Strange issue with BSET

DanielAllsopp · 10 February 2021, 16:18

So I have a struct setup like so:

Code:

                  rsset   0
testWord              rs.w    1
sizeOf            rs.l    0

structTest:
  ds.b            sizeOf

and I load it into memory like this

Code:

lea structTest,a1

What I basically want to do is set bit 0 of testWord, but I'm getting strange results, i.e. the value is ending up as $100 instead of $1 if I access the location directly:

Code:

clr.w testWord(a1)
bset #$0,testWord(a1)    ;testWord now equals $100

but if I do it via a data register:

Code:

clr.w d1
move.w d1, testWord(a1)
bset #$0,d1
move.w d1, testWord(a1)    ;testWord now equals $1

I'm losing the plot here trying to figure out why that is? Can anyone enlighten me?

roondar · 10 February 2021, 16:25

BSET/BCLR/BCHG/BTST only work on a single byte when used on a value in memory

thomas · 10 February 2021, 16:29

Obviously bset in memory is acting on bytes, not words. I don't know if seomthing like bset.w exists, probably not. In this case you have to take into account that it acts on bytes and skip the high-order byte:

bset #$0,testWord+1(a1)

DanielAllsopp · 10 February 2021, 16:32

Ah, well that makes sense now then. As always, thanks for the help guys, I'll build my low level knowledge up slowly but surely

ross · 10 February 2021, 16:37

Quote:

Originally Posted by thomas

Obviously bset in memory is acting on bytes, not words. I don't know if seomthing like bset.w exists,

Sure it exist: or.w

Antiriad_UK · 10 February 2021, 17:32

And the classic "btst #14,dmaconr" which only works by "accident" because it does modulo 8 hiding the mistake from everyone which I saw in loads of old code

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Strange keyboard issue	malko	support.WinUAE	3	01 October 2019 19:15
Strange Keyboard Issue	TheTonyOne	support.FS-UAE	4	07 January 2015 01:48
Strange Issue	mai	support.WinUAE	6	06 January 2010 12:54
Strange Graphics issue	asm1	support.Hardware	5	04 March 2009 18:12
Strange 3.5 floppy issue ...	falconsfan	support.Hardware	1	28 April 2007 09:13

10 February 2021, 16:18	#1
DanielAllsopp Registered User Join Date: Feb 2018 Location: Northumberland, UK Posts: 272	Strange issue with BSET So I have a struct setup like so: Code: rsset 0 testWord rs.w 1 sizeOf rs.l 0 structTest: ds.b sizeOf and I load it into memory like this Code: lea structTest,a1 What I basically want to do is set bit 0 of testWord, but I'm getting strange results, i.e. the value is ending up as $100 instead of $1 if I access the location directly: Code: clr.w testWord(a1) bset #$0,testWord(a1) ;testWord now equals $100 but if I do it via a data register: Code: clr.w d1 move.w d1, testWord(a1) bset #$0,d1 move.w d1, testWord(a1) ;testWord now equals $1 I'm losing the plot here trying to figure out why that is? Can anyone enlighten me?

10 February 2021, 16:25	#2
roondar Registered User Join Date: Jul 2015 Location: The Netherlands Posts: 3,430	BSET/BCLR/BCHG/BTST only work on a single byte when used on a value in memory

10 February 2021, 16:29	#3
thomas Registered User Join Date: Jan 2002 Location: Germany Posts: 7,021	Obviously bset in memory is acting on bytes, not words. I don't know if seomthing like bset.w exists, probably not. In this case you have to take into account that it acts on bytes and skip the high-order byte: bset #$0,testWord+1(a1)

10 February 2021, 16:32	#4
DanielAllsopp Registered User Join Date: Feb 2018 Location: Northumberland, UK Posts: 272	Ah, well that makes sense now then. As always, thanks for the help guys, I'll build my low level knowledge up slowly but surely

10 February 2021, 17:32	#6
Antiriad_UK OCS forever! Join Date: Mar 2019 Location: Birmingham, UK Posts: 418	And the classic "btst #14,dmaconr" which only works by "accident" because it does modulo 8 hiding the mistake from everyone which I saw in loads of old code

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)