i2p/haiku-beta6
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

We use the same strategy for computing the system time as on x86 now.

Browse Source 
The time base conversion factor is the 32 bit value
  2^32 * 1000000 / time base frequency,
so the system time can be computed by
  system time = time base * conversion factor / 2^32.
The expression in system_time() looks more complicated now, but is
actually much faster (factor 2.5 on my Mac mini). I'm positively
surprised, how good the assembly looks, that GCC 4 generates. There's
not that much potential for optimization by hand-coding the function.


git-svn-id: file:///srv/svn/repos/haiku/haiku/trunk@15863 a95241bf-73f2-0310-859d-f6bbb57e9c96
This commit is contained in:
Ingo Weinhold
2006-01-07 23:05:56 +00:00
parent 8baf8813c0
commit 262e0a636b
4 changed files with 12 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files
headers/private/kernel/arch/ppc/arch_cpu.h
+1 -1
View File
@@ -105,7 +105,7 @@ extern void reset_dbats(void);
//extern void setl2cr(unsigned int val);
extern long long get_time_base(void);
void __ppc_setup_system_time(vint64 *cvFactor);
void __ppc_setup_system_time(vint32 *cvFactor);
// defined in libroot: os/arch/system_time.c
int64 __ppc_get_time_base(void);
// defined in libroot: os/arch/system_time_asm.S
headers/private/kernel/arch/ppc/arch_real_time_data.h
+1 -1
View File
@@ -15,7 +15,7 @@ struct ppc_real_time_data {
struct arch_real_time_data {
struct ppc_real_time_data data[2];
vint64 system_time_conversion_factor;
vint32 system_time_conversion_factor;
vint32 version;
// Since there're no cheap atomic_{set,get,add}64() on PPC 32 (i.e. one
// that doesn't involve a syscall), we can't have just a single