From 2e04ef76916d1e29a077ea9d0f2003c8fd86724d Mon Sep 17 00:00:00 2001 From: Rusty Russell Date: Thu, 30 Jul 2009 16:03:45 -0600 Subject: lguest: fix comment style I don't really notice it (except to begrudge the extra vertical space), but Ingo does. And he pointed out that one excuse of lguest is as a teaching tool, it should set a good example. Signed-off-by: Rusty Russell Cc: Ingo Molnar --- drivers/lguest/core.c | 114 +++++---- drivers/lguest/hypercalls.c | 141 +++++++---- drivers/lguest/interrupts_and_traps.c | 288 +++++++++++++++-------- drivers/lguest/lg.h | 23 +- drivers/lguest/lguest_device.c | 150 +++++++----- drivers/lguest/lguest_user.c | 137 +++++++---- drivers/lguest/page_tables.c | 427 ++++++++++++++++++++++------------ drivers/lguest/segments.c | 106 ++++++--- drivers/lguest/x86/core.c | 372 +++++++++++++++++++---------- drivers/lguest/x86/switcher_32.S | 18 +- 10 files changed, 1168 insertions(+), 608 deletions(-) (limited to 'drivers') diff --git a/drivers/lguest/core.c b/drivers/lguest/core.c index a6974e9b8eb..cd058bc903f 100644 --- a/drivers/lguest/core.c +++ b/drivers/lguest/core.c @@ -1,6 +1,8 @@ -/*P:400 This contains run_guest() which actually calls into the Host<->Guest +/*P:400 + * This contains run_guest() which actually calls into the Host<->Guest * Switcher and analyzes the return, such as determining if the Guest wants the - * Host to do something. This file also contains useful helper routines. :*/ + * Host to do something. This file also contains useful helper routines. +:*/ #include #include #include @@ -24,7 +26,8 @@ static struct page **switcher_page; /* This One Big lock protects all inter-guest data structures. */ DEFINE_MUTEX(lguest_lock); -/*H:010 We need to set up the Switcher at a high virtual address. Remember the +/*H:010 + * We need to set up the Switcher at a high virtual address. Remember the * Switcher is a few hundred bytes of assembler code which actually changes the * CPU to run the Guest, and then changes back to the Host when a trap or * interrupt happens. @@ -33,7 +36,8 @@ DEFINE_MUTEX(lguest_lock); * Host since it will be running as the switchover occurs. * * Trying to map memory at a particular address is an unusual thing to do, so - * it's not a simple one-liner. */ + * it's not a simple one-liner. + */ static __init int map_switcher(void) { int i, err; @@ -47,8 +51,10 @@ static __init int map_switcher(void) * easy. */ - /* We allocate an array of struct page pointers. map_vm_area() wants - * this, rather than just an array of pages. */ + /* + * We allocate an array of struct page pointers. map_vm_area() wants + * this, rather than just an array of pages. + */ switcher_page = kmalloc(sizeof(switcher_page[0])*TOTAL_SWITCHER_PAGES, GFP_KERNEL); if (!switcher_page) { @@ -56,8 +62,10 @@ static __init int map_switcher(void) goto out; } - /* Now we actually allocate the pages. The Guest will see these pages, - * so we make sure they're zeroed. */ + /* + * Now we actually allocate the pages. The Guest will see these pages, + * so we make sure they're zeroed. + */ for (i = 0; i < TOTAL_SWITCHER_PAGES; i++) { unsigned long addr = get_zeroed_page(GFP_KERNEL); if (!addr) { @@ -67,19 +75,23 @@ static __init int map_switcher(void) switcher_page[i] = virt_to_page(addr); } - /* First we check that the Switcher won't overlap the fixmap area at + /* + * First we check that the Switcher won't overlap the fixmap area at * the top of memory. It's currently nowhere near, but it could have - * very strange effects if it ever happened. */ + * very strange effects if it ever happened. + */ if (SWITCHER_ADDR + (TOTAL_SWITCHER_PAGES+1)*PAGE_SIZE > FIXADDR_START){ err = -ENOMEM; printk("lguest: mapping switcher would thwack fixmap\n"); goto free_pages; } - /* Now we reserve the "virtual memory area" we want: 0xFFC00000 + /* + * Now we reserve the "virtual memory area" we want: 0xFFC00000 * (SWITCHER_ADDR). We might not get it in theory, but in practice * it's worked so far. The end address needs +1 because __get_vm_area - * allocates an extra guard page, so we need space for that. */ + * allocates an extra guard page, so we need space for that. + */ switcher_vma = __get_vm_area(TOTAL_SWITCHER_PAGES * PAGE_SIZE, VM_ALLOC, SWITCHER_ADDR, SWITCHER_ADDR + (TOTAL_SWITCHER_PAGES+1) * PAGE_SIZE); @@ -89,11 +101,13 @@ static __init int map_switcher(void) goto free_pages; } - /* This code actually sets up the pages we've allocated to appear at + /* + * This code actually sets up the pages we've allocated to appear at * SWITCHER_ADDR. map_vm_area() takes the vma we allocated above, the * kind of pages we're mapping (kernel pages), and a pointer to our * array of struct pages. It increments that pointer, but we don't - * care. */ + * care. + */ pagep = switcher_page; err = map_vm_area(switcher_vma, PAGE_KERNEL_EXEC, &pagep); if (err) { @@ -101,8 +115,10 @@ static __init int map_switcher(void) goto free_vma; } - /* Now the Switcher is mapped at the right address, we can't fail! - * Copy in the compiled-in Switcher code (from _switcher.S). */ + /* + * Now the Switcher is mapped at the right address, we can't fail! + * Copy in the compiled-in Switcher code (from _switcher.S). + */ memcpy(switcher_vma->addr, start_switcher_text, end_switcher_text - start_switcher_text); @@ -124,8 +140,7 @@ out: } /*:*/ -/* Cleaning up the mapping when the module is unloaded is almost... - * too easy. */ +/* Cleaning up the mapping when the module is unloaded is almost... too easy. */ static void unmap_switcher(void) { unsigned int i; @@ -151,16 +166,19 @@ static void unmap_switcher(void) * But we can't trust the Guest: it might be trying to access the Launcher * code. We have to check that the range is below the pfn_limit the Launcher * gave us. We have to make sure that addr + len doesn't give us a false - * positive by overflowing, too. */ + * positive by overflowing, too. + */ bool lguest_address_ok(const struct lguest *lg, unsigned long addr, unsigned long len) { return (addr+len) / PAGE_SIZE < lg->pfn_limit && (addr+len >= addr); } -/* This routine copies memory from the Guest. Here we can see how useful the +/* + * This routine copies memory from the Guest. Here we can see how useful the * kill_lguest() routine we met in the Launcher can be: we return a random - * value (all zeroes) instead of needing to return an error. */ + * value (all zeroes) instead of needing to return an error. + */ void __lgread(struct lg_cpu *cpu, void *b, unsigned long addr, unsigned bytes) { if (!lguest_address_ok(cpu->lg, addr, bytes) @@ -181,9 +199,11 @@ void __lgwrite(struct lg_cpu *cpu, unsigned long addr, const void *b, } /*:*/ -/*H:030 Let's jump straight to the the main loop which runs the Guest. +/*H:030 + * Let's jump straight to the the main loop which runs the Guest. * Remember, this is called by the Launcher reading /dev/lguest, and we keep - * going around and around until something interesting happens. */ + * going around and around until something interesting happens. + */ int run_guest(struct lg_cpu *cpu, unsigned long __user *user) { /* We stop running once the Guest is dead. */ @@ -195,8 +215,10 @@ int run_guest(struct lg_cpu *cpu, unsigned long __user *user) if (cpu->hcall) do_hypercalls(cpu); - /* It's possible the Guest did a NOTIFY hypercall to the - * Launcher, in which case we return from the read() now. */ + /* + * It's possible the Guest did a NOTIFY hypercall to the + * Launcher, in which case we return from the read() now. + */ if (cpu->pending_notify) { if (!send_notify_to_eventfd(cpu)) { if (put_user(cpu->pending_notify, user)) @@ -209,29 +231,39 @@ int run_guest(struct lg_cpu *cpu, unsigned long __user *user) if (signal_pending(current)) return -ERESTARTSYS; - /* Check if there are any interrupts which can be delivered now: + /* + * Check if there are any interrupts which can be delivered now: * if so, this sets up the hander to be executed when we next - * run the Guest. */ + * run the Guest. + */ irq = interrupt_pending(cpu, &more); if (irq < LGUEST_IRQS) try_deliver_interrupt(cpu, irq, more); - /* All long-lived kernel loops need to check with this horrible + /* + * All long-lived kernel loops need to check with this horrible * thing called the freezer. If the Host is trying to suspend, - * it stops us. */ + * it stops us. + */ try_to_freeze(); - /* Just make absolutely sure the Guest is still alive. One of - * those hypercalls could have been fatal, for example. */ + /* + * Just make absolutely sure the Guest is still alive. One of + * those hypercalls could have been fatal, for example. + */ if (cpu->lg->dead) break; - /* If the Guest asked to be stopped, we sleep. The Guest's - * clock timer will wake us. */ + /* + * If the Guest asked to be stopped, we sleep. The Guest's + * clock timer will wake us. + */ if (cpu->halted) { set_current_state(TASK_INTERRUPTIBLE); - /* Just before we sleep, make sure no interrupt snuck in - * which we should be doing. */ + /* + * Just before we sleep, make sure no interrupt snuck in + * which we should be doing. + */ if (interrupt_pending(cpu, &more) < LGUEST_IRQS) set_current_state(TASK_RUNNING); else @@ -239,8 +271,10 @@ int run_guest(struct lg_cpu *cpu, unsigned long __user *user) continue; } - /* OK, now we're ready to jump into the Guest. First we put up - * the "Do Not Disturb" sign: */ + /* + * OK, now we're ready to jump into the Guest. First we put up + * the "Do Not Disturb" sign: + */ local_irq_disable(); /* Actually run the Guest until something happens. */ @@ -327,8 +361,10 @@ static void __exit fini(void) } /*:*/ -/* The Host side of lguest can be a module. This is a nice way for people to - * play with it. */ +/* + * The Host side of lguest can be a module. This is a nice way for people to + * play with it. + */ module_init(init); module_exit(fini); MODULE_LICENSE("GPL"); diff --git a/drivers/lguest/hypercalls.c b/drivers/lguest/hypercalls.c index c29ffa19cb7..787ab4bc09f 100644 --- a/drivers/lguest/hypercalls.c +++ b/drivers/lguest/hypercalls.c @@ -1,8 +1,10 @@ -/*P:500 Just as userspace programs request kernel operations through a system +/*P:500 + * Just as userspace programs request kernel operations through a system * call, the Guest requests Host operations through a "hypercall". You might * notice this nomenclature doesn't really follow any logic, but the name has * been around for long enough that we're stuck with it. As you'd expect, this - * code is basically a one big switch statement. :*/ + * code is basically a one big switch statement. +:*/ /* Copyright (C) 2006 Rusty Russell IBM Corporation @@ -28,30 +30,41 @@ #include #include "lg.h" -/*H:120 This is the core hypercall routine: where the Guest gets what it wants. - * Or gets killed. Or, in the case of LHCALL_SHUTDOWN, both. */ +/*H:120 + * This is the core hypercall routine: where the Guest gets what it wants. + * Or gets killed. Or, in the case of LHCALL_SHUTDOWN, both. + */ static void do_hcall(struct lg_cpu *cpu, struct hcall_args *args) { switch (args->arg0) { case LHCALL_FLUSH_ASYNC: - /* This call does nothing, except by breaking out of the Guest - * it makes us process all the asynchronous hypercalls. */ + /* + * This call does nothing, except by breaking out of the Guest + * it makes us process all the asynchronous hypercalls. + */ break; case LHCALL_SEND_INTERRUPTS: - /* This call does nothing too, but by breaking out of the Guest - * it makes us process any pending interrupts. */ + /* + * This call does nothing too, but by breaking out of the Guest + * it makes us process any pending interrupts. + */ break; case LHCALL_LGUEST_INIT: - /* You can't get here unless you're already initialized. Don't - * do that. */ + /* + * You can't get here unless you're already initialized. Don't + * do that. + */ kill_guest(cpu, "already have lguest_data"); break; case LHCALL_SHUTDOWN: { - /* Shutdown is such a trivial hypercall that we do it in four - * lines right here. */ char msg[128]; - /* If the lgread fails, it will call kill_guest() itself; the - * kill_guest() with the message will be ignored. */ + /* + * Shutdown is such a trivial hypercall that we do it in four + * lines right here. + * + * If the lgread fails, it will call kill_guest() itself; the + * kill_guest() with the message will be ignored. + */ __lgread(cpu, msg, args->arg1, sizeof(msg)); msg[sizeof(msg)-1] = '\0'; kill_guest(cpu, "CRASH: %s", msg); @@ -60,16 +73,17 @@ static void do_hcall(struct lg_cpu *cpu, struct hcall_args *args) break; } case LHCALL_FLUSH_TLB: - /* FLUSH_TLB comes in two flavors, depending on the - * argument: */ + /* FLUSH_TLB comes in two flavors, depending on the argument: */ if (args->arg1) guest_pagetable_clear_all(cpu); else guest_pagetable_flush_user(cpu); break; - /* All these calls simply pass the arguments through to the right - * routines. */ + /* + * All these calls simply pass the arguments through to the right + * routines. + */ case LHCALL_NEW_PGTABLE: guest_new_pagetable(cpu, args->arg1); break; @@ -112,15 +126,16 @@ static void do_hcall(struct lg_cpu *cpu, struct hcall_args *args) kill_guest(cpu, "Bad hypercall %li\n", args->arg0); } } -/*:*/ -/*H:124 Asynchronous hypercalls are easy: we just look in the array in the +/*H:124 + * Asynchronous hypercalls are easy: we just look in the array in the * Guest's "struct lguest_data" to see if any new ones are marked "ready". * * We are careful to do these in order: obviously we respect the order the * Guest put them in the ring, but we also promise the Guest that they will * happen before any normal hypercall (which is why we check this before - * checking for a normal hcall). */ + * checking for a normal hcall). + */ static void do_async_hcalls(struct lg_cpu *cpu) { unsigned int i; @@ -133,22 +148,28 @@ static void do_async_hcalls(struct lg_cpu *cpu) /* We process "struct lguest_data"s hcalls[] ring once. */ for (i = 0; i < ARRAY_SIZE(st); i++) { struct hcall_args args; - /* We remember where we were up to from last time. This makes + /* + * We remember where we were up to from last time. This makes * sure that the hypercalls are done in the order the Guest - * places them in the ring. */ + * places them in the ring. + */ unsigned int n = cpu->next_hcall; /* 0xFF means there's no call here (yet). */ if (st[n] == 0xFF) break; - /* OK, we have hypercall. Increment the "next_hcall" cursor, - * and wrap back to 0 if we reach the end. */ + /* + * OK, we have hypercall. Increment the "next_hcall" cursor, + * and wrap back to 0 if we reach the end. + */ if (++cpu->next_hcall == LHCALL_RING_SIZE) cpu->next_hcall = 0; - /* Copy the hypercall arguments into a local copy of - * the hcall_args struct. */ + /* + * Copy the hypercall arguments into a local copy of the + * hcall_args struct. + */ if (copy_from_user(&args, &cpu->lg->lguest_data->hcalls[n], sizeof(struct hcall_args))) { kill_guest(cpu, "Fetching async hypercalls"); @@ -164,19 +185,25 @@ static void do_async_hcalls(struct lg_cpu *cpu) break; } - /* Stop doing hypercalls if they want to notify the Launcher: - * it needs to service this first. */ + /* + * Stop doing hypercalls if they want to notify the Launcher: + * it needs to service this first. + */ if (cpu->pending_notify) break; } } -/* Last of all, we look at what happens first of all. The very first time the - * Guest makes a hypercall, we end up here to set things up: */ +/* + * Last of all, we look at what happens first of all. The very first time the + * Guest makes a hypercall, we end up here to set things up: + */ static void initialize(struct lg_cpu *cpu) { - /* You can't do anything until you're initialized. The Guest knows the - * rules, so we're unforgiving here. */ + /* + * You can't do anything until you're initialized. The Guest knows the + * rules, so we're unforgiving here. + */ if (cpu->hcall->arg0 != LHCALL_LGUEST_INIT) { kill_guest(cpu, "hypercall %li before INIT", cpu->hcall->arg0); return; @@ -185,32 +212,40 @@ static void initialize(struct lg_cpu *cpu) if (lguest_arch_init_hypercalls(cpu)) kill_guest(cpu, "bad guest page %p", cpu->lg->lguest_data); - /* The Guest tells us where we're not to deliver interrupts by putting - * the range of addresses into "struct lguest_data". */ + /* + * The Guest tells us where we're not to deliver interrupts by putting + * the range of addresses into "struct lguest_data". + */ if (get_user(cpu->lg->noirq_start, &cpu->lg->lguest_data->noirq_start) || get_user(cpu->lg->noirq_end, &cpu->lg->lguest_data->noirq_end)) kill_guest(cpu, "bad guest page %p", cpu->lg->lguest_data); - /* We write the current time into the Guest's data page once so it can - * set its clock. */ + /* + * We write the current time into the Guest's data page once so it can + * set its clock. + */ write_timestamp(cpu); /* page_tables.c will also do some setup. */ page_table_guest_data_init(cpu); - /* This is the one case where the above accesses might have been the + /* + * This is the one case where the above accesses might have been the * first write to a Guest page. This may have caused a copy-on-write * fault, but the old page might be (read-only) in the Guest - * pagetable. */ + * pagetable. + */ guest_pagetable_clear_all(cpu); } /*:*/ -/*M:013 If a Guest reads from a page (so creates a mapping) that it has never +/*M:013 + * If a Guest reads from a page (so creates a mapping) that it has never * written to, and then the Launcher writes to it (ie. the output of a virtual * device), the Guest will still see the old page. In practice, this never * happens: why would the Guest read a page which it has never written to? But - * a similar scenario might one day bite us, so it's worth mentioning. :*/ + * a similar scenario might one day bite us, so it's worth mentioning. +:*/ /*H:100 * Hypercalls @@ -229,17 +264,22 @@ void do_hypercalls(struct lg_cpu *cpu) return; } - /* The Guest has initialized. + /* + * The Guest has initialized. * - * Look in the hypercall ring for the async hypercalls: */ + * Look in the hypercall ring for the async hypercalls: + */ do_async_hcalls(cpu); - /* If we stopped reading the hypercall ring because the Guest did a + /* + * If we stopped reading the hypercall ring because the Guest did a * NOTIFY to the Launcher, we want to return now. Otherwise we do - * the hypercall. */ + * the hypercall. + */ if (!cpu->pending_notify) { do_hcall(cpu, cpu->hcall); - /* Tricky point: we reset the hcall pointer to mark the + /* + * Tricky point: we reset the hcall pointer to mark the * hypercall as "done". We use the hcall pointer rather than * the trap number to indicate a hypercall is pending. * Normally it doesn't matter: the Guest will run again and @@ -248,13 +288,16 @@ void do_hypercalls(struct lg_cpu *cpu) * However, if we are signalled or the Guest sends I/O to the * Launcher, the run_guest() loop will exit without running the * Guest. When it comes back it would try to re-run the - * hypercall. Finding that bug sucked. */ + * hypercall. Finding that bug sucked. + */ cpu->hcall = NULL; } } -/* This routine supplies the Guest with time: it's used for wallclock time at - * initial boot and as a rough time source if the TSC isn't available. */ +/* + * This routine supplies the Guest with time: it's used for wallclock time at + * initial boot and as a rough time source if the TSC isn't available. + */ void write_timestamp(struct lg_cpu *cpu) { struct timespec now; diff --git a/drivers/lguest/interrupts_and_traps.c b/drivers/lguest/interrupts_and_traps.c index 0e9067b0d50..18648180db0 100644 --- a/drivers/lguest/interrupts_and_traps.c +++ b/drivers/lguest/interrupts_and_traps.c @@ -1,4 +1,5 @@ -/*P:800 Interrupts (traps) are complicated enough to earn their own file. +/*P:800 + * Interrupts (traps) are complicated enough to earn their own file. * There are three classes of interrupts: * * 1) Real hardware interrupts which occur while we're running the Guest, @@ -10,7 +11,8 @@ * just like real hardware would deliver them. Traps from the Guest can be set * up to go directly back into the Guest, but sometimes the Host wants to see * them first, so we also have a way of "reflecting" them into the Guest as if - * they had been delivered to it directly. :*/ + * they had been delivered to it directly. +:*/ #include #include #include @@ -26,8 +28,10 @@ static unsigned long idt_address(u32 lo, u32 hi) return (lo & 0x0000FFFF) | (hi & 0xFFFF0000); } -/* The "type" of the interrupt handler is a 4 bit field: we only support a - * couple of types. */ +/* + * The "type" of the interrupt handler is a 4 bit field: we only support a + * couple of types. + */ static int idt_type(u32 lo, u32 hi) { return (hi >> 8) & 0xF; @@ -39,8 +43,10 @@ static bool idt_present(u32 lo, u32 hi) return (hi & 0x8000); } -/* We need a helper to "push" a value onto the Guest's stack, since that's a - * big part of what delivering an interrupt does. */ +/* + * We need a helper to "push" a value onto the Guest's stack, since that's a + * big part of what delivering an interrupt does. + */ static void push_guest_stack(struct lg_cpu *cpu, unsigned long *gstack, u32 val) { /* Stack grows upwards: move stack then write value. */ @@ -48,7 +54,8 @@ static void push_guest_stack(struct lg_cpu *cpu, unsigned long *gstack, u32 val) lgwrite(cpu, *gstack, u32, val); } -/*H:210 The set_guest_interrupt() routine actually delivers the interrupt or +/*H:210 + * The set_guest_interrupt() routine actually delivers the interrupt or * trap. The mechanics of delivering traps and interrupts to the Guest are the * same, except some traps have an "error code" which gets pushed onto the * stack as well: the caller tells us if this is one. @@ -59,7 +66,8 @@ static void push_guest_stack(struct lg_cpu *cpu, unsigned long *gstack, u32 val) * * We set up the stack just like the CPU does for a real interrupt, so it's * identical for the Guest (and the standard "iret" instruction will undo - * it). */ + * it). + */ static void set_guest_interrupt(struct lg_cpu *cpu, u32 lo, u32 hi, bool has_err) { @@ -67,20 +75,26 @@ static void set_guest_interrupt(struct lg_cpu *cpu, u32 lo, u32 hi, u32 eflags, ss, irq_enable; unsigned long virtstack; - /* There are two cases for interrupts: one where the Guest is already + /* + * There are two cases for interrupts: one where the Guest is already * in the kernel, and a more complex one where the Guest is in - * userspace. We check the privilege level to find out. */ + * userspace. We check the privilege level to find out. + */ if ((cpu->regs->ss&0x3) != GUEST_PL) { - /* The Guest told us their kernel stack with the SET_STACK - * hypercall: both the virtual address and the segment */ + /* + * The Guest told us their kernel stack with the SET_STACK + * hypercall: both the virtual address and the segment. + */ virtstack = cpu->esp1; ss = cpu->ss1; origstack = gstack = guest_pa(cpu, virtstack); - /* We push the old stack segment and pointer onto the new + /* + * We push the old stack segment and pointer onto the new * stack: when the Guest does an "iret" back from the interrupt * handler the CPU will notice they're dropping privilege - * levels and expect these here. */ + * levels and expect these here. + */ push_guest_stack(cpu, &gstack, cpu->regs->ss); push_guest_stack(cpu, &gstack, cpu->regs->esp); } else { @@ -91,18 +105,22 @@ static void set_guest_interrupt(struct lg_cpu *cpu, u32 lo, u32 hi, origstack = gstack = guest_pa(cpu, virtstack); } - /* Remember that we never let the Guest actually disable interrupts, so + /* + * Remember that we never let the Guest actually disable interrupts, so * the "Interrupt Flag" bit is always set. We copy that bit from the * Guest's "irq_enabled" field into the eflags word: we saw the Guest - * copy it back in "lguest_iret". */ + * copy it back in "lguest_iret". + */ eflags = cpu->regs->eflags; if (get_user(irq_enable, &cpu->lg->lguest_data->irq_enabled) == 0 && !(irq_enable & X86_EFLAGS_IF)) eflags &= ~X86_EFLAGS_IF; - /* An interrupt is expected to push three things on the stack: the old + /* + * An interrupt is expected to push three things on the stack: the old * "eflags" word, the old code segment, and the old instruction - * pointer. */ + * pointer. + */ push_guest_stack(cpu, &gstack, eflags); push_guest_stack(cpu, &gstack, cpu->regs->cs); push_guest_stack(cpu, &gstack, cpu->regs->eip); @@ -111,15 +129,19 @@ static void set_guest_interrupt(struct lg_cpu *cpu, u32 lo, u32 hi, if (has_err) push_guest_stack(cpu, &gstack, cpu->regs->errcode); - /* Now we've pushed all the old state, we change the stack, the code - * segment and the address to execute. */ + /* + * Now we've pushed all the old state, we change the stack, the code + * segment and the address to execute. + */ cpu->regs->ss = ss; cpu->regs->esp = virtstack + (gstack - origstack); cpu->regs->cs = (__KERNEL_CS|GUEST_PL); cpu->regs->eip = idt_address(lo, hi); - /* There are two kinds of interrupt handlers: 0xE is an "interrupt - * gate" which expects interrupts to be disabled on entry. */ + /* + * There are two kinds of interrupt handlers: 0xE is an "interrupt + * gate" which expects interrupts to be disabled on entry. + */ if (idt_type(lo, hi) == 0xE) if (put_user(0, &cpu->lg->lguest_data->irq_enabled)) kill_guest(cpu, "Disabling interrupts"); @@ -130,7 +152,8 @@ static void set_guest_interrupt(struct lg_cpu *cpu, u32 lo, u32 hi, * * interrupt_pending() returns the first pending interrupt which isn't blocked * by the Guest. It is called before every entry to the Guest, and just before - * we go to sleep when the Guest has halted itself. */ + * we go to sleep when the Guest has halted itself. + */ unsigned int interrupt_pending(struct lg_cpu *cpu, bool *more) { unsigned int irq; @@ -140,8 +163,10 @@ unsigned int interrupt_pending(struct lg_cpu *cpu, bool *more) if (!cpu->lg->lguest_data) return LGUEST_IRQS; - /* Take our "irqs_pending" array and remove any interrupts the Guest - * wants blocked: the result ends up in "blk". */ + /* + * Take our "irqs_pending" array and remove any interrupts the Guest + * wants blocked: the result ends up in "blk". + */ if (copy_from_user(&blk, cpu->lg->lguest_data->blocked_interrupts, sizeof(blk))) return LGUEST_IRQS; @@ -154,16 +179,20 @@ unsigned int interrupt_pending(struct lg_cpu *cpu, bool *more) return irq; } -/* This actually diverts the Guest to running an interrupt handler, once an - * interrupt has been identified by interrupt_pending(). */ +/* + * This actually diverts the Guest to running an interrupt handler, once an + * interrupt has been identified by interrupt_pending(). + */ void try_deliver_interrupt(struct lg_cpu *cpu, unsigned int irq, bool more) { struct desc_struct *idt; BUG_ON(irq >= LGUEST_IRQS); - /* They may be in the middle of an iret, where they asked us never to - * deliver interrupts. */ + /* + * They may be in the middle of an iret, where they asked us never to + * deliver interrupts. + */ if (cpu->regs->eip >= cpu->lg->noirq_start && (cpu->regs->eip < cpu->lg->noirq_end)) return; @@ -187,29 +216,37 @@ void try_deliver_interrupt(struct lg_cpu *cpu, unsigned int irq, bool more) } } - /* Look at the IDT entry the Guest gave us for this interrupt. The + /* + * Look at the IDT entry the Guest gave us for this interrupt. The * first 32 (FIRST_EXTERNAL_VECTOR) entries are for traps, so we skip - * over them. */ + * over them. + */ idt = &cpu->arch.idt[FIRST_EXTERNAL_VECTOR+irq]; /* If they don't have a handler (yet?), we just ignore it */ if (idt_present(idt->a, idt->b)) { /* OK, mark it no longer pending and deliver it. */ clear_bit(irq, cpu->irqs_pending); - /* set_guest_interrupt() takes the interrupt descriptor and a + /* + * set_guest_interrupt() takes the interrupt descriptor and a * flag to say whether this interrupt pushes an error code onto - * the stack as well: virtual interrupts never do. */ + * the stack as well: virtual interrupts never do. + */ set_guest_interrupt(cpu, idt->a, idt->b, false); } - /* Every time we deliver an interrupt, we update the timestamp in the + /* + * Every time we deliver an interrupt, we update the timestamp in the * Guest's lguest_data struct. It would be better for the Guest if we * did this more often, but it can actually be quite slow: doing it * here is a compromise which means at least it gets updated every - * timer interrupt. */ + * timer interrupt. + */ write_timestamp(cpu); - /* If there are no other interrupts we want to deliver, clear - * the pending flag. */ + /* + * If there are no other interrupts we want to deliver, clear + * the pending flag. + */ if (!more) put_user(0, &cpu->lg->lguest_data->irq_pending); } @@ -217,24 +254,29 @@ void try_deliver_interrupt(struct lg_cpu *cpu, unsigned int irq, bool more) /* And this is the routine when we want to set an interrupt for the Guest. */ void set_interrupt(struct lg_cpu *cpu, unsigned int irq) { - /* Next time the Guest runs, the core code will see if it can deliver - * this interrupt. */ + /* + * Next time the Guest runs, the core code will see if it can deliver + * this interrupt. + */ set_bit(irq, cpu->irqs_pending); - /* Make sure it sees it; it might be asleep (eg. halted), or - * running the Guest right now, in which case kick_process() - * will knock it out. */ + /* + * Make sure it sees it; it might be asleep (eg. halted), or running + * the Guest right now, in which case kick_process() will knock it out. + */ if (!wake_up_process(cpu->tsk)) kick_process(cpu->tsk); } /*:*/ -/* Linux uses trap 128 for system calls. Plan9 uses 64, and Ron Minnich sent +/* + * Linux uses trap 128 for system calls. Plan9 uses 64, and Ron Minnich sent * me a patch, so we support that too. It'd be a big step for lguest if half * the Plan 9 user base were to start using it. * * Actually now I think of it, it's possible that Ron *is* half the Plan 9 - * userbase. Oh well. */ + * userbase. Oh well. + */ static bool could_be_syscall(unsigned int num) { /* Normal Linux SYSCALL_VECTOR or reserved vector? */ @@ -274,9 +316,11 @@ void free_interrupts(void) clear_bit(syscall_vector, used_vectors); } -/*H:220 Now we've got the routines to deliver interrupts, delivering traps like +/*H:220 + * Now we've got the routines to deliver interrupts, delivering traps like * page fault is easy. The only trick is that Intel decided that some traps - * should have error codes: */ + * should have error codes: + */ static bool has_err(unsigned int trap) { return (trap == 8 || (trap >= 10 && trap <= 14) || trap == 17); @@ -285,13 +329,17 @@ static bool has_err(unsigned int trap) /* deliver_trap() returns true if it could deliver the trap. */ bool deliver_trap(struct lg_cpu *cpu, unsigned int num) { - /* Trap numbers are always 8 bit, but we set an impossible trap number - * for traps inside the Switcher, so check that here. */ + /* + * Trap numbers are always 8 bit, but we set an impossible trap number + * for traps inside the Switcher, so check that here. + */ if (num >= ARRAY_SIZE(cpu->arch.idt)) return false; - /* Early on the Guest hasn't set the IDT entries (or maybe it put a - * bogus one in): if we fail here, the Guest will be killed. */ + /* + * Early on the Guest hasn't set the IDT entries (or maybe it put a + * bogus one in): if we fail here, the Guest will be killed. + */ if (!idt_present(cpu->arch.idt[num].a, cpu->arch.idt[num].b)) return false; set_guest_interrupt(cpu, cpu->arch.idt[num].a, @@ -299,7 +347,8 @@ bool deliver_trap(struct lg_cpu *cpu, unsigned int num) return true; } -/*H:250 Here's the hard part: returning to the Host every time a trap happens +/*H:250 + * Here's the hard part: returning to the Host every time a trap happens * and then calling deliver_trap() and re-entering the Guest is slow. * Particularly because Guest userspace system calls are traps (usually trap * 128). @@ -311,69 +360,87 @@ bool deliver_trap(struct lg_cpu *cpu, unsigned int num) * the other hypervisors would beat it up at lunchtime. * * This routine indicates if a particular trap number could be delivered - * directly. */ + * directly. + */ static bool direct_trap(unsigned int num) { - /* Hardware interrupts don't go to the Guest at all (except system - * call). */ + /* + * Hardware interrupts don't go to the Guest at all (except system + * call). + */ if (num >= FIRST_EXTERNAL_VECTOR && !could_be_syscall(num)) return false; - /* The Host needs to see page faults (for shadow paging and to save the + /* + * The Host needs to see page faults (for shadow paging and to save the * fault address), general protection faults (in/out emulation) and * device not available (TS handling), invalid opcode fault (kvm hcall), - * and of course, the hypercall trap. */ + * and of course, the hypercall trap. + */ return num != 14 && num != 13 && num != 7 && num != 6 && num != LGUEST_TRAP_ENTRY; } /*:*/ -/*M:005 The Guest has the ability to turn its interrupt gates into trap gates, +/*M:005 + * The Guest has the ability to turn its interrupt gates into trap gates, * if it is careful. The Host will let trap gates can go directly to the * Guest, but the Guest needs the interrupts atomically disabled for an * interrupt gate. It can do this by pointing the trap gate at instructions - * within noirq_start and noirq_end, where it can safely disable interrupts. */ + * within noirq_start and noirq_end, where it can safely disable interrupts. + */ -/*M:006 The Guests do not use the sysenter (fast system call) instruction, +/*M:006 + * The Guests do not use the sysenter (fast system call) instruction, * because it's hardcoded to enter privilege level 0 and so can't go direct. * It's about twice as fast as the older "int 0x80" system call, so it might * still be worthwhile to handle it in the Switcher and lcall down to the * Guest. The sysenter semantics are hairy tho: search for that keyword in - * entry.S :*/ + * entry.S +:*/ -/*H:260 When we make traps go directly into the Guest, we need to make sure +/*H:260 + * When we make traps go directly into the Guest, we need to make sure * the kernel stack is valid (ie. mapped in the page tables). Otherwise, the * CPU trying to deliver the trap will fault while trying to push the interrupt * words on the stack: this is called a double fault, and it forces us to kill * the Guest. * - * Which is deeply unfair, because (literally!) it wasn't the Guests' fault. */ + * Which is deeply unfair, because (literally!) it wasn't the Guests' fault. + */ void pin_stack_pages(struct lg_cpu *cpu) { unsigned int i; - /* Depending on the CONFIG_4KSTACKS option, the Guest can have one or - * two pages of stack space. */ + /* + * Depending on the CONFIG_4KSTACKS option, the Guest can have one or + * two pages of stack space. + */ for (i = 0; i < cpu->lg->stack_pages; i++) - /* The stack grows *upwards*, so the address we're given is the + /* + * The stack grows *upwards*, so the address we're given is the * start of the page after the kernel stack. Subtract one to * get back onto the first stack page, and keep subtracting to - * get to the rest of the stack pages. */ + * get to the rest of the stack pages. + */ pin_page(cpu, cpu->esp1 - 1 - i * PAGE_SIZE); } -/* Direct traps also mean that we need to know whenever the Guest wants to use +/* + * Direct traps also mean that we need to know whenever the Guest wants to use * a different kernel stack, so we can change the IDT entries to use that * stack. The IDT entries expect a virtual address, so unlike most addresses * the Guest gives us, the "esp" (stack pointer) value here is virtual, not * physical. * * In Linux each process has its own kernel stack, so this happens a lot: we - * change stacks on each context switch. */ + * change stacks on each context switch. + */ void guest_set_stack(struct lg_cpu *cpu, u32 seg, u32 esp, unsigned int pages) { - /* You are not allowed have a stack segment with privilege level 0: bad - * Guest! */ + /* + * You're not allowed a stack segment with privilege level 0: bad Guest! + */ if ((seg & 0x3) != GUEST_PL) kill_guest(cpu, "bad stack segment %i", seg); /* We only expect one or two stack pages. */ @@ -387,11 +454,15 @@ void guest_set_stack(struct lg_cpu *cpu, u32 seg, u32 esp, unsigned int pages) pin_stack_pages(cpu); } -/* All this reference to mapping stacks leads us neatly into the other complex - * part of the Host: page table handling. */ +/* + * All this reference to mapping stacks leads us neatly into the other complex + * part of the Host: page table handling. + */ -/*H:235 This is the routine which actually checks the Guest's IDT entry and - * transfers it into the entry in "struct lguest": */ +/*H:235 + * This is the routine which actually checks the Guest's IDT entry and + * transfers it into the entry in "struct lguest": + */ static void set_trap(struct lg_cpu *cpu, struct desc_struct *trap, unsigned int num, u32 lo, u32 hi) { @@ -407,30 +478,38 @@ static void set_trap(struct lg_cpu *cpu, struct desc_struct *trap, if (type != 0xE && type != 0xF) kill_guest(cpu, "bad IDT type %i", type); - /* We only copy the handler address, present bit, privilege level and + /* + * We only copy the handler address, present bit, privilege level and * type. The privilege level controls where the trap can be triggered * manually with an "int" instruction. This is usually GUEST_PL, - * except for system calls which userspace can use. */ + * except for system calls which userspace can use. + */ trap->a = ((__KERNEL_CS|GUEST_PL)<<16) | (lo&0x0000FFFF); trap->b = (hi&0xFFFFEF00); } -/*H:230 While we're here, dealing with delivering traps and interrupts to the +/*H:230 + * While we're here, dealing with delivering traps and interrupts to the * Guest, we might as well complete the picture: how the Guest tells us where * it wants them to go. This would be simple, except making traps fast * requires some tricks. * * We saw the Guest setting Interrupt Descriptor Table (IDT) entries with the - * LHCALL_LOAD_IDT_ENTRY hypercall before: that comes here. */ + * LHCALL_LOAD_IDT_ENTRY hypercall before: that comes here. + */ void load_guest_idt_entry(struct lg_cpu *cpu, unsigned int num, u32 lo, u32 hi) { - /* Guest never handles: NMI, doublefault, spurious interrupt or - * hypercall. We ignore when it tries to set them. */ + /* + * Guest never handles: NMI, doublefault, spurious interrupt or + * hypercall. We ignore when it tries to set them. + */ if (num == 2 || num == 8 || num == 15 || num == LGUEST_TRAP_ENTRY) return; - /* Mark the IDT as changed: next time the Guest runs we'll know we have - * to copy this again. */ + /* + * Mark the IDT as changed: next time the Guest runs we'll know we have + * to copy this again. + */ cpu->changed |= CHANGED_IDT; /* Check that the Guest doesn't try to step outside the bounds. */ @@ -440,9 +519,11 @@ void load_guest_idt_entry(struct lg_cpu *cpu, unsigned int num, u32 lo, u32 hi) set_trap(cpu, &cpu->arch.idt[num], num, lo, hi); } -/* The default entry for each interrupt points into the Switcher routines which +/* + * The default entry for each interrupt points into the Switcher routines which * simply return to the Host. The run_guest() loop will then call - * deliver_trap() to bounce it back into the Guest. */ + * deliver_trap() to bounce it back into the Guest. + */ static void default_idt_entry(struct desc_struct *idt, int trap, const unsigned long handler, @@ -451,13 +532,17 @@ static void default_idt_entry(struct desc_struct *idt, /* A present interrupt gate. */ u32 flags = 0x8e00; - /* Set the privilege level on the entry for the hypercall: this allows - * the Guest to use the "int" instruction to trigger it. */ + /* + * Set the privilege level on the entry for the hypercall: this allows + * the Guest to use the "int" instruction to trigger it. + */ if (trap == LGUEST_TRAP_ENTRY) flags |= (GUEST_PL << 13); else if (base) - /* Copy priv. level from what Guest asked for. This allows - * debug (int 3) traps from Guest userspace, for example. */ + /* + * Copy privilege level from what Guest asked for. This allows + * debug (int 3) traps from Guest userspace, for example. + */ flags |= (base->b & 0x6000); /* Now pack it into the IDT entry in its weird format. */ @@ -475,16 +560,20 @@ void setup_default_idt_entries(struct lguest_ro_state *state, default_idt_entry(&state->guest_idt[i], i, def[i], NULL); } -/*H:240 We don't use the IDT entries in the "struct lguest" directly, instead +/*H:240 + * We don't use the IDT entries in the "struct lguest" directly, instead * we copy them into the IDT which we've set up for Guests on this CPU, just - * before we run the Guest. This routine does that copy. */ + * before we run the Guest. This routine does that copy. + */ void copy_traps(const struct lg_cpu *cpu, struct desc_struct *idt, const unsigned long *def) { unsigned int i; - /* We can simply copy the direct traps, otherwise we use the default - * ones in the Switcher: they will return to the Host. */ + /* + * We can simply copy the direct traps, otherwise we use the default + * ones in the Switcher: they will return to the Host. + */ for (i = 0; i < ARRAY_SIZE(cpu->arch.idt); i++) { const struct desc_struct *gidt = &cpu->arch.idt[i]; @@ -492,14 +581,16 @@ void copy_traps(const struct lg_cpu *cpu, struct desc_struct *idt, if (!direct_trap(i)) continue; - /* Only trap gates (type 15) can go direct to the Guest. + /* + * Only trap gates (type 15) can go direct to the Guest. * Interrupt gates (type 14) disable interrupts as they are * entered, which we never let the Guest do. Not present * entries (type 0x0) also can't go direct, of course. * * If it can't go direct, we still need to copy the priv. level: * they might want to give userspace access to a software - * interrupt. */ + * interrupt. + */ if (idt_type(gidt->a, gidt->b) == 0xF) idt[i] = *gidt; else @@ -518,7 +609,8 @@ void copy_traps(const struct lg_cpu *cpu, struct desc_struct *idt, * the next timer interrupt (in nanoseconds). We use the high-resolution timer * infrastructure to set a callback at that time. * - * 0 means "turn off the clock". */ + * 0 means "turn off the clock". + */ void guest_set_clockevent(struct lg_cpu *cpu, unsigned long delta) { ktime_t expires; @@ -529,9 +621,11 @@ void guest_set_clockevent(struct lg_cpu *cpu, unsigned long delta) return; } - /* We use wallclock time here, so the Guest might not be running for + /* + * We use wallclock time here, so the Guest might not be running for * all the time between now and the timer interrupt it asked for. This - * is almost always the right thing to do. */ + * is almost always the right thing to do. + */ expires = ktime_add_ns(ktime_get_real(), delta); hrtimer_start(&cpu->hrt, expires, HRTIMER_MODE_ABS); } diff --git a/drivers/lguest/lg.h b/drivers/lguest/lg.h index 01c59192379..74c0db691b5 100644 --- a/drivers/lguest/lg.h +++ b/drivers/lguest/lg.h @@ -54,13 +54,13 @@ struct lg_cpu { unsigned long pending_notify; /* pfn from LHCALL_NOTIFY */ - /* At end of a page shared mapped over lguest_pages in guest. */ + /* At end of a page shared mapped over lguest_pages in guest. */ unsigned long regs_page; struct lguest_regs *regs; struct lguest_pages *last_pages; - int cpu_pgd; /* which pgd this cpu is currently using */ + int cpu_pgd; /* Which pgd this cpu is currently using */ /* If a hypercall was asked for, this points to the arguments. */ struct hcall_args *hcall; @@ -96,8 +96,11 @@ struct lguest unsigned int nr_cpus; u32 pfn_limit; - /* This provides the offset to the base of guest-physical - * memory in the Launcher. */ + + /* + * This provides the offset to the base of guest-physical memory in the + * Launcher. + */ void __user *mem_base; unsigned long kernel_address; @@ -122,11 +125,13 @@ bool lguest_address_ok(const struct lguest *lg, void __lgread(struct lg_cpu *, void *, unsigned long, unsigned); void __lgwrite(struct lg_cpu *, unsigned long, const void *, unsigned); -/*H:035 Using memory-copy operations like that is usually inconvient, so we +/*H:035 + * Using memory-copy operations like that is usually inconvient, so we * have the following helper macros which read and write a specific type (often * an unsigned long). * - * This reads into a variable of the given type then returns that. */ + * This reads into a variable of the given type then returns that. + */ #define lgread(cpu, addr, type) \ ({ type _v; __lgread((cpu), &_v, (addr), sizeof(_v)); _v; }) @@ -140,9 +145,11 @@ void __lgwrite(struct lg_cpu *, unsigned long, const void *, unsigned); int run_guest(struct lg_cpu *cpu, unsigned long __user *user); -/* Helper macros to obtain the first 12 or the last 20 bits, this is only the +/* + * Helper macros to obtain the first 12 or the last 20 bits, this is only the * first step in the migration to the kernel types. pte_pfn is already defined - * in the kernel. */ + * in the kernel. + */ #define pgd_flags(x) (pgd_val(x) & ~PAGE_MASK) #define pgd_pfn(x) (pgd_val(x) >> PAGE_SHIFT) #define pmd_flags(x) (pmd_val(x) & ~PAGE_MASK) diff --git a/drivers/lguest/lguest_device.c b/drivers/lguest/lguest_device.c index e082cdac88b..cc000e79c3d 100644 --- a/drivers/lguest/lguest_device.c +++ b/drivers/lguest/lguest_device.c @@ -1,10 +1,12 @@ -/*P:050 Lguest guests use a very simple method to describe devices. It's a +/*P:050 + * Lguest guests use a very simple method to describe devices. It's a * series of device descriptors contained just above the top of normal Guest * memory. * * We use the standard "virtio" device infrastructure, which provides us with a * console, a network and a block driver. Each one expects some configuration - * information and a "virtqueue" or two to send and receive data. :*/ + * information and a "virtqueue" or two to send and receive data. +:*/ #include #include #include @@ -20,8 +22,10 @@ /* The pointer to our (page) of device descriptions. */ static void *lguest_devices; -/* For Guests, device memory can be used as normal memory, so we cast away the - * __iomem to quieten sparse. */ +/* + * For Guests, device memory can be used as normal memory, so we cast away the + * __iomem to quieten sparse. + */ static inline void *lguest_map(unsigned long phys_addr, unsigned long pages) { return (__force void *)ioremap_cache(phys_addr, PAGE_SIZE*pages); @@ -32,8 +36,10 @@ static inline void lguest_unmap(void *addr) iounmap((__force void __iomem *)addr); } -/*D:100 Each lguest device is just a virtio device plus a pointer to its entry - * in the lguest_devices page. */ +/*D:100 + * Each lguest device is just a virtio device plus a pointer to its entry + * in the lguest_devices page. + */ struct lguest_device { struct virtio_device vdev; @@ -41,9 +47,11 @@ struct lguest_device { struct lguest_device_desc *desc; }; -/* Since the virtio infrastructure hands us a pointer to the virtio_device all +/* + * Since the virtio infrastructure hands us a pointer to the virtio_device all * the time, it helps to have a curt macro to get a pointer to the struct - * lguest_device it's enclosed in. */ + * lguest_device it's enclosed in. + */ #define to_lgdev(vd) container_of(vd, struct lguest_device, vdev) /*D:130 @@ -55,7 +63,8 @@ struct lguest_device { * the driver will look at them during setup. * * A convenient routine to return the device's virtqueue config array: - * immediately after the descriptor. */ + * immediately after the descriptor. + */ static struct lguest_vqconfig *lg_vq(const struct lguest_device_desc *desc) { return (void *)(desc + 1); @@ -98,10 +107,12 @@ static u32 lg_get_features(struct virtio_device *vdev) return features; } -/* The virtio core takes the features the Host offers, and copies the - * ones supported by the driver into the vdev->features array. Once - * that's all sorted out, this routine is called so we can tell the - * Host which features we understand and accept. */ +/* + * The virtio core takes the features the Host offers, and copies the ones + * supported by the driver into the vdev->features array. Once that's all + * sorted out, this routine is called so we can tell the Host which features we + * understand and accept. + */ static void lg_finalize_features(struct virtio_device *vdev) { unsigned int i, bits; @@ -112,10 +123,11 @@ static void lg_finalize_features(struct virtio_device *vdev) /* Give virtio_ring a chance to accept features. */ vring_transport_features(vdev); - /* The vdev->feature array is a Linux bitmask: this isn't the - * same as a the simple array of bits used by lguest devices - * for features. So we do this slow, manual conversion which is - * completely general. */ + /* + * The vdev->feature array is a Linux bitmask: this isn't the same as a + * the simple array of bits used by lguest devices for features. So we + * do this slow, manual conversion which is completely general. + */ memset(out_features, 0, desc->feature_len); bits = min_t(unsigned, desc->feature_len, sizeof(vdev->features)) * 8; for (i = 0; i < bits; i++) { @@ -146,15 +158,19 @@ static void lg_set(struct virtio_device *vdev, unsigned int offset, memcpy(lg_config(desc) + offset, buf, len); } -/* The operations to get and set the status word just access the status field - * of the device descriptor. */ +/* + * The operations to get and set the status word just access the status field + * of the device descriptor. + */ static u8 lg_get_status(struct virtio_device *vdev) { return to_lgdev(vdev)->desc->status; } -/* To notify on status updates, we (ab)use the NOTIFY hypercall, with the - * descriptor address of the device. A zero status means "reset". */ +/* + * To notify on status updates, we (ab)use the NOTIFY hypercall, with the + * descriptor address of the device. A zero status means "reset". + */ static void set_status(struct virtio_device *vdev, u8 status) { unsigned long offset = (void *)to_lgdev(vdev)->desc - lguest_devices; @@ -200,13 +216,17 @@ struct lguest_vq_info void *pages; }; -/* When the virtio_ring code wants to prod the Host, it calls us here and we +/* + * When the virtio_ring code wants to prod the Host, it calls us here and we * make a hypercall. We hand the physical address of the virtqueue so the Host - * knows which virtqueue we're talking about. */ + * knows which virtqueue we're talking about. + */ static void lg_notify(struct virtqueue *vq) { - /* We store our virtqueue information in the "priv" pointer of the - * virtqueue structure. */ + /* + * We store our virtqueue information in the "priv" pointer of the + * virtqueue structure. + */ struct lguest_vq_info *lvq = vq->priv; kvm_hypercall1(LHCALL_NOTIFY, lvq->config.pfn << PAGE_SHIFT); @@ -215,7 +235,8 @@ static void lg_notify(struct virtqueue *vq) /* An extern declaration inside a C file is bad form. Don't do it. */ extern void lguest_setup_irq(unsigned int irq); -/* This routine finds the first virtqueue described in the configuration of +/* + * This routine finds the first virtqueue described in the configuration of * this device and sets it up. * * This is kind of an ugly duckling. It'd be nicer to have a standard @@ -225,7 +246,8 @@ extern void lguest_setup_irq(unsigned int irq); * simpler for the Host to simply tell us where the pages are. * * So we provide drivers with a "find the Nth virtqueue and set it up" - * function. */ + * function. + */ static struct virtqueue *lg_find_vq(struct virtio_device *vdev, unsigned index, void (*callback)(struct virtqueue *vq), @@ -244,9 +266,11 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev, if (!lvq) return ERR_PTR(-ENOMEM); - /* Make a copy of the "struct lguest_vqconfig" entry, which sits after + /* + * Make a copy of the "struct lguest_vqconfig" entry, which sits after * the descriptor. We need a copy because the config space might not - * be aligned correctly. */ + * be aligned correctly. + */ memcpy(&lvq->config, lg_vq(ldev->desc)+index, sizeof(lvq->config)); printk("Mapping virtqueue %i addr %lx\n", index, @@ -261,8 +285,10 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev, goto free_lvq; } - /* OK, tell virtio_ring.c to set up a virtqueue now we know its size - * and we've got a pointer to its pages. */ + /* + * OK, tell virtio_ring.c to set up a virtqueue now we know its size + * and we've got a pointer to its pages. + */ vq = vring_new_virtqueue(lvq->config.num, LGUEST_VRING_ALIGN, vdev, lvq->pages, lg_notify, callback, name); if (!vq) { @@ -273,18 +299,23 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev, /* Make sure the interrupt is allocated. */ lguest_setup_irq(lvq->config.irq); - /* Tell the interrupt for this virtqueue to go to the virtio_ring - * interrupt handler. */ - /* FIXME: We used to have a flag for the Host to tell us we could use + /* + * Tell the interrupt for this virtqueue to go to the virtio_ring + * interrupt handler. + * + * FIXME: We used to have a flag for the Host to tell us we could use * the interrupt as a source of randomness: it'd be nice to have that - * back.. */ + * back. + */ err = request_irq(lvq->config.irq, vring_interrupt, IRQF_SHARED, dev_name(&vdev->dev), vq); if (err) goto destroy_vring; - /* Last of all we hook up our 'struct lguest_vq_info" to the - * virtqueue's priv pointer. */ + /* + * Last of all we hook up our 'struct lguest_vq_info" to the + * virtqueue's priv pointer. + */ vq->priv = lvq; return vq; @@ -358,11 +389,14 @@ static struct virtio_config_ops lguest_config_ops = { .del_vqs = lg_del_vqs, }; -/* The root device for the lguest virtio devices. This makes them appear as - * /sys/devices/lguest/0,1,2 not /sys/devices/0,1,2. */ +/* + * The root device for the lguest virtio devices. This makes them appear as + * /sys/devices/lguest/0,1,2 not /sys/devices/0,1,2. + */ static struct device *lguest_root; -/*D:120 This is the core of the lguest bus: actually adding a new device. +/*D:120 + * This is the core of the lguest bus: actually adding a new device. * It's a separate function because it's neater that way, and because an * earlier version of the code supported hotplug and unplug. They were removed * early on because they were never used. @@ -371,14 +405,14 @@ static struct device *lguest_root; * * It's worth reading this carefully: we start with a pointer to the new device * descriptor in the "lguest_devices" page, and the offset into the device - * descriptor page so we can uniquely identify it if things go badly wrong. */ + * descriptor page so we can uniquely identify it if things go badly wrong. + */ static void add_lguest_device(struct lguest_device_desc *d, unsigned int offset) { struct lguest_device *ldev; - /* Start with zeroed memory; Linux's device layer seems to count on - * it. */ + /* Start with zeroed memory; Linux's device layer counts on it. */ ldev = kzalloc(sizeof(*ldev), GFP_KERNEL); if (!ldev) { printk(KERN_EMERG "Cannot allocate lguest dev %u type %u\n", @@ -390,15 +424,19 @@ static void add_lguest_device(struct lguest_device_desc *d, ldev->vdev.dev.parent = lguest_root; /* We have a unique device index thanks to the dev_index counter. */ ldev->vdev.id.device = d->type; - /* We have a simple set of routines for querying the device's - * configuration information and setting its status. */ + /* + * We have a simple set of routines for querying the device's + * configuration information and setting its status. + */ ldev->vdev.config = &lguest_config_ops; /* And we remember the device's descriptor for lguest_config_ops. */ ldev->desc = d; - /* register_virtio_device() sets up the generic fields for the struct + /* + * register_virtio_device() sets up the generic fields for the struct * virtio_device and calls device_register(). This makes the bus - * infrastructure look for a matching driver. */ + * infrastructure look for a matching driver. + */ if (register_virtio_device(&ldev->vdev) != 0) { printk(KERN_ERR "Failed to register lguest dev %u type %u\n", offset, d->type); @@ -406,8 +444,10 @@ static void add_lguest_device(struct lguest_device_desc *d, } } -/*D:110 scan_devices() simply iterates through the device page. The type 0 is - * reserved to mean "end of devices". */ +/*D:110 + * scan_devices() simply iterates through the device page. The type 0 is + * reserved to mean "end of devices". + */ static void scan_devices(void) { unsigned int i; @@ -426,7 +466,8 @@ static void scan_devices(void) } } -/*D:105 Fairly early in boot, lguest_devices_init() is called to set up the +/*D:105 + * Fairly early in boot, lguest_devices_init() is called to set up the * lguest device infrastructure. We check that we are a Guest by checking * pv_info.name: there are other ways of checking, but this seems most * obvious to me. @@ -437,7 +478,8 @@ static void scan_devices(void) * correct sysfs incantation). * * Finally we call scan_devices() which adds all the devices found in the - * lguest_devices page. */ + * lguest_devices page. + */ static int __init lguest_devices_init(void) { if (strcmp(pv_info.name, "lguest") != 0) @@ -456,11 +498,13 @@ static int __init lguest_devices_init(void) /* We do this after core stuff, but before the drivers. */ postcore_initcall(lguest_devices_init); -/*D:150 At this point in the journey we used to now wade through the lguest +/*D:150 + * At this point in the journey we used to now wade through the lguest * devices themselves: net, block and console. Since they're all now virtio * devices rather than lguest-specific, I've decided to ignore them. Mostly, * they're kind of boring. But this does mean you'll never experience the * thrill of reading the forbidden love scene buried deep in the block driver. * * "make Launcher" beckons, where we answer questions like "Where do Guests - * come from?", and "What do you do when someone asks for optimization?". */ + * come from?", and "What do you do when someone asks for optimization?". + */ diff --git a/drivers/lguest/lguest_user.c b/drivers/lguest/lguest_user.c index 407722a8e0c..7e92017103d 100644 --- a/drivers/lguest/lguest_user.c +++ b/drivers/lguest/lguest_user.c @@ -1,8 +1,10 @@ -/*P:200 This contains all the /dev/lguest code, whereby the userspace launcher +/*P:200 + * This contains all the /dev/lguest code, whereby the userspace launcher * controls and communicates with the Guest. For example, the first write will * tell us the Guest's memory layout, pagetable, entry point and kernel address * offset. A read will run the Guest until something happens, such as a signal - * or the Guest doing a NOTIFY out to the Launcher. :*/ + * or the Guest doing a NOTIFY out to the Launcher. +:*/ #include #include #include @@ -37,8 +39,10 @@ static int add_eventfd(struct lguest *lg, unsigned long addr, int fd) if (!addr) return -EINVAL; - /* Replace the old array with the new one, carefully: others can - * be accessing it at the same time */ + /* + * Replace the old array with the new one, carefully: others can + * be accessing it at the same time. + */ new = kmalloc(sizeof(*new) + sizeof(new->map[0]) * (old->num + 1), GFP_KERNEL); if (!new) @@ -61,8 +65,10 @@ static int add_eventfd(struct lguest *lg, unsigned long addr, int fd) /* Now put new one in place. */ rcu_assign_pointer(lg->eventfds, new); - /* We're not in a big hurry. Wait until noone's looking at old - * version, then delete it. */ + /* + * We're not in a big hurry. Wait until noone's looking at old + * version, then delete it. + */ synchronize_rcu(); kfree(old); @@ -87,8 +93,10 @@ static int attach_eventfd(struct lguest *lg, const unsigned long __user *input) return err; } -/*L:050 Sending an interrupt is done by writing LHREQ_IRQ and an interrupt - * number to /dev/lguest. */ +/*L:050 + * Sending an interrupt is done by writing LHREQ_IRQ and an interrupt + * number to /dev/lguest. + */ static int user_send_irq(struct lg_cpu *cpu, const unsigned long __user *input) { unsigned long irq; @@ -102,8 +110,10 @@ static int user_send_irq(struct lg_cpu *cpu, const unsigned long __user *input) return 0; } -/*L:040 Once our Guest is initialized, the Launcher makes it run by reading - * from /dev/lguest. */ +/*L:040 + * Once our Guest is initialized, the Launcher makes it run by reading + * from /dev/lguest. + */ static ssize_t read(struct file *file, char __user *user, size_t size,loff_t*o) { struct lguest *lg = file->private_data; @@ -139,8 +149,10 @@ static ssize_t read(struct file *file, char __user *user, size_t size,loff_t*o) return len; } - /* If we returned from read() last time because the Guest sent I/O, - * clear the flag. */ + /* + * If we returned from read() last time because the Guest sent I/O, + * clear the flag. + */ if (cpu->pending_notify) cpu->pending_notify = 0; @@ -148,8 +160,10 @@ static ssize_t read(struct file *file, char __user *user, size_t size,loff_t*o) return run_guest(cpu, (unsigned long __user *)user); } -/*L:025 This actually initializes a CPU. For the moment, a Guest is only - * uniprocessor, so "id" is always 0. */ +/*L:025 + * This actually initializes a CPU. For the moment, a Guest is only + * uniprocessor, so "id" is always 0. + */ static int lg_cpu_start(struct lg_cpu *cpu, unsigned id, unsigned long start_ip) { /* We have a limited number the number of CPUs in the lguest struct. */ @@ -164,8 +178,10 @@ static int lg_cpu_start(struct lg_cpu *cpu, unsigned id, unsigned long start_ip) /* Each CPU has a timer it can set. */ init_clockdev(cpu); - /* We need a complete page for the Guest registers: they are accessible - * to the Guest and we can only grant it access to whole pages. */ + /* + * We need a complete page for the Guest registers: they are accessible + * to the Guest and we can only grant it access to whole pages. + */ cpu->regs_page = get_zeroed_page(GFP_KERNEL); if (!cpu->regs_page) return -ENOMEM; @@ -173,29 +189,38 @@ static int lg_cpu_start(struct lg_cpu *cpu, unsigned id, unsigned long start_ip) /* We actually put the registers at the bottom of the page. */ cpu->regs = (void *)cpu->regs_page + PAGE_SIZE - sizeof(*cpu->regs); - /* Now we initialize the Guest's registers, handing it the start - * address. */ + /* + * Now we initialize the Guest's registers, handing it the start + * address. + */ lguest_arch_setup_regs(cpu, start_ip); - /* We keep a pointer to the Launcher task (ie. current task) for when - * other Guests want to wake this one (eg. console input). */ + /* + * We keep a pointer to the Launcher task (ie. current task) for when + * other Guests want to wake this one (eg. console input). + */ cpu->tsk = current; - /* We need to keep a pointer to the Launcher's memory map, because if + /* + * We need to keep a pointer to the Launcher's memory map, because if * the Launcher dies we need to clean it up. If we don't keep a - * reference, it is destroyed before close() is called. */ + * reference, it is destro