6502/cc65
1
0
Fork 0
mirror of https://github.com/cc65/cc65.git synced 2025-01-12 17:30:50 +00:00
Code Issues Projects Releases Wiki Activity
cc65/libsrc/c128/gettime.s

120 lines
3.0 KiB
ArmAsm
Raw Normal View History

Added a systime implementation by Stefan Haubenthal with a few changes. The routine reads the TOD clock of CIA1 on the C64 and C128. Since systime was a dummy routine common for all CBMs before, this change adds an individual dummy routine for all other CBM systems. CBM510/610 do also have a TOD clock, so a similar function as in the C64 could be used ... git-svn-id: svn://svn.cc65.org/cc65/trunk@3974 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-07-27 18:44:37 +00:00
;
; Stefan Haubenthal, 27.7.2009
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take a struct timezone argument that doesn't make sense - even the man pages says "The use of the timezone structure is obsolete; the tz argument should normally be specified as NULL."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
; Oliver Schmidt, 14.8.2018
Added a systime implementation by Stefan Haubenthal with a few changes. The routine reads the TOD clock of CIA1 on the C64 and C128. Since systime was a dummy routine common for all CBMs before, this change adds an individual dummy routine for all other CBM systems. CBM510/610 do also have a TOD clock, so a similar function as in the C64 could be used ... git-svn-id: svn://svn.cc65.org/cc65/trunk@3974 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-07-27 18:44:37 +00:00
;
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
; int clock_gettime (clockid_t clk_id, struct timespec *tp);
Added a systime implementation by Stefan Haubenthal with a few changes. The routine reads the TOD clock of CIA1 on the C64 and C128. Since systime was a dummy routine common for all CBMs before, this change adds an individual dummy routine for all other CBM systems. CBM510/610 do also have a TOD clock, so a similar function as in the C64 could be used ... git-svn-id: svn://svn.cc65.org/cc65/trunk@3974 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-07-27 18:44:37 +00:00
;
.include "time.inc"
.include "c128.inc"
Removed (pretty inconsistently used) tab chars from source code base.
2013-05-09 13:56:54 +02:00
.include "get_tv.inc"
Added a systime implementation by Stefan Haubenthal with a few changes. The routine reads the TOD clock of CIA1 on the C64 and C128. Since systime was a dummy routine common for all CBMs before, this change adds an individual dummy routine for all other CBM systems. CBM510/610 do also have a TOD clock, so a similar function as in the C64 could be used ... git-svn-id: svn://svn.cc65.org/cc65/trunk@3974 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-07-27 18:44:37 +00:00
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
.constructor inittime
.importzp sreg, tmp1, tmp2
.import pushax, pusheax, tosmul0ax, steaxspidx, incsp1, return0
Removed (pretty inconsistently used) tab chars from source code base.
2013-05-09 13:56:54 +02:00
.import _get_tv
Added a systime implementation by Stefan Haubenthal with a few changes. The routine reads the TOD clock of CIA1 on the C64 and C128. Since systime was a dummy routine common for all CBMs before, this change adds an individual dummy routine for all other CBM systems. CBM510/610 do also have a TOD clock, so a similar function as in the C64 could be used ... git-svn-id: svn://svn.cc65.org/cc65/trunk@3974 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-07-27 18:44:37 +00:00
The CIA TOD needs a write to the 1/10 sec register to start. Shortened the code by using initialized data. git-svn-id: svn://svn.cc65.org/cc65/trunk@4224 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-09-24 19:40:36 +00:00
;----------------------------------------------------------------------------
Added a systime implementation by Stefan Haubenthal with a few changes. The routine reads the TOD clock of CIA1 on the C64 and C128. Since systime was a dummy routine common for all CBMs before, this change adds an individual dummy routine for all other CBM systems. CBM510/610 do also have a TOD clock, so a similar function as in the C64 could be used ... git-svn-id: svn://svn.cc65.org/cc65/trunk@3974 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-07-27 18:44:37 +00:00
.code
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
.proc _clock_gettime
jsr pushax
jsr pushax
Removed (pretty inconsistently used) tab chars from source code base.
2013-05-09 13:56:54 +02:00
lda CIA1_TODHR
bpl AM
and #%01111111
sed
clc
adc #$12
cld
AM: jsr BCD2dec
sta TM + tm::tm_hour
lda CIA1_TODMIN
jsr BCD2dec
sta TM + tm::tm_min
lda CIA1_TODSEC
jsr BCD2dec
sta TM + tm::tm_sec
lda #<TM
ldx #>TM
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
jsr _mktime
ldy #timespec::tv_sec
jsr steaxspidx ; Pops address pushed by 2. pushax
lda #<(100 * 1000 * 1000 / $10000)
ldx #>(100 * 1000 * 1000 / $10000)
sta sreg
stx sreg+1
lda #<(100 * 1000 * 1000)
ldx #>(100 * 1000 * 1000)
jsr pusheax
lda CIA1_TOD10
ldx #>$0000
jsr tosmul0ax
ldy #timespec::tv_nsec
jsr steaxspidx ; Pops address pushed by 1. pushax
jsr incsp1
jmp return0
Added a systime implementation by Stefan Haubenthal with a few changes. The routine reads the TOD clock of CIA1 on the C64 and C128. Since systime was a dummy routine common for all CBMs before, this change adds an individual dummy routine for all other CBM systems. CBM510/610 do also have a TOD clock, so a similar function as in the C64 could be used ... git-svn-id: svn://svn.cc65.org/cc65/trunk@3974 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-07-27 18:44:37 +00:00
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
.endproc
;----------------------------------------------------------------------------
Added a systime implementation by Stefan Haubenthal with a few changes. The routine reads the TOD clock of CIA1 on the C64 and C128. Since systime was a dummy routine common for all CBMs before, this change adds an individual dummy routine for all other CBM systems. CBM510/610 do also have a TOD clock, so a similar function as in the C64 could be used ... git-svn-id: svn://svn.cc65.org/cc65/trunk@3974 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-07-27 18:44:37 +00:00
; dec = (((BCD>>4)*10) + (BCD&0xf))
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
.proc BCD2dec
tax
Removed (pretty inconsistently used) tab chars from source code base.
2013-05-09 13:56:54 +02:00
and #%00001111
sta tmp1
txa
Added a systime implementation by Stefan Haubenthal with a few changes. The routine reads the TOD clock of CIA1 on the C64 and C128. Since systime was a dummy routine common for all CBMs before, this change adds an individual dummy routine for all other CBM systems. CBM510/610 do also have a TOD clock, so a similar function as in the C64 could be used ... git-svn-id: svn://svn.cc65.org/cc65/trunk@3974 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-07-27 18:44:37 +00:00
and #%11110000 ; *16
lsr ; *8
sta tmp2
lsr
lsr ; *2
adc tmp2 ; = *10
adc tmp1
rts
.endproc
The CIA TOD needs a write to the 1/10 sec register to start. Shortened the code by using initialized data. git-svn-id: svn://svn.cc65.org/cc65/trunk@4224 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-09-24 19:40:36 +00:00
;----------------------------------------------------------------------------
; Constructor that writes to the 1/10 sec register of the TOD to kick it
; into action. If this is not done, the clock hangs. We will read the register
; and write it again, ignoring a possible change in between.
Renamed INITBSS to INIT and INIT to ONCE. The way we want to use the INITBSS segment - and especially the fact that it won't have the type bss on all ROM based targets - means that the name INITBSS is misleading. After all INIT is the best name from my perspective as it serves several purposes and therefore needs a rather generic name. Unfortunately this means that the current INIT segment needs to be renamed too. Looking for a short (ideally 4 letter) name I came up with ONCE as it contains all code (and data) accessed only once during initialization.
2016-03-06 21:26:22 +01:00
.segment "ONCE"
The CIA TOD needs a write to the 1/10 sec register to start. Shortened the code by using initialized data. git-svn-id: svn://svn.cc65.org/cc65/trunk@4224 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-09-24 19:40:36 +00:00
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
.proc inittime
The CIA TOD needs a write to the 1/10 sec register to start. Shortened the code by using initialized data. git-svn-id: svn://svn.cc65.org/cc65/trunk@4224 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-09-24 19:40:36 +00:00
lda CIA1_TOD10
sta CIA1_TOD10
Removed (pretty inconsistently used) tab chars from source code base.
2013-05-09 13:56:54 +02:00
jsr _get_tv
cmp #TV::PAL
bne @60Hz
lda CIA1_CRA
ora #$80
sta CIA1_CRA
@60Hz: rts
The CIA TOD needs a write to the 1/10 sec register to start. Shortened the code by using initialized data. git-svn-id: svn://svn.cc65.org/cc65/trunk@4224 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-09-24 19:40:36 +00:00
.endproc
;----------------------------------------------------------------------------
; TM struct with date set to 1970-01-01
.data
Removed (pretty inconsistently used) tab chars from source code base.
2013-05-09 13:56:54 +02:00
TM: .word 0 ; tm_sec
The CIA TOD needs a write to the 1/10 sec register to start. Shortened the code by using initialized data. git-svn-id: svn://svn.cc65.org/cc65/trunk@4224 b7a2c559-68d2-44c3-8de9-860c34a00d81
2009-09-24 19:40:36 +00:00
.word 0 ; tm_min
.word 0 ; tm_hour
.word 1 ; tm_mday
.word 0 ; tm_mon
.word 70 ; tm_year
.word 0 ; tm_wday
.word 0 ; tm_yday
.word 0 ; tm_isdst
Reference in New Issue Copy Permalink