;
+
+ /* Which device owns me. */
+ struct device *dev;
+
+ /* The configuration for this queue. */
+ struct lguest_vqconfig config;
+
+ /* The actual ring of buffers. */
+ struct vring vring;
+
+ /* Last available index we saw. */
+ u16 last_avail_idx;
+
+ /* The routine to call when the Guest pings us. */
+ void (*handle_output)(int fd, struct virtqueue *me);
+};
+
+/* Since guest is UP and we don't run at the same time, we don't need barriers.
+ * But I include them in the code in case others copy it. */
+#define wmb()
+
+/* Convert an iovec element to the given type.
+ *
+ * This is a fairly ugly trick: we need to know the size of the type and
+ * alignment requirement to check the pointer is kosher. It's also nice to
+ * have the name of the type in case we report failure.
+ *
+ * Typing those three things all the time is cumbersome and error prone, so we
+ * have a macro which sets them all up and passes to the real function. */
+#define convert(iov, type) \
+ ((type *)_convert((iov), sizeof(type), __alignof__(type), #type))
+
+static void *_convert(struct iovec *iov, size_t size, size_t align,
+ const char *name)
+{
+ if (iov->iov_len != size)
+ errx(1, "Bad iovec size %zu for %s", iov->iov_len, name);
+ if ((unsigned long)iov->iov_base % align != 0)
+ errx(1, "Bad alignment %p for %s", iov->iov_base, name);
+ return iov->iov_base;
+}
+
+/* The virtio configuration space is defined to be little-endian. x86 is
+ * little-endian too, but it's nice to be explicit so we have these helpers. */
+#define cpu_to_le16(v16) (v16)
+#define cpu_to_le32(v32) (v32)
+#define cpu_to_le64(v64) (v64)
+#define le16_to_cpu(v16) (v16)
+#define le32_to_cpu(v32) (v32)
+#define le64_to_cpu(v32) (v64)
+
+/*L:100 The Launcher code itself takes us out into userspace, that scary place
+ * where pointers run wild and free! Unfortunately, like most userspace
+ * programs, it's quite boring (which is why everyone likes to hack on the
+ * kernel!). Perhaps if you make up an Lguest Drinking Game at this point, it
+ * will get you through this section. Or, maybe not.
+ *
+ * The Launcher sets up a big chunk of memory to be the Guest's "physical"
+ * memory and stores it in "guest_base". In other words, Guest physical ==
+ * Launcher virtual with an offset.
+ *
+ * This can be tough to get your head around, but usually it just means that we
+ * use these trivial conversion functions when the Guest gives us it's
+ * "physical" addresses: */
+static void *from_guest_phys(unsigned long addr)
+{
+ return guest_base + addr;
+}
+
+static unsigned long to_guest_phys(const void *addr)
+{
+ return (addr - guest_base);
+}
+
+/*L:130
+ * Loading the Kernel.
+ *
+ * We start with couple of simple helper routines. open_or_die() avoids
+ * error-checking code cluttering the callers: */
static int open_or_die(const char *name, int flags)
{
int fd = open(name, flags);
return fd;
}
-static void *map_zeroed_pages(unsigned long addr, unsigned int num)
+/* map_zeroed_pages() takes a number of pages. */
+static void *map_zeroed_pages(unsigned int num)
{
- static int fd = -1;
+ int fd = open_or_die("/dev/zero", O_RDONLY);
+ void *addr;
- if (fd == -1)
- fd = open_or_die("/dev/zero", O_RDONLY);
+ /* We use a private mapping (ie. if we write to the page, it will be
+ * copied). */
+ addr = mmap(NULL, getpagesize() * num,
+ PROT_READ|PROT_WRITE|PROT_EXEC, MAP_PRIVATE, fd, 0);
+ if (addr == MAP_FAILED)
+ err(1, "Mmaping %u pages of /dev/zero", num);
- if (mmap((void *)addr, getpagesize() * num,
- PROT_READ|PROT_WRITE|PROT_EXEC, MAP_FIXED|MAP_PRIVATE, fd, 0)
- != (void *)addr)
- err(1, "Mmaping %u pages of /dev/zero @%p", num, (void *)addr);
- return (void *)addr;
+ return addr;
}
-/* Find magic string marking entry point, return entry point. */
-static unsigned long entry_point(void *start, void *end,
- unsigned long page_offset)
+/* Get some more pages for a device. */
+static void *get_pages(unsigned int num)
{
- void *p;
+ void *addr = from_guest_phys(guest_limit);
+
+ guest_limit += num * getpagesize();
+ if (guest_limit > guest_max)
+ errx(1, "Not enough memory for devices");
+ return addr;
+}
+/* To find out where to start we look for the magic Guest string, which marks
+ * the code we see in lguest_asm.S. This is a hack which we are currently
+ * plotting to replace with the normal Linux entry point. */
+static unsigned long entry_point(const void *start, const void *end)
+{
+ const void *p;
+
+ /* The scan gives us the physical starting address. We boot with
+ * pagetables set up with virtual and physical the same, so that's
+ * OK. */
for (p = start; p < end; p++)
if (memcmp(p, "GenuineLguest", strlen("GenuineLguest")) == 0)
- return (long)p + strlen("GenuineLguest") + page_offset;
+ return to_guest_phys(p + strlen("GenuineLguest"));
- err(1, "Is this image a genuine lguest?");
+ errx(1, "Is this image a genuine lguest?");
}
-/* Returns the entry point */
-static unsigned long map_elf(int elf_fd, const Elf32_Ehdr *ehdr,
- unsigned long *page_offset)
+/* This routine is used to load the kernel or initrd. It tries mmap, but if
+ * that fails (Plan 9's kernel file isn't nicely aligned on page boundaries),
+ * it falls back to reading the memory in. */
+static void map_at(int fd, void *addr, unsigned long offset, unsigned long len)
{
- void *addr;
+ ssize_t r;
+
+ /* We map writable even though for some segments are marked read-only.
+ * The kernel really wants to be writable: it patches its own
+ * instructions.
+ *
+ * MAP_PRIVATE means that the page won't be copied until a write is
+ * done to it. This allows us to share untouched memory between
+ * Guests. */
+ if (mmap(addr, len, PROT_READ|PROT_WRITE|PROT_EXEC,
+ MAP_FIXED|MAP_PRIVATE, fd, offset) != MAP_FAILED)
+ return;
+
+ /* pread does a seek and a read in one shot: saves a few lines. */
+ r = pread(fd, addr, len, offset);
+ if (r != len)
+ err(1, "Reading offset %lu len %lu gave %zi", offset, len, r);
+}
+
+/* This routine takes an open vmlinux image, which is in ELF, and maps it into
+ * the Guest memory. ELF = Embedded Linking Format, which is the format used
+ * by all modern binaries on Linux including the kernel.
+ *
+ * The ELF headers give *two* addresses: a physical address, and a virtual
+ * address. We use the physical address; the Guest will map itself to the
+ * virtual address.
+ *
+ * We return the starting address. */
+static unsigned long map_elf(int elf_fd, const Elf32_Ehdr *ehdr)
+{
+ void *start = (void *)-1, *end = NULL;
Elf32_Phdr phdr[ehdr->e_phnum];
unsigned int i;
- unsigned long start = -1UL, end = 0;
- /* Sanity checks. */
+ /* Sanity checks on the main ELF header: an x86 executable with a
+ * reasonable number of correctly-sized program headers. */
if (ehdr->e_type != ET_EXEC
|| ehdr->e_machine != EM_386
|| ehdr->e_phentsize != sizeof(Elf32_Phdr)
|| ehdr->e_phnum < 1 || ehdr->e_phnum > 65536U/sizeof(Elf32_Phdr))
errx(1, "Malformed elf header");
+ /* An ELF executable contains an ELF header and a number of "program"
+ * headers which indicate which parts ("segments") of the program to
+ * load where. */
+
+ /* We read in all the program headers at once: */
if (lseek(elf_fd, ehdr->e_phoff, SEEK_SET) < 0)
err(1, "Seeking to program headers");
if (read(elf_fd, phdr, sizeof(phdr)) != sizeof(phdr))
err(1, "Reading program headers");
- *page_offset = 0;
- /* We map the loadable segments at virtual addresses corresponding
- * to their physical addresses (our virtual == guest physical). */
+ /* Try all the headers: there are usually only three. A read-only one,
+ * a read-write one, and a "note" section which isn't loadable. */
for (i = 0; i < ehdr->e_phnum; i++) {
+ /* If this isn't a loadable segment, we ignore it */
if (phdr[i].p_type != PT_LOAD)
continue;
verbose("Section %i: size %i addr %p\n",
i, phdr[i].p_memsz, (void *)phdr[i].p_paddr);
- /* We expect linear address space. */
- if (!*page_offset)
- *page_offset = phdr[i].p_vaddr - phdr[i].p_paddr;
- else if (*page_offset != phdr[i].p_vaddr - phdr[i].p_paddr)
- errx(1, "Page offset of section %i different", i);
-
- if (phdr[i].p_paddr < start)
- start = phdr[i].p_paddr;
- if (phdr[i].p_paddr + phdr[i].p_filesz > end)
- end = phdr[i].p_paddr + phdr[i].p_filesz;
-
- /* We map everything private, writable. */
- addr = mmap((void *)phdr[i].p_paddr,
- phdr[i].p_filesz,
- PROT_READ|PROT_WRITE|PROT_EXEC,
- MAP_FIXED|MAP_PRIVATE,
- elf_fd, phdr[i].p_offset);
- if (addr != (void *)phdr[i].p_paddr)
- err(1, "Mmaping vmlinux seg %i gave %p not %p",
- i, addr, (void *)phdr[i].p_paddr);
- }
-
- return entry_point((void *)start, (void *)end, *page_offset);
-}
-
-/* This is amazingly reliable. */
-static unsigned long intuit_page_offset(unsigned char *img, unsigned long len)
-{
- unsigned int i, possibilities[256] = { 0 };
+ /* We track the first and last address we mapped, so we can
+ * tell entry_point() where to scan. */
+ if (from_guest_phys(phdr[i].p_paddr) < start)
+ start = from_guest_phys(phdr[i].p_paddr);
+ if (from_guest_phys(phdr[i].p_paddr) + phdr[i].p_filesz > end)
+ end=from_guest_phys(phdr[i].p_paddr)+phdr[i].p_filesz;
- for (i = 0; i + 4 < len; i++) {
- /* mov 0xXXXXXXXX,%eax */
- if (img[i] == 0xA1 && ++possibilities[img[i+4]] > 3)
- return (unsigned long)img[i+4] << 24;
+ /* We map this section of the file at its physical address. */
+ map_at(elf_fd, from_guest_phys(phdr[i].p_paddr),
+ phdr[i].p_offset, phdr[i].p_filesz);
}
- errx(1, "could not determine page offset");
+
+ return entry_point(start, end);
}
-static unsigned long unpack_bzimage(int fd, unsigned long *page_offset)
+/*L:160 Unfortunately the entire ELF image isn't compressed: the segments
+ * which need loading are extracted and compressed raw. This denies us the
+ * information we need to make a fully-general loader. */
+static unsigned long unpack_bzimage(int fd)
{
gzFile f;
int ret, len = 0;
- void *img = (void *)0x100000;
-
+ /* A bzImage always gets loaded at physical address 1M. This is
+ * actually configurable as CONFIG_PHYSICAL_START, but as the comment
+ * there says, "Don't change this unless you know what you are doing".
+ * Indeed. */
+ void *img = from_guest_phys(0x100000);
+
+ /* gzdopen takes our file descriptor (carefully placed at the start of
+ * the GZIP header we found) and returns a gzFile. */
f = gzdopen(fd, "rb");
+ /* We read it into memory in 64k chunks until we hit the end. */
while ((ret = gzread(f, img + len, 65536)) > 0)
len += ret;
if (ret < 0)
err(1, "reading image from bzImage");
verbose("Unpacked size %i addr %p\n", len, img);
- *page_offset = intuit_page_offset(img, len);
- return entry_point(img, img + len, *page_offset);
+ return entry_point(img, img + len);
}
-static unsigned long load_bzimage(int fd, unsigned long *page_offset)
+/*L:150 A bzImage, unlike an ELF file, is not meant to be loaded. You're
+ * supposed to jump into it and it will unpack itself. We can't do that
+ * because the Guest can't run the unpacking code, and adding features to
+ * lguest kills puppies, so we don't want to.
+ *
+ * The bzImage is formed by putting the decompressing code in front of the
+ * compressed kernel code. So we can simple scan through it looking for the
+ * first "gzip" header, and start decompressing from there. */
+static unsigned long load_bzimage(int fd)
{
unsigned char c;
int state = 0;
- /* Ugly brute force search for gzip header. */
+ /* GZIP header is 0x1F 0x8B <method> <flags>... <compressed-by>. */
while (read(fd, &c, 1) == 1) {
switch (state) {
case 0:
state++;
break;
case 9:
+ /* Seek back to the start of the gzip header. */
lseek(fd, -10, SEEK_CUR);
- if (c != 0x03) /* Compressed under UNIX. */
+ /* One final check: "compressed under UNIX". */
+ if (c != 0x03)
state = -1;
else
- return unpack_bzimage(fd, page_offset);
+ return unpack_bzimage(fd);
}
}
errx(1, "Could not find kernel in bzImage");
}
-static unsigned long load_kernel(int fd, unsigned long *page_offset)
+/*L:140 Loading the kernel is easy when it's a "vmlinux", but most kernels
+ * come wrapped up in the self-decompressing "bzImage" format. With some funky
+ * coding, we can load those, too. */
+static unsigned long load_kernel(int fd)
{
Elf32_Ehdr hdr;
+ /* Read in the first few bytes. */
if (read(fd, &hdr, sizeof(hdr)) != sizeof(hdr))
err(1, "Reading kernel");
+ /* If it's an ELF file, it starts with "\177ELF" */
if (memcmp(hdr.e_ident, ELFMAG, SELFMAG) == 0)
- return map_elf(fd, &hdr, page_offset);
+ return map_elf(fd, &hdr);
- return load_bzimage(fd, page_offset);
+ /* Otherwise we assume it's a bzImage, and try to unpack it */
+ return load_bzimage(fd);
}
+/* This is a trivial little helper to align pages. Andi Kleen hated it because
+ * it calls getpagesize() twice: "it's dumb code."
+ *
+ * Kernel guys get really het up about optimization, even when it's not
+ * necessary. I leave this code as a reaction against that. */
static inline unsigned long page_align(unsigned long addr)
{
+ /* Add upwards and truncate downwards. */
return ((addr + getpagesize()-1) & ~(getpagesize()-1));
}
-/* initrd gets loaded at top of memory: return length. */
+/*L:180 An "initial ram disk" is a disk image loaded into memory along with
+ * the kernel which the kernel can use to boot from without needing any
+ * drivers. Most distributions now use this as standard: the initrd contains
+ * the code to load the appropriate driver modules for the current machine.
+ *
+ * Importantly, James Morris works for RedHat, and Fedora uses initrds for its
+ * kernels. He sent me this (and tells me when I break it). */
static unsigned long load_initrd(const char *name, unsigned long mem)
{
int ifd;
struct stat st;
unsigned long len;
- void *iaddr;
ifd = open_or_die(name, O_RDONLY);
+ /* fstat() is needed to get the file size. */
if (fstat(ifd, &st) < 0)
err(1, "fstat() on initrd '%s'", name);
+ /* We map the initrd at the top of memory, but mmap wants it to be
+ * page-aligned, so we round the size up for that. */
len = page_align(st.st_size);
- iaddr = mmap((void *)mem - len, st.st_size,
- PROT_READ|PROT_EXEC|PROT_WRITE,
- MAP_FIXED|MAP_PRIVATE, ifd, 0);
- if (iaddr != (void *)mem - len)
- err(1, "Mmaping initrd '%s' returned %p not %p",
- name, iaddr, (void *)mem - len);
+ map_at(ifd, from_guest_phys(mem - len), 0, st.st_size);
+ /* Once a file is mapped, you can close the file descriptor. It's a
+ * little odd, but quite useful. */
close(ifd);
- verbose("mapped initrd %s size=%lu @ %p\n", name, st.st_size, iaddr);
+ verbose("mapped initrd %s size=%lu @ %p\n", name, len, (void*)mem-len);
+
+ /* We return the initrd size. */
return len;
}
+/* Once we know how much memory we have, we can construct simple linear page
+ * tables which set virtual == physical which will get the Guest far enough
+ * into the boot to create its own.
+ *
+ * We lay them out of the way, just below the initrd (which is why we need to
+ * know its size). */
static unsigned long setup_pagetables(unsigned long mem,
- unsigned long initrd_size,
- unsigned long page_offset)
+ unsigned long initrd_size)
{
- u32 *pgdir, *linear;
+ unsigned long *pgdir, *linear;
unsigned int mapped_pages, i, linear_pages;
- unsigned int ptes_per_page = getpagesize()/sizeof(u32);
+ unsigned int ptes_per_page = getpagesize()/sizeof(void *);
- /* If we can map all of memory above page_offset, we do so. */
- if (mem <= -page_offset)
- mapped_pages = mem/getpagesize();
- else
- mapped_pages = -page_offset/getpagesize();
+ mapped_pages = mem/getpagesize();
- /* Each linear PTE page can map ptes_per_page pages. */
+ /* Each PTE page can map ptes_per_page pages: how many do we need? */
linear_pages = (mapped_pages + ptes_per_page-1)/ptes_per_page;
- /* We lay out top-level then linear mapping immediately below initrd */
- pgdir = (void *)mem - initrd_size - getpagesize();
+ /* We put the toplevel page directory page at the top of memory. */
+ pgdir = from_guest_phys(mem) - initrd_size - getpagesize();
+
+ /* Now we use the next linear_pages pages as pte pages */
linear = (void *)pgdir - linear_pages*getpagesize();
+ /* Linear mapping is easy: put every page's address into the mapping in
+ * order. PAGE_PRESENT contains the flags Present, Writable and
+ * Executable. */
for (i = 0; i < mapped_pages; i++)
linear[i] = ((i * getpagesize()) | PAGE_PRESENT);
- /* Now set up pgd so that this memory is at page_offset */
+ /* The top level points to the linear page table pages above. */
for (i = 0; i < mapped_pages; i += ptes_per_page) {
- pgdir[(i + page_offset/getpagesize())/ptes_per_page]
- = (((u32)linear + i*sizeof(u32)) | PAGE_PRESENT);
+ pgdir[i/ptes_per_page]
+ = ((to_guest_phys(linear) + i*sizeof(void *))
+ | PAGE_PRESENT);
}
- verbose("Linear mapping of %u pages in %u pte pages at %p\n",
- mapped_pages, linear_pages, linear);
+ verbose("Linear mapping of %u pages in %u pte pages at %#lx\n",
+ mapped_pages, linear_pages, to_guest_phys(linear));
- return (unsigned long)pgdir;
+ /* We return the top level (guest-physical) address: the kernel needs
+ * to know where it is. */
+ return to_guest_phys(pgdir);
}
+/* Simple routine to roll all the commandline arguments together with spaces
+ * between them. */
static void concat(char *dst, char *args[])
{
unsigned int i, len = 0;
dst[len] = '\0';
}
-static int tell_kernel(u32 pgdir, u32 start, u32 page_offset)
+/* This is where we actually tell the kernel to initialize the Guest. We saw
+ * the arguments it expects when we looked at initialize() in lguest_user.c:
+ * the base of guest "physical" memory, the top physical page to allow, the
+ * top level pagetable and the entry point for the Guest. */
+static int tell_kernel(unsigned long pgdir, unsigned long start)
{
- u32 args[] = { LHREQ_INITIALIZE,
- LGUEST_GUEST_TOP/getpagesize(), /* Just below us */
- pgdir, start, page_offset };
+ unsigned long args[] = { LHREQ_INITIALIZE,
+ (unsigned long)guest_base,
+ guest_limit / getpagesize(), pgdir, start };
int fd;
+ verbose("Guest: %p - %p (%#lx)\n",
+ guest_base, guest_base + guest_limit, guest_limit);
fd = open_or_die("/dev/lguest", O_RDWR);
if (write(fd, args, sizeof(args)) < 0)
err(1, "Writing to /dev/lguest");
+
+ /* We return the /dev/lguest file descriptor to control this Guest */
return fd;
}
+/*:*/
-static void set_fd(int fd, struct device_list *devices)
+static void add_device_fd(int fd)
{
- FD_SET(fd, &devices->infds);
- if (fd > devices->max_infd)
- devices->max_infd = fd;
+ FD_SET(fd, &devices.infds);
+ if (fd > devices.max_infd)
+ devices.max_infd = fd;
}
-/* When input arrives, we tell the kernel to kick lguest out with -EAGAIN. */
-static void wake_parent(int pipefd, int lguest_fd, struct device_list *devices)
+/*L:200
+ * The Waker.
+ *
+ * With a console and network devices, we can have lots of input which we need
+ * to process. We could try to tell the kernel what file descriptors to watch,
+ * but handing a file descriptor mask through to the kernel is fairly icky.
+ *
+ * Instead, we fork off a process which watches the file descriptors and writes
+ * the LHREQ_BREAK command to the /dev/lguest filedescriptor to tell the Host
+ * loop to stop running the Guest. This causes it to return from the
+ * /dev/lguest read with -EAGAIN, where it will write to /dev/lguest to reset
+ * the LHREQ_BREAK and wake us up again.
+ *
+ * This, of course, is merely a different *kind* of icky.
+ */
+static void wake_parent(int pipefd, int lguest_fd)
{
- set_fd(pipefd, devices);
+ /* Add the pipe from the Launcher to the fdset in the device_list, so
+ * we watch it, too. */
+ add_device_fd(pipefd);
for (;;) {
- fd_set rfds = devices->infds;
- u32 args[] = { LHREQ_BREAK, 1 };
+ fd_set rfds = devices.infds;
+ unsigned long args[] = { LHREQ_BREAK, 1 };
- select(devices->max_infd+1, &rfds, NULL, NULL, NULL);
+ /* Wait until input is ready from one of the devices. */
+ select(devices.max_infd+1, &rfds, NULL, NULL, NULL);
+ /* Is it a message from the Launcher? */
if (FD_ISSET(pipefd, &rfds)) {
- int ignorefd;
- if (read(pipefd, &ignorefd, sizeof(ignorefd)) == 0)
+ int fd;
+ /* If read() returns 0, it means the Launcher has
+ * exited. We silently follow. */
+ if (read(pipefd, &fd, sizeof(fd)) == 0)
exit(0);
- FD_CLR(ignorefd, &devices->infds);
- } else
+ /* Otherwise it's telling us to change what file
+ * descriptors we're to listen to. */
+ if (fd >= 0)
+ FD_SET(fd, &devices.infds);
+ else
+ FD_CLR(-fd - 1, &devices.infds);
+ } else /* Send LHREQ_BREAK command. */
write(lguest_fd, args, sizeof(args));
}
}
-static int setup_waker(int lguest_fd, struct device_list *device_list)
+/* This routine just sets up a pipe to the Waker process. */
+static int setup_waker(int lguest_fd)
{
int pipefd[2], child;
+ /* We create a pipe to talk to the waker, and also so it knows when the
+ * Launcher dies (and closes pipe). */
pipe(pipefd);
child = fork();
if (child == -1)
err(1, "forking");
if (child == 0) {
+ /* Close the "writing" end of our copy of the pipe */
close(pipefd[1]);
- wake_parent(pipefd[0], lguest_fd, device_list);
+ wake_parent(pipefd[0], lguest_fd);
}
+ /* Close the reading end of our copy of the pipe. */
close(pipefd[0]);
+ /* Here is the fd used to talk to the waker. */
return pipefd[1];
}
+/*L:210
+ * Device Handling.
+ *
+ * When the Guest sends DMA to us, it sends us an array of addresses and sizes.
+ * We need to make sure it's not trying to reach into the Launcher itself, so
+ * we have a convenient routine which check it and exits with an error message
+ * if something funny is going on:
+ */
static void *_check_pointer(unsigned long addr, unsigned int size,
unsigned int line)
{
- if (addr >= LGUEST_GUEST_TOP || addr + size >= LGUEST_GUEST_TOP)
- errx(1, "%s:%i: Invalid address %li", __FILE__, line, addr);
- return (void *)addr;
+ /* We have to separately check addr and addr+size, because size could
+ * be huge and addr + size might wrap around. */
+ if (addr >= guest_limit || addr + size >= guest_limit)
+ errx(1, "%s:%i: Invalid address %#lx", __FILE__, line, addr);
+ /* We return a pointer for the caller's convenience, now we know it's
+ * safe to use. */
+ return from_guest_phys(addr);
}
+/* A macro which transparently hands the line number to the real function. */
#define check_pointer(addr,size) _check_pointer(addr, size, __LINE__)
-/* Returns pointer to dma->used_len */
-static u32 *dma2iov(unsigned long dma, struct iovec iov[], unsigned *num)
+/* This function returns the next descriptor in the chain, or vq->vring.num. */
+static unsigned next_desc(struct virtqueue *vq, unsigned int i)
{
- unsigned int i;
- struct lguest_dma *udma;
+ unsigned int next;
- udma = check_pointer(dma, sizeof(*udma));
- for (i = 0; i < LGUEST_MAX_DMA_SECTIONS; i++) {
- if (!udma->len[i])
- break;
+ /* If this descriptor says it doesn't chain, we're done. */
+ if (!(vq->vring.desc[i].flags & VRING_DESC_F_NEXT))
+ return vq->vring.num;
- iov[i].iov_base = check_pointer(udma->addr[i], udma->len[i]);
- iov[i].iov_len = udma->len[i];
- }
- *num = i;
- return &udma->used_len;
+ /* Check they're not leading us off end of descriptors. */
+ next = vq->vring.desc[i].next;
+ /* Make sure compiler knows to grab that: we don't want it changing! */
+ wmb();
+
+ if (next >= vq->vring.num)
+ errx(1, "Desc next is %u", next);
+
+ return next;
}
-static u32 *get_dma_buffer(int fd, void *key,
- struct iovec iov[], unsigned int *num, u32 *irq)
+/* This looks in the virtqueue and for the first available buffer, and converts
+ * it to an iovec for convenient access. Since descriptors consist of some
+ * number of output then some number of input descriptors, it's actually two
+ * iovecs, but we pack them into one and note how many of each there were.
+ *
+ * This function returns the descriptor number found, or vq->vring.num (which
+ * is never a valid descriptor number) if none was found. */
+static unsigned get_vq_desc(struct virtqueue *vq,
+ struct iovec iov[],
+ unsigned int *out_num, unsigned int *in_num)
{
- u32 buf[] = { LHREQ_GETDMA, (u32)key };
- unsigned long udma;
- u32 *res;
+ unsigned int i, head;
+
+ /* Check it isn't doing very strange things with descriptor numbers. */
+ if ((u16)(vq->vring.avail->idx - vq->last_avail_idx) > vq->vring.num)
+ errx(1, "Guest moved used index from %u to %u",
+ vq->last_avail_idx, vq->vring.avail->idx);
+
+ /* If there's nothing new since last we looked, return invalid. */
+ if (vq->vring.avail->idx == vq->last_avail_idx)
+ return vq->vring.num;
+
+ /* Grab the next descriptor number they're advertising, and increment
+ * the index we've seen. */
+ head = vq->vring.avail->ring[vq->last_avail_idx++ % vq->vring.num];
+
+ /* If their number is silly, that's a fatal mistake. */
+ if (head >= vq->vring.num)
+ errx(1, "Guest says index %u is available", head);
+
+ /* When we start there are none of either input nor output. */
+ *out_num = *in_num = 0;
+
+ i = head;
+ do {
+ /* Grab the first descriptor, and check it's OK. */
+ iov[*out_num + *in_num].iov_len = vq->vring.desc[i].len;
+ iov[*out_num + *in_num].iov_base
+ = check_pointer(vq->vring.desc[i].addr,
+ vq->vring.desc[i].len);
+ /* If this is an input descriptor, increment that count. */
+ if (vq->vring.desc[i].flags & VRING_DESC_F_WRITE)
+ (*in_num)++;
+ else {
+ /* If it's an output descriptor, they're all supposed
+ * to come before any input descriptors. */
+ if (*in_num)
+ errx(1, "Descriptor has out after in");
+ (*out_num)++;
+ }
+
+ /* If we've got too many, that implies a descriptor loop. */
+ if (*out_num + *in_num > vq->vring.num)
+ errx(1, "Looped descriptor");
+ } while ((i = next_desc(vq, i)) != vq->vring.num);
- udma = write(fd, buf, sizeof(buf));
- if (udma == (unsigned long)-1)
- return NULL;
+ return head;
+}
- /* Kernel stashes irq in ->used_len. */
- res = dma2iov(udma, iov, num);
- *irq = *res;
- return res;
+/* Once we've used one of their buffers, we tell them about it. We'll then
+ * want to send them an interrupt, using trigger_irq(). */
+static void add_used(struct virtqueue *vq, unsigned int head, int len)
+{
+ struct vring_used_elem *used;
+
+ /* Get a pointer to the next entry in the used ring. */
+ used = &vq->vring.used->ring[vq->vring.used->idx % vq->vring.num];
+ used->id = head;
+ used->len = len;
+ /* Make sure buffer is written before we update index. */
+ wmb();
+ vq->vring.used->idx++;
}
-static void trigger_irq(int fd, u32 irq)
+/* This actually sends the interrupt for this virtqueue */
+static void trigger_irq(int fd, struct virtqueue *vq)
{
- u32 buf[] = { LHREQ_IRQ, irq };
+ unsigned long buf[] = { LHREQ_IRQ, vq->config.irq };
+
+ if (vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)
+ return;
+
+ /* Send the Guest an interrupt tell them we used something up. */
if (write(fd, buf, sizeof(buf)) != 0)
- err(1, "Triggering irq %i", irq);
+ err(1, "Triggering irq %i", vq->config.irq);
}
-static void discard_iovec(struct iovec *iov, unsigned int *num)
+/* And here's the combo meal deal. Supersize me! */
+static void add_used_and_trigger(int fd, struct virtqueue *vq,
+ unsigned int head, int len)
{
- static char discard_buf[1024];
- *num = 1;
- iov->iov_base = discard_buf;
- iov->iov_len = sizeof(discard_buf);
+ add_used(vq, head, len);
+ trigger_irq(fd, vq);
}
+/* Here is the input terminal setting we save, and the routine to restore them
+ * on exit so the user can see what they type next. */
static struct termios orig_term;
static void restore_term(void)
{
tcsetattr(STDIN_FILENO, TCSANOW, &orig_term);
}
+/* We associate some data with the console for our exit hack. */
struct console_abort
{
+ /* How many times have they hit ^C? */
int count;
+ /* When did they start? */
struct timeval start;
};
-/* We DMA input to buffer bound at start of console page. */
+/* This is the routine which handles console input (ie. stdin). */
static bool handle_console_input(int fd, struct device *dev)
{
- u32 irq = 0, *lenp;
int len;
- unsigned int num;
- struct iovec iov[LGUEST_MAX_DMA_SECTIONS];
+ unsigned int head, in_num, out_num;
+ struct iovec iov[dev->vq->vring.num];
struct console_abort *abort = dev->priv;
- lenp = get_dma_buffer(fd, dev->mem, iov, &num, &irq);
- if (!lenp) {
- warn("console: no dma buffer!");
- discard_iovec(iov, &num);
- }
+ /* First we need a console buffer from the Guests's input virtqueue. */
+ head = get_vq_desc(dev->vq, iov, &out_num, &in_num);
+
+ /* If they're not ready for input, stop listening to this file
+ * descriptor. We'll start again once they add an input buffer. */
+ if (head == dev->vq->vring.num)
+ return false;
- len = readv(dev->fd, iov, num);
+ if (out_num)
+ errx(1, "Output buffers in console in queue?");
+
+ /* This is why we convert to iovecs: the readv() call uses them, and so
+ * it reads straight into the Guest's buffer. */
+ len = readv(dev->fd, iov, in_num);
if (len <= 0) {
+ /* This implies that the console is closed, is /dev/null, or
+ * something went terribly wrong. */
warnx("Failed to get console input, ignoring console.");
- len = 0;
+ /* Put the input terminal back. */
+ restore_term();
+ /* Remove callback from input vq, so it doesn't restart us. */
+ dev->vq->handle_output = NULL;
+ /* Stop listening to this fd: don't call us again. */
+ return false;
}
- if (lenp) {
- *lenp = len;
- trigger_irq(fd, irq);
- }
+ /* Tell the Guest about the new input. */
+ add_used_and_trigger(fd, dev->vq, head, len);
- /* Three ^C within one second? Exit. */
+ /* Three ^C within one second? Exit.
+ *
+ * This is such a hack, but works surprisingly well. Each ^C has to be
+ * in a buffer by itself, so they can't be too fast. But we check that
+ * we get three within about a second, so they can't be too slow. */
if (len == 1 && ((char *)iov[0].iov_base)[0] == 3) {
if (!abort->count++)
gettimeofday(&abort->start, NULL);
struct timeval now;
gettimeofday(&now, NULL);
if (now.tv_sec <= abort->start.tv_sec+1) {
- /* Make sure waker is not blocked in BREAK */
- u32 args[] = { LHREQ_BREAK, 0 };
+ unsigned long args[] = { LHREQ_BREAK, 0 };
+ /* Close the fd so Waker will know it has to
+ * exit. */
close(waker_fd);
+ /* Just in case waker is blocked in BREAK, send
+ * unbreak now. */
write(fd, args, sizeof(args));
exit(2);
}
abort->count = 0;
}
} else
+ /* Any other key resets the abort counter. */
abort->count = 0;
- if (!len) {
- restore_term();
- return false;
- }
+ /* Everything went OK! */
return true;
}
-static u32 handle_console_output(int fd, const struct iovec *iov,
- unsigned num, struct device*dev)
-{
- return writev(STDOUT_FILENO, iov, num);
-}
-
-static u32 handle_tun_output(int fd, const struct iovec *iov,
- unsigned num, struct device *dev)
+/* Handling output for console is simple: we just get all the output buffers
+ * and write them to stdout. */
+static void handle_console_output(int fd, struct virtqueue *vq)
{
- /* Now we've seen output, we should warn if we can't get buffers. */
- *(bool *)dev->priv = true;
- return writev(dev->fd, iov, num);
+ unsigned int head, out, in;
+ int len;
+ struct iovec iov[vq->vring.num];
+
+ /* Keep getting output buffers from the Guest until we run out. */
+ while ((head = get_vq_desc(vq, iov, &out, &in)) != vq->vring.num) {
+ if (in)
+ errx(1, "Input buffers in output queue?");
+ len = writev(STDOUT_FILENO, iov, out);
+ add_used_and_trigger(fd, vq, head, len);
+ }
}
-static unsigned long peer_offset(unsigned int peernum)
+/* Handling output for network is also simple: we get all the output buffers
+ * and write them (ignoring the first element) to this device's file descriptor
+ * (stdout). */
+static void handle_net_output(int fd, struct virtqueue *vq)
{
- return 4 * peernum;
+ unsigned int head, out, in;
+ int len;
+ struct iovec iov[vq->vring.num];
+
+ /* Keep getting output buffers from the Guest until we run out. */
+ while ((head = get_vq_desc(vq, iov, &out, &in)) != vq->vring.num) {
+ if (in)
+ errx(1, "Input buffers in output queue?");
+ /* Check header, but otherwise ignore it (we said we supported
+ * no features). */
+ (void)convert(&iov[0], struct virtio_net_hdr);
+ len = writev(vq->dev->fd, iov+1, out-1);
+ add_used_and_trigger(fd, vq, head, len);
+ }
}
+/* This is where we handle a packet coming in from the tun device to our
+ * Guest. */
static bool handle_tun_input(int fd, struct device *dev)
{
- u32 irq = 0, *lenp;
+ unsigned int head, in_num, out_num;
int len;
- unsigned num;
- struct iovec iov[LGUEST_MAX_DMA_SECTIONS];
-
- lenp = get_dma_buffer(fd, dev->mem+peer_offset(NET_PEERNUM), iov, &num,
- &irq);
- if (!lenp) {
- if (*(bool *)dev->priv)
+ struct iovec iov[dev->vq->vring.num];
+ struct virtio_net_hdr *hdr;
+
+ /* First we need a network buffer from the Guests's recv virtqueue. */
+ head = get_vq_desc(dev->vq, iov, &out_num, &in_num);
+ if (head == dev->vq->vring.num) {
+ /* Now, it's expected that if we try to send a packet too
+ * early, the Guest won't be ready yet. Wait until the device
+ * status says it's ready. */
+ /* FIXME: Actually want DRIVER_ACTIVE here. */
+ if (dev->desc->status & VIRTIO_CONFIG_S_DRIVER_OK)
warn("network: no dma buffer!");
- discard_iovec(iov, &num);
- }
+ /* We'll turn this back on if input buffers are registered. */
+ return false;
+ } else if (out_num)
+ errx(1, "Output buffers in network recv queue?");
- len = readv(dev->fd, iov, num);
+ /* First element is the header: we set it to 0 (no features). */
+ hdr = convert(&iov[0], struct virtio_net_hdr);
+ hdr->flags = 0;
+ hdr->gso_type = VIRTIO_NET_HDR_GSO_NONE;
+
+ /* Read the packet from the device directly into the Guest's buffer. */
+ len = readv(dev->fd, iov+1, in_num-1);
if (len <= 0)
err(1, "reading network");
- if (lenp) {
- *lenp = len;
- trigger_irq(fd, irq);
- }
+
+ /* Tell the Guest about the new packet. */
+ add_used_and_trigger(fd, dev->vq, head, sizeof(*hdr) + len);
+
verbose("tun input packet len %i [%02x %02x] (%s)\n", len,
- ((u8 *)iov[0].iov_base)[0], ((u8 *)iov[0].iov_base)[1],
- lenp ? "sent" : "discarded");
+ ((u8 *)iov[1].iov_base)[0], ((u8 *)iov[1].iov_base)[1],
+ head != dev->vq->vring.num ? "sent" : "discarded");
+
+ /* All good. */
return true;
}
-static u32 handle_block_output(int fd, const struct iovec *iov,
- unsigned num, struct device *dev)
+/* This callback ensures we try again, in case we stopped console or net
+ * delivery because Guest didn't have any buffers. */
+static void enable_fd(int fd, struct virtqueue *vq)
{
- struct lguest_block_page *p = dev->mem;
- u32 irq, *lenp;
- unsigned int len, reply_num;
- struct iovec reply[LGUEST_MAX_DMA_SECTIONS];
- off64_t device_len, off = (off64_t)p->sector * 512;
-
- device_len = *(off64_t *)dev->priv;
-
- if (off >= device_len)
- err(1, "Bad offset %llu vs %llu", off, device_len);
- if (lseek64(dev->fd, off, SEEK_SET) != off)
- err(1, "Bad seek to sector %i", p->sector);
-
- verbose("Block: %s at offset %llu\n", p->type ? "WRITE" : "READ", off);
-
- lenp = get_dma_buffer(fd, dev->mem, reply, &reply_num, &irq);
- if (!lenp)
- err(1, "Block request didn't give us a dma buffer");
-
- if (p->type) {
- len = writev(dev->fd, iov, num);
- if (off + len > device_len) {
- ftruncate(dev->fd, device_len);
- errx(1, "Write past end %llu+%u", off, len);
- }
- *lenp = 0;
- } else {
- len = readv(dev->fd, reply, reply_num);
- *lenp = len;
- }
-
- p->result = 1 + (p->bytes != len);
- trigger_irq(fd, irq);
- return 0;
+ add_device_fd(vq->dev->fd);
+ /* Tell waker to listen to it again */
+ write(waker_fd, &vq->dev->fd, sizeof(vq->dev->fd));
}
-static void handle_output(int fd, unsigned long dma, unsigned long key,
- struct device_list *devices)
+/* This is the generic routine we call when the Guest uses LHCALL_NOTIFY. */
+static void handle_output(int fd, unsigned long addr)
{
struct device *i;
- u32 *lenp;
- struct iovec iov[LGUEST_MAX_DMA_SECTIONS];
- unsigned num = 0;
-
- lenp = dma2iov(dma, iov, &num);
- for (i = devices->dev; i; i = i->next) {
- if (i->handle_output && key == i->watch_key) {
- *lenp = i->handle_output(fd, iov, num, i);
- return;
+ struct virtqueue *vq;
+
+ /* Check each virtqueue. */
+ for (i = devices.dev; i; i = i->next) {
+ for (vq = i->vq; vq; vq = vq->next) {
+ if (vq->config.pfn == addr/getpagesize()
+ && vq->handle_output) {
+ verbose("Output to %s\n", vq->dev->name);
+ vq->handle_output(fd, vq);
+ return;
+ }
}
}
- warnx("Pending dma %p, key %p", (void *)dma, (void *)key);
+
+ /* Early console write is done using notify on a nul-terminated string
+ * in Guest memory. */
+ if (addr >= guest_limit)
+ errx(1, "Bad NOTIFY %#lx", addr);
+
+ write(STDOUT_FILENO, from_guest_phys(addr),
+ strnlen(from_guest_phys(addr), guest_limit - addr));
}
-static void handle_input(int fd, struct device_list *devices)
+/* This is called when the waker wakes us up: check for incoming file
+ * descriptors. */
+static void handle_input(int fd)
{
+ /* select() wants a zeroed timeval to mean "don't wait". */
struct timeval poll = { .tv_sec = 0, .tv_usec = 0 };
for (;;) {
struct device *i;
- fd_set fds = devices->infds;
+ fd_set fds = devices.infds;
- if (select(devices->max_infd+1, &fds, NULL, NULL, &poll) == 0)
+ /* If nothing is ready, we're done. */
+ if (select(devices.max_infd+1, &fds, NULL, NULL, &poll) == 0)
break;
- for (i = devices->dev; i; i = i->next) {
+ /* Otherwise, call the device(s) which have readable
+ * file descriptors and a method of handling them. */
+ for (i = devices.dev; i; i = i->next) {
if (i->handle_input && FD_ISSET(i->fd, &fds)) {
- if (!i->handle_input(fd, i)) {
- FD_CLR(i->fd, &devices->infds);
- /* Tell waker to ignore it too... */
- write(waker_fd, &i->fd, sizeof(i->fd));
- }
+ int dev_fd;
+ if (i->handle_input(fd, i))
+ continue;
+
+ /* If handle_input() returns false, it means we
+ * should no longer service it. Networking and
+ * console do this when there's no input
+ * buffers to deliver into. Console also uses
+ * it when it discovers that stdin is
+ * closed. */
+ FD_CLR(i->fd, &devices.infds);
+ /* Tell waker to ignore it too, by sending a
+ * negative fd number (-1, since 0 is a valid
+ * FD number). */
+ dev_fd = -i->fd - 1;
+ write(waker_fd, &dev_fd, sizeof(dev_fd));
}
}
}
}
-static struct lguest_device_desc *new_dev_desc(u16 type, u16 features,
- u16 num_pages)
+/*L:190
+ * Device Setup
+ *
+ * All devices need a descriptor so the Guest knows it exists, and a "struct
+ * device" so the Launcher can keep track of it. We have common helper
+ * routines to allocate them.
+ *
+ * This routine allocates a new "struct lguest_device_desc" from descriptor
+ * table just above the Guest's normal memory. It returns a pointer to that
+ * descriptor. */
+static struct lguest_device_desc *new_dev_desc(u16 type)
{
- static unsigned long top = LGUEST_GUEST_TOP;
- struct lguest_device_desc *desc;
+ struct lguest_device_desc *d;
- desc = malloc(sizeof(*desc));
- desc->type = type;
- desc->num_pages = num_pages;
- desc->features = features;
- desc->status = 0;
- if (num_pages) {
- top -= num_pages*getpagesize();
- map_zeroed_pages(top, num_pages);
- desc->pfn = top / getpagesize();
- } else
- desc->pfn = 0;
- return desc;
+ /* We only have one page for all the descriptors. */
+ if (devices.desc_used + sizeof(*d) > getpagesize())
+ errx(1, "Too many devices");
+
+ /* We don't need to set config_len or status: page is 0 already. */
+ d = (void *)devices.descpage + devices.desc_used;
+ d->type = type;
+ devices.desc_used += sizeof(*d);
+
+ return d;
+}
+
+/* Each device descriptor is followed by some configuration information.
+ * The first byte is a "status" byte for the Guest to report what's happening.
+ * After that are fields: u8 type, u8 len, [... len bytes...].
+ *
+ * This routine adds a new field to an existing device's descriptor. It only
+ * works for the last device, but that's OK because that's how we use it. */
+static void add_desc_field(struct device *dev, u8 type, u8 len, const void *c)
+{
+ /* This is the last descriptor, right? */
+ assert(devices.descpage + devices.desc_used
+ == (u8 *)(dev->desc + 1) + dev->desc->config_len);
+
+ /* We only have one page of device descriptions. */
+ if (devices.desc_used + 2 + len > getpagesize())
+ errx(1, "Too many devices");
+
+ /* Copy in the new config header: type then length. */
+ devices.descpage[devices.desc_used++] = type;
+ devices.descpage[devices.desc_used++] = len;
+ memcpy(devices.descpage + devices.desc_used, c, len);
+ devices.desc_used += len;
+
+ /* Update the device descriptor length: two byte head then data. */
+ dev->desc->config_len += 2 + len;
+}
+
+/* This routine adds a virtqueue to a device. We specify how many descriptors
+ * the virtqueue is to have. */
+static void add_virtqueue(struct device *dev, unsigned int num_descs,
+ void (*handle_output)(int fd, struct virtqueue *me))
+{
+ unsigned int pages;
+ struct virtqueue **i, *vq = malloc(sizeof(*vq));
+ void *p;
+
+ /* First we need some pages for this virtqueue. */
+ pages = (vring_size(num_descs) + getpagesize() - 1) / getpagesize();
+ p = get_pages(pages);
+
+ /* Initialize the configuration. */
+ vq->config.num = num_descs;
+ vq->config.irq = devices.next_irq++;
+ vq->config.pfn = to_guest_phys(p) / getpagesize();
+
+ /* Initialize the vring. */
+ vring_init(&vq->vring, num_descs, p);
+
+ /* Add the configuration information to this device's descriptor. */
+ add_desc_field(dev, VIRTIO_CONFIG_F_VIRTQUEUE,
+ sizeof(vq->config), &vq->config);
+
+ /* Add to tail of list, so dev->vq is first vq, dev->vq->next is
+ * second. */
+ for (i = &dev->vq; *i; i = &(*i)->next);
+ *i = vq;
+
+ /* Link virtqueue back to device. */
+ vq->dev = dev;
+
+ /* Set up handler. */
+ vq->handle_output = handle_output;
+ if (!handle_output)
+ vq->vring.used->flags = VRING_USED_F_NO_NOTIFY;
}
-static struct device *new_device(struct device_list *devices,
- u16 type, u16 num_pages, u16 features,
- int fd,
- bool (*handle_input)(int, struct device *),
- unsigned long watch_off,
- u32 (*handle_output)(int,
- const struct iovec *,
- unsigned,
- struct device *))
+/* This routine does all the creation and setup of a new device, including
+ * caling new_dev_desc() to allocate the descriptor and device memory. */
+static struct device *new_device(const char *name, u16 type, int fd,
+ bool (*handle_input)(int, struct device *))
{
struct device *dev = malloc(sizeof(*dev));
- /* Append to device list. */
- *devices->lastdev = dev;
+ /* Append to device list. Prepending to a single-linked list is
+ * easier, but the user expects the devices to be arranged on the bus
+ * in command-line order. The first network device on the command line
+ * is eth0, the first block device /dev/lgba, etc. */
+ *devices.lastdev = dev;
dev->next = NULL;
- devices->lastdev = &dev->next;
+ devices.lastdev = &dev->next;
+ /* Now we populate the fields one at a time. */
dev->fd = fd;
+ /* If we have an input handler for this file descriptor, then we add it
+ * to the device_list's fdset and maxfd. */
if (handle_input)
- set_fd(dev->fd, devices);
- dev->desc = new_dev_desc(type, features, num_pages);
- dev->mem = (void *)(dev->desc->pfn * getpagesize());
+ add_device_fd(dev->fd);
+ dev->desc = new_dev_desc(type);
dev->handle_input = handle_input;
- dev->watch_key = (unsigned long)dev->mem + watch_off;
- dev->handle_output = handle_output;
+ dev->name = name;
return dev;
}
-static void setup_console(struct device_list *devices)
+/* Our first setup routine is the console. It's a fairly simple device, but
+ * UNIX tty handling makes it uglier than it could be. */
+static void setup_console(void)
{
struct device *dev;
+ /* If we can save the initial standard input settings... */
if (tcgetattr(STDIN_FILENO, &orig_term) == 0) {
struct termios term = orig_term;
+ /* Then we turn off echo, line buffering and ^C etc. We want a
+ * raw input stream to the Guest. */
term.c_lflag &= ~(ISIG|ICANON|ECHO);
tcsetattr(STDIN_FILENO, TCSANOW, &term);
+ /* If we exit gracefully, the original settings will be
+ * restored so the user can see what they're typing. */
atexit(restore_term);
}
- /* We don't currently require a page for the console. */
- dev = new_device(devices, LGUEST_DEVICE_T_CONSOLE, 0, 0,
- STDIN_FILENO, handle_console_input,
- LGUEST_CONSOLE_DMA_KEY, handle_console_output);
+ dev = new_device("console", VIRTIO_ID_CONSOLE,
+ STDIN_FILENO, handle_console_input);
+ /* We store the console state in dev->priv, and initialize it. */
dev->priv = malloc(sizeof(struct console_abort));
((struct console_abort *)dev->priv)->count = 0;
- verbose("device %p: console\n",
- (void *)(dev->desc->pfn * getpagesize()));
-}
-static void setup_block_file(const char *filename, struct device_list *devices)
-{
- int fd;
- struct device *dev;
- off64_t *device_len;
- struct lguest_block_page *p;
-
- fd = open_or_die(filename, O_RDWR|O_LARGEFILE|O_DIRECT);
- dev = new_device(devices, LGUEST_DEVICE_T_BLOCK, 1,
- LGUEST_DEVICE_F_RANDOMNESS,
- fd, NULL, 0, handle_block_output);
- device_len = dev->priv = malloc(sizeof(*device_len));
- *device_len = lseek64(fd, 0, SEEK_END);
- p = dev->mem;
-
- p->num_sectors = *device_len/512;
- verbose("device %p: block %i sectors\n",
- (void *)(dev->desc->pfn * getpagesize()), p->num_sectors);
-}
-
-/* We use fnctl locks to reserve network slots (autocleanup!) */
-static unsigned int find_slot(int netfd, const char *filename)
-{
- struct flock fl;
-
- fl.l_type = F_WRLCK;
- fl.l_whence = SEEK_SET;
- fl.l_len = 1;
- for (fl.l_start = 0;
- fl.l_start < getpagesize()/sizeof(struct lguest_net);
- fl.l_start++) {
- if (fcntl(netfd, F_SETLK, &fl) == 0)
- return fl.l_start;
- }
- errx(1, "No free slots in network file %s", filename);
-}
-
-static void setup_net_file(const char *filename,
- struct device_list *devices)
-{
- int netfd;
- struct device *dev;
-
- netfd = open(filename, O_RDWR, 0);
- if (netfd < 0) {
- if (errno == ENOENT) {
- netfd = open(filename, O_RDWR|O_CREAT, 0600);
- if (netfd >= 0) {
- char page[getpagesize()];
- memset(page, 0, sizeof(page));
- write(netfd, page, sizeof(page));
- }
- }
- if (netfd < 0)
- err(1, "cannot open net file '%s'", filename);
- }
-
- dev = new_device(devices, LGUEST_DEVICE_T_NET, 1,
- find_slot(netfd, filename)|LGUEST_NET_F_NOCSUM,
- -1, NULL, 0, NULL);
+ /* The console needs two virtqueues: the input then the output. When
+ * they put something the input queue, we make sure we're listening to
+ * stdin. When they put something in the output queue, we write it to
+ * stdout. */
+ add_virtqueue(dev, VIRTQUEUE_NUM, enable_fd);
+ add_virtqueue(dev, VIRTQUEUE_NUM, handle_console_output);
- /* We overwrite the /dev/zero mapping with the actual file. */
- if (mmap(dev->mem, getpagesize(), PROT_READ|PROT_WRITE,
- MAP_FIXED|MAP_SHARED, netfd, 0) != dev->mem)
- err(1, "could not mmap '%s'", filename);
- verbose("device %p: shared net %s, peer %i\n",
- (void *)(dev->desc->pfn * getpagesize()), filename,
- dev->desc->features & ~LGUEST_NET_F_NOCSUM);
+ verbose("device %u: console\n", devices.device_num++);
}
+/*:*/
+
+/*M:010 Inter-guest networking is an interesting area. Simplest is to have a
+ * --sharenet=<name> option which opens or creates a named pipe. This can be
+ * used to send packets to another guest in a 1:1 manner.
+ *
+ * More sopisticated is to use one of the tools developed for project like UML
+ * to do networking.
+ *
+ * Faster is to do virtio bonding in kernel. Doing this 1:1 would be
+ * completely generic ("here's my vring, attach to your vring") and would work
+ * for any traffic. Of course, namespace and permissions issues need to be
+ * dealt with. A more sophisticated "multi-channel" virtio_net.c could hide
+ * multiple inter-guest channels behind one interface, although it would
+ * require some manner of hotplugging new virtio channels.
+ *
+ * Finally, we could implement a virtio network switch in the kernel. :*/
static u32 str2ip(const char *ipaddr)
{
return (byte[0] << 24) | (byte[1] << 16) | (byte[2] << 8) | byte[3];
}
-/* adapted from libbridge */
+/* This code is "adapted" from libbridge: it attaches the Host end of the
+ * network device to the bridge device specified by the command line.
+ *
+ * This is yet another James Morris contribution (I'm an IP-level guy, so I
+ * dislike bridging), and I just try not to break it. */
static void add_to_bridge(int fd, const char *if_name, const char *br_name)
{
int ifidx;
err(1, "can't add %s to bridge %s", if_name, br_name);
}
+/* This sets up the Host end of the network device with an IP address, brings
+ * it up so packets will flow, the copies the MAC address into the hwaddr
+ * pointer. */
static void configure_device(int fd, const char *devname, u32 ipaddr,
unsigned char hwaddr[6])
{
struct ifreq ifr;
struct sockaddr_in *sin = (struct sockaddr_in *)&ifr.ifr_addr;
+ /* Don't read these incantations. Just cut & paste them like I did! */
memset(&ifr, 0, sizeof(ifr));
strcpy(ifr.ifr_name, devname);
sin->sin_family = AF_INET;
if (ioctl(fd, SIOCSIFFLAGS, &ifr) != 0)
err(1, "Bringing interface %s up", devname);
+ /* SIOC stands for Socket I/O Control. G means Get (vs S for Set
+ * above). IF means Interface, and HWADDR is hardware address.
+ * Simple! */
if (ioctl(fd, SIOCGIFHWADDR, &ifr) != 0)
err(1, "getting hw address for %s", devname);
-
memcpy(hwaddr, ifr.ifr_hwaddr.sa_data, 6);
}
-static void setup_tun_net(const char *arg, struct device_list *devices)
+/*L:195 Our network is a Host<->Guest network. This can either use bridging or
+ * routing, but the principle is the same: it uses the "tun" device to inject
+ * packets into the Host as if they came in from a normal network card. We
+ * just shunt packets between the Guest and the tun device. */
+static void setup_tun_net(const char *arg)
{
struct device *dev;
struct ifreq ifr;
int netfd, ipfd;
u32 ip;
const char *br_name = NULL;
+ u8 hwaddr[6];
+ /* We open the /dev/net/tun device and tell it we want a tap device. A
+ * tap device is like a tun device, only somehow different. To tell
+ * the truth, I completely blundered my way through this code, but it
+ * works now! */
netfd = open_or_die("/dev/net/tun", O_RDWR);
memset(&ifr, 0, sizeof(ifr));
ifr.ifr_flags = IFF_TAP | IFF_NO_PI;
strcpy(ifr.ifr_name, "tap%d");
if (ioctl(netfd, TUNSETIFF, &ifr) != 0)
err(1, "configuring /dev/net/tun");
+ /* We don't need checksums calculated for packets coming in this
+ * device: trust us! */
ioctl(netfd, TUNSETNOCSUM, 1);
- /* You will be peer 1: we should create enough jitter to randomize */
- dev = new_device(devices, LGUEST_DEVICE_T_NET, 1,
- NET_PEERNUM|LGUEST_DEVICE_F_RANDOMNESS, netfd,
- handle_tun_input, peer_offset(0), handle_tun_output);
- dev->priv = malloc(sizeof(bool));
- *(bool *)dev->priv = false;
+ /* First we create a new network device. */
+ dev = new_device("net", VIRTIO_ID_NET, netfd, handle_tun_input);
+
+ /* Network devices need a receive and a send queue, just like
+ * console. */
+ add_virtqueue(dev, VIRTQUEUE_NUM, enable_fd);
+ add_virtqueue(dev, VIRTQUEUE_NUM, handle_net_output);
+ /* We need a socket to perform the magic network ioctls to bring up the
+ * tap interface, connect to the bridge etc. Any socket will do! */
ipfd = socket(PF_INET, SOCK_DGRAM, IPPROTO_IP);
if (ipfd < 0)
err(1, "opening IP socket");
+ /* If the command line was --tunnet=bridge:<name> do bridging. */
if (!strncmp(BRIDGE_PFX, arg, strlen(BRIDGE_PFX))) {
ip = INADDR_ANY;
br_name = arg + strlen(BRIDGE_PFX);
add_to_bridge(ipfd, ifr.ifr_name, br_name);
- } else
+ } else /* It is an IP address to set up the device with */
ip = str2ip(arg);
- /* We are peer 0, ie. first slot. */
- configure_device(ipfd, ifr.ifr_name, ip, dev->mem);
+ /* Set up the tun device, and get the mac address for the interface. */
+ configure_device(ipfd, ifr.ifr_name, ip, hwaddr);
- /* Set "promisc" bit: we want every single packet. */
- *((u8 *)dev->mem) |= 0x1;
+ /* Tell Guest what MAC address to use. */
+ add_desc_field(dev, VIRTIO_CONFIG_NET_MAC_F, sizeof(hwaddr), hwaddr);
+ /* We don't seed the socket any more; setup is done. */
close(ipfd);
- verbose("device %p: tun net %u.%u.%u.%u\n",
- (void *)(dev->desc->pfn * getpagesize()),
- (u8)(ip>>24), (u8)(ip>>16), (u8)(ip>>8), (u8)ip);
+ verbose("device %u: tun net %u.%u.%u.%u\n",
+ devices.device_num++,
+ (u8)(ip>>24),(u8)(ip>>16),(u8)(ip>>8),(u8)ip);
if (br_name)
verbose("attached to bridge: %s\n", br_name);
}
-/* Now we know how much memory we have, we copy in device descriptors */
-static void map_device_descriptors(struct device_list *devs, unsigned long mem)
+
+/*
+ * Block device.
+ *
+ * Serving a block device is really easy: the Guest asks for a block number and
+ * we read or write that position in the file.
+ *
+ * Unfortunately, this is amazingly slow: the Guest waits until the read is
+ * finished before running anything else, even if it could be doing useful
+ * work. We could use async I/O, except it's reputed to suck so hard that
+ * characters actually go missing from your code when you try to use it.
+ *
+ * So we farm the I/O out to thread, and communicate with it via a pipe. */
+
+/* This hangs off device->priv, with the data. */
+struct vblk_info
{
- struct device *i;
- unsigned int num;
- struct lguest_device_desc *descs;
-
- /* Device descriptor array sits just above top of normal memory */
- descs = map_zeroed_pages(mem, 1);
-
- for (i = devs->dev, num = 0; i; i = i->next, num++) {
- if (num == LGUEST_MAX_DEVICES)
- errx(1, "too many devices");
- verbose("Device %i: %s\n", num,
- i->desc->type == LGUEST_DEVICE_T_NET ? "net"
- : i->desc->type == LGUEST_DEVICE_T_CONSOLE ? "console"
- : i->desc->type == LGUEST_DEVICE_T_BLOCK ? "block"
- : "unknown");
- descs[num] = *i->desc;
- free(i->desc);
- i->desc = &descs[num];
+ /* The size of the file. */
+ off64_t len;
+
+ /* The file descriptor for the file. */
+ int fd;
+
+ /* IO thread listens on this file descriptor [0]. */
+ int workpipe[2];
+
+ /* IO thread writes to this file descriptor to mark it done, then
+ * Launcher triggers interrupt to Guest. */
+ int done_fd;
+};
+
+/* This is the core of the I/O thread. It returns true if it did something. */
+static bool service_io(struct device *dev)
+{
+ struct vblk_info *vblk = dev->priv;
+ unsigned int head, out_num, in_num, wlen;
+ int ret;
+ struct virtio_blk_inhdr *in;
+ struct virtio_blk_outhdr *out;
+ struct iovec iov[dev->vq->vring.num];
+ off64_t off;
+
+ head = get_vq_desc(dev->vq, iov, &out_num, &in_num);
+ if (head == dev->vq->vring.num)
+ return false;
+
+ if (out_num == 0 || in_num == 0)
+ errx(1, "Bad virtblk cmd %u out=%u in=%u",
+ head, out_num, in_num);
+
+ out = convert(&iov[0], struct virtio_blk_outhdr);
+ in = convert(&iov[out_num+in_num-1], struct virtio_blk_inhdr);
+ off = out->sector * 512;
+
+ /* This is how we implement barriers. Pretty poor, no? */
+ if (out->type & VIRTIO_BLK_T_BARRIER)
+ fdatasync(vblk->fd);
+
+ if (out->type & VIRTIO_BLK_T_SCSI_CMD) {
+ fprintf(stderr, "Scsi commands unsupported\n");
+ in->status = VIRTIO_BLK_S_UNSUPP;
+ wlen = sizeof(in);
+ } else if (out->type & VIRTIO_BLK_T_OUT) {
+ /* Write */
+
+ /* Move to the right location in the block file. This can fail
+ * if they try to write past end. */
+ if (lseek64(vblk->fd, off, SEEK_SET) != off)
+ err(1, "Bad seek to sector %llu", out->sector);
+
+ ret = writev(vblk->fd, iov+1, out_num-1);
+ verbose("WRITE to sector %llu: %i\n", out->sector, ret);
+
+ /* Grr... Now we know how long the descriptor they sent was, we
+ * make sure they didn't try to write over the end of the block
+ * file (possibly extending it). */
+ if (ret > 0 && off + ret > vblk->len) {
+ /* Trim it back to the correct length */
+ ftruncate64(vblk->fd, vblk->len);
+ /* Die, bad Guest, die. */
+ errx(1, "Write past end %llu+%u", off, ret);
+ }
+ wlen = sizeof(in);
+ in->status = (ret >= 0 ? VIRTIO_BLK_S_OK : VIRTIO_BLK_S_IOERR);
+ } else {
+ /* Read */
+
+ /* Move to the right location in the block file. This can fail
+ * if they try to read past end. */
+ if (lseek64(vblk->fd, off, SEEK_SET) != off)
+ err(1, "Bad seek to sector %llu", out->sector);
+
+ ret = readv(vblk->fd, iov+1, in_num-1);
+ verbose("READ from sector %llu: %i\n", out->sector, ret);
+ if (ret >= 0) {
+ wlen = sizeof(in) + ret;
+ in->status = VIRTIO_BLK_S_OK;
+ } else {
+ wlen = sizeof(in);
+ in->status = VIRTIO_BLK_S_IOERR;
+ }
+ }
+
+ /* We can't trigger an IRQ, because we're not the Launcher. It does
+ * that when we tell it we're done. */
+ add_used(dev->vq, head, wlen);
+ return true;
+}
+
+/* This is the thread which actually services the I/O. */
+static int io_thread(void *_dev)
+{
+ struct device *dev = _dev;
+ struct vblk_info *vblk = dev->priv;
+ char c;
+
+ /* Close other side of workpipe so we get 0 read when main dies. */
+ close(vblk->workpipe[1]);
+ /* Close the other side of the done_fd pipe. */
+ close(dev->fd);
+
+ /* When this read fails, it means Launcher died, so we follow. */
+ while (read(vblk->workpipe[0], &c, 1) == 1) {
+ /* We acknowledge each request immediately, to reduce latency,
+ * rather than waiting until we've done them all. I haven't
+ * measured to see if it makes any difference. */
+ while (service_io(dev))
+ write(vblk->done_fd, &c, 1);
}
+ return 0;
+}
+
+/* When the thread says some I/O is done, we interrupt the Guest. */
+static bool handle_io_finish(int fd, struct device *dev)
+{
+ char c;
+
+ /* If child died, presumably it printed message. */
+ if (read(dev->fd, &c, 1) != 1)
+ exit(1);
+
+ /* It did some work, so trigger the irq. */
+ trigger_irq(fd, dev->vq);
+ return true;
+}
+
+/* When the Guest submits some I/O, we wake the I/O thread. */
+static void handle_virtblk_output(int fd, struct virtqueue *vq)
+{
+ struct vblk_info *vblk = vq->dev->priv;
+ char c = 0;
+
+ /* Wake up I/O thread and tell it to go to work! */
+ if (write(vblk->workpipe[1], &c, 1) != 1)
+ /* Presumably it indicated why it died. */
+ exit(1);
}
-static void __attribute__((noreturn))
-run_guest(int lguest_fd, struct device_list *device_list)
+/* This creates a virtual block device. */
+static void setup_block_file(const char *filename)
+{
+ int p[2];
+ struct device *dev;
+ struct vblk_info *vblk;
+ void *stack;
+ u64 cap;
+ unsigned int val;
+
+ /* This is the pipe the I/O thread will use to tell us I/O is done. */
+ pipe(p);
+
+ /* The device responds to return from I/O thread. */
+ dev = new_device("block", VIRTIO_ID_BLOCK, p[0], handle_io_finish);
+
+ /* The device has a virtqueue. */
+ add_virtqueue(dev, VIRTQUEUE_NUM, handle_virtblk_output);
+
+ /* Allocate the room for our own bookkeeping */
+ vblk = dev->priv = malloc(sizeof(*vblk));
+
+ /* First we open the file and store the length. */
+ vblk->fd = open_or_die(filename, O_RDWR|O_LARGEFILE);
+ vblk->len = lseek64(vblk->fd, 0, SEEK_END);
+
+ /* Tell Guest how many sectors this device has. */
+ cap = cpu_to_le64(vblk->len / 512);
+ add_desc_field(dev, VIRTIO_CONFIG_BLK_F_CAPACITY, sizeof(cap), &cap);
+
+ /* Tell Guest not to put in too many descriptors at once: two are used
+ * for the in and out elements. */
+ val = cpu_to_le32(VIRTQUEUE_NUM - 2);
+ add_desc_field(dev, VIRTIO_CONFIG_BLK_F_SEG_MAX, sizeof(val), &val);
+
+ /* The I/O thread writes to this end of the pipe when done. */
+ vblk->done_fd = p[1];
+
+ /* This is how we tell the I/O thread about more work. */
+ pipe(vblk->workpipe);
+
+ /* Create stack for thread and run it */
+ stack = malloc(32768);
+ if (clone(io_thread, stack + 32768, CLONE_VM, dev) == -1)
+ err(1, "Creating clone");
+
+ /* We don't need to keep the I/O thread's end of the pipes open. */
+ close(vblk->done_fd);
+ close(vblk->workpipe[0]);
+
+ verbose("device %u: virtblock %llu sectors\n",
+ devices.device_num, cap);
+}
+/* That's the end of device setup. */
+
+/*L:220 Finally we reach the core of the Launcher, which runs the Guest, serves
+ * its input and output, and finally, lays it to rest. */
+static void __attribute__((noreturn)) run_guest(int lguest_fd)
{
for (;;) {
- u32 args[] = { LHREQ_BREAK, 0 };
- unsigned long arr[2];
+ unsigned long args[] = { LHREQ_BREAK, 0 };
+ unsigned long notify_addr;
int readval;
/* We read from the /dev/lguest device to run the Guest. */
- readval = read(lguest_fd, arr, sizeof(arr));
+ readval = read(lguest_fd, ¬ify_addr, sizeof(notify_addr));
- if (readval == sizeof(arr)) {
- handle_output(lguest_fd, arr[0], arr[1], device_list);
+ /* One unsigned long means the Guest did HCALL_NOTIFY */
+ if (readval == sizeof(notify_addr)) {
+ verbose("Notify on address %#lx\n", notify_addr);
+ handle_output(lguest_fd, notify_addr);
continue;
+ /* ENOENT means the Guest died. Reading tells us why. */
} else if (errno == ENOENT) {
char reason[1024] = { 0 };
read(lguest_fd, reason, sizeof(reason)-1);
errx(1, "%s", reason);
+ /* EAGAIN means the waker wanted us to look at some input.
+ * Anything else means a bug or incompatible change. */
} else if (errno != EAGAIN)
err(1, "Running guest failed");
- handle_input(lguest_fd, device_list);
+
+ /* Service input, then unset the BREAK which releases
+ * the Waker. */
+ handle_input(lguest_fd);
if (write(lguest_fd, args, sizeof(args)) < 0)
err(1, "Resetting break");
}
}
+/*
+ * This is the end of the Launcher.
+ *
+ * But wait! We've seen I/O from the Launcher, and we've seen I/O from the
+ * Drivers. If we were to see the Host kernel I/O code, our understanding
+ * would be complete... :*/
static struct option opts[] = {
{ "verbose", 0, NULL, 'v' },
- { "sharenet", 1, NULL, 's' },
{ "tunnet", 1, NULL, 't' },
{ "block", 1, NULL, 'b' },
{ "initrd", 1, NULL, 'i' },
static void usage(void)
{
errx(1, "Usage: lguest [--verbose] "
- "[--sharenet=<filename>|--tunnet=(<ipaddr>|bridge:<bridgename>)\n"
+ "[--tunnet=(<ipaddr>|bridge:<bridgename>)\n"
"|--block=<filename>|--initrd=<filename>]...\n"
"<mem-in-mb> vmlinux [args...]");
}
+/*L:105 The main routine is where the real work begins: */
int main(int argc, char *argv[])
{
- unsigned long mem, pgdir, start, page_offset, initrd_size = 0;
- int c, lguest_fd;
- struct device_list device_list;
- void *boot = (void *)0;
+ /* Memory, top-level pagetable, code startpoint and size of the
+ * (optional) initrd. */
+ unsigned long mem = 0, pgdir, start, initrd_size = 0;
+ /* A temporary and the /dev/lguest file descriptor. */
+ int i, c, lguest_fd;
+ /* The boot information for the Guest. */
+ void *boot;
+ /* If they specify an initrd file to load. */
const char *initrd_name = NULL;
- device_list.max_infd = -1;
- device_list.dev = NULL;
- device_list.lastdev = &device_list.dev;
- FD_ZERO(&device_list.infds);
+ /* First we initialize the device list. Since console and network
+ * device receive input from a file descriptor, we keep an fdset
+ * (infds) and the maximum fd number (max_infd) with the head of the
+ * list. We also keep a pointer to the last device, for easy appending
+ * to the list. Finally, we keep the next interrupt number to hand out
+ * (1: remember that 0 is used by the timer). */
+ FD_ZERO(&devices.infds);
+ devices.max_infd = -1;
+ devices.lastdev = &devices.dev;
+ devices.next_irq = 1;
+
+ /* We need to know how much memory so we can set up the device
+ * descriptor and memory pages for the devices as we parse the command
+ * line. So we quickly look through the arguments to find the amount
+ * of memory now. */
+ for (i = 1; i < argc; i++) {
+ if (argv[i][0] != '-') {
+ mem = atoi(argv[i]) * 1024 * 1024;
+ /* We start by mapping anonymous pages over all of
+ * guest-physical memory range. This fills it with 0,
+ * and ensures that the Guest won't be killed when it
+ * tries to access it. */
+ guest_base = map_zeroed_pages(mem / getpagesize()
+ + DEVICE_PAGES);
+ guest_limit = mem;
+ guest_max = mem + DEVICE_PAGES*getpagesize();
+ devices.descpage = get_pages(1);
+ break;
+ }
+ }
+ /* The options are fairly straight-forward */
while ((c = getopt_long(argc, argv, "v", opts, NULL)) != EOF) {
switch (c) {
case 'v':
verbose = true;
break;
- case 's':
- setup_net_file(optarg, &device_list);
- break;
case 't':
- setup_tun_net(optarg, &device_list);
+ setup_tun_net(optarg);
break;
case 'b':
- setup_block_file(optarg, &device_list);
+ setup_block_file(optarg);
break;
case 'i':
initrd_name = optarg;
usage();
}
}
+ /* After the other arguments we expect memory and kernel image name,
+ * followed by command line arguments for the kernel. */
if (optind + 2 > argc)
usage();
- /* We need a console device */
- setup_console(&device_list);
+ verbose("Guest base is at %p\n", guest_base);
- /* First we map /dev/zero over all of guest-physical memory. */
- mem = atoi(argv[optind]) * 1024 * 1024;
- map_zeroed_pages(0, mem / getpagesize());
+ /* We always have a console device */
+ setup_console();
/* Now we load the kernel */
- start = load_kernel(open_or_die(argv[optind+1], O_RDONLY),
- &page_offset);
+ start = load_kernel(open_or_die(argv[optind+1], O_RDONLY));
- /* Write the device descriptors into memory. */
- map_device_descriptors(&device_list, mem);
+ /* Boot information is stashed at physical address 0 */
+ boot = from_guest_phys(0);
- /* Map the initrd image if requested */
+ /* Map the initrd image if requested (at top of physical memory) */
if (initrd_name) {
initrd_size = load_initrd(initrd_name, mem);
+ /* These are the location in the Linux boot header where the
+ * start and size of the initrd are expected to be found. */
*(unsigned long *)(boot+0x218) = mem - initrd_size;
*(unsigned long *)(boot+0x21c) = initrd_size;
+ /* The bootloader type 0xFF means "unknown"; that's OK. */
*(unsigned char *)(boot+0x210) = 0xFF;
}
- /* Set up the initial linar pagetables. */
- pgdir = setup_pagetables(mem, initrd_size, page_offset);
+ /* Set up the initial linear pagetables, starting below the initrd. */
+ pgdir = setup_pagetables(mem, initrd_size);
- /* E820 memory map: ours is a simple, single region. */
+ /* The Linux boot header contains an "E820" memory map: ours is a
+ * simple, single region. */
*(char*)(boot+E820NR) = 1;
*((struct e820entry *)(boot+E820MAP))
= ((struct e820entry) { 0, mem, E820_RAM });
- /* Command line pointer and command line (at 4096) */
- *(void **)(boot + 0x228) = boot + 4096;
+ /* The boot header contains a command line pointer: we put the command
+ * line after the boot header (at address 4096) */
+ *(u32 *)(boot + 0x228) = 4096;
concat(boot + 4096, argv+optind+2);
- /* Paravirt type: 1 == lguest */
+
+ /* The guest type value of "1" tells the Guest it's under lguest. */
*(int *)(boot + 0x23c) = 1;
- lguest_fd = tell_kernel(pgdir, start, page_offset);
- waker_fd = setup_waker(lguest_fd, &device_list);
+ /* We tell the kernel to initialize the Guest: this returns the open
+ * /dev/lguest file descriptor. */
+ lguest_fd = tell_kernel(pgdir, start);
+
+ /* We fork off a child process, which wakes the Launcher whenever one
+ * of the input file descriptors needs attention. Otherwise we would
+ * run the Guest until it tries to output something. */
+ waker_fd = setup_waker(lguest_fd);
- run_guest(lguest_fd, &device_list);
+ /* Finally, run the Guest. This doesn't return. */
+ run_guest(lguest_fd);
}
+/*:*/
+
+/*M:999
+ * Mastery is done: you now know everything I do.
+ *
+ * But surely you have seen code, features and bugs in your wanderings which
+ * you now yearn to attack? That is the real game, and I look forward to you
+ * patching and forking lguest into the Your-Name-Here-visor.
+ *
+ * Farewell, and good coding!
+ * Rusty Russell.
+ */