arch_bpf_trampoline_size() provides JIT size of the BPF trampoline
before the buffer for JIT'ing it is allocated. The total number of
instructions emitted for BPF trampoline JIT code depends on where
the final image is located. So, the size arrived at with the dummy
pass in arch_bpf_trampoline_size() can vary from the actual size
needed in arch_prepare_bpf_trampoline(). When the instructions
accounted in arch_bpf_trampoline_size() is less than the number of
instructions emitted during the actual JIT compile of the trampoline,
the below warning is produced:
WARNING: CPU: 8 PID: 204190 at arch/powerpc/net/bpf_jit_comp.c:981 __arch_prepare_bpf_trampoline.isra.0+0xd2c/0xdcc
which is:
/* Make sure the trampoline generation logic doesn't overflow */
if (image && WARN_ON_ONCE(&image[ctx->idx] >
(u32 *)rw_image_end - BPF_INSN_SAFETY)) {
So, during the dummy pass, instead of providing some arbitrary image
location, account for maximum possible instructions if and when there
is a dependency with image location for JIT'ing.
Fixes: d243b62b7bd3 ("powerpc64/bpf: Add support for bpf trampolines")
Reported-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Closes: https://lore.kernel.org/all/6168bfc8-659f-4b5a-a6fb-90a916dde3b3@linux.ibm.com/
Cc: stable@vger.kernel.org # v6.13+
Acked-by: Naveen N Rao (AMD) <naveen@kernel.org>
Tested-by: Venkat Rao Bagalkote <venkat88@linux.ibm.com>
Signed-off-by: Hari Bathini <hbathini@linux.ibm.com>
Signed-off-by: Madhavan Srinivasan <maddy@linux.ibm.com>
Link: https://patch.msgid.link/20250422082609.949301-1-hbathini@linux.ibm.com
EMIT(PPC_INST_BRANCH_COND | (((cond) & 0x3ff) << 16) | (offset & 0xfffc)); \
} while (0)
-/* Sign-extended 32-bit immediate load */
+/*
+ * Sign-extended 32-bit immediate load
+ *
+ * If this is a dummy pass (!image), account for
+ * maximum possible instructions.
+ */
#define PPC_LI32(d, i) do { \
+ if (!image) \
+ ctx->idx += 2; \
+ else { \
if ((int)(uintptr_t)(i) >= -32768 && \
(int)(uintptr_t)(i) < 32768) \
EMIT(PPC_RAW_LI(d, i)); \
EMIT(PPC_RAW_LIS(d, IMM_H(i))); \
if (IMM_L(i)) \
EMIT(PPC_RAW_ORI(d, d, IMM_L(i))); \
- } } while(0)
+ } \
+ } } while (0)
#ifdef CONFIG_PPC64
+/* If dummy pass (!image), account for maximum possible instructions */
#define PPC_LI64(d, i) do { \
+ if (!image) \
+ ctx->idx += 5; \
+ else { \
if ((long)(i) >= -2147483648 && \
(long)(i) < 2147483648) \
PPC_LI32(d, i); \
if ((uintptr_t)(i) & 0x000000000000ffffULL) \
EMIT(PPC_RAW_ORI(d, d, (uintptr_t)(i) & \
0xffff)); \
- } } while (0)
+ } \
+ } } while (0)
#define PPC_LI_ADDR PPC_LI64
#ifndef CONFIG_PPC_KERNEL_PCREL
EMIT(PPC_RAW_ADDI(_R3, _R1, regs_off));
if (!p->jited)
PPC_LI_ADDR(_R4, (unsigned long)p->insnsi);
- if (!create_branch(&branch_insn, (u32 *)&ro_image[ctx->idx], (unsigned long)p->bpf_func,
- BRANCH_SET_LINK)) {
- if (image)
- image[ctx->idx] = ppc_inst_val(branch_insn);
+ /* Account for max possible instructions during dummy pass for size calculation */
+ if (image && !create_branch(&branch_insn, (u32 *)&ro_image[ctx->idx],
+ (unsigned long)p->bpf_func,
+ BRANCH_SET_LINK)) {
+ image[ctx->idx] = ppc_inst_val(branch_insn);
ctx->idx++;
} else {
EMIT(PPC_RAW_LL(_R12, _R25, offsetof(struct bpf_prog, bpf_func)));
bpf_trampoline_restore_tail_call_cnt(image, ctx, func_frame_offset, r4_off);
/* Reserve space to patch branch instruction to skip fexit progs */
- im->ip_after_call = &((u32 *)ro_image)[ctx->idx];
+ if (ro_image) /* image is NULL for dummy pass */
+ im->ip_after_call = &((u32 *)ro_image)[ctx->idx];
EMIT(PPC_RAW_NOP());
}
}
if (flags & BPF_TRAMP_F_CALL_ORIG) {
- im->ip_epilogue = &((u32 *)ro_image)[ctx->idx];
+ if (ro_image) /* image is NULL for dummy pass */
+ im->ip_epilogue = &((u32 *)ro_image)[ctx->idx];
PPC_LI_ADDR(_R3, im);
ret = bpf_jit_emit_func_call_rel(image, ro_image, ctx,
(unsigned long)__bpf_tramp_exit);
struct bpf_tramp_links *tlinks, void *func_addr)
{
struct bpf_tramp_image im;
- void *image;
int ret;
- /*
- * Allocate a temporary buffer for __arch_prepare_bpf_trampoline().
- * This will NOT cause fragmentation in direct map, as we do not
- * call set_memory_*() on this buffer.
- *
- * We cannot use kvmalloc here, because we need image to be in
- * module memory range.
- */
- image = bpf_jit_alloc_exec(PAGE_SIZE);
- if (!image)
- return -ENOMEM;
-
- ret = __arch_prepare_bpf_trampoline(&im, image, image + PAGE_SIZE, image,
- m, flags, tlinks, func_addr);
- bpf_jit_free_exec(image);
-
+ ret = __arch_prepare_bpf_trampoline(&im, NULL, NULL, NULL, m, flags, tlinks, func_addr);
return ret;
}
u64 func_addr;
u32 true_cond;
u32 tmp_idx;
- int j;
if (i && (BPF_CLASS(code) == BPF_ALU64 || BPF_CLASS(code) == BPF_ALU) &&
(BPF_CLASS(prevcode) == BPF_ALU64 || BPF_CLASS(prevcode) == BPF_ALU) &&
* 16 byte instruction that uses two 'struct bpf_insn'
*/
case BPF_LD | BPF_IMM | BPF_DW: /* dst = (u64) imm */
- tmp_idx = ctx->idx;
PPC_LI32(dst_reg_h, (u32)insn[i + 1].imm);
PPC_LI32(dst_reg, (u32)insn[i].imm);
- /* padding to allow full 4 instructions for later patching */
- if (!image)
- for (j = ctx->idx - tmp_idx; j < 4; j++)
- EMIT(PPC_RAW_NOP());
/* Adjust for two bpf instructions */
addrs[++i] = ctx->idx * 4;
break;
#ifdef CONFIG_PPC_KERNEL_PCREL
reladdr = func_addr - local_paca->kernelbase;
- if (reladdr < (long)SZ_8G && reladdr >= -(long)SZ_8G) {
+ /*
+ * If fimage is NULL (the initial pass to find image size),
+ * account for the maximum no. of instructions possible.
+ */
+ if (!fimage) {
+ ctx->idx += 7;
+ return 0;
+ } else if (reladdr < (long)SZ_8G && reladdr >= -(long)SZ_8G) {
EMIT(PPC_RAW_LD(_R12, _R13, offsetof(struct paca_struct, kernelbase)));
/* Align for subsequent prefix instruction */
if (!IS_ALIGNED((unsigned long)fimage + CTX_NIA(ctx), 8))
u64 imm64;
u32 true_cond;
u32 tmp_idx;
- int j;
/*
* addrs[] maps a BPF bytecode address into a real offset from
case BPF_LD | BPF_IMM | BPF_DW: /* dst = (u64) imm */
imm64 = ((u64)(u32) insn[i].imm) |
(((u64)(u32) insn[i+1].imm) << 32);
- tmp_idx = ctx->idx;
PPC_LI64(dst_reg, imm64);
- /* padding to allow full 5 instructions for later patching */
- if (!image)
- for (j = ctx->idx - tmp_idx; j < 5; j++)
- EMIT(PPC_RAW_NOP());
/* Adjust for two bpf instructions */
addrs[++i] = ctx->idx * 4;
break;
projects / thirdparty / kernel / linux.git / commitdiff
