Sending INIT/SIPI to APs from ap_init() resets them into 16-bit mode
to loop within sipi_entry().
To drive the APs into 32-bit mode, the SIPI vector needs:
1. A GDT descriptor reachable from 16-bit code (gdt32_descr).
2. A 32-bit entrypoint reachable from 16-bit code (ap_start32).
3. The locations of GDT and the 32-bit entrypoint.
Setting these up at compile time (like on non-EFI builds) is not
possible since EFI builds with -shared -fPIC and efistart64.S cannot
reference any absolute addresses.
Relative addressing is unavailable on 16-bit mode.
Moreover, EFI may not load the 32-bit entrypoint to be reachable from
16-bit mode.
To overcome these problems,
1. Fill the GDT descriptor at runtime after relocating
[sipi_entry-sipi_end] to lowmem. Since sipi_entry does not know the
address of this descriptor, use the last two bytes of SIPI page to
communicate it.
2. Place a call gate in the GDT to point to ap_start32.
3. Popluate sipi_entry() to lcall to ap_start32.
With this, the APs can transition to 32-bit mode and loop at a known
location.
Signed-off-by: Varad Gautam <varad.gautam@xxxxxxxx>
---
lib/x86/smp.c | 56 ++++++++++++++++++++++++++++++++++++++++++++
x86/efi/efistart64.S | 29 ++++++++++++++++++++++-
2 files changed, 84 insertions(+), 1 deletion(-)
diff --git a/lib/x86/smp.c b/lib/x86/smp.c
index 1a68557..33251cb 100644
--- a/lib/x86/smp.c
+++ b/lib/x86/smp.c
@@ -6,6 +6,7 @@
#include "apic.h"
#include "fwcfg.h"
#include "desc.h"
+#include "asm/page.h"
#define IPI_VECTOR 0x20
@@ -148,16 +149,71 @@ void smp_reset_apic(void)
atomic_inc(&active_cpus);
}
+#ifdef CONFIG_EFI
+extern u8 gdt32_descr, gdt32, gdt32_end;
+extern u8 ap_start32;
+#endif
+
void ap_init(void)
{
u8 *dst_addr = 0;
size_t sipi_sz = (&sipi_end - &sipi_entry) + 1;
+ assert(sipi_sz < PAGE_SIZE);
+
asm volatile("cld");
/* Relocate SIPI vector to dst_addr so it can run in 16-bit mode. */
+ memset(dst_addr, 0, PAGE_SIZE);
memcpy(dst_addr, &sipi_entry, sipi_sz);
+#ifdef CONFIG_EFI
+ volatile struct descriptor_table_ptr *gdt32_descr_rel;
+ idt_entry_t *gate_descr;
+ u16 *gdt32_descr_reladdr = (u16 *) (PAGE_SIZE - sizeof(u16));
+
+ /*
+ * gdt32_descr for CONFIG_EFI needs to be filled here dynamically
+ * since compile time calculation of offsets is not allowed when
+ * building with -shared, and rip-relative addressing is not supported
+ * in 16-bit mode.
+ *
+ * Use the last two bytes of SIPI page to store relocated gdt32_descr
+ * addr.
+ */
+ *gdt32_descr_reladdr = (&gdt32_descr - &sipi_entry);
+
+ gdt32_descr_rel = (struct descriptor_table_ptr *) ((u64) *gdt32_descr_reladdr);
+ gdt32_descr_rel->limit = (u16) (&gdt32_end - &gdt32 - 1);
+ gdt32_descr_rel->base = (ulong) ((u32) (&gdt32 - &sipi_entry));
+
+ /*
+ * EFI may not load the 32-bit AP entrypoint (ap_start32) low enough
+ * to be reachable from the SIPI vector. Since we build with -shared, this
+ * location needs to be fetched at runtime, and rip-relative addressing is
+ * not supported in 16-bit mode.
+ * To perform 16-bit -> 32-bit far jump, our options are:
+ * - ljmpl $cs, $label : unusable since $label is not known at build time.
+ * - push $cs; push $label; lret : requires an intermediate trampoline since
+ * $label must still be within 0 - 0xFFFF for 16-bit far return to work.
+ * - lcall into a call-gate : best suited.
+ *
+ * Set up call gate to ap_start32 within GDT.
+ *
+ * gdt32 layout:
+ *
+ * Entry | Segment
+ * 0 | NULL descr
+ * 1 | Code segment descr
+ * 2 | Data segment descr
+ * 3 | Call gate descr
+ */
+ gate_descr = (idt_entry_t *) ((u8 *)(&gdt32 - &sipi_entry)
+ + 3 * sizeof(gdt_entry_t));
+ set_idt_entry_t(gate_descr, sizeof(gdt_entry_t), (void *) &ap_start32,
+ 0x8 /* sel */, 0xc /* type */, 0 /* dpl */);
+#endif
+
/* INIT */
apic_icr_write(APIC_DEST_ALLBUT | APIC_DEST_PHYSICAL | APIC_DM_INIT | APIC_INT_ASSERT, 0);
diff --git a/x86/efi/efistart64.S b/x86/efi/efistart64.S
index ea3d1c0..9a0cf98 100644
--- a/x86/efi/efistart64.S
+++ b/x86/efi/efistart64.S
@@ -57,7 +57,34 @@ load_gdt_tss:
.globl sipi_entry
sipi_entry:
- jmp sipi_entry
+ mov %cr0, %eax
+ or $1, %eax
+ mov %eax, %cr0
+
+ /* Retrieve relocated gdt32_descr address at (PAGE_SIZE - 2). */
+ mov (PAGE_SIZE - 2), %ebx
+ lgdtl (%ebx)
+
+ lcall $0x18, $0x0
+
+.globl gdt32
+gdt32:
+ .quad 0
+ .quad 0x00cf9b000000ffff // flat 32-bit code segment
+ .quad 0x00cf93000000ffff // flat 32-bit data segment
+ .quad 0 // call gate to 32-bit AP entrypoint
+.globl gdt32_end
+gdt32_end:
+
+.globl gdt32_descr
+gdt32_descr:
+ .word 0
+ .long 0
.globl sipi_end
sipi_end:
+
+.code32
+.globl ap_start32
+ap_start32:
+ jmp ap_start32
--
2.32.0
