The acpi claim's mediated access and the Global Lock, and acpiserver loading the machine's tables through them by Japabu · Pull Request #749 · ToyOSOrg/ToyOS

Merge origin/main (8506640f1: #747, #748, #751) into the mediated access's branch

Japabu

and

claude

committed

18cf656

The ACPI server loads the machine's tables through the mediated access, and evaluates \_S5

Japabu

and

claude

committed

4284062

A DSDT whose FADT could not be read is refused as unread, by the kernel's name for it

Japabu

and

claude

committed

2484aa3

The mediated read passes in runtime-services data, where one machine keeps its tables

Japabu

and

claude

committed

253d4f5

The mediated read passes at an unlisted address the range registers type uncacheable

Japabu

and

claude

committed

88f38a2

Answer review round 3: the bound under 1 MiB is the policy's, registers that are off refuse, and a held Global Lock is denied

Japabu

and

claude

committed

e3449c6

Merge remote-tracking branch 'origin/main' into wt/toyos-acpis1

Japabu

committed

dbdc05f

The boot processor's range registers type an unlisted address, read once at boot

Japabu

and

claude

committed

c62f75d

Merge remote-tracking branch 'origin/main' into wt/toyos-acpis1

Japabu

committed

87bf03b

The ACPI module's header says whose range registers type an unlisted address

Japabu

and

claude

committed

3a18a85

Answer review round 4: each CPU's range registers are compared with the boot processor's, and one that differs stops the unlisted read

Japabu

and

claude

committed

9b928ca

The range registers' decision is written as CI's clippy asks

Japabu

and

claude

committed

edfd522

The claim's issue carries all three readings of the range registers

Japabu

and

claude

committed

54e08c6

Navigation Menu

Repository navigation

The acpi claim's mediated access and the Global Lock, and acpiserver loading the machine's tables through them - #749

The acpi claim's mediated access and the Global Lock, and acpiserver loading the machine's tables through them#749
Japabu merged 17 commits into
mainfrom
wt/toyos-acpis1

Commits on Oct 7, 2026

The acpi claim's mediated access: firmware memory, ports and configuration space read and written by the kernel, and the Global Lock

Merge remote-tracking branch 'origin/main' into wt/toyos-acpis1

Answer review round 1: no configuration write, whole pages and every BAR refused, the lock word typed, the stop obeyed

The ECAM test clones the memory it decides twice

Commits on Oct 8, 2026

Merge origin/main (8506640f1: #747, #748, #751) into the mediated access's branch

The ACPI server loads the machine's tables through the mediated access, and evaluates \_S5

A DSDT whose FADT could not be read is refused as unread, by the kernel's name for it

The mediated read passes in runtime-services data, where one machine keeps its tables

The mediated read passes at an unlisted address the range registers type uncacheable

Answer review round 3: the bound under 1 MiB is the policy's, registers that are off refuse, and a held Global Lock is denied

Merge remote-tracking branch 'origin/main' into wt/toyos-acpis1

The boot processor's range registers type an unlisted address, read once at boot

Merge remote-tracking branch 'origin/main' into wt/toyos-acpis1

The ACPI module's header says whose range registers type an unlisted address

Answer review round 4: each CPU's range registers are compared with the boot processor's, and one that differs stops the unlisted read

The range registers' decision is written as CI's clippy asks

The claim's issue carries all three readings of the range registers