ldls: a second driver, and everything it found

e1000 -- Linux's, unmodified -- probes, opens, gets an address by DHCP and
carries ping traffic in a SchOS process. The kernel was driving that NIC and
gave it up, which makes it the first real handover as well as the second
driver.

None of what it needed was a missing feature. Every one was a call that
succeeded, returned zero and did nothing:

  _find_next_bit          "bit 0 is set"          -- for_each_cpu never ended
  netif_carrier_off       carrier already up      -- TCTL.EN never set
  __x86_indirect_thunk_r8 the jump discarded      -- 255 rx buffers, none made
  dma_map_page_attrs      bus address 0           -- NIC aimed at physical 0

so the rule from memcpy is now general: a stub returning zero is unsafe
wherever zero is a meaningful answer.

The loader grew ko_fixup_percpu(). The stack guard turned out to be one
per-CPU variable among several; `current` cannot be defused because the driver
dereferences it. Taking the %gs off instead leaves the same opcode reading the
same width from storage the host supplies -- a per-CPU variable in a
single-threaded host is just a variable. Eight bytes become seven and a nop,
and the displacement is not always last, so there is now just enough of an x86
decoder in ko.c to know how long the instruction is.

ldls/kpi/thunk.c is fifteen lines of assembly and the highest-value file here:
retpolines are control flow, and a stub cannot stand in for a jump.

The kernel side:

  - ioperm.c, the TSS I/O permission bitmap. A driver's outl is an instruction,
    not a call, and e1000's reset path picks it deliberately. IOPL=3 would hand
    over the PIC and the disk with it, so it is per-port. Writing it found that
    the I/O map base had been written at offset 64 rather than 102, leaving the
    CPU checking ports 0-831 against the TSS's own RSP0.
  - claim masks the device's IRQ line: 450,000 a second, otherwise.
  - net_detach(), the missing half of net_attach_late(). An interface left in
    the stack keeps its route and sends every frame to a driver that refuses
    it, which looks exactly like a userspace driver that does not work.
  - e1000_release() gives up the interface, not just the hardware.

8139too still passes 4/4; ldls-check still passes 27; four CPUs, no panic.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01VQ7Qx8E1q1M1bbCq8QLHS2

ldls/ko.h

@@ -225,6 +225,35 @@ int ko_relocate(struct ko *k, ko_resolver r, void *ctx);
 // should not be editing it.
 int ko_defuse_gs(struct ko *k, const char *sym);
 
+// Turn every *other* `%gs:`-relative reference into an ordinary one.
+//
+// The stack guard is one per-CPU variable; there are others, and `current` is
+// the one every driver touches. `mov %gs:const_current_task(%rip), %r8` reads
+// a pointer out of the kernel's per-CPU area, and a host that is not that
+// kernel has no such area -- so on SchOS it reads a kernel address from user
+// mode and faults.
+//
+// Defusing is not available here: the guard's *value* is arbitrary, so zeroing
+// a register is a correct answer, and `current`'s is not -- the driver will
+// dereference it.
+//
+// What works instead is that **a per-CPU variable in a single-threaded host is
+// just a variable**. Taking the `%gs` off leaves the same instruction reading
+// the same width in the same direction, from storage the embedder supplies for
+// that name. Eight bytes become seven and a nop, and the driver never learns
+// the difference.
+//
+// `percpu` is asked for an address for each name and should return the same
+// one every time it is asked for that name -- and should initialise it, since
+// the driver now reads its *contents*: zero is right for a counter and wrong
+// for `current`. Returning 0 refuses the object.
+//
+// Call after ko_relocate() and after ko_defuse_gs(), for the same reason.
+// Returns how many it rewrote, or -1.
+int ko_fixup_percpu(struct ko *k,
+                    unsigned long long (*percpu)(void *ctx, const char *name),
+                    void *ctx);
+
 // A value from the `.modinfo` section -- "license", "description", "author",
 // "alias". Returns 0 if the object has no such key.
 //