| 
									
										
										
										
											2009-07-30 16:03:45 -06:00
										 |  |  | /*P:900 | 
					
						
							| 
									
										
										
										
											2009-07-30 16:03:45 -06:00
										 |  |  |  * 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. | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:02 -07:00
										 |  |  |  * | 
					
						
							|  |  |  |  * You have now completed Preparation.  If this has whet your appetite; if you
 | 
					
						
							|  |  |  |  * are feeling invigorated and refreshed then the next, more challenging stage | 
					
						
							| 
									
										
										
										
											2009-07-30 16:03:45 -06:00
										 |  |  |  * can be found in "make Guest". | 
					
						
							|  |  |  |  :*/ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 
 | 
					
						
							| 
									
										
										
										
											2009-07-30 16:03:45 -06:00
										 |  |  | /*M:012 | 
					
						
							|  |  |  |  * Lguest is meant to be simple: my rule of thumb is that 1% more LOC must | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  |  * 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? | 
					
						
							| 
									
										
										
										
											2009-07-30 16:03:45 -06:00
										 |  |  |  * And who would write the verse documenting it? | 
					
						
							|  |  |  | :*/ | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 
 | 
					
						
							| 
									
										
										
										
											2009-07-30 16:03:45 -06:00
										 |  |  | /*M:011 | 
					
						
							|  |  |  |  * Lguest64 handles NMI.  This gave me NMI envy (until I looked at their | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  |  * code).  It's worth doing though, since it would let us use oprofile in the | 
					
						
							| 
									
										
										
										
											2009-07-30 16:03:45 -06:00
										 |  |  |  * Host when a Guest is running. | 
					
						
							|  |  |  | :*/ | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | /*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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | #include <linux/linkage.h> | 
					
						
							|  |  |  | #include <asm/asm-offsets.h> | 
					
						
							| 
									
										
										
										
											2007-08-09 20:57:13 +10:00
										 |  |  | #include <asm/page.h> | 
					
						
							| 
									
										
										
										
											2007-10-22 11:03:28 +10:00
										 |  |  | #include <asm/segment.h> | 
					
						
							|  |  |  | #include <asm/lguest.h> | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | // 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | .text | 
					
						
							|  |  |  | ENTRY(start_switcher_text) | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | // 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | ENTRY(switch_to_guest) | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// 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 | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	pushl	%es | 
					
						
							|  |  |  | 	pushl	%ds | 
					
						
							|  |  |  | 	pushl	%gs | 
					
						
							|  |  |  | 	pushl	%fs | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	pushl	%ebp | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// The Host's stack is done, now save it away | 
					
						
							|  |  |  | 	// In our "struct lguest_pages" at offset | 
					
						
							|  |  |  | 	// Distilled into asm-offsets.h | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	%esp, LGUEST_PAGES_host_sp(%eax) | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// All saved and there's now five steps before us: | 
					
						
							|  |  |  | 	// Stack, GDT, IDT, TSS | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// Then last of all the page tables are flipped. | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	%eax, %edx | 
					
						
							|  |  |  | 	addl	$LGUEST_PAGES_regs, %edx | 
					
						
							|  |  |  | 	movl	%edx, %esp | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// The Guest's GDT we so carefully | 
					
						
							|  |  |  | 	// Placed in the "struct lguest_pages" before | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	lgdt	LGUEST_PAGES_guest_gdt_desc(%eax) | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// The Guest's IDT we did partially | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// Copy to "struct lguest_pages" as well. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	lidt	LGUEST_PAGES_guest_idt_desc(%eax) | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// The TSS entry which controls traps | 
					
						
							|  |  |  | 	// Must be loaded up with "ltr" now: | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// The GDT entry that TSS uses  | 
					
						
							|  |  |  | 	// Changes type when we load it: damn Intel! | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// For after we switch over our page tables | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// That entry will be read-only: we'd crash. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	$(GDT_ENTRY_TSS*8), %edx | 
					
						
							|  |  |  | 	ltr	%dx | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// Look back now, before we take this last step! | 
					
						
							|  |  |  | 	// The Host's TSS entry was also marked used;
 | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// Let's clear it again for our return. | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// The GDT descriptor of the Host | 
					
						
							|  |  |  | 	// Points to the table after two "size" bytes | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	(LGUEST_PAGES_host_gdt_desc+2)(%eax), %edx | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// Clear "used" from type field (byte 5, bit 2) | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	andb	$0xFD, (GDT_ENTRY_TSS*8 + 5)(%edx) | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	%ebx, %cr3 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// The page table change did one tricky thing: | 
					
						
							|  |  |  | 	// The Guest's register page has been mapped | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// Writable under our %esp (stack) -- | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// We can simply pop off all Guest regs. | 
					
						
							| 
									
										
										
										
											2007-10-22 11:03:29 +10:00
										 |  |  | 	popl	%eax | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	popl	%ebx | 
					
						
							|  |  |  | 	popl	%ecx | 
					
						
							|  |  |  | 	popl	%edx | 
					
						
							|  |  |  | 	popl	%esi | 
					
						
							|  |  |  | 	popl	%edi | 
					
						
							|  |  |  | 	popl	%ebp | 
					
						
							|  |  |  | 	popl	%gs | 
					
						
							|  |  |  | 	popl	%fs | 
					
						
							|  |  |  | 	popl	%ds | 
					
						
							|  |  |  | 	popl	%es | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	addl	$8, %esp | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 
 | 
					
						
							|  |  |  | 	// The last five stack slots hold return address | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// And everything needed to switch privilege | 
					
						
							|  |  |  | 	// From Switcher's level 0 to Guest's 1, | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// And the stack where the Guest had last left it. | 
					
						
							|  |  |  | 	// Interrupts are turned back on: we are Guest. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	iret | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2008-03-28 11:05:53 -05:00
										 |  |  | // We tread two paths to switch back to the Host | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | // Yet both must save Guest state and restore Host | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | // So we put the routine in a macro. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | #define SWITCH_TO_HOST							\ | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	/* We save the Guest state: all registers first			\ | 
					
						
							|  |  |  | 	 * Laid out just as "struct lguest_regs" defines */		\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	pushl	%es;							\
 | 
					
						
							|  |  |  | 	pushl	%ds;							\
 | 
					
						
							|  |  |  | 	pushl	%fs;							\
 | 
					
						
							|  |  |  | 	pushl	%gs;							\
 | 
					
						
							|  |  |  | 	pushl	%ebp;							\
 | 
					
						
							|  |  |  | 	pushl	%edi;							\
 | 
					
						
							|  |  |  | 	pushl	%esi;							\
 | 
					
						
							|  |  |  | 	pushl	%edx;							\
 | 
					
						
							|  |  |  | 	pushl	%ecx;							\
 | 
					
						
							|  |  |  | 	pushl	%ebx;							\
 | 
					
						
							| 
									
										
										
										
											2007-10-22 11:03:29 +10:00
										 |  |  | 	pushl	%eax;							\
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	/* 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. */			\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	$(LGUEST_DS), %eax;					\
 | 
					
						
							|  |  |  | 	movl	%eax, %ds;						\
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	/* So where are we?  Which CPU, which struct?			\ | 
					
						
							| 
									
										
										
										
											2007-08-09 20:57:13 +10:00
										 |  |  | 	 * 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". */		\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	%esp, %eax;						\
 | 
					
						
							| 
									
										
										
										
											2007-08-09 20:57:13 +10:00
										 |  |  | 	andl	$(~(1 << PAGE_SHIFT - 1)), %eax;			\
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	/* Save our trap number: the switch will obscure it		\ | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	 * (In the Host the Guest regs are not mapped here)		\ | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	 * %ebx holds it safe for deliver_to_host */			\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	LGUEST_PAGES_regs_trapnum(%eax), %ebx;			\
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	/* 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) */		\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	LGUEST_PAGES_host_cr3(%eax), %edx;			\
 | 
					
						
							|  |  |  | 	movl	%edx, %cr3;						\
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	/* 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. */			\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	andb	$0xFD, (LGUEST_PAGES_guest_gdt+GDT_ENTRY_TSS*8+5)(%eax); \
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	/* Switch to Host's GDT, IDT. */				\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	lgdt	LGUEST_PAGES_host_gdt_desc(%eax);			\
 | 
					
						
							|  |  |  | 	lidt	LGUEST_PAGES_host_idt_desc(%eax);			\
 | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	/* Restore the Host's stack where its saved regs lie */		\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	LGUEST_PAGES_host_sp(%eax), %esp;			\
 | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	/* Last the TSS: our Host is returned */			\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	$(GDT_ENTRY_TSS*8), %edx;				\
 | 
					
						
							|  |  |  | 	ltr	%dx;							\
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	/* Restore now the regs saved right at the first. */		\ | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	popl	%ebp;							\
 | 
					
						
							|  |  |  | 	popl	%fs;							\
 | 
					
						
							|  |  |  | 	popl	%gs;							\
 | 
					
						
							|  |  |  | 	popl	%ds;							\
 | 
					
						
							|  |  |  | 	popl	%es | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | // The first path is trod when the Guest has trapped: | 
					
						
							|  |  |  | // (Which trap it was has been pushed on the stack). | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | // We need only switch back, and the Host will decode | 
					
						
							|  |  |  | // Why we came home, and what needs to be done. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | return_to_host: | 
					
						
							|  |  |  | 	SWITCH_TO_HOST | 
					
						
							|  |  |  | 	iret | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | // We are lead to the second path like so: | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | // An interrupt, with some cause external | 
					
						
							|  |  |  | // Has ajerked us rudely from the Guest's code | 
					
						
							|  |  |  | // Again we must return home to the Host | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | deliver_to_host: | 
					
						
							|  |  |  | 	SWITCH_TO_HOST | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// But now we must go home via that place | 
					
						
							|  |  |  | 	// Where that interrupt was supposed to go | 
					
						
							|  |  |  | 	// Had we not been ensconced, running the Guest. | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// Here we see the trickness of run_guest_once(): | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	movl	(LGUEST_PAGES_host_idt_desc+2)(%eax), %edx | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// 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: | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	leal	(%edx,%ebx,8), %eax | 
					
						
							|  |  |  | 	movzwl	(%eax),%edx | 
					
						
							|  |  |  | 	movl	4(%eax), %eax | 
					
						
							|  |  |  | 	xorw	%ax, %ax | 
					
						
							|  |  |  | 	orl	%eax, %edx | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// Now the address of the handler's in %edx | 
					
						
							| 
									
										
										
										
											2007-10-25 15:02:50 +10:00
										 |  |  | 	// We call it now: its "iret" drops us home. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | 	jmp	*%edx | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | // 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | .macro IRQ_STUB N TARGET | 
					
						
							|  |  |  | 	.data; .long 1f; .text; 1:
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  |  // Trap eight, ten through fourteen and seventeen | 
					
						
							|  |  |  |  // Supply an error number.  Else zero. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  |  .if (\N <> 8) && (\N < 10 || \N > 14) && (\N <> 17) | 
					
						
							|  |  |  | 	pushl	$0 | 
					
						
							|  |  |  |  .endif | 
					
						
							|  |  |  | 	pushl	$\N | 
					
						
							|  |  |  | 	jmp	\TARGET | 
					
						
							|  |  |  | 	ALIGN | 
					
						
							|  |  |  | .endm | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | // This macro creates numerous entries | 
					
						
							|  |  |  | // Using GAS macros which out-power C's. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | .macro IRQ_STUBS FIRST LAST TARGET | 
					
						
							|  |  |  |  irq=\FIRST | 
					
						
							|  |  |  |  .rept \LAST-\FIRST+1 | 
					
						
							|  |  |  | 	IRQ_STUB irq \TARGET | 
					
						
							|  |  |  |   irq=irq+1 | 
					
						
							|  |  |  |  .endr | 
					
						
							|  |  |  | .endm | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | // 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | .data | 
					
						
							|  |  |  | .global default_idt_entries
 | 
					
						
							|  |  |  | default_idt_entries: | 
					
						
							|  |  |  | .text | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | 	// 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | handle_nmi: | 
					
						
							|  |  |  | 	addl	$8, %esp | 
					
						
							|  |  |  | 	iret | 
					
						
							|  |  |  | 
 | 
					
						
							| 
									
										
										
										
											2007-07-26 10:41:04 -07:00
										 |  |  | // 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. | 
					
						
							| 
									
										
										
										
											2007-07-19 01:49:23 -07:00
										 |  |  | ENTRY(end_switcher_text) |