Page Fault Vector Typo #123
Conversation
|
Nice catch! Thanks a million for the PR! Do you want to add yourself in the Acknowledgments section? |
The signature for the RSDP is "RSD PTR ", and it is not an SDT. https://uefi.org/specs/ACPI/6.5/05_ACPI_Software_Programming_Model.html?highlight=rsdt#root-system-description-table-rsdt
I'm alright thanks. All credit to y'all, great work here |
RSDP Typo
Femtoseconds conversion
Fixed the description of the APIC address (the physaddr is not shifted, but the lower bits are masked), and fixed a typo with the location. On my ``qemu-system-x86_64``, the LAPIC base is 0xFEE00000 in physical memory.
APIC Address Incorrect
|
That's it from me, you can check the intel manual for the local apic address as well, As a side note, let me know if it is not standard to merge pull requests modular like this. If you can't tell, I am new to this... |
02_Architecture/08_Timers.md
Outdated
| - Main counter value: `0xF0`. | ||
|
|
||
| We can read the main counter at any time, which is measured in timer ticks. We can convert these ticks into realtime by multiplying them with the timer period in the general capabilities register. Bits 63:32 of the general capabilities register contain the number of femtoseconds for each tick. A nanosecond is 1000 femtoseconds, and 1 second is 1'000'000'000 femtoseconds. | ||
| We can read the main counter at any time, which is measured in timer ticks. We can convert these ticks into realtime by multiplying them with the timer period in the general capabilities register. Bits 63:32 of the general capabilities register contain the number of femtoseconds for each tick. A nanosecond is 1'000'000 femtoseconds, and 1 second is 10<sup>15</sup> femtoseconds. |
There was a problem hiding this comment.
Although it could be better to have it as a power, this is not correctly rendered in our pdf conversion with pandoc.
Luckily github markdown supports latex style formula, so you probbly just need to replace it with: $10^15$ and this works also with the pdf conversion. Can you make this change please?
I forgot to answer that question, in my opinion it depends on the project mantainer usually (there are some that have strict requirements even for commit messages), in our case we don't have any strict requirement, and usually we sqash the commits into a single one, so this is not an issue for me. Maybe for small changes what I can suggest, is that you don't need to create PR against the branch with your changes, but you can commit directly there. If you have more questions about github commits you can reach us out on discord (check the Readme for the invite link) |
|
Sounds great. Thanks again for your work! |
You need to create a entry in the symbol table to reference a C-object itself, e.g. for self-referencing in a struct
2 participants
GP is vector 13 (0xD), while PF is 14 (0xE) per Intel SDM vol 3A.
https://wiki.osdev.org/Interrupt_Descriptor_Table