18 files changed, 429 insertions, 304 deletions

diff --git a/arch/riscv/Kconfig b/arch/riscv/Kconfig
index d8efbaa78d67..fa7dc03459e7 100644
--- a/arch/riscv/Kconfig
+++ b/arch/riscv/Kconfig
@@ -64,6 +64,8 @@ config RISCV
 	select SPARSEMEM_STATIC if 32BIT
 	select ARCH_WANT_DEFAULT_TOPDOWN_MMAP_LAYOUT if MMU
 	select HAVE_ARCH_MMAP_RND_BITS if MMU
+	select ARCH_HAS_GCOV_PROFILE_ALL
+	select HAVE_COPY_THREAD_TLS
 
 config ARCH_MMAP_RND_BITS_MIN
 	default 18 if 64BIT
diff --git a/arch/riscv/boot/dts/sifive/fu540-c000.dtsi b/arch/riscv/boot/dts/sifive/fu540-c000.dtsi
index 70a1891e7cd0..a2e3d54e830c 100644
--- a/arch/riscv/boot/dts/sifive/fu540-c000.dtsi
+++ b/arch/riscv/boot/dts/sifive/fu540-c000.dtsi
@@ -54,6 +54,7 @@
 			reg = <1>;
 			riscv,isa = "rv64imafdc";
 			tlb-split;
+			next-level-cache = <&l2cache>;
 			cpu1_intc: interrupt-controller {
 				#interrupt-cells = <1>;
 				compatible = "riscv,cpu-intc";
@@ -77,6 +78,7 @@
 			reg = <2>;
 			riscv,isa = "rv64imafdc";
 			tlb-split;
+			next-level-cache = <&l2cache>;
 			cpu2_intc: interrupt-controller {
 				#interrupt-cells = <1>;
 				compatible = "riscv,cpu-intc";
@@ -100,6 +102,7 @@
 			reg = <3>;
 			riscv,isa = "rv64imafdc";
 			tlb-split;
+			next-level-cache = <&l2cache>;
 			cpu3_intc: interrupt-controller {
 				#interrupt-cells = <1>;
 				compatible = "riscv,cpu-intc";
@@ -123,6 +126,7 @@
 			reg = <4>;
 			riscv,isa = "rv64imafdc";
 			tlb-split;
+			next-level-cache = <&l2cache>;
 			cpu4_intc: interrupt-controller {
 				#interrupt-cells = <1>;
 				compatible = "riscv,cpu-intc";
@@ -253,6 +257,17 @@
 			#pwm-cells = <3>;
 			status = "disabled";
 		};
+		l2cache: cache-controller@2010000 {
+			compatible = "sifive,fu540-c000-ccache", "cache";
+			cache-block-size = <64>;
+			cache-level = <2>;
+			cache-sets = <1024>;
+			cache-size = <2097152>;
+			cache-unified;
+			interrupt-parent = <&plic0>;
+			interrupts = <1 2 3>;
+			reg = <0x0 0x2010000 0x0 0x1000>;
+		};
 
 	};
 };
diff --git a/arch/riscv/include/asm/asm-prototypes.h b/arch/riscv/include/asm/asm-prototypes.h
index dd62b691c443..27e005fca584 100644
--- a/arch/riscv/include/asm/asm-prototypes.h
+++ b/arch/riscv/include/asm/asm-prototypes.h
@@ -5,4 +5,8 @@
 #include <linux/ftrace.h>
 #include <asm-generic/asm-prototypes.h>
 
+long long __lshrti3(long long a, int b);
+long long __ashrti3(long long a, int b);
+long long __ashlti3(long long a, int b);
+
 #endif /* _ASM_RISCV_PROTOTYPES_H */
diff --git a/arch/riscv/include/asm/csr.h b/arch/riscv/include/asm/csr.h
index 0a62d2d68455..435b65532e29 100644
--- a/arch/riscv/include/asm/csr.h
+++ b/arch/riscv/include/asm/csr.h
@@ -116,9 +116,9 @@
 # define SR_PIE		SR_MPIE
 # define SR_PP		SR_MPP
 
-# define IRQ_SOFT	IRQ_M_SOFT
-# define IRQ_TIMER	IRQ_M_TIMER
-# define IRQ_EXT	IRQ_M_EXT
+# define RV_IRQ_SOFT		IRQ_M_SOFT
+# define RV_IRQ_TIMER	IRQ_M_TIMER
+# define RV_IRQ_EXT		IRQ_M_EXT
 #else /* CONFIG_RISCV_M_MODE */
 # define CSR_STATUS	CSR_SSTATUS
 # define CSR_IE		CSR_SIE
@@ -133,15 +133,15 @@
 # define SR_PIE		SR_SPIE
 # define SR_PP		SR_SPP
 
-# define IRQ_SOFT	IRQ_S_SOFT
-# define IRQ_TIMER	IRQ_S_TIMER
-# define IRQ_EXT	IRQ_S_EXT
+# define RV_IRQ_SOFT		IRQ_S_SOFT
+# define RV_IRQ_TIMER	IRQ_S_TIMER
+# define RV_IRQ_EXT		IRQ_S_EXT
 #endif /* CONFIG_RISCV_M_MODE */
 
 /* IE/IP (Supervisor/Machine Interrupt Enable/Pending) flags */
-#define IE_SIE		(_AC(0x1, UL) << IRQ_SOFT)
-#define IE_TIE		(_AC(0x1, UL) << IRQ_TIMER)
-#define IE_EIE		(_AC(0x1, UL) << IRQ_EXT)
+#define IE_SIE		(_AC(0x1, UL) << RV_IRQ_SOFT)
+#define IE_TIE		(_AC(0x1, UL) << RV_IRQ_TIMER)
+#define IE_EIE		(_AC(0x1, UL) << RV_IRQ_EXT)
 
 #ifndef __ASSEMBLY__
 
diff --git a/arch/riscv/include/asm/perf_event.h b/arch/riscv/include/asm/perf_event.h
index aefbfaa6a781..0234048b12bc 100644
--- a/arch/riscv/include/asm/perf_event.h
+++ b/arch/riscv/include/asm/perf_event.h
@@ -82,4 +82,8 @@ struct riscv_pmu {
 	int		irq;
 };
 
+#ifdef CONFIG_PERF_EVENTS
+#define perf_arch_bpf_user_pt_regs(regs) (struct user_regs_struct *)regs
+#endif
+
 #endif /* _ASM_RISCV_PERF_EVENT_H */
diff --git a/arch/riscv/include/asm/pgtable.h b/arch/riscv/include/asm/pgtable.h
index 36ae01761352..f66b87314fa2 100644
--- a/arch/riscv/include/asm/pgtable.h
+++ b/arch/riscv/include/asm/pgtable.h
@@ -94,6 +94,10 @@ extern pgd_t swapper_pg_dir[];
 #define VMALLOC_END      (PAGE_OFFSET - 1)
 #define VMALLOC_START    (PAGE_OFFSET - VMALLOC_SIZE)
 
+#define BPF_JIT_REGION_SIZE	(SZ_128M)
+#define BPF_JIT_REGION_START	(PAGE_OFFSET - BPF_JIT_REGION_SIZE)
+#define BPF_JIT_REGION_END	(VMALLOC_END)
+
 /*
  * Roughly size the vmemmap space to be large enough to fit enough
  * struct pages to map half the virtual address space. Then
diff --git a/arch/riscv/include/asm/sifive_l2_cache.h b/arch/riscv/include/asm/sifive_l2_cache.h
deleted file mode 100644
index 04f6748fc50b..000000000000
--- a/arch/riscv/include/asm/sifive_l2_cache.h
+++ /dev/null
@@ -1,16 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-/*
- * SiFive L2 Cache Controller header file
- *
- */
-
-#ifndef _ASM_RISCV_SIFIVE_L2_CACHE_H
-#define _ASM_RISCV_SIFIVE_L2_CACHE_H
-
-extern int register_sifive_l2_error_notifier(struct notifier_block *nb);
-extern int unregister_sifive_l2_error_notifier(struct notifier_block *nb);
-
-#define SIFIVE_L2_ERR_TYPE_CE 0
-#define SIFIVE_L2_ERR_TYPE_UE 1
-
-#endif /* _ASM_RISCV_SIFIVE_L2_CACHE_H */
diff --git a/arch/riscv/include/asm/vmalloc.h b/arch/riscv/include/asm/vmalloc.h
new file mode 100644
index 000000000000..ff9abc00d139
--- /dev/null
+++ b/arch/riscv/include/asm/vmalloc.h
@@ -0,0 +1,4 @@
+#ifndef _ASM_RISCV_VMALLOC_H
+#define _ASM_RISCV_VMALLOC_H
+
+#endif /* _ASM_RISCV_VMALLOC_H */
diff --git a/arch/riscv/include/uapi/asm/bpf_perf_event.h b/arch/riscv/include/uapi/asm/bpf_perf_event.h
new file mode 100644
index 000000000000..6cb1c2823288
--- /dev/null
+++ b/arch/riscv/include/uapi/asm/bpf_perf_event.h
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
+#ifndef _UAPI__ASM_BPF_PERF_EVENT_H__
+#define _UAPI__ASM_BPF_PERF_EVENT_H__
+
+#include <asm/ptrace.h>
+
+typedef struct user_regs_struct bpf_user_pt_regs_t;
+
+#endif /* _UAPI__ASM_BPF_PERF_EVENT_H__ */
diff --git a/arch/riscv/kernel/entry.S b/arch/riscv/kernel/entry.S
index e163b7b64c86..bad4d85b5e91 100644
--- a/arch/riscv/kernel/entry.S
+++ b/arch/riscv/kernel/entry.S
@@ -155,7 +155,7 @@ _save_context:
 	REG_L x2,  PT_SP(sp)
 	.endm
 
-#if !IS_ENABLED(CONFIG_PREEMPT)
+#if !IS_ENABLED(CONFIG_PREEMPTION)
 .set resume_kernel, restore_all
 #endif
 
@@ -305,7 +305,7 @@ restore_all:
 	sret
 #endif
 
-#if IS_ENABLED(CONFIG_PREEMPT)
+#if IS_ENABLED(CONFIG_PREEMPTION)
 resume_kernel:
 	REG_L s0, TASK_TI_PREEMPT_COUNT(tp)
 	bnez s0, restore_all
diff --git a/arch/riscv/kernel/ftrace.c b/arch/riscv/kernel/ftrace.c
index b94d8db5ddcc..c40fdcdeb950 100644
--- a/arch/riscv/kernel/ftrace.c
+++ b/arch/riscv/kernel/ftrace.c
@@ -142,7 +142,7 @@ void prepare_ftrace_return(unsigned long *parent, unsigned long self_addr,
 	 */
 	old = *parent;
 
-	if (function_graph_enter(old, self_addr, frame_pointer, parent))
+	if (!function_graph_enter(old, self_addr, frame_pointer, parent))
 		*parent = return_hooker;
 }
 
diff --git a/arch/riscv/kernel/head.S b/arch/riscv/kernel/head.S
index 797802c73dee..a4242be66966 100644
--- a/arch/riscv/kernel/head.S
+++ b/arch/riscv/kernel/head.S
@@ -80,7 +80,9 @@ _start_kernel:
 
 #ifdef CONFIG_SMP
 	li t0, CONFIG_NR_CPUS
-	bgeu a0, t0, .Lsecondary_park
+	blt a0, t0, .Lgood_cores
+	tail .Lsecondary_park
+.Lgood_cores:
 #endif
 
 	/* Pick one hart to run the main boot sequence */
@@ -209,11 +211,6 @@ relocate:
 	tail smp_callin
 #endif
 
-.align 2
-.Lsecondary_park:
-	/* We lack SMP support or have too many harts, so park this hart */
-	wfi
-	j .Lsecondary_park
 END(_start)
 
 #ifdef CONFIG_RISCV_M_MODE
@@ -251,7 +248,7 @@ ENTRY(reset_regs)
 #ifdef CONFIG_FPU
 	csrr	t0, CSR_MISA
 	andi	t0, t0, (COMPAT_HWCAP_ISA_F | COMPAT_HWCAP_ISA_D)
-	bnez	t0, .Lreset_regs_done
+	beqz	t0, .Lreset_regs_done
 
 	li	t1, SR_FS
 	csrs	CSR_STATUS, t1
@@ -295,6 +292,13 @@ ENTRY(reset_regs)
 END(reset_regs)
 #endif /* CONFIG_RISCV_M_MODE */
 
+.section ".text", "ax",@progbits
+.align 2
+.Lsecondary_park:
+	/* We lack SMP support or have too many harts, so park this hart */
+	wfi
+	j .Lsecondary_park
+
 __PAGE_ALIGNED_BSS
 	/* Empty zero page */
 	.balign PAGE_SIZE
diff --git a/arch/riscv/kernel/irq.c b/arch/riscv/kernel/irq.c
index 3f07a91d5afb..345c4f2eba13 100644
--- a/arch/riscv/kernel/irq.c
+++ b/arch/riscv/kernel/irq.c
@@ -23,11 +23,11 @@ asmlinkage __visible void __irq_entry do_IRQ(struct pt_regs *regs)
 
 	irq_enter();
 	switch (regs->cause & ~CAUSE_IRQ_FLAG) {
-	case IRQ_TIMER:
+	case RV_IRQ_TIMER:
 		riscv_timer_interrupt();
 		break;
 #ifdef CONFIG_SMP
-	case IRQ_SOFT:
+	case RV_IRQ_SOFT:
 		/*
 		 * We only use software interrupts to pass IPIs, so if a non-SMP
 		 * system gets one, then we don't know what to do.
@@ -35,7 +35,7 @@ asmlinkage __visible void __irq_entry do_IRQ(struct pt_regs *regs)
 		riscv_software_interrupt();
 		break;
 #endif
-	case IRQ_EXT:
+	case RV_IRQ_EXT:
 		handle_arch_irq(regs);
 		break;
 	default:
diff --git a/arch/riscv/kernel/process.c b/arch/riscv/kernel/process.c
index 95a3031e5c7c..817cf7b0974c 100644
--- a/arch/riscv/kernel/process.c
+++ b/arch/riscv/kernel/process.c
@@ -99,8 +99,8 @@ int arch_dup_task_struct(struct task_struct *dst, struct task_struct *src)
 	return 0;
 }
 
-int copy_thread(unsigned long clone_flags, unsigned long usp,
-	unsigned long arg, struct task_struct *p)
+int copy_thread_tls(unsigned long clone_flags, unsigned long usp,
+	unsigned long arg, struct task_struct *p, unsigned long tls)
 {
 	struct pt_regs *childregs = task_pt_regs(p);
 
@@ -121,7 +121,7 @@ int copy_thread(unsigned long clone_flags, unsigned long usp,
 		if (usp) /* User fork */
 			childregs->sp = usp;
 		if (clone_flags & CLONE_SETTLS)
-			childregs->tp = childregs->a5;
+			childregs->tp = tls;
 		childregs->a0 = 0; /* Return value of fork() */
 		p->thread.ra = (unsigned long)ret_from_fork;
 	}
diff --git a/arch/riscv/kernel/vdso/Makefile b/arch/riscv/kernel/vdso/Makefile
index 49a5852fd07d..33b16f4212f7 100644
--- a/arch/riscv/kernel/vdso/Makefile
+++ b/arch/riscv/kernel/vdso/Makefile
@@ -58,7 +58,8 @@ quiet_cmd_vdsold = VDSOLD  $@
       cmd_vdsold = $(CC) $(KBUILD_CFLAGS) $(call cc-option, -no-pie) -nostdlib -nostartfiles $(SYSCFLAGS_$(@F)) \
                            -Wl,-T,$(filter-out FORCE,$^) -o $@.tmp && \
                    $(CROSS_COMPILE)objcopy \
-                           $(patsubst %, -G __vdso_%, $(vdso-syms)) $@.tmp $@
+                           $(patsubst %, -G __vdso_%, $(vdso-syms)) $@.tmp $@ && \
+                   rm $@.tmp
 
 # install commands for the unstripped file
 quiet_cmd_vdso_install = INSTALL $@
diff --git a/arch/riscv/lib/tishift.S b/arch/riscv/lib/tishift.S
index 15f9d54c7db6..ef90075c4b0a 100644
--- a/arch/riscv/lib/tishift.S
+++ b/arch/riscv/lib/tishift.S
@@ -4,34 +4,73 @@
  */
 
 #include <linux/linkage.h>
+#include <asm-generic/export.h>
 
-ENTRY(__lshrti3)
+SYM_FUNC_START(__lshrti3)
 	beqz	a2, .L1
 	li	a5,64
 	sub	a5,a5,a2
-	addi	sp,sp,-16
 	sext.w	a4,a5
 	blez	a5, .L2
 	sext.w	a2,a2
-	sll	a4,a1,a4
 	srl	a0,a0,a2
-	srl	a1,a1,a2
+	sll	a4,a1,a4
+	srl	a2,a1,a2
 	or	a0,a0,a4
-	sd	a1,8(sp)
-	sd	a0,0(sp)
-	ld	a0,0(sp)
-	ld	a1,8(sp)
-	addi	sp,sp,16
-	ret
+	mv	a1,a2
 .L1:
 	ret
 .L2:
-	negw	a4,a4
-	srl	a1,a1,a4
-	sd	a1,0(sp)
-	sd	zero,8(sp)
-	ld	a0,0(sp)
-	ld	a1,8(sp)
-	addi	sp,sp,16
+	negw	a0,a4
+	li	a2,0
+	srl	a0,a1,a0
+	mv	a1,a2
+	ret
+SYM_FUNC_END(__lshrti3)
+EXPORT_SYMBOL(__lshrti3)
+
+SYM_FUNC_START(__ashrti3)
+	beqz	a2, .L3
+	li	a5,64
+	sub	a5,a5,a2
+	sext.w	a4,a5
+	blez	a5, .L4
+	sext.w	a2,a2
+	srl	a0,a0,a2
+	sll	a4,a1,a4
+	sra	a2,a1,a2
+	or	a0,a0,a4
+	mv	a1,a2
+.L3:
+	ret
+.L4:
+	negw	a0,a4
+	srai	a2,a1,0x3f
+	sra	a0,a1,a0
+	mv	a1,a2
+	ret
+SYM_FUNC_END(__ashrti3)
+EXPORT_SYMBOL(__ashrti3)
+
+SYM_FUNC_START(__ashlti3)
+	beqz	a2, .L5
+	li	a5,64
+	sub	a5,a5,a2
+	sext.w	a4,a5
+	blez	a5, .L6
+	sext.w	a2,a2
+	sll	a1,a1,a2
+	srl	a4,a0,a4
+	sll	a2,a0,a2
+	or	a1,a1,a4
+	mv	a0,a2
+.L5:
+	ret
+.L6:
+	negw	a1,a4
+	li	a2,0
+	sll	a1,a0,a1
+	mv	a0,a2
 	ret
-ENDPROC(__lshrti3)
+SYM_FUNC_END(__ashlti3)
+EXPORT_SYMBOL(__ashlti3)
diff --git a/arch/riscv/mm/init.c b/arch/riscv/mm/init.c
index 69f6678db7f3..965a8cf4829c 100644
--- a/arch/riscv/mm/init.c
+++ b/arch/riscv/mm/init.c
@@ -99,13 +99,13 @@ static void __init setup_initrd(void)
 		pr_info("initrd not found or empty");
 		goto disable;
 	}
-	if (__pa(initrd_end) > PFN_PHYS(max_low_pfn)) {
+	if (__pa_symbol(initrd_end) > PFN_PHYS(max_low_pfn)) {
 		pr_err("initrd extends beyond end of memory");
 		goto disable;
 	}
 
 	size = initrd_end - initrd_start;
-	memblock_reserve(__pa(initrd_start), size);
+	memblock_reserve(__pa_symbol(initrd_start), size);
 	initrd_below_start_ok = 1;
 
 	pr_info("Initial ramdisk at: 0x%p (%lu bytes)\n",
@@ -124,8 +124,8 @@ void __init setup_bootmem(void)
 {
 	struct memblock_region *reg;
 	phys_addr_t mem_size = 0;
-	phys_addr_t vmlinux_end = __pa(&_end);
-	phys_addr_t vmlinux_start = __pa(&_start);
+	phys_addr_t vmlinux_end = __pa_symbol(&_end);
+	phys_addr_t vmlinux_start = __pa_symbol(&_start);
 
 	/* Find the memory region containing the kernel */
 	for_each_memblock(memory, reg) {
@@ -445,7 +445,7 @@ static void __init setup_vm_final(void)
 
 	/* Setup swapper PGD for fixmap */
 	create_pgd_mapping(swapper_pg_dir, FIXADDR_START,
-			   __pa(fixmap_pgd_next),
+			   __pa_symbol(fixmap_pgd_next),
 			   PGDIR_SIZE, PAGE_TABLE);
 
 	/* Map all memory banks */
@@ -474,7 +474,7 @@ static void __init setup_vm_final(void)
 	clear_fixmap(FIX_PMD);
 
 	/* Move to swapper page table */
-	csr_write(CSR_SATP, PFN_DOWN(__pa(swapper_pg_dir)) | SATP_MODE);
+	csr_write(CSR_SATP, PFN_DOWN(__pa_symbol(swapper_pg_dir)) | SATP_MODE);
 	local_flush_tlb_all();
 }
 #else
diff --git a/arch/riscv/net/bpf_jit_comp.c b/arch/riscv/net/bpf_jit_comp.c
index 7fbf56aab661..483f4ad7f4dc 100644
--- a/arch/riscv/net/bpf_jit_comp.c
+++ b/arch/riscv/net/bpf_jit_comp.c
@@ -120,6 +120,11 @@ static bool seen_reg(int reg, struct rv_jit_context *ctx)
 	return false;
 }
 
+static void mark_fp(struct rv_jit_context *ctx)
+{
+	__set_bit(RV_CTX_F_SEEN_S5, &ctx->flags);
+}
+
 static void mark_call(struct rv_jit_context *ctx)
 {
 	__set_bit(RV_CTX_F_SEEN_CALL, &ctx->flags);
@@ -456,6 +461,11 @@ static u32 rv_amoadd_d(u8 rd, u8 rs2, u8 rs1, u8 aq, u8 rl)
 	return rv_amo_insn(0, aq, rl, rs2, rs1, 3, rd, 0x2f);
 }
 
+static u32 rv_auipc(u8 rd, u32 imm31_12)
+{
+	return rv_u_insn(imm31_12, rd, 0x17);
+}
+
 static bool is_12b_int(s64 val)
 {
 	return -(1 << 11) <= val && val < (1 << 11);
@@ -479,27 +489,7 @@ static bool is_32b_int(s64 val)
 static int is_12b_check(int off, int insn)
 {
 	if (!is_12b_int(off)) {
-		pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
-		       insn, (int)off);
-		return -1;
-	}
-	return 0;
-}
-
-static int is_13b_check(int off, int insn)
-{
-	if (!is_13b_int(off)) {
-		pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
-		       insn, (int)off);
-		return -1;
-	}
-	return 0;
-}
-
-static int is_21b_check(int off, int insn)
-{
-	if (!is_21b_int(off)) {
-		pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
+		pr_err("bpf-jit: insn=%d 12b < offset=%d not supported yet!\n",
 		       insn, (int)off);
 		return -1;
 	}
@@ -545,10 +535,13 @@ static void emit_imm(u8 rd, s64 val, struct rv_jit_context *ctx)
 		emit(rv_addi(rd, rd, lower), ctx);
 }
 
-static int rv_offset(int bpf_to, int bpf_from, struct rv_jit_context *ctx)
+static int rv_offset(int insn, int off, struct rv_jit_context *ctx)
 {
-	int from = ctx->offset[bpf_from] - 1, to = ctx->offset[bpf_to];
+	int from, to;
 
+	off++; /* BPF branch is from PC+1, RV is from PC */
+	from = (insn > 0) ? ctx->offset[insn - 1] : 0;
+	to = (insn + off > 0) ? ctx->offset[insn + off - 1] : 0;
 	return (to - from) << 2;
 }
 
@@ -559,7 +552,7 @@ static int epilogue_offset(struct rv_jit_context *ctx)
 	return (to - from) << 2;
 }
 
-static void __build_epilogue(u8 reg, struct rv_jit_context *ctx)
+static void __build_epilogue(bool is_tail_call, struct rv_jit_context *ctx)
 {
 	int stack_adjust = ctx->stack_size, store_offset = stack_adjust - 8;
 
@@ -596,8 +589,114 @@ static void __build_epilogue(u8 reg, struct rv_jit_context *ctx)
 
 	emit(rv_addi(RV_REG_SP, RV_REG_SP, stack_adjust), ctx);
 	/* Set return value. */
-	emit(rv_addi(RV_REG_A0, RV_REG_A5, 0), ctx);
-	emit(rv_jalr(RV_REG_ZERO, reg, 0), ctx);
+	if (!is_tail_call)
+		emit(rv_addi(RV_REG_A0, RV_REG_A5, 0), ctx);
+	emit(rv_jalr(RV_REG_ZERO, is_tail_call ? RV_REG_T3 : RV_REG_RA,
+		     is_tail_call ? 4 : 0), /* skip TCC init */
+	     ctx);
+}
+
+/* return -1 or inverted cond */
+static int invert_bpf_cond(u8 cond)
+{
+	switch (cond) {
+	case BPF_JEQ:
+		return BPF_JNE;
+	case BPF_JGT:
+		return BPF_JLE;
+	case BPF_JLT:
+		return BPF_JGE;
+	case BPF_JGE:
+		return BPF_JLT;
+	case BPF_JLE:
+		return BPF_JGT;
+	case BPF_JNE:
+		return BPF_JEQ;
+	case BPF_JSGT:
+		return BPF_JSLE;
+	case BPF_JSLT:
+		return BPF_JSGE;
+	case BPF_JSGE:
+		return BPF_JSLT;
+	case BPF_JSLE:
+		return BPF_JSGT;
+	}
+	return -1;
+}
+
+static void emit_bcc(u8 cond, u8 rd, u8 rs, int rvoff,
+		     struct rv_jit_context *ctx)
+{
+	switch (cond) {
+	case BPF_JEQ:
+		emit(rv_beq(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JGT:
+		emit(rv_bltu(rs, rd, rvoff >> 1), ctx);
+		return;
+	case BPF_JLT:
+		emit(rv_bltu(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JGE:
+		emit(rv_bgeu(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JLE:
+		emit(rv_bgeu(rs, rd, rvoff >> 1), ctx);
+		return;
+	case BPF_JNE:
+		emit(rv_bne(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JSGT:
+		emit(rv_blt(rs, rd, rvoff >> 1), ctx);
+		return;
+	case BPF_JSLT:
+		emit(rv_blt(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JSGE:
+		emit(rv_bge(rd, rs, rvoff >> 1), ctx);
+		return;
+	case BPF_JSLE:
+		emit(rv_bge(rs, rd, rvoff >> 1), ctx);
+	}
+}
+
+static void emit_branch(u8 cond, u8 rd, u8 rs, int rvoff,
+			struct rv_jit_context *ctx)
+{
+	s64 upper, lower;
+
+	if (is_13b_int(rvoff)) {
+		emit_bcc(cond, rd, rs, rvoff, ctx);
+		return;
+	}
+
+	/* Adjust for jal */
+	rvoff -= 4;
+
+	/* Transform, e.g.:
+	 *   bne rd,rs,foo
+	 * to
+	 *   beq rd,rs,<.L1>
+	 *   (auipc foo)
+	 *   jal(r) foo
+	 * .L1
+	 */
+	cond = invert_bpf_cond(cond);
+	if (is_21b_int(rvoff)) {
+		emit_bcc(cond, rd, rs, 8, ctx);
+		emit(rv_jal(RV_REG_ZERO, rvoff >> 1), ctx);
+		return;
+	}
+
+	/* 32b No need for an additional rvoff adjustment, since we
+	 * get that from the auipc at PC', where PC = PC' + 4.
+	 */
+	upper = (rvoff + (1 << 11)) >> 12;
+	lower = rvoff & 0xfff;
+
+	emit_bcc(cond, rd, rs, 12, ctx);
+	emit(rv_auipc(RV_REG_T1, upper), ctx);
+	emit(rv_jalr(RV_REG_ZERO, RV_REG_T1, lower), ctx);
 }
 
 static void emit_zext_32(u8 reg, struct rv_jit_context *ctx)
@@ -627,18 +726,14 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 		return -1;
 	emit(rv_lwu(RV_REG_T1, off, RV_REG_A1), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	if (is_13b_check(off, insn))
-		return -1;
-	emit(rv_bgeu(RV_REG_A2, RV_REG_T1, off >> 1), ctx);
+	emit_branch(BPF_JGE, RV_REG_A2, RV_REG_T1, off, ctx);
 
 	/* if (TCC-- < 0)
 	 *     goto out;
 	 */
 	emit(rv_addi(RV_REG_T1, tcc, -1), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	if (is_13b_check(off, insn))
-		return -1;
-	emit(rv_blt(tcc, RV_REG_ZERO, off >> 1), ctx);
+	emit_branch(BPF_JSLT, tcc, RV_REG_ZERO, off, ctx);
 
 	/* prog = array->ptrs[index];
 	 * if (!prog)
@@ -651,18 +746,15 @@ static int emit_bpf_tail_call(int insn, struct rv_jit_context *ctx)
 		return -1;
 	emit(rv_ld(RV_REG_T2, off, RV_REG_T2), ctx);
 	off = (tc_ninsn - (ctx->ninsns - start_insn)) << 2;
-	if (is_13b_check(off, insn))
-		return -1;
-	emit(rv_beq(RV_REG_T2, RV_REG_ZERO, off >> 1), ctx);
+	emit_branch(BPF_JEQ, RV_REG_T2, RV_REG_ZERO, off, ctx);
 
 	/* goto *(prog->bpf_func + 4); */
 	off = offsetof(struct bpf_prog, bpf_func);
 	if (is_12b_check(off, insn))
 		return -1;
 	emit(rv_ld(RV_REG_T3, off, RV_REG_T2), ctx);
-	emit(rv_addi(RV_REG_T3, RV_REG_T3, 4), ctx);
 	emit(rv_addi(RV_REG_TCC, RV_REG_T1, 0), ctx);
-	__build_epilogue(RV_REG_T3, ctx);
+	__build_epilogue(true, ctx);
 	return 0;
 }
 
@@ -687,13 +779,6 @@ static void init_regs(u8 *rd, u8 *rs, const struct bpf_insn *insn,
 		*rs = bpf_to_rv_reg(insn->src_reg, ctx);
 }
 
-static int rv_offset_check(int *rvoff, s16 off, int insn,
-			   struct rv_jit_context *ctx)
-{
-	*rvoff = rv_offset(insn + off, insn, ctx);
-	return is_13b_check(*rvoff, insn);
-}
-
 static void emit_zext_32_rd_rs(u8 *rd, u8 *rs, struct rv_jit_context *ctx)
 {
 	emit(rv_addi(RV_REG_T2, *rd, 0), ctx);
@@ -726,13 +811,57 @@ static void emit_sext_32_rd(u8 *rd, struct rv_jit_context *ctx)
 	*rd = RV_REG_T2;
 }
 
+static void emit_jump_and_link(u8 rd, s64 rvoff, bool force_jalr,
+			       struct rv_jit_context *ctx)
+{
+	s64 upper, lower;
+
+	if (rvoff && is_21b_int(rvoff) && !force_jalr) {
+		emit(rv_jal(rd, rvoff >> 1), ctx);
+		return;
+	}
+
+	upper = (rvoff + (1 << 11)) >> 12;
+	lower = rvoff & 0xfff;
+	emit(rv_auipc(RV_REG_T1, upper), ctx);
+	emit(rv_jalr(rd, RV_REG_T1, lower), ctx);
+}
+
+static bool is_signed_bpf_cond(u8 cond)
+{
+	return cond == BPF_JSGT || cond == BPF_JSLT ||
+		cond == BPF_JSGE || cond == BPF_JSLE;
+}
+
+static int emit_call(bool fixed, u64 addr, struct rv_jit_context *ctx)
+{
+	s64 off = 0;
+	u64 ip;
+	u8 rd;
+
+	if (addr && ctx->insns) {
+		ip = (u64)(long)(ctx->insns + ctx->ninsns);
+		off = addr - ip;
+		if (!is_32b_int(off)) {
+			pr_err("bpf-jit: target call addr %pK is out of range\n",
+			       (void *)addr);
+			return -ERANGE;
+		}
+	}
+
+	emit_jump_and_link(RV_REG_RA, off, !fixed, ctx);
+	rd = bpf_to_rv_reg(BPF_REG_0, ctx);
+	emit(rv_addi(rd, RV_REG_A0, 0), ctx);
+	return 0;
+}
+
 static int emit_insn(const struct bpf_insn *insn, struct rv_jit_context *ctx,
 		     bool extra_pass)
 {
 	bool is64 = BPF_CLASS(insn->code) == BPF_ALU64 ||
 		    BPF_CLASS(insn->code) == BPF_JMP;
+	int s, e, rvoff, i = insn - ctx->prog->insnsi;
 	struct bpf_prog_aux *aux = ctx->prog->aux;
-	int rvoff, i = insn - ctx->prog->insnsi;
 	u8 rd = -1, rs = -1, code = insn->code;
 	s16 off = insn->off;
 	s32 imm = insn->imm;
@@ -1000,214 +1129,110 @@ out_be:
 
 	/* JUMP off */
 	case BPF_JMP | BPF_JA:
-		rvoff = rv_offset(i + off, i, ctx);
-		if (!is_21b_int(rvoff)) {
-			pr_err("bpf-jit: insn=%d offset=%d not supported yet!\n",
-			       i, rvoff);
-			return -1;
-		}
-
-		emit(rv_jal(RV_REG_ZERO, rvoff >> 1), ctx);
+		rvoff = rv_offset(i, off, ctx);
+		emit_jump_and_link(RV_REG_ZERO, rvoff, false, ctx);
 		break;
 
 	/* IF (dst COND src) JUMP off */
 	case BPF_JMP | BPF_JEQ | BPF_X:
 	case BPF_JMP32 | BPF_JEQ | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_beq(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGT | BPF_X:
 	case BPF_JMP32 | BPF_JGT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bltu(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLT | BPF_X:
 	case BPF_JMP32 | BPF_JLT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bltu(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGE | BPF_X:
 	case BPF_JMP32 | BPF_JGE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bgeu(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLE | BPF_X:
 	case BPF_JMP32 | BPF_JLE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bgeu(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JNE | BPF_X:
 	case BPF_JMP32 | BPF_JNE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bne(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGT | BPF_X:
 	case BPF_JMP32 | BPF_JSGT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_blt(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLT | BPF_X:
 	case BPF_JMP32 | BPF_JSLT | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_blt(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGE | BPF_X:
 	case BPF_JMP32 | BPF_JSGE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bge(rd, rs, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLE | BPF_X:
 	case BPF_JMP32 | BPF_JSLE | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_sext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_bge(rs, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSET | BPF_X:
 	case BPF_JMP32 | BPF_JSET | BPF_X:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		if (!is64)
-			emit_zext_32_rd_rs(&rd, &rs, ctx);
-		emit(rv_and(RV_REG_T1, rd, rs), ctx);
-		emit(rv_bne(RV_REG_T1, RV_REG_ZERO, rvoff >> 1), ctx);
+		rvoff = rv_offset(i, off, ctx);
+		if (!is64) {
+			s = ctx->ninsns;
+			if (is_signed_bpf_cond(BPF_OP(code)))
+				emit_sext_32_rd_rs(&rd, &rs, ctx);
+			else
+				emit_zext_32_rd_rs(&rd, &rs, ctx);
+			e = ctx->ninsns;
+
+			/* Adjust for extra insns */
+			rvoff -= (e - s) << 2;
+		}
+
+		if (BPF_OP(code) == BPF_JSET) {
+			/* Adjust for and */
+			rvoff -= 4;
+			emit(rv_and(RV_REG_T1, rd, rs), ctx);
+			emit_branch(BPF_JNE, RV_REG_T1, RV_REG_ZERO, rvoff,
+				    ctx);
+		} else {
+			emit_branch(BPF_OP(code), rd, rs, rvoff, ctx);
+		}
 		break;
 
 	/* IF (dst COND imm) JUMP off */
 	case BPF_JMP | BPF_JEQ | BPF_K:
 	case BPF_JMP32 | BPF_JEQ | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_beq(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGT | BPF_K:
 	case BPF_JMP32 | BPF_JGT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bltu(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLT | BPF_K:
 	case BPF_JMP32 | BPF_JLT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bltu(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JGE | BPF_K:
 	case BPF_JMP32 | BPF_JGE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bgeu(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JLE | BPF_K:
 	case BPF_JMP32 | BPF_JLE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bgeu(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JNE | BPF_K:
 	case BPF_JMP32 | BPF_JNE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_bne(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGT | BPF_K:
 	case BPF_JMP32 | BPF_JSGT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_blt(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLT | BPF_K:
 	case BPF_JMP32 | BPF_JSLT | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_blt(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSGE | BPF_K:
 	case BPF_JMP32 | BPF_JSGE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_bge(rd, RV_REG_T1, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSLE | BPF_K:
 	case BPF_JMP32 | BPF_JSLE | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
-		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_sext_32_rd(&rd, ctx);
-		emit(rv_bge(RV_REG_T1, rd, rvoff >> 1), ctx);
-		break;
 	case BPF_JMP | BPF_JSET | BPF_K:
 	case BPF_JMP32 | BPF_JSET | BPF_K:
-		if (rv_offset_check(&rvoff, off, i, ctx))
-			return -1;
+		rvoff = rv_offset(i, off, ctx);
+		s = ctx->ninsns;
 		emit_imm(RV_REG_T1, imm, ctx);
-		if (!is64)
-			emit_zext_32_rd_t1(&rd, ctx);
-		emit(rv_and(RV_REG_T1, rd, RV_REG_T1), ctx);
-		emit(rv_bne(RV_REG_T1, RV_REG_ZERO, rvoff >> 1), ctx);
+		if (!is64) {
+			if (is_signed_bpf_cond(BPF_OP(code)))
+				emit_sext_32_rd(&rd, ctx);
+			else
+				emit_zext_32_rd_t1(&rd, ctx);
+		}
+		e = ctx->ninsns;
+
+		/* Adjust for extra insns */
+		rvoff -= (e - s) << 2;
+
+		if (BPF_OP(code) == BPF_JSET) {
+			/* Adjust for and */
+			rvoff -= 4;
+			emit(rv_and(RV_REG_T1, rd, RV_REG_T1), ctx);
+			emit_branch(BPF_JNE, RV_REG_T1, RV_REG_ZERO, rvoff,
+				    ctx);
+		} else {
+			emit_branch(BPF_OP(code), rd, RV_REG_T1, rvoff, ctx);
+		}
 		break;
 
 	/* function call */
 	case BPF_JMP | BPF_CALL:
 	{
 		bool fixed;
-		int i, ret;
+		int ret;
 		u64 addr;
 
 		mark_call(ctx);
@@ -1215,20 +1240,9 @@ out_be:
 					    &fixed);
 		if (ret < 0)
 			return ret;
-		if (fixed) {
-			emit_imm(RV_REG_T1, addr, ctx);
-		} else {
-			i = ctx->ninsns;
-			emit_imm(RV_REG_T1, addr, ctx);
-			for (i = ctx->ninsns - i; i < 8; i++) {
-				/* nop */
-				emit(rv_addi(RV_REG_ZERO, RV_REG_ZERO, 0),
-				     ctx);
-			}
-		}
-		emit(rv_jalr(RV_REG_RA, RV_REG_T1, 0), ctx);
-		rd = bpf_to_rv_reg(BPF_REG_0, ctx);
-		emit(rv_addi(rd, RV_REG_A0, 0), ctx);
+		ret = emit_call(fixed, addr, ctx);
+		if (ret)
+			return ret;
 		break;
 	}
 	/* tail call */
@@ -1243,9 +1257,7 @@ out_be:
 			break;
 
 		rvoff = epilogue_offset(ctx);
-		if (is_21b_check(rvoff, i))
-			return -1;
-		emit(rv_jal(RV_REG_ZERO, rvoff >> 1), ctx);
+		emit_jump_and_link(RV_REG_ZERO, rvoff, false, ctx);
 		break;
 
 	/* dst = imm64 */
@@ -1426,6 +1438,10 @@ static void build_prologue(struct rv_jit_context *ctx)
 {
 	int stack_adjust = 0, store_offset, bpf_stack_adjust;
 
+	bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, 16);
+	if (bpf_stack_adjust)
+		mark_fp(ctx);
+
 	if (seen_reg(RV_REG_RA, ctx))
 		stack_adjust += 8;
 	stack_adjust += 8; /* RV_REG_FP */
@@ -1443,7 +1459,6 @@ static void build_prologue(struct rv_jit_context *ctx)
 		stack_adjust += 8;
 
 	stack_adjust = round_up(stack_adjust, 16);
-	bpf_stack_adjust = round_up(ctx->prog->aux->stack_depth, 16);
 	stack_adjust += bpf_stack_adjust;
 
 	store_offset = stack_adjust - 8;
@@ -1502,10 +1517,10 @@ static void build_prologue(struct rv_jit_context *ctx)
 
 static void build_epilogue(struct rv_jit_context *ctx)
 {
-	__build_epilogue(RV_REG_RA, ctx);
+	__build_epilogue(false, ctx);
 }
 
-static int build_body(struct rv_jit_context *ctx, bool extra_pass)
+static int build_body(struct rv_jit_context *ctx, bool extra_pass, int *offset)
 {
 	const struct bpf_prog *prog = ctx->prog;
 	int i;
@@ -1517,12 +1532,12 @@ static int build_body(struct rv_jit_context *ctx, bool extra_pass)
 		ret = emit_insn(insn, ctx, extra_pass);
 		if (ret > 0) {
 			i++;
-			if (ctx->insns == NULL)
-				ctx->offset[i] = ctx->ninsns;
+			if (offset)
+				offset[i] = ctx->ninsns;
 			continue;
 		}
-		if (ctx->insns == NULL)
-			ctx->offset[i] = ctx->ninsns;
+		if (offset)
+			offset[i] = ctx->ninsns;
 		if (ret)
 			return ret;
 	}
@@ -1548,9 +1563,10 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 {
 	bool tmp_blinded = false, extra_pass = false;
 	struct bpf_prog *tmp, *orig_prog = prog;
+	int pass = 0, prev_ninsns = 0, i;
 	struct rv_jit_data *jit_data;
+	unsigned int image_size = 0;
 	struct rv_jit_context *ctx;
-	unsigned int image_size;
 
 	if (!prog->jit_requested)
 		return orig_prog;
@@ -1587,33 +1603,59 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 		prog = orig_prog;
 		goto out_offset;
 	}
+	for (i = 0; i < prog->len; i++) {
+		prev_ninsns += 32;
+		ctx->offset[i] = prev_ninsns;
+	}
 
-	/* First pass generates the ctx->offset, but does not emit an image. */
-	if (build_body(ctx, extra_pass)) {
-		prog = orig_prog;
-		goto out_offset;
+	for (i = 0; i < 16; i++) {
+		pass++;
+		ctx->ninsns = 0;
+		if (build_body(ctx, extra_pass, ctx->offset)) {
+			prog = orig_prog;
+			goto out_offset;
+		}
+		build_prologue(ctx);
+		ctx->epilogue_offset = ctx->ninsns;
+		build_epilogue(ctx);
+
+		if (ctx->ninsns == prev_ninsns) {
+			if (jit_data->header)
+				break;
+
+			image_size = sizeof(u32) * ctx->ninsns;
+			jit_data->header =
+				bpf_jit_binary_alloc(image_size,
+						     &jit_data->image,
+						     sizeof(u32),
+						     bpf_fill_ill_insns);
+			if (!jit_data->header) {
+				prog = orig_prog;
+				goto out_offset;
+			}
+
+			ctx->insns = (u32 *)jit_data->image;
+			/* Now, when the image is allocated, the image
+			 * can potentially shrink more (auipc/jalr ->
+			 * jal).
+			 */
+		}
+		prev_ninsns = ctx->ninsns;
 	}
-	build_prologue(ctx);
-	ctx->epilogue_offset = ctx->ninsns;
-	build_epilogue(ctx);
 
-	/* Allocate image, now that we know the size. */
-	image_size = sizeof(u32) * ctx->ninsns;
-	jit_data->header = bpf_jit_binary_alloc(image_size, &jit_data->image,
-						sizeof(u32),
-						bpf_fill_ill_insns);
-	if (!jit_data->header) {
+	if (i == 16) {
+		pr_err("bpf-jit: image did not converge in <%d passes!\n", i);
+		bpf_jit_binary_free(jit_data->header);
 		prog = orig_prog;
 		goto out_offset;
 	}
 
-	/* Second, real pass, that acutally emits the image. */
-	ctx->insns = (u32 *)jit_data->image;
 skip_init_ctx:
+	pass++;
 	ctx->ninsns = 0;
 
 	build_prologue(ctx);
-	if (build_body(ctx, extra_pass)) {
+	if (build_body(ctx, extra_pass, NULL)) {
 		bpf_jit_binary_free(jit_data->header);
 		prog = orig_prog;
 		goto out_offset;
@@ -1621,7 +1663,7 @@ skip_init_ctx:
 	build_epilogue(ctx);
 
 	if (bpf_jit_enable > 1)
-		bpf_jit_dump(prog->len, image_size, 2, ctx->insns);
+		bpf_jit_dump(prog->len, image_size, pass, ctx->insns);
 
 	prog->bpf_func = (void *)ctx->insns;
 	prog->jited = 1;
@@ -1641,3 +1683,16 @@ out:
 					   tmp : orig_prog);
 	return prog;
 }
+
+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_EXEC, 0, NUMA_NO_NODE,
+				    __builtin_return_address(0));
+}
+
+void bpf_jit_free_exec(void *addr)
+{
+	return vfree(addr);
+}