Displaying 4 results from an estimated 4 matches for "__vsyscall".
Did you mean:
__syscall
2007 Apr 18
0
[RFC/PATCH PV_OPS X86_64 14/17] paravirt_ops - vsyscall
...static __always_inline void do_vgettimeo
usec = __xtime.tv_nsec / 1000;
if (__vxtime.mode != VXTIME_HPET) {
- t = get_cycles_sync();
+ t = vget_cycles_sync();
if (t < __vxtime.last_tsc)
t = __vxtime.last_tsc;
usec += ((t - __vxtime.last_tsc) *
@@ -147,8 +147,8 @@ time_t __vsyscall(1) vtime(time_t *t)
long __vsyscall(2)
vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
{
- unsigned int dummy, p;
- unsigned long j = 0;
+ unsigned int p;
+ unsigned long dummy, j = 0;
/* Fast cache - only recompute value once per jiffies and avoid
relatively costly...
2007 Apr 18
0
[RFC/PATCH PV_OPS X86_64 14/17] paravirt_ops - vsyscall
...static __always_inline void do_vgettimeo
usec = __xtime.tv_nsec / 1000;
if (__vxtime.mode != VXTIME_HPET) {
- t = get_cycles_sync();
+ t = vget_cycles_sync();
if (t < __vxtime.last_tsc)
t = __vxtime.last_tsc;
usec += ((t - __vxtime.last_tsc) *
@@ -147,8 +147,8 @@ time_t __vsyscall(1) vtime(time_t *t)
long __vsyscall(2)
vgetcpu(unsigned *cpu, unsigned *node, struct getcpu_cache *tcache)
{
- unsigned int dummy, p;
- unsigned long j = 0;
+ unsigned int p;
+ unsigned long dummy, j = 0;
/* Fast cache - only recompute value once per jiffies and avoid
relatively costly...
2007 Apr 18
0
[PATCH] paravirt_ops x86_64 , take 2
...)
* addition of the startup_paravirt function, to kick off guests (not
tested)
* fixed problems with patching
* added a new field, vsyscall_page in the paravirt_ops struct, which
allows the kernel to map a vsyscall_page on its own
* fixed vsyscall functions to avoid calling paravirt_ops functions.
__vsyscall_0 is the page to be mapped for the host. (set and get cpu not
yet tested.)
* fixed cpuid calls.
* added substitute for the swapgs instruction. (Notice that I'm not
saying it works ;-) )
In my TODO list, you can find:
* putting swapgs to work
* making sure legacy mode binaries work
* merging...
2007 Apr 18
0
[PATCH] paravirt_ops x86_64 , take 2
...)
* addition of the startup_paravirt function, to kick off guests (not
tested)
* fixed problems with patching
* added a new field, vsyscall_page in the paravirt_ops struct, which
allows the kernel to map a vsyscall_page on its own
* fixed vsyscall functions to avoid calling paravirt_ops functions.
__vsyscall_0 is the page to be mapped for the host. (set and get cpu not
yet tested.)
* fixed cpuid calls.
* added substitute for the swapgs instruction. (Notice that I'm not
saying it works ;-) )
In my TODO list, you can find:
* putting swapgs to work
* making sure legacy mode binaries work
* merging...