summaryrefslogtreecommitdiff
path: root/drivers/lguest/x86/switcher_32.S
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/lguest/x86/switcher_32.S')
-rw-r--r--drivers/lguest/x86/switcher_32.S388
1 files changed, 0 insertions, 388 deletions
diff --git a/drivers/lguest/x86/switcher_32.S b/drivers/lguest/x86/switcher_32.S
deleted file mode 100644
index 40634b0db9f7..000000000000
--- a/drivers/lguest/x86/switcher_32.S
+++ /dev/null
@@ -1,388 +0,0 @@
-/*P:900
- * This is the Switcher: code which sits at 0xFFC00000 (or 0xFFE00000) astride
- * both the Host and Guest to do the low-level Guest<->Host switch. It is as
- * simple as it can be made, but it's naturally very specific to x86.
- *
- * You have now completed Preparation. If this has whet your appetite; if you
- * are feeling invigorated and refreshed then the next, more challenging stage
- * can be found in "make Guest".
- :*/
-
-/*M:012
- * Lguest is meant to be simple: my rule of thumb is that 1% more LOC must
- * gain at least 1% more performance. Since neither LOC nor performance can be
- * measured beforehand, it generally means implementing a feature then deciding
- * if it's worth it. And once it's implemented, who can say no?
- *
- * This is why I haven't implemented this idea myself. I want to, but I
- * haven't. You could, though.
- *
- * The main place where lguest performance sucks is Guest page faulting. When
- * a Guest userspace process hits an unmapped page we switch back to the Host,
- * walk the page tables, find it's not mapped, switch back to the Guest page
- * fault handler, which calls a hypercall to set the page table entry, then
- * finally returns to userspace. That's two round-trips.
- *
- * If we had a small walker in the Switcher, we could quickly check the Guest
- * page table and if the page isn't mapped, immediately reflect the fault back
- * into the Guest. This means the Switcher would have to know the top of the
- * Guest page table and the page fault handler address.
- *
- * For simplicity, the Guest should only handle the case where the privilege
- * level of the fault is 3 and probably only not present or write faults. It
- * should also detect recursive faults, and hand the original fault to the
- * Host (which is actually really easy).
- *
- * Two questions remain. Would the performance gain outweigh the complexity?
- * And who would write the verse documenting it?
-:*/
-
-/*M:011
- * Lguest64 handles NMI. This gave me NMI envy (until I looked at their
- * code). It's worth doing though, since it would let us use oprofile in the
- * Host when a Guest is running.
-:*/
-
-/*S:100
- * Welcome to the Switcher itself!
- *
- * This file contains the low-level code which changes the CPU to run the Guest
- * code, and returns to the Host when something happens. Understand this, and
- * you understand the heart of our journey.
- *
- * Because this is in assembler rather than C, our tale switches from prose to
- * verse. First I tried limericks:
- *
- * There once was an eax reg,
- * To which our pointer was fed,
- * It needed an add,
- * Which asm-offsets.h had
- * But this limerick is hurting my head.
- *
- * Next I tried haikus, but fitting the required reference to the seasons in
- * every stanza was quickly becoming tiresome:
- *
- * The %eax reg
- * Holds "struct lguest_pages" now:
- * Cherry blossoms fall.
- *
- * Then I started with Heroic Verse, but the rhyming requirement leeched away
- * the content density and led to some uniquely awful oblique rhymes:
- *
- * These constants are coming from struct offsets
- * For use within the asm switcher text.
- *
- * Finally, I settled for something between heroic hexameter, and normal prose
- * with inappropriate linebreaks. Anyway, it aint no Shakespeare.
- */
-
-// Not all kernel headers work from assembler
-// But these ones are needed: the ENTRY() define
-// And constants extracted from struct offsets
-// To avoid magic numbers and breakage:
-// Should they change the compiler can't save us
-// Down here in the depths of assembler code.
-#include <linux/linkage.h>
-#include <asm/asm-offsets.h>
-#include <asm/page.h>
-#include <asm/segment.h>
-#include <asm/lguest.h>
-
-// We mark the start of the code to copy
-// It's placed in .text tho it's never run here
-// You'll see the trick macro at the end
-// Which interleaves data and text to effect.
-.text
-ENTRY(start_switcher_text)
-
-// When we reach switch_to_guest we have just left
-// The safe and comforting shores of C code
-// %eax has the "struct lguest_pages" to use
-// Where we save state and still see it from the Guest
-// And %ebx holds the Guest shadow pagetable:
-// Once set we have truly left Host behind.
-ENTRY(switch_to_guest)
- // We told gcc all its regs could fade,
- // Clobbered by our journey into the Guest
- // We could have saved them, if we tried
- // But time is our master and cycles count.
-
- // Segment registers must be saved for the Host
- // We push them on the Host stack for later
- pushl %es
- pushl %ds
- pushl %gs
- pushl %fs
- // But the compiler is fickle, and heeds
- // No warning of %ebp clobbers
- // When frame pointers are used. That register
- // Must be saved and restored or chaos strikes.
- pushl %ebp
- // The Host's stack is done, now save it away
- // In our "struct lguest_pages" at offset
- // Distilled into asm-offsets.h
- movl %esp, LGUEST_PAGES_host_sp(%eax)
-
- // All saved and there's now five steps before us:
- // Stack, GDT, IDT, TSS
- // Then last of all the page tables are flipped.
-
- // Yet beware that our stack pointer must be
- // Always valid lest an NMI hits
- // %edx does the duty here as we juggle
- // %eax is lguest_pages: our stack lies within.
- movl %eax, %edx
- addl $LGUEST_PAGES_regs, %edx
- movl %edx, %esp
-
- // The Guest's GDT we so carefully
- // Placed in the "struct lguest_pages" before
- lgdt LGUEST_PAGES_guest_gdt_desc(%eax)
-
- // The Guest's IDT we did partially
- // Copy to "struct lguest_pages" as well.
- lidt LGUEST_PAGES_guest_idt_desc(%eax)
-
- // The TSS entry which controls traps
- // Must be loaded up with "ltr" now:
- // The GDT entry that TSS uses
- // Changes type when we load it: damn Intel!
- // For after we switch over our page tables
- // That entry will be read-only: we'd crash.
- movl $(GDT_ENTRY_TSS*8), %edx
- ltr %dx
-
- // Look back now, before we take this last step!
- // The Host's TSS entry was also marked used;
- // Let's clear it again for our return.
- // The GDT descriptor of the Host
- // Points to the table after two "size" bytes
- movl (LGUEST_PAGES_host_gdt_desc+2)(%eax), %edx
- // Clear "used" from type field (byte 5, bit 2)
- andb $0xFD, (GDT_ENTRY_TSS*8 + 5)(%edx)
-
- // Once our page table's switched, the Guest is live!
- // The Host fades as we run this final step.
- // Our "struct lguest_pages" is now read-only.
- movl %ebx, %cr3
-
- // The page table change did one tricky thing:
- // The Guest's register page has been mapped
- // Writable under our %esp (stack) --
- // We can simply pop off all Guest regs.
- popl %eax
- popl %ebx
- popl %ecx
- popl %edx
- popl %esi
- popl %edi
- popl %ebp
- popl %gs
- popl %fs
- popl %ds
- popl %es
-
- // Near the base of the stack lurk two strange fields
- // Which we fill as we exit the Guest
- // These are the trap number and its error
- // We can simply step past them on our way.
- addl $8, %esp
-
- // The last five stack slots hold return address
- // And everything needed to switch privilege
- // From Switcher's level 0 to Guest's 1,
- // And the stack where the Guest had last left it.
- // Interrupts are turned back on: we are Guest.
- iret
-
-// We tread two paths to switch back to the Host
-// Yet both must save Guest state and restore Host
-// So we put the routine in a macro.
-#define SWITCH_TO_HOST \
- /* We save the Guest state: all registers first \
- * Laid out just as "struct lguest_regs" defines */ \
- pushl %es; \
- pushl %ds; \
- pushl %fs; \
- pushl %gs; \
- pushl %ebp; \
- pushl %edi; \
- pushl %esi; \
- pushl %edx; \
- pushl %ecx; \
- pushl %ebx; \
- pushl %eax; \
- /* Our stack and our code are using segments \
- * Set in the TSS and IDT \
- * Yet if we were to touch data we'd use \
- * Whatever data segment the Guest had. \
- * Load the lguest ds segment for now. */ \
- movl $(LGUEST_DS), %eax; \
- movl %eax, %ds; \
- /* So where are we? Which CPU, which struct? \
- * The stack is our clue: our TSS starts \
- * It at the end of "struct lguest_pages". \
- * Or we may have stumbled while restoring \
- * Our Guest segment regs while in switch_to_guest, \
- * The fault pushed atop that part-unwound stack. \
- * If we round the stack down to the page start \
- * We're at the start of "struct lguest_pages". */ \
- movl %esp, %eax; \
- andl $(~(1 << PAGE_SHIFT - 1)), %eax; \
- /* Save our trap number: the switch will obscure it \
- * (In the Host the Guest regs are not mapped here) \
- * %ebx holds it safe for deliver_to_host */ \
- movl LGUEST_PAGES_regs_trapnum(%eax), %ebx; \
- /* The Host GDT, IDT and stack! \
- * All these lie safely hidden from the Guest: \
- * We must return to the Host page tables \
- * (Hence that was saved in struct lguest_pages) */ \
- movl LGUEST_PAGES_host_cr3(%eax), %edx; \
- movl %edx, %cr3; \
- /* As before, when we looked back at the Host \
- * As we left and marked TSS unused \
- * So must we now for the Guest left behind. */ \
- andb $0xFD, (LGUEST_PAGES_guest_gdt+GDT_ENTRY_TSS*8+5)(%eax); \
- /* Switch to Host's GDT, IDT. */ \
- lgdt LGUEST_PAGES_host_gdt_desc(%eax); \
- lidt LGUEST_PAGES_host_idt_desc(%eax); \
- /* Restore the Host's stack where its saved regs lie */ \
- movl LGUEST_PAGES_host_sp(%eax), %esp; \
- /* Last the TSS: our Host is returned */ \
- movl $(GDT_ENTRY_TSS*8), %edx; \
- ltr %dx; \
- /* Restore now the regs saved right at the first. */ \
- popl %ebp; \
- popl %fs; \
- popl %gs; \
- popl %ds; \
- popl %es
-
-// The first path is trod when the Guest has trapped:
-// (Which trap it was has been pushed on the stack).
-// We need only switch back, and the Host will decode
-// Why we came home, and what needs to be done.
-return_to_host:
- SWITCH_TO_HOST
- iret
-
-// We are lead to the second path like so:
-// An interrupt, with some cause external
-// Has ajerked us rudely from the Guest's code
-// Again we must return home to the Host
-deliver_to_host:
- SWITCH_TO_HOST
- // But now we must go home via that place
- // Where that interrupt was supposed to go
- // Had we not been ensconced, running the Guest.
- // Here we see the trickness of run_guest_once():
- // The Host stack is formed like an interrupt
- // With EIP, CS and EFLAGS layered.
- // Interrupt handlers end with "iret"
- // And that will take us home at long long last.
-
- // But first we must find the handler to call!
- // The IDT descriptor for the Host
- // Has two bytes for size, and four for address:
- // %edx will hold it for us for now.
- movl (LGUEST_PAGES_host_idt_desc+2)(%eax), %edx
- // We now know the table address we need,
- // And saved the trap's number inside %ebx.
- // Yet the pointer to the handler is smeared
- // Across the bits of the table entry.
- // What oracle can tell us how to extract
- // From such a convoluted encoding?
- // I consulted gcc, and it gave
- // These instructions, which I gladly credit:
- leal (%edx,%ebx,8), %eax
- movzwl (%eax),%edx
- movl 4(%eax), %eax
- xorw %ax, %ax
- orl %eax, %edx
- // Now the address of the handler's in %edx
- // We call it now: its "iret" drops us home.
- jmp *%edx
-
-// Every interrupt can come to us here
-// But we must truly tell each apart.
-// They number two hundred and fifty six
-// And each must land in a different spot,
-// Push its number on stack, and join the stream.
-
-// And worse, a mere six of the traps stand apart
-// And push on their stack an addition:
-// An error number, thirty two bits long
-// So we punish the other two fifty
-// And make them push a zero so they match.
-
-// Yet two fifty six entries is long
-// And all will look most the same as the last
-// So we create a macro which can make
-// As many entries as we need to fill.
-
-// Note the change to .data then .text:
-// We plant the address of each entry
-// Into a (data) table for the Host
-// To know where each Guest interrupt should go.
-.macro IRQ_STUB N TARGET
- .data; .long 1f; .text; 1:
- // Trap eight, ten through fourteen and seventeen
- // Supply an error number. Else zero.
- .if (\N <> 8) && (\N < 10 || \N > 14) && (\N <> 17)
- pushl $0
- .endif
- pushl $\N
- jmp \TARGET
- ALIGN
-.endm
-
-// This macro creates numerous entries
-// Using GAS macros which out-power C's.
-.macro IRQ_STUBS FIRST LAST TARGET
- irq=\FIRST
- .rept \LAST-\FIRST+1
- IRQ_STUB irq \TARGET
- irq=irq+1
- .endr
-.endm
-
-// Here's the marker for our pointer table
-// Laid in the data section just before
-// Each macro places the address of code
-// Forming an array: each one points to text
-// Which handles interrupt in its turn.
-.data
-.global default_idt_entries
-default_idt_entries:
-.text
- // The first two traps go straight back to the Host
- IRQ_STUBS 0 1 return_to_host
- // We'll say nothing, yet, about NMI
- IRQ_STUB 2 handle_nmi
- // Other traps also return to the Host
- IRQ_STUBS 3 31 return_to_host
- // All interrupts go via their handlers
- IRQ_STUBS 32 127 deliver_to_host
- // 'Cept system calls coming from userspace
- // Are to go to the Guest, never the Host.
- IRQ_STUB 128 return_to_host
- IRQ_STUBS 129 255 deliver_to_host
-
-// The NMI, what a fabulous beast
-// Which swoops in and stops us no matter that
-// We're suspended between heaven and hell,
-// (Or more likely between the Host and Guest)
-// When in it comes! We are dazed and confused
-// So we do the simplest thing which one can.
-// Though we've pushed the trap number and zero
-// We discard them, return, and hope we live.
-handle_nmi:
- addl $8, %esp
- iret
-
-// We are done; all that's left is Mastery
-// And "make Mastery" is a journey long
-// Designed to make your fingers itch to code.
-
-// Here ends the text, the file and poem.
-ENTRY(end_switcher_text)