Previous Thread
Next Thread
Print Thread
Page 40 of 61 1 2 38 39 40 41 42 60 61
Joined: Mar 2001
Posts: 16,613
R
Very Senior Member
Offline
Very Senior Member
R
Joined: Mar 2001
Posts: 16,613
Hmm. I don't see anything obviously wrong with the card device (although it should use one of those newfangled input mergers for the two chips' IRQs instead of rolling its own). Might be a bug in the 6840.

Joined: May 2009
Posts: 1,879
J
Very Senior Member
Offline
Very Senior Member
J
Joined: May 2009
Posts: 1,879
That wouldn't really shock me. The 6840 has issues with the Fairlight CMI IIx driver as well.

Joined: Feb 2014
Posts: 504
G
Senior Member
Offline
Senior Member
G
Joined: Feb 2014
Posts: 504
I thought I'd chuck some LOG statements into 6840ptm.cpp and uncomment #define VERBOSE 1


Code
void ptm6840_device::update_interrupts()
{
        int new_state = ((m_status_reg & TIMER1_IRQ) && (m_control_reg[0] & INTERRUPT_EN)) ||
                                        ((m_status_reg & TIMER2_IRQ) && (m_control_reg[1] & INTERRUPT_EN)) ||
                                        ((m_status_reg & TIMER3_IRQ) && (m_control_reg[2] & INTERRUPT_EN));

        LOG("UPDATEINTER newstate=%d time=%f %d\n",new_state, machine().time().as_double(), m_irq);

        if (new_state != m_irq)
        {
                m_irq = new_state;

                if (m_irq)
                {
                        m_status_reg |= ANY_IRQ;
                }
                else
                {
                        m_status_reg &= ~ANY_IRQ;
                }

                m_irq_cb(m_irq);

                LOG("INTERRUPTS AFTER M_IRQ_CB: m_irq_cb time=%f %d\n", machine().time().as_double(), m_irq);

        }
}


So after it loads up and starts recording there's interrupts being generated continuously,

Code

[:sl2:midi:midi_ptm] UPDATEINTER newstate=0 time=30.329903 0
[:sl2:midi:midi_ptm] 30.329903: **ptm6840 t1 output 0 **
[:sl2:midi:midi_ptm] Timer #1 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #1 reload_count: clock = 1021800.000000  count = 332
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000325
[:sl2:midi:midi_ptm] **ptm6840 t2 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER newstate=1 time=30.330194 0
[:sl2:midi:midi_ptm] INTERRUPTS AFTER M_IRQ_CB: m_irq_cb time=30.330194 1
[:sl2:midi:midi_ptm] 30.330194: **ptm6840 t2 output 0 **
[:sl2:midi:midi_ptm] Timer #2 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #2 reload_count: clock = 1021800.000000  count = 10624
[:sl2:midi:midi_ptm] Timer #2 reload_count: output = 0.010397
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000325
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout** newstate=1 time=30.330228 1
[:sl2:midi:midi_ptm] 30.330228: **ptm6840 t1 output 1 **
[:sl2:midi:midi_ptm] Timer #1 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #1 reload_count: clock = 1021800.000000  count = 332
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000325
[:sl2:midi:midi_ptm] ':maincpu' (25F7): Status read = 0087
[:sl2:midi:midi_ptm] Timer #2 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #2 read counter: 10574
[:sl2:midi:midi_ptm] UPDATEINTER newstate=0 time=30.330242 1
[:sl2:midi:midi_ptm] INTERRUPTS AFTER M_IRQ_CB: m_irq_cb time=30.330242 0
[:sl2:midi:midi_ptm] ':maincpu' (25FA): Counter 2 read = 0029
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER newstate=0 time=30.330553 0
[:sl2:midi:midi_ptm] 30.330553: **ptm6840 t1 output 0 **
[:sl2:midi:midi_ptm] Timer #1 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #1 reload_count: clock = 1021800.000000  count = 332
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000325
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout**


and then at some point, after typing a bunch of midi keys, it will stop generating interrupts.

Code
[:sl2:midi:midi_ptm] UPDATEINTER newstate=0 time=30.350698 0
[:sl2:midi:midi_ptm] 30.350698: **ptm6840 t1 output 0 **
[:sl2:midi:midi_ptm] Timer #1 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #1 reload_count: clock = 1021800.000000  count = 332
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000325
[:sl2:midi:midi_ptm] **ptm6840 t2 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER newstate=1 time=30.350988 0

last interrupt gets generated here at 30.350988 and newstate seems to get stuck at 1 so it never toggles the 

[:sl2:midi:midi_ptm] INTERRUPTS AFTER M_IRQ_CB: m_irq_cb time=30.350988 1
[:sl2:midi:midi_ptm] 30.350988: **ptm6840 t2 output 0 **
[:sl2:midi:midi_ptm] Timer #2 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #2 reload_count: clock = 1021800.000000  count = 10624
[:sl2:midi:midi_ptm] Timer #2 reload_count: output = 0.010397
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER newstate=1 time=30.351023 1
[:sl2:midi:midi_ptm] 30.351023: **ptm6840 t1 output 1 **
[:sl2:midi:midi_ptm] Timer #1 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #1 reload_count: clock = 1021800.000000  count = 332
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000325
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER newstate=1 time=30.351348 1


The last time it changes the Control registers in the log is at time=20.41 seconds, setting control reg #2 to d3 which does have interrupts enabled (bit 6).

Code
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000101
[:sl2:midi:midi_ptm] Control register #2 selected
[:sl2:midi:midi_ptm] operation mode   = 010 continuous mode
[:sl2:midi:midi_ptm] value            = 00D3
[:sl2:midi:midi_ptm] t3divisor        = 1
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER newstate=0 time=20.411291 0
[:sl2:midi:midi_ptm] 20.411291: **ptm6840 t1 output 0 **
[:sl2:midi:midi_ptm] Timer #1 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #1 reload_count: clock = 1021800.000000  count = 103
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000101
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER newstate=0 time=20.411392 0
[:sl2:midi:midi_ptm] 20.411392: **ptm6840 t1 output 1 **
[:sl2:midi:midi_ptm] Timer #1 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #1 reload_count: clock = 1021800.000000  count = 103
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000101
[:sl2:midi:midi_ptm] msbbuf1 = 01
[:sl2:midi:midi_ptm] UPDATEINTER newstate=0 time=20.411446 0
[:sl2:midi:midi_ptm] ':maincpu' (23AB): Counter #1 latch = 014B
[:sl2:midi:midi_ptm] msbbuf2 = 29
[:sl2:midi:midi_ptm] UPDATEINTER newstate=0 time=20.411462 0
[:sl2:midi:midi_ptm] ':maincpu' (23B6): Counter #2 latch = 297F
[:sl2:midi:midi_ptm] Control register #1 selected
[:sl2:midi:midi_ptm] operation mode   = 010 continuous mode
[:sl2:midi:midi_ptm] value            = 0092
[:sl2:midi:midi_ptm] t3divisor        = 1
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout**



Adding a line to print the status register it gets stuck at 87 and stays there:
Code
        LOG("UPDATEINTER newstate=%d time=%f %d\n",new_state, machine().time().as_double(), m_irq);
        LOG("UPDATEINTER time=%f statusreg=%x\n",machine().time().as_double(), m_status_reg);


[:sl2:midi:midi_ptm] UPDATEINTER time=23.440963 statusreg=5
[:sl2:midi:midi_ptm] 23.440963: **ptm6840 t1 output 1 **
[:sl2:midi:midi_ptm] Timer #1 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #1 reload_count: clock = 1021800.000000  count = 332
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000325
[:sl2:midi:midi_ptm] **ptm6840 t2 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER newstate=1 time=23.440966 0
[:sl2:midi:midi_ptm] UPDATEINTER time=23.440966 statusreg=7
[:sl2:midi:midi_ptm] INTERRUPTS AFTER M_IRQ_CB: m_irq_cb time=23.440966 1
[:sl2:midi:midi_ptm] 23.440966: **ptm6840 t2 output 0 **
[:sl2:midi:midi_ptm] Timer #2 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #2 reload_count: clock = 1021800.000000  count = 10624
[:sl2:midi:midi_ptm] Timer #2 reload_count: output = 0.010397
[:sl2:midi:midi_ptm] **ptm6840 t1 timeout**
[:sl2:midi:midi_ptm] UPDATEINTER newstate=1 time=23.441288 1
[:sl2:midi:midi_ptm] UPDATEINTER time=23.441288 statusreg=87
[:sl2:midi:midi_ptm] 23.441288: **ptm6840 t1 output 0 **
[:sl2:midi:midi_ptm] Timer #1 internal clock freq 1021800.000000 
[:sl2:midi:midi_ptm] Timer #1 reload_count: clock = 1021800.000000  count = 332
[:sl2:midi:midi_ptm] Timer #1 reload_count: output = 0.000325




edit:

and if I go into the Debug Memory View and locate the MC6840 PTM/:sl2:midi:midi_ptm/0/m_status_reg and after it gets stuck,
manually set the low nibble to 0, it gets unstuck.

I can then play a few more keys before it gets stuck again, and setting the low nibble to 0 gets it unstuck again.


Also, another way to unstick it is to set the MC6840 PTM/:sl2:midi:midi_ptm/0/m_irq to 0 (gets stuck on 1)



edit:

doing some hacky mangling of ptm6840_device:timeout() stops the freezing, but playback is erratic, going superfast at times 8-)

Code
void ptm6840_device::timeout(int idx)
{
        LOG("**ptm6840 t%d timeout**\n", idx + 1);
        LOG("TIMEOUT time=%f statusreg=%x\n",machine().time().as_double(), m_status_reg);

        // Set the interrupt flag
        m_status_reg |= (1 << idx);
        m_status_read_since_int &= ~(1 << idx);
        update_interrupts();
        m_status_reg &= ~(1 << idx);    //  if I clear the status reg bits here it seems to stop the freezing



edit:

also noticed that the routine for m_ptm_irq in src/devices/bus/a2midi.cpp had a copy paste typo:


Code
WRITE_LINE_MEMBER( a2bus_midi_device::ptm_irq_w )
{
        m_ptm_irq = state ? true : false;     // should be m_ptm_irq?  ( I suppose it doesn't make too much difference )

        if (m_acia_irq || m_ptm_irq)
        {
                raise_slot_irq();
        }
        else
        {
                lower_slot_irq();
        }
}

Last edited by Golden Child; 05/31/19 11:07 PM.
Joined: Sep 2013
Posts: 14
S
Spk Offline
Member
Offline
Member
S
Joined: Sep 2013
Posts: 14
Hi RB

I've sent an apparent Apple IIe Vulcan ROM file to messdrivers. I found it on the Net. Not sure if it's real or useful. Just FYI in case the email was junked.

Cheers

Joined: Mar 2001
Posts: 16,613
R
Very Senior Member
Offline
Very Senior Member
R
Joined: Mar 2001
Posts: 16,613
Hi Spk! I haven't received the email and it's not in my Spam folder so I'm not sure what's going on.

Joined: Sep 2013
Posts: 14
S
Spk Offline
Member
Offline
Member
S
Joined: Sep 2013
Posts: 14
Mmm that is odd. I've resent. If it still doesn't go through I'll PM you a link

Joined: Mar 2001
Posts: 16,613
R
Very Senior Member
Offline
Very Senior Member
R
Joined: Mar 2001
Posts: 16,613
Got the resend, thanks!

Joined: May 2006
Posts: 130
F
Senior Member
Offline
Senior Member
F
Joined: May 2006
Posts: 130
Just going to post a brief status report here on my Apple II softlist work:

I had completely misunderstood significant things about softlists, so some side work I'd been doing for 3.5" disks was not really necessary and was backed out before it went beyond initial testing. Other than the embarrassment, it's not a bad situation as it means the end result I'm working towards will be a lot cleaner.

Honestly, part of the problem is that I've only learned what I need to know about getting these right by trial and error. I've been refining my tools and processes as I go, but some of the early data my first-generation tools generated wasn't great and requires some hand-massaging before I merge it in. This slows the process because I'm double checking my work for each entry as I go-- I'm too deep in to just throw out the first pass work and start over.

Nevertheless, we're getting there.

Joined: Mar 2001
Posts: 16,613
R
Very Senior Member
Offline
Very Senior Member
R
Joined: Mar 2001
Posts: 16,613
Sounds great. Keep up the good work!

Joined: Feb 2004
Posts: 2,194
Very Senior Member
Offline
Very Senior Member
Joined: Feb 2004
Posts: 2,194
Originally Posted by Firehawke
Just going to post a brief status report here on my Apple II softlist work:

You're doing great work, no-one's unhappy with the speed of progress. That's the way things go with these kinds of tools - you try and automate part of your workflow, find out where it's lacking, and tune it up. You don't know which parts will need tuning until you try to use it for real.

Page 40 of 61 1 2 38 39 40 41 42 60 61

Link Copied to Clipboard
Who's Online Now
2 members (Lord Nightmare, 1 invisible), 33 guests, and 2 robots.
Key: Admin, Global Mod, Mod
ShoutChat
Comment Guidelines: Do post respectful and insightful comments. Don't flame, hate, spam.
Forum Statistics
Forums9
Topics8,836
Posts116,223
Members4,921
Most Online890
Jan 17th, 2020
Powered by UBB.threads™ PHP Forum Software 7.7.5