Page 3 of 4

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:01 pm
by DavidCooper
Sorry - that was misinformation. It doesn't work! I ran the wrong version by mistake, the one that I'd modified.

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:05 pm
by DavidCooper
By the way, I meant to tell you that your clear_screen routine seems to cause the screen mode to change, resulting in a strange trail of dots along the top pixel-line of the screen instead of the name of your OS.

Bochs reports something about 640 x 480.

Don't understand that, because graphics screens appear at A0000h. Weird!

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:14 pm
by DavidCooper
The debug code (posting to the screen before the jump to the kernel) must be causing a crash due to ES not being set to 16.

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:20 pm
by antoniovazquezblanco
DavidCooper wrote:The debug code (posting to the screen before the jump to the kernel) must be causing a crash due to ES not being set to 16.
it isn' that....

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:21 pm
by DavidCooper
Yes, I've just used nops to replace the code which posts to the screen before the jump to the kernel (using latest version of your disk image), and again it works perfectly.

Your kernel code is now correct.

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:25 pm
by antoniovazquezblanco
DavidCooper wrote:Yes, I've just replaced the code posting to the screen before the jump to the kernel with nops and it works perfectly.
Not for me ¬¬

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:28 pm
by DavidCooper
antoniovazquezblanco wrote:
DavidCooper wrote:Yes, I've just replaced the code posting to the screen before the jump to the kernel with nops and it works perfectly.
Not for me ¬¬
Have you made another disk image available with that code removed? If so I'll check it. Have you tried it in Bochs or are you just using Qemu?

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:32 pm
by antoniovazquezblanco

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:36 pm
by DavidCooper
Okay. That one's just crashed in Bochs. I'm about to explore it.

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:40 pm
by DavidCooper
¡Ya está allí! (I hope that means: "It's still there!") Get rid of the code between the two jumps!

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:42 pm
by antoniovazquezblanco
DavidCooper wrote:Ya esta alli! Get rid of the code between the two jumps!
I don't get it? What codee? :P

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:45 pm
by DavidCooper
The debug code posting to the screen just before the call to the kernel is still there in the latest disk image. Unless the wrong one got loaded... Let me try it again...

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:48 pm
by DavidCooper
Here's what I'm looking at, and the bit in bold is the bit that you're supposed to have removed.
234 5 124 0 0 49 192 142 216 142 208 188 0 124 104 0 16 7 187 0

0 180 2 176 1 181 0 177 2 182 0 178 0 205 19 115 9 232 77 0

15 131 212 255 235 62 250 228 146 12 2 230 146 251 184 128 128 142 192 190

183 124 49 255 185 10 0 102 243 165 15 1 22 177 124 250 15 32 192 102

131 200 1 15 34 192 234 91 124 8 0 199 5 0 128 11 0 80 31 77

31
234 0 0 1 0 8 0 190 144 124 232 16 0 244 235 253 80 82 184

0 0 178 0 249 205 19 90 88 195 96 180 14 172 60 0 116 4 205 16

235 247 97 195 91 33 93 32 69 114 114 111 114 32 98 111 111 116 105 110

103 32 70 105 114 101 98 105 114 100 32 79 83 46 46 46 0 39 0 0

8 8 0 0 0 0 0 0 0 0 0 255 255 0 0 0 154 207 0 255

255 0 0 0 146 207 0 0 0 0 0 0 0 0 0 0 0 0 0 0
disk.dir
000 001
000 102
The 234 0 0 1 0 8 0 after it is the jump to the kernel.
The 234 91 124 8 0 before it is the jump to set CS. Didn't that have a prefix in front of it before? Let me check that...

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:50 pm
by antoniovazquezblanco
DavidCooper wrote:Here's what I'm looking at, and the bit in bold is the bit that you're supposed to have removed.
234 5 124 0 0 49 192 142 216 142 208 188 0 124 104 0 16 7 187 0

0 180 2 176 1 181 0 177 2 182 0 178 0 205 19 115 9 232 77 0

15 131 212 255 235 62 250 228 146 12 2 230 146 251 184 128 128 142 192 190

183 124 49 255 185 10 0 102 243 165 15 1 22 177 124 250 15 32 192 102

131 200 1 15 34 192 234 91 124 8 0 199 5 0 128 11 0 80 31 77

31
234 0 0 1 0 8 0 190 144 124 232 16 0 244 235 253 80 82 184

0 0 178 0 249 205 19 90 88 195 96 180 14 172 60 0 116 4 205 16

235 247 97 195 91 33 93 32 69 114 114 111 114 32 98 111 111 116 105 110

103 32 70 105 114 101 98 105 114 100 32 79 83 46 46 46 0 39 0 0

8 8 0 0 0 0 0 0 0 0 0 255 255 0 0 0 154 207 0 255

255 0 0 0 146 207 0 0 0 0 0 0 0 0 0 0 0 0 0 0
disk.dir Mach.Ed2
000 001 000 000
000 102 034 039
Are you shure you've downloaded the latest link? It doesn't appear for me!

Re: Bug jumping from bootsect to kernel

Posted: Wed Apr 27, 2011 3:56 pm
by DavidCooper
antoniovazquezblanco wrote:Are you sure you've downloaded the latest link? It doesn't appear for me!
I've just downloaded it again from the same link and it has a different time stamped on it, so I'll try running it again. While I'm doing that, have a quick look up this page at my posts as I've edited some of them a few times to add information which you may have missed.