[PATCH bpf-next 1/2] arm64/bpf: remove 128MB limit for BPF JIT programs

From: Alan Maguire
Date: Wed Nov 03 2021 - 05:50:49 EST


From: Russell King <russell.king@xxxxxxxxxx>

commit 91fc957c9b1d ("arm64/bpf: don't allocate BPF JIT programs in module memory")

...restricts BPF JIT program allocation to a 128MB region to ensure
BPF programs are still in branching range of each other. However
this restriction should not apply to the aarch64 JIT, since
BPF_JMP | BPF_CALL are implemented as a 64-bit move into a register
and then a BLR instruction - which has the effect of being able to call
anything without proximity limitation. Removing the contiguous JIT
region requires explicitly searching the bpf exception tables first
in fixup_exception(), since they are formatted differently from
the rest of the exception tables. Previously we used the fact that
the JIT memory was contiguous to identify the fact that the context
for the exception (the program counter) is a BPF program.

The practical reason to relax this restriction on JIT memory is that 128MB
of JIT memory can be quickly exhausted, especially where PAGE_SIZE is 64KB -
one page is needed per program. In cases where seccomp filters are applied
to multiple VMs on VM launch - such filters are classic BPF but converted to
BPF - this can severely limit the number of VMs that can be launched. In a
world where we support BPF JIT always on, turning off the JIT isn't always
an option either.

Fixes: 91fc957c9b1d ("arm64/bpf: don't allocate BPF JIT programs in module memory")

Suggested-by: Ard Biesheuvel <ard.biesheuvel@xxxxxxxxxx>
Signed-off-by: Russell King <russell.king@xxxxxxxxxx>
Tested-by: Alan Maguire <alan.maguire@xxxxxxxxxx>
---
arch/arm64/include/asm/extable.h | 9 ---------
arch/arm64/include/asm/memory.h | 5 +----
arch/arm64/kernel/traps.c | 2 +-
arch/arm64/mm/extable.c | 13 +++++++++----
arch/arm64/mm/ptdump.c | 2 --
arch/arm64/net/bpf_jit_comp.c | 10 ++++++----
6 files changed, 17 insertions(+), 24 deletions(-)

diff --git a/arch/arm64/include/asm/extable.h b/arch/arm64/include/asm/extable.h
index b15eb4a..840a35e 100644
--- a/arch/arm64/include/asm/extable.h
+++ b/arch/arm64/include/asm/extable.h
@@ -22,15 +22,6 @@ struct exception_table_entry

#define ARCH_HAS_RELATIVE_EXTABLE

-static inline bool in_bpf_jit(struct pt_regs *regs)
-{
- if (!IS_ENABLED(CONFIG_BPF_JIT))
- return false;
-
- return regs->pc >= BPF_JIT_REGION_START &&
- regs->pc < BPF_JIT_REGION_END;
-}
-
#ifdef CONFIG_BPF_JIT
int arm64_bpf_fixup_exception(const struct exception_table_entry *ex,
struct pt_regs *regs);
diff --git a/arch/arm64/include/asm/memory.h b/arch/arm64/include/asm/memory.h
index f1745a8..0588632 100644
--- a/arch/arm64/include/asm/memory.h
+++ b/arch/arm64/include/asm/memory.h
@@ -44,11 +44,8 @@
#define _PAGE_OFFSET(va) (-(UL(1) << (va)))
#define PAGE_OFFSET (_PAGE_OFFSET(VA_BITS))
#define KIMAGE_VADDR (MODULES_END)
-#define BPF_JIT_REGION_START (_PAGE_END(VA_BITS_MIN))
-#define BPF_JIT_REGION_SIZE (SZ_128M)
-#define BPF_JIT_REGION_END (BPF_JIT_REGION_START + BPF_JIT_REGION_SIZE)
#define MODULES_END (MODULES_VADDR + MODULES_VSIZE)
-#define MODULES_VADDR (BPF_JIT_REGION_END)
+#define MODULES_VADDR (_PAGE_END(VA_BITS_MIN))
#define MODULES_VSIZE (SZ_128M)
#define VMEMMAP_START (-(UL(1) << (VA_BITS - VMEMMAP_SHIFT)))
#define VMEMMAP_END (VMEMMAP_START + VMEMMAP_SIZE)
diff --git a/arch/arm64/kernel/traps.c b/arch/arm64/kernel/traps.c
index b03e383..fe0cd05 100644
--- a/arch/arm64/kernel/traps.c
+++ b/arch/arm64/kernel/traps.c
@@ -988,7 +988,7 @@ static int bug_handler(struct pt_regs *regs, unsigned int esr)
static int reserved_fault_handler(struct pt_regs *regs, unsigned int esr)
{
pr_err("%s generated an invalid instruction at %pS!\n",
- in_bpf_jit(regs) ? "BPF JIT" : "Kernel text patching",
+ "Kernel text patching",
(void *)instruction_pointer(regs));

/* We cannot handle this */
diff --git a/arch/arm64/mm/extable.c b/arch/arm64/mm/extable.c
index aa00601..60a8b6a 100644
--- a/arch/arm64/mm/extable.c
+++ b/arch/arm64/mm/extable.c
@@ -9,14 +9,19 @@
int fixup_exception(struct pt_regs *regs)
{
const struct exception_table_entry *fixup;
+ unsigned long addr;

- fixup = search_exception_tables(instruction_pointer(regs));
- if (!fixup)
- return 0;
+ addr = instruction_pointer(regs);

- if (in_bpf_jit(regs))
+ /* Search the BPF tables first, these are formatted differently */
+ fixup = search_bpf_extables(addr);
+ if (fixup)
return arm64_bpf_fixup_exception(fixup, regs);

+ fixup = search_exception_tables(addr);
+ if (!fixup)
+ return 0;
+
regs->pc = (unsigned long)&fixup->fixup + fixup->fixup;
return 1;
}
diff --git a/arch/arm64/mm/ptdump.c b/arch/arm64/mm/ptdump.c
index 1c40353..9bc4066 100644
--- a/arch/arm64/mm/ptdump.c
+++ b/arch/arm64/mm/ptdump.c
@@ -41,8 +41,6 @@ enum address_markers_idx {
{ 0 /* KASAN_SHADOW_START */, "Kasan shadow start" },
{ KASAN_SHADOW_END, "Kasan shadow end" },
#endif
- { BPF_JIT_REGION_START, "BPF start" },
- { BPF_JIT_REGION_END, "BPF end" },
{ MODULES_VADDR, "Modules start" },
{ MODULES_END, "Modules end" },
{ VMALLOC_START, "vmalloc() area" },
diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c
index 41c23f4..465c44d 100644
--- a/arch/arm64/net/bpf_jit_comp.c
+++ b/arch/arm64/net/bpf_jit_comp.c
@@ -1136,12 +1136,14 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
return prog;
}

+u64 bpf_jit_alloc_exec_limit(void)
+{
+ return VMALLOC_END - VMALLOC_START;
+}
+
void *bpf_jit_alloc_exec(unsigned long size)
{
- return __vmalloc_node_range(size, PAGE_SIZE, BPF_JIT_REGION_START,
- BPF_JIT_REGION_END, GFP_KERNEL,
- PAGE_KERNEL, 0, NUMA_NO_NODE,
- __builtin_return_address(0));
+ return vmalloc(size);
}

void bpf_jit_free_exec(void *addr)
--
1.8.3.1