xhci.c's XHCI_XFER_CONFIG_DESC_FULL case walked the descriptor set inline, in
the event loop, where nothing could test it. The walk is now
usb_config_parse() (kernel/src/usb/usb_config.c), returning what it found;
the case acts on it as before -- the same messages, the same decisions: the
first Interface descriptor only, a BOT interface's bulk endpoints (the last of
each direction), a boot keyboard's first interrupt IN endpoint, bLength 0
ending the walk, nothing read past len. One difference, in debug output only:
the bulk endpoints are logged IN then OUT, not in descriptor order.
kernel/tests/test_usb_config.c, 7 checks, pins the walk as moved: a BOT stick
as QEMU presents it, a boot keyboard with its HID descriptor, a mouse before a
drive (the first interface only), a drive with bulk IN only, bLength 0, an
endpoint cut by len, an interface cut before its protocol byte. All pass.
Built for amd64, aarch64, riscv64; the boots come with the protocol (a data
stick enumerates through this path). The fixes for #170 and #171 follow.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RzZgrf7V7Zk2iczBXDRbm
Captain Bob, 2026-11-12: "Build an appropriate test harness! Love the idea."
The kernel had no tests below the boots, so defects QEMU cannot provoke could
not be tested first. Plan: docs/v4.0.0/plans/2026-11-12-kernel-host-tests.md.
kernel/tests/: one host program per kernel source, built with that source as
it is, the kernel's flags (-std=c99 -Wall -Wextra -Werror), and fakes --
the console captured line by line, a deterministic RNG that can fail, and a RAM
disk whose reads, writes or flushes fail at chosen 1 KiB blocks.
make -C kernel/tests test runs them; SRC_<test>= runs a test against other
code, which is how a test is seen to fail against the code before a fix.
test_gpt.c, 12 checks: a table written and read back; 16, 17 and 18 used
entries against max 16 and an entry out of range after the 16th (#180); the
RNG, the first write, a later write and a flush failing (#168). Against gpt.c
before #168 and #180: 4 failures, each defect caught. Now: 0. Both runs in
logs/.../kernel-tests/. Not yet a protocol stage: to rule (the plan, step 5).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_017RzZgrf7V7Zk2iczBXDRbm
Claude