3b26575b4a
Three things, and the first one is the freeze.
**The loader jumped out from under itself.** A machine with a lot of RAM
stopped at "Starting the kernel" with the splash still up: no panic, no reset,
nothing. The firmware's own exception handler had the answer the whole time,
once a machine was found that printed it:
X64 Exception Type - 0E(#PF - Page-Fault)
ExceptionData - 0x10 I:1 P:0
RIP - 0000000140004DEE <- the loader's own code, at 5.4 GB
RAX - 0000000000001000 <- the page tables it had just installed
RCX - 00000000001000F0 <- the kernel it was about to jump to
I:1 P:0 is an instruction fetch from a page that is not present. The loader
built page tables covering four gigabytes, loaded cr3 -- and the firmware had
placed the loader image *above* four gigabytes, so the next instruction fetch
was unmapped. It reproduces exactly on `-m 8G` and never on `-m 1G`, which is
why every test here passed.
The PDPT is one page whether four entries are filled or all 512, so all 512
are: 512 GB of identity map, replaced by kvmalloc() a few hundred instructions
later.
**And 3 GB was never the limit, it was a circle.** PHYSTOP_MAX was 0xC0000000
because kvmalloc() mapped exactly four 1 GB pages, and kvmalloc mapped four
because that was the cap -- each holding the other up, throwing away thirteen
gigabytes of a sixteen gigabyte machine. kvmalloc now maps as many gigabytes as
the machine has RAM and tells ioremap where the direct map stopped, instead of
ioremap assuming it stopped at four.
Freeing memory becomes two passes, and the ordering is forced rather than
chosen: kinit1() runs *before* kvmalloc because kvmalloc calls kalloc for its
own page tables, and freeing a page means writing into it -- so kinit1 can only
touch what the loader mapped. kinit_high() adds the rest afterwards.
`phystop` and the new `lowtop` are now two different numbers: the DMA arena
stays below 4 GB because a device is not obliged to reach the top of memory,
and an RTL8139's buffer address register is thirty-two bits wide.
**And the 0 MB.** The System Information screen reported "0 MB usable" on a
machine with sixteen gigabytes in it, because memory_mb() had its own 16 KB
guess at how big a memory map gets, and returned zero when GetMemoryMap said
EFI_BUFFER_TOO_SMALL. That is the same bug as the previous commit's, in a
second copy, and this one failed silently. There is one buffer and one
map_read() now.
Verified at 1 GB, 8 GB and 16 GB: all three boot, no traps, no panics, socktest
and memfdtest green in each. 8 GB and 16 GB did not boot at all before -- and
16 GB now reports 17408 MB with "mem: 13312 MB above 4 GB added to the
allocator", where the old kernel would have used three. Both display
configurations still clean.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
68 lines
2.4 KiB
Text
68 lines
2.4 KiB
Text
# Build output. Everything here is reproducible from `make`.
|
|
build/
|
|
schos.img
|
|
schos.iso
|
|
*.o
|
|
*.d
|
|
*.asm
|
|
*.sym
|
|
|
|
# The staged root filesystem. Assembled by the build from cmd/, lib/ and
|
|
# ports/ -- tracking it would mean committing a second copy of every binary
|
|
# on every build.
|
|
sysroot/
|
|
|
|
# Upstream tarballs, fetched not authored: musl, toybox, mksh, dash, the DOOM
|
|
# WADs and usb.ids. 61 MB of other people's release artefacts, all of which
|
|
# the build downloads again if they are missing. The patches that matter are
|
|
# under ports/*/patches/ and those *are* tracked.
|
|
dist/
|
|
|
|
# Test and screenshot output.
|
|
*.log
|
|
*.ppm
|
|
*.png.tmp
|
|
|
|
# Claude Code's per-machine settings (permissions granted on this host).
|
|
# .claude/memory/ IS tracked -- it is the project memory, and committing it is
|
|
# how it survives a directory rename and reaches the other machines.
|
|
.claude/settings.local.json
|
|
|
|
# Emacs lock files: dangling symlinks like .#README -> user@host.pid, left
|
|
# behind when a session ends badly. One was committed in the initial import.
|
|
.#*
|
|
*~
|
|
\#*\#
|
|
|
|
# Build output under projects/ -- these are programs that run on SchOS, and
|
|
# their binaries were tracked by accident.
|
|
projects/schcom/schcom
|
|
projects/schcom/schcom-demo
|
|
|
|
# A built ISO dropped next to the site for download.html. 62 MB; not source.
|
|
docs/Website/releases/
|
|
|
|
# Kernel modules, ever. docs/plan/ldls.md rule 1: LDLS ships the layer and
|
|
# never the thing it loads, and that applies to this tree as well as to an
|
|
# image. The loader's test input is built from ldls/test/refmod.c, which is
|
|
# ours; a .ko here would be someone else's GPL code in an Apache-2.0 repo.
|
|
*.ko
|
|
*.ko.xz
|
|
|
|
# Core dumps from host-side test harnesses (tools/*test.c run on the build
|
|
# machine). One of these got committed once; the pattern is cheaper than
|
|
# remembering.
|
|
core
|
|
core.*
|
|
|
|
# Reference material: somebody else's source tree, read and never shipped.
|
|
#
|
|
# help/ is where a third-party kernel gets cloned when it is worth reading --
|
|
# Zircon, at the time of writing. It is BSD-licensed and this project is
|
|
# Apache-2.0, so the licences are compatible and that is *not* the reason it is
|
|
# ignored: it is ignored because a copy of another project's source inside this
|
|
# one is a copy that will rot, that nobody will update, and that makes the
|
|
# provenance of every file around it a question. Clone it, read it, learn the
|
|
# facts, write our own expression -- the rule docs/architecture/ldls.md section
|
|
# 3 already states for Linux drivers.
|
|
help/
|