1
0
mirror of https://github.com/cc65/cc65.git synced 2025-01-10 19:29:45 +00:00
cc65/libsrc/common/time.s

78 lines
1.5 KiB
ArmAsm
Raw Normal View History

;
; Ullrich von Bassewitz, 12.11.2002
;
; time_t __fastcall__ time (time_t* timep);
;
.export _time
.import pusha, ldeaxi
Replaced _systime with clock_gettime. We want to add the capability to not only get the time but also set the time, but there's no "setter" for the "getter" time(). The first ones that come into mind are gettimeofday() and settimeofday(). However, they take a struct timezone argument that doesn't make sense - even the man pages says "The use of the timezone structure is obsolete; the tz argument should normally be specified as NULL." And POSIX says "Applications should use the clock_gettime() function instead of the obsolescent gettimeofday() function." The ...timeofday() functions work with microseconds while the clock_...time() functions work with nanoseconds. Given that we expect our targets to support only 1/10 of seconds the microseconds look preferable at first sight. However, already microseconds require the cc65 data type 'long' so it's not such a relevant difference to nanoseconds. Additionally clock_getres() seems useful. In order to avoid code duplication clock_gettime() takes over the role of the actual time getter from _systime(). So time() now calls clock_gettime() instead of _systime(). For some reason beyond my understanding _systime() was mentioned in time.h. _systime() worked exactly like e.g. _sysremove() and those _sys...() functions are all considered internal. The only reason I could see would be a performance gain of bypassing the time() wrapper. However, all known _systime() implementations internally called mktime(). And mktime() is implemented in C using an iterative algorithm so I really can't see what would be left to gain here. From that perspective I decided to just remove _systime().
2018-08-15 15:59:11 +02:00
.importzp ptr1, sreg, tmp1, tmp2
Replaced _systime with clock_gettime. We want to add the capability to not only get the time but also set the time, but there's no "setter" for the "getter" time(). The first ones that come into mind are gettimeofday() and settimeofday(). However, they take a struct timezone argument that doesn't make sense - even the man pages says "The use of the timezone structure is obsolete; the tz argument should normally be specified as NULL." And POSIX says "Applications should use the clock_gettime() function instead of the obsolescent gettimeofday() function." The ...timeofday() functions work with microseconds while the clock_...time() functions work with nanoseconds. Given that we expect our targets to support only 1/10 of seconds the microseconds look preferable at first sight. However, already microseconds require the cc65 data type 'long' so it's not such a relevant difference to nanoseconds. Additionally clock_getres() seems useful. In order to avoid code duplication clock_gettime() takes over the role of the actual time getter from _systime(). So time() now calls clock_gettime() instead of _systime(). For some reason beyond my understanding _systime() was mentioned in time.h. _systime() worked exactly like e.g. _sysremove() and those _sys...() functions are all considered internal. The only reason I could see would be a performance gain of bypassing the time() wrapper. However, all known _systime() implementations internally called mktime(). And mktime() is implemented in C using an iterative algorithm so I really can't see what would be left to gain here. From that perspective I decided to just remove _systime().
2018-08-15 15:59:11 +02:00
.include "time.inc"
.code
.proc _time
pha
txa
pha ; Save timep
Replaced _systime with clock_gettime. We want to add the capability to not only get the time but also set the time, but there's no "setter" for the "getter" time(). The first ones that come into mind are gettimeofday() and settimeofday(). However, they take a struct timezone argument that doesn't make sense - even the man pages says "The use of the timezone structure is obsolete; the tz argument should normally be specified as NULL." And POSIX says "Applications should use the clock_gettime() function instead of the obsolescent gettimeofday() function." The ...timeofday() functions work with microseconds while the clock_...time() functions work with nanoseconds. Given that we expect our targets to support only 1/10 of seconds the microseconds look preferable at first sight. However, already microseconds require the cc65 data type 'long' so it's not such a relevant difference to nanoseconds. Additionally clock_getres() seems useful. In order to avoid code duplication clock_gettime() takes over the role of the actual time getter from _systime(). So time() now calls clock_gettime() instead of _systime(). For some reason beyond my understanding _systime() was mentioned in time.h. _systime() worked exactly like e.g. _sysremove() and those _sys...() functions are all considered internal. The only reason I could see would be a performance gain of bypassing the time() wrapper. However, all known _systime() implementations internally called mktime(). And mktime() is implemented in C using an iterative algorithm so I really can't see what would be left to gain here. From that perspective I decided to just remove _systime().
2018-08-15 15:59:11 +02:00
; Get the time (machine dependent)
.assert timespec::tv_sec = 0, error
lda #CLOCK_REALTIME
jsr pusha
Replaced _systime with clock_gettime. We want to add the capability to not only get the time but also set the time, but there's no "setter" for the "getter" time(). The first ones that come into mind are gettimeofday() and settimeofday(). However, they take a struct timezone argument that doesn't make sense - even the man pages says "The use of the timezone structure is obsolete; the tz argument should normally be specified as NULL." And POSIX says "Applications should use the clock_gettime() function instead of the obsolescent gettimeofday() function." The ...timeofday() functions work with microseconds while the clock_...time() functions work with nanoseconds. Given that we expect our targets to support only 1/10 of seconds the microseconds look preferable at first sight. However, already microseconds require the cc65 data type 'long' so it's not such a relevant difference to nanoseconds. Additionally clock_getres() seems useful. In order to avoid code duplication clock_gettime() takes over the role of the actual time getter from _systime(). So time() now calls clock_gettime() instead of _systime(). For some reason beyond my understanding _systime() was mentioned in time.h. _systime() worked exactly like e.g. _sysremove() and those _sys...() functions are all considered internal. The only reason I could see would be a performance gain of bypassing the time() wrapper. However, all known _systime() implementations internally called mktime(). And mktime() is implemented in C using an iterative algorithm so I really can't see what would be left to gain here. From that perspective I decided to just remove _systime().
2018-08-15 15:59:11 +02:00
lda #<time
ldx #>time
jsr _clock_gettime
; _clock_gettime returns 0 on success and -1 on error. Check that.
inx ; Did _clock_gettime return -1?
bne @L2 ; Jump if not
; We had an error so invalidate time. A contains $FF.
ldy #3
@L1: sta time,y
dey
bpl @L1
; Restore timep and check if it is NULL
@L2: pla
2013-10-02 20:55:01 +01:00
sta ptr1+1
pla
2013-10-02 20:55:01 +01:00
sta ptr1 ; Restore timep
ora ptr1+1 ; timep == 0?
beq @L4
; timep is not NULL, store the result there
ldy #3
@L3: lda time,y
sta (ptr1),y
dey
bpl @L3
Replaced _systime with clock_gettime. We want to add the capability to not only get the time but also set the time, but there's no "setter" for the "getter" time(). The first ones that come into mind are gettimeofday() and settimeofday(). However, they take a struct timezone argument that doesn't make sense - even the man pages says "The use of the timezone structure is obsolete; the tz argument should normally be specified as NULL." And POSIX says "Applications should use the clock_gettime() function instead of the obsolescent gettimeofday() function." The ...timeofday() functions work with microseconds while the clock_...time() functions work with nanoseconds. Given that we expect our targets to support only 1/10 of seconds the microseconds look preferable at first sight. However, already microseconds require the cc65 data type 'long' so it's not such a relevant difference to nanoseconds. Additionally clock_getres() seems useful. In order to avoid code duplication clock_gettime() takes over the role of the actual time getter from _systime(). So time() now calls clock_gettime() instead of _systime(). For some reason beyond my understanding _systime() was mentioned in time.h. _systime() worked exactly like e.g. _sysremove() and those _sys...() functions are all considered internal. The only reason I could see would be a performance gain of bypassing the time() wrapper. However, all known _systime() implementations internally called mktime(). And mktime() is implemented in C using an iterative algorithm so I really can't see what would be left to gain here. From that perspective I decided to just remove _systime().
2018-08-15 15:59:11 +02:00
; Load the final result.
@L4: lda time+3
Replaced _systime with clock_gettime. We want to add the capability to not only get the time but also set the time, but there's no "setter" for the "getter" time(). The first ones that come into mind are gettimeofday() and settimeofday(). However, they take a struct timezone argument that doesn't make sense - even the man pages says "The use of the timezone structure is obsolete; the tz argument should normally be specified as NULL." And POSIX says "Applications should use the clock_gettime() function instead of the obsolescent gettimeofday() function." The ...timeofday() functions work with microseconds while the clock_...time() functions work with nanoseconds. Given that we expect our targets to support only 1/10 of seconds the microseconds look preferable at first sight. However, already microseconds require the cc65 data type 'long' so it's not such a relevant difference to nanoseconds. Additionally clock_getres() seems useful. In order to avoid code duplication clock_gettime() takes over the role of the actual time getter from _systime(). So time() now calls clock_gettime() instead of _systime(). For some reason beyond my understanding _systime() was mentioned in time.h. _systime() worked exactly like e.g. _sysremove() and those _sys...() functions are all considered internal. The only reason I could see would be a performance gain of bypassing the time() wrapper. However, all known _systime() implementations internally called mktime(). And mktime() is implemented in C using an iterative algorithm so I really can't see what would be left to gain here. From that perspective I decided to just remove _systime().
2018-08-15 15:59:11 +02:00
sta sreg+1
lda time+2
sta sreg
ldx time+1
lda time
rts
.endproc
Replaced _systime with clock_gettime. We want to add the capability to not only get the time but also set the time, but there's no "setter" for the "getter" time(). The first ones that come into mind are gettimeofday() and settimeofday(). However, they take a struct timezone argument that doesn't make sense - even the man pages says "The use of the timezone structure is obsolete; the tz argument should normally be specified as NULL." And POSIX says "Applications should use the clock_gettime() function instead of the obsolescent gettimeofday() function." The ...timeofday() functions work with microseconds while the clock_...time() functions work with nanoseconds. Given that we expect our targets to support only 1/10 of seconds the microseconds look preferable at first sight. However, already microseconds require the cc65 data type 'long' so it's not such a relevant difference to nanoseconds. Additionally clock_getres() seems useful. In order to avoid code duplication clock_gettime() takes over the role of the actual time getter from _systime(). So time() now calls clock_gettime() instead of _systime(). For some reason beyond my understanding _systime() was mentioned in time.h. _systime() worked exactly like e.g. _sysremove() and those _sys...() functions are all considered internal. The only reason I could see would be a performance gain of bypassing the time() wrapper. However, all known _systime() implementations internally called mktime(). And mktime() is implemented in C using an iterative algorithm so I really can't see what would be left to gain here. From that perspective I decided to just remove _systime().
2018-08-15 15:59:11 +02:00
; ------------------------------------------------------------------------
; Data
Replaced _systime with clock_gettime. We want to add the capability to not only get the time but also set the time, but there's no "setter" for the "getter" time(). The first ones that come into mind are gettimeofday() and settimeofday(). However, they take a struct timezone argument that doesn't make sense - even the man pages says "The use of the timezone structure is obsolete; the tz argument should normally be specified as NULL." And POSIX says "Applications should use the clock_gettime() function instead of the obsolescent gettimeofday() function." The ...timeofday() functions work with microseconds while the clock_...time() functions work with nanoseconds. Given that we expect our targets to support only 1/10 of seconds the microseconds look preferable at first sight. However, already microseconds require the cc65 data type 'long' so it's not such a relevant difference to nanoseconds. Additionally clock_getres() seems useful. In order to avoid code duplication clock_gettime() takes over the role of the actual time getter from _systime(). So time() now calls clock_gettime() instead of _systime(). For some reason beyond my understanding _systime() was mentioned in time.h. _systime() worked exactly like e.g. _sysremove() and those _sys...() functions are all considered internal. The only reason I could see would be a performance gain of bypassing the time() wrapper. However, all known _systime() implementations internally called mktime(). And mktime() is implemented in C using an iterative algorithm so I really can't see what would be left to gain here. From that perspective I decided to just remove _systime().
2018-08-15 15:59:11 +02:00
.bss
time: .tag timespec