Donate to e Foundation | Murena handsets with /e/OS | Own a part of Murena! Learn more

Commit 633872b9 authored by Rusty Russell's avatar Rusty Russell
Browse files

lguest: tidy up documentation 



After Adrian Bunk's "make async_hcall static" moved things around, update
comments to match (aka "make Guest").

Signed-off-by: default avatarRusty Russell <rusty@rustcorp.com.au>
parent fad23fc7

arch/x86/lguest/boot.c

+21 −22
Original line number Original line Diff line number Diff line
@@ -93,27 +93,7 @@ struct lguest_data lguest_data = { 
};

};

static cycle_t clock_base;

static cycle_t clock_base;





/*G:035 Notice the lazy_hcall() above, rather than hcall().  This is our first

/*G:037 async_hcall() is pretty simple: I'm quite proud of it really.  We have a

 * real optimization trick!

 *

 * When lazy_mode is set, it means we're allowed to defer all hypercalls and do

 * them as a batch when lazy_mode is eventually turned off.  Because hypercalls

 * are reasonably expensive, batching them up makes sense.  For example, a

 * large munmap might update dozens of page table entries: that code calls

 * paravirt_enter_lazy_mmu(), does the dozen updates, then calls

 * lguest_leave_lazy_mode().

 *

 * So, when we're in lazy mode, we call async_hypercall() to store the call for

 * future processing.  When lazy mode is turned off we issue a hypercall to

 * flush the stored calls.

 */

static void lguest_leave_lazy_mode(void)

{

	paravirt_leave_lazy(paravirt_get_lazy_mode());

	hcall(LHCALL_FLUSH_ASYNC, 0, 0, 0);

}



/* async_hcall() is pretty simple: I'm quite proud of it really.  We have a

 * ring buffer of stored hypercalls which the Host will run though next time we

 * ring buffer of stored hypercalls which the Host will run though next time we

 * do a normal hypercall.  Each entry in the ring has 4 slots for the hypercall

 * do a normal hypercall.  Each entry in the ring has 4 slots for the hypercall

 * arguments, and a "hcall_status" word which is 0 if the call is ready to go,

 * arguments, and a "hcall_status" word which is 0 if the call is ready to go,
@@ -151,6 +131,18 @@ static void async_hcall(unsigned long call, unsigned long arg1, 
	local_irq_restore(flags);

	local_irq_restore(flags);

}

}





/*G:035 Notice the lazy_hcall() above, rather than hcall().  This is our first

 * real optimization trick!

 *

 * When lazy_mode is set, it means we're allowed to defer all hypercalls and do

 * them as a batch when lazy_mode is eventually turned off.  Because hypercalls

 * are reasonably expensive, batching them up makes sense.  For example, a

 * large munmap might update dozens of page table entries: that code calls

 * paravirt_enter_lazy_mmu(), does the dozen updates, then calls

 * lguest_leave_lazy_mode().

 *

 * So, when we're in lazy mode, we call async_hcall() to store the call for

 * future processing. */

static void lazy_hcall(unsigned long call,

static void lazy_hcall(unsigned long call,

		       unsigned long arg1,

		       unsigned long arg1,

		       unsigned long arg2,

		       unsigned long arg2,
@@ -161,7 +153,14 @@ static void lazy_hcall(unsigned long call, 
	else

	else

		async_hcall(call, arg1, arg2, arg3);

		async_hcall(call, arg1, arg2, arg3);

}

}

/*:*/



/* When lazy mode is turned off reset the per-cpu lazy mode variable and then

 * issue a hypercall to flush any stored calls. */

static void lguest_leave_lazy_mode(void)

{

	paravirt_leave_lazy(paravirt_get_lazy_mode());

	hcall(LHCALL_FLUSH_ASYNC, 0, 0, 0);

}





/*G:033

/*G:033

 * After that diversion we return to our first native-instruction

 * After that diversion we return to our first native-instruction