diff --git a/lib/cpus/aarch64/cortex_a710.S b/lib/cpus/aarch64/cortex_a710.S
index 4d5d949..aea62ae 100644
--- a/lib/cpus/aarch64/cortex_a710.S
+++ b/lib/cpus/aarch64/cortex_a710.S
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2021, Arm Limited. All rights reserved.
+ * Copyright (c) 2021-2022, Arm Limited. All rights reserved.
  *
  * SPDX-License-Identifier: BSD-3-Clause
  */
@@ -10,6 +10,7 @@
 #include <cortex_a710.h>
 #include <cpu_macros.S>
 #include <plat_macros.S>
+#include "wa_cve_2022_23960_bhb_vector.S"
 
 /* Hardware handled coherency */
 #if HW_ASSISTED_COHERENCY == 0
@@ -21,6 +22,10 @@
 #error "Cortex A710 supports only AArch64. Compile with CTX_INCLUDE_AARCH32_REGS=0"
 #endif
 
+#if WORKAROUND_CVE_2022_23960
+	wa_cve_2022_23960_bhb_vector_table CORTEX_A710_BHB_LOOP_COUNT, cortex_a710
+#endif /* WORKAROUND_CVE_2022_23960 */
+
 /* --------------------------------------------------
  * Errata Workaround for Cortex-A710 Erratum 1987031.
  * This applies to revision r0p0, r1p0 and r2p0 of Cortex-A710. It is still
@@ -305,6 +310,15 @@
 	b       cpu_rev_var_ls
 endfunc check_errata_2282622
 
+func check_errata_cve_2022_23960
+#if WORKAROUND_CVE_2022_23960
+	mov	x0, #ERRATA_APPLIES
+#else
+	mov	x0, #ERRATA_MISSING
+#endif
+	ret
+endfunc check_errata_cve_2022_23960
+
 	/* ----------------------------------------------------
 	 * HW will do the cache maintenance while powering down
 	 * ----------------------------------------------------
@@ -344,6 +358,7 @@
 	report_errata ERRATA_A710_2267065, cortex_a710, 2267065
 	report_errata ERRATA_A710_2136059, cortex_a710, 2136059
 	report_errata ERRATA_A710_2282622, cortex_a710, 2282622
+	report_errata WORKAROUND_CVE_2022_23960, cortex_a710, cve_2022_23960
 
 	ldp	x8, x30, [sp], #16
 	ret
@@ -404,6 +419,15 @@
 	bl	errata_a710_2282622_wa
 #endif
 
+#if IMAGE_BL31 && WORKAROUND_CVE_2022_23960
+	/*
+	 * The Cortex-A710 generic vectors are overridden to apply errata
+         * mitigation on exception entry from lower ELs.
+         */
+	adr	x0, wa_cve_vbar_cortex_a710
+	msr	vbar_el3, x0
+#endif /* IMAGE_BL31 && WORKAROUND_CVE_2022_23960 */
+
 	isb
 	ret	x19
 endfunc cortex_a710_reset_func
diff --git a/lib/cpus/aarch64/cortex_a77.S b/lib/cpus/aarch64/cortex_a77.S
index 8c8f4d3..e7365e2 100644
--- a/lib/cpus/aarch64/cortex_a77.S
+++ b/lib/cpus/aarch64/cortex_a77.S
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2018-2021, ARM Limited and Contributors. All rights reserved.
+ * Copyright (c) 2018-2022, ARM Limited and Contributors. All rights reserved.
  *
  * SPDX-License-Identifier: BSD-3-Clause
  */
@@ -10,6 +10,7 @@
 #include <cortex_a77.h>
 #include <cpu_macros.S>
 #include <plat_macros.S>
+#include "wa_cve_2022_23960_bhb_vector.S"
 
 /* Hardware handled coherency */
 #if HW_ASSISTED_COHERENCY == 0
@@ -21,6 +22,10 @@
 #error "Cortex-A77 supports only AArch64. Compile with CTX_INCLUDE_AARCH32_REGS=0"
 #endif
 
+#if WORKAROUND_CVE_2022_23960
+	wa_cve_2022_23960_bhb_vector_table CORTEX_A77_BHB_LOOP_COUNT, cortex_a77
+#endif /* WORKAROUND_CVE_2022_23960 */
+
 	/* --------------------------------------------------
 	 * Errata Workaround for Cortex A77 Errata #1508412.
 	 * This applies only to revision <= r1p0 of Cortex A77.
@@ -194,6 +199,15 @@
 	b	cpu_rev_var_ls
 endfunc check_errata_1791578
 
+func check_errata_cve_2022_23960
+#if WORKAROUND_CVE_2022_23960
+	mov	x0, #ERRATA_APPLIES
+#else
+	mov	x0, #ERRATA_MISSING
+#endif
+	ret
+endfunc check_errata_cve_2022_23960
+
 	/* -------------------------------------------------
 	 * The CPU Ops reset function for Cortex-A77.
 	 * Shall clobber: x0-x19
@@ -224,6 +238,16 @@
 	bl	errata_a77_1791578_wa
 #endif
 
+#if IMAGE_BL31 && WORKAROUND_CVE_2022_23960
+	/*
+	 * The Cortex-A77 generic vectors are overridden to apply errata
+         * mitigation on exception entry from lower ELs.
+	 */
+	adr	x0, wa_cve_vbar_cortex_a77
+	msr	vbar_el3, x0
+#endif /* IMAGE_BL31 && WORKAROUND_CVE_2022_23960 */
+
+	isb
 	ret	x19
 endfunc cortex_a77_reset_func
 
@@ -261,6 +285,7 @@
 	report_errata ERRATA_A77_1925769, cortex_a77, 1925769
 	report_errata ERRATA_A77_1946167, cortex_a77, 1946167
 	report_errata ERRATA_A77_1791578, cortex_a77, 1791578
+	report_errata WORKAROUND_CVE_2022_23960, cortex_a77, cve_2022_23960
 
 	ldp	x8, x30, [sp], #16
 	ret
diff --git a/lib/cpus/aarch64/cortex_a78.S b/lib/cpus/aarch64/cortex_a78.S
index a1288ba..1a6f848 100644
--- a/lib/cpus/aarch64/cortex_a78.S
+++ b/lib/cpus/aarch64/cortex_a78.S
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2021, ARM Limited. All rights reserved.
+ * Copyright (c) 2019-2022, ARM Limited. All rights reserved.
  *
  * SPDX-License-Identifier: BSD-3-Clause
  */
@@ -10,12 +10,16 @@
 #include <cortex_a78.h>
 #include <cpu_macros.S>
 #include <plat_macros.S>
+#include "wa_cve_2022_23960_bhb_vector.S"
 
 /* Hardware handled coherency */
 #if HW_ASSISTED_COHERENCY == 0
 #error "cortex_a78 must be compiled with HW_ASSISTED_COHERENCY enabled"
 #endif
 
+#if WORKAROUND_CVE_2022_23960
+	wa_cve_2022_23960_bhb_vector_table CORTEX_A78_BHB_LOOP_COUNT, cortex_a78
+#endif /* WORKAROUND_CVE_2022_23960 */
 
 /* --------------------------------------------------
  * Errata Workaround for A78 Erratum 1688305.
@@ -263,6 +267,15 @@
 	b	cpu_rev_var_range
 endfunc check_errata_2242635
 
+func check_errata_cve_2022_23960
+#if WORKAROUND_CVE_2022_23960
+	mov	x0, #ERRATA_APPLIES
+#else
+	mov	x0, #ERRATA_MISSING
+#endif
+	ret
+endfunc check_errata_cve_2022_23960
+
 	/* -------------------------------------------------
 	 * The CPU Ops reset function for Cortex-A78
 	 * -------------------------------------------------
@@ -327,6 +340,15 @@
 	msr	CPUAMCNTENSET1_EL0, x0
 #endif
 
+#if IMAGE_BL31 && WORKAROUND_CVE_2022_23960
+	/*
+	 * The Cortex-A78 generic vectors are overridden to apply errata
+	 * mitigation on exception entry from lower ELs.
+	 */
+	adr	x0, wa_cve_vbar_cortex_a78
+	msr	vbar_el3, x0
+#endif /* IMAGE_BL31 && WORKAROUND_CVE_2022_23960 */
+
 	isb
 	ret	x19
 endfunc cortex_a78_reset_func
@@ -368,6 +390,7 @@
 	report_errata ERRATA_A78_1952683, cortex_a78, 1952683
 	report_errata ERRATA_A78_2132060, cortex_a78, 2132060
 	report_errata ERRATA_A78_2242635, cortex_a78, 2242635
+	report_errata WORKAROUND_CVE_2022_23960, cortex_a78, cve_2022_23960
 
 	ldp	x8, x30, [sp], #16
 	ret
diff --git a/lib/cpus/aarch64/cortex_x2.S b/lib/cpus/aarch64/cortex_x2.S
index 2ecfbbb..9586a5b 100644
--- a/lib/cpus/aarch64/cortex_x2.S
+++ b/lib/cpus/aarch64/cortex_x2.S
@@ -10,6 +10,7 @@
 #include <cortex_x2.h>
 #include <cpu_macros.S>
 #include <plat_macros.S>
+#include "wa_cve_2022_23960_bhb_vector.S"
 
 /* Hardware handled coherency */
 #if HW_ASSISTED_COHERENCY == 0
@@ -21,6 +22,10 @@
 #error "Cortex X2 supports only AArch64. Compile with CTX_INCLUDE_AARCH32_REGS=0"
 #endif
 
+#if WORKAROUND_CVE_2022_23960
+	wa_cve_2022_23960_bhb_vector_table CORTEX_X2_BHB_LOOP_COUNT, cortex_x2
+#endif /* WORKAROUND_CVE_2022_23960 */
+
 	/* --------------------------------------------------
 	 * Errata Workaround for Cortex X2 Errata #2002765.
 	 * This applies to revisions r0p0, r1p0, and r2p0 and
@@ -222,6 +227,16 @@
 	mov	x1, #0x20
 	b	cpu_rev_var_ls
 endfunc check_errata_2216384
+
+func check_errata_cve_2022_23960
+#if WORKAROUND_CVE_2022_23960
+	mov	x0, #ERRATA_APPLIES
+#else
+	mov	x0, #ERRATA_MISSING
+#endif
+	ret
+endfunc check_errata_cve_2022_23960
+
 	/* ----------------------------------------------------
 	 * HW will do the cache maintenance while powering down
 	 * ----------------------------------------------------
@@ -258,6 +273,7 @@
 	report_errata ERRATA_X2_2017096, cortex_x2, 2017096
 	report_errata ERRATA_X2_2081180, cortex_x2, 2081180
 	report_errata ERRATA_X2_2216384, cortex_x2, 2216384
+	report_errata WORKAROUND_CVE_2022_23960, cortex_x2, cve_2022_23960
 
 	ldp	x8, x30, [sp], #16
 	ret
@@ -305,6 +321,16 @@
 	bl	errata_x2_2216384_wa
 #endif
 
+#if IMAGE_BL31 && WORKAROUND_CVE_2022_23960
+	/*
+	 * The Cortex-X2 generic vectors are overridden to apply errata
+         * mitigation on exception entry from lower ELs.
+         */
+	adr	x0, wa_cve_vbar_cortex_x2
+	msr	vbar_el3, x0
+#endif /* IMAGE_BL31 && WORKAROUND_CVE_2022_23960 */
+
+	isb
 	ret x19
 endfunc cortex_x2_reset_func
 
diff --git a/lib/cpus/aarch64/neoverse_n1.S b/lib/cpus/aarch64/neoverse_n1.S
index 9c97cf6..b75b0c1 100644
--- a/lib/cpus/aarch64/neoverse_n1.S
+++ b/lib/cpus/aarch64/neoverse_n1.S
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2017-2021, Arm Limited and Contributors. All rights reserved.
+ * Copyright (c) 2017-2022, Arm Limited and Contributors. All rights reserved.
  *
  * SPDX-License-Identifier: BSD-3-Clause
  */
@@ -8,8 +8,8 @@
 #include <asm_macros.S>
 #include <cpuamu.h>
 #include <cpu_macros.S>
-#include <context.h>
 #include <neoverse_n1.h>
+#include "wa_cve_2022_23960_bhb_vector.S"
 
 /* Hardware handled coherency */
 #if HW_ASSISTED_COHERENCY == 0
@@ -23,6 +23,10 @@
 
 	.global neoverse_n1_errata_ic_trap_handler
 
+#if WORKAROUND_CVE_2022_23960
+	wa_cve_2022_23960_bhb_vector_table NEOVERSE_N1_BHB_LOOP_COUNT, neoverse_n1
+#endif /* WORKAROUND_CVE_2022_23960 */
+
 /* --------------------------------------------------
  * Errata Workaround for Neoverse N1 Erratum 1043202.
  * This applies to revision r0p0 and r1p0 of Neoverse N1.
@@ -464,6 +468,15 @@
 	b	cpu_rev_var_range
 endfunc check_errata_1946160
 
+func check_errata_cve_2022_23960
+#if WORKAROUND_CVE_2022_23960
+	mov	x0, #ERRATA_APPLIES
+#else
+	mov	x0, #ERRATA_MISSING
+#endif
+	ret
+endfunc check_errata_cve_2022_23960
+
 func neoverse_n1_reset_func
 	mov	x19, x30
 
@@ -575,6 +588,15 @@
 	bl	errata_dsu_936184_wa
 #endif
 
+#if IMAGE_BL31 && WORKAROUND_CVE_2022_23960
+	/*
+	 * The Neoverse-N1 generic vectors are overridden to apply errata
+         * mitigation on exception entry from lower ELs.
+	 */
+	adr	x0, wa_cve_vbar_neoverse_n1
+	msr	vbar_el3, x0
+#endif /* IMAGE_BL31 && WORKAROUND_CVE_2022_23960 */
+
 	isb
 	ret	x19
 endfunc neoverse_n1_reset_func
@@ -624,6 +646,7 @@
 	report_errata ERRATA_N1_1868343, neoverse_n1, 1868343
 	report_errata ERRATA_N1_1946160, neoverse_n1, 1946160
 	report_errata ERRATA_DSU_936184, neoverse_n1, dsu_936184
+	report_errata WORKAROUND_CVE_2022_23960, neoverse_n1, cve_2022_23960
 
 	ldp	x8, x30, [sp], #16
 	ret
diff --git a/lib/cpus/aarch64/neoverse_n2.S b/lib/cpus/aarch64/neoverse_n2.S
index 621aded..b93f2a6 100644
--- a/lib/cpus/aarch64/neoverse_n2.S
+++ b/lib/cpus/aarch64/neoverse_n2.S
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2020-2021, Arm Limited. All rights reserved.
+ * Copyright (c) 2020-2022, Arm Limited. All rights reserved.
  *
  * SPDX-License-Identifier: BSD-3-Clause
  */
@@ -8,6 +8,7 @@
 #include <asm_macros.S>
 #include <cpu_macros.S>
 #include <neoverse_n2.h>
+#include "wa_cve_2022_23960_bhb_vector.S"
 
 /* Hardware handled coherency */
 #if HW_ASSISTED_COHERENCY == 0
@@ -19,6 +20,10 @@
 #error "Neoverse-N2 supports only AArch64. Compile with CTX_INCLUDE_AARCH32_REGS=0"
 #endif
 
+#if WORKAROUND_CVE_2022_23960
+	wa_cve_2022_23960_bhb_vector_table NEOVERSE_N2_BHB_LOOP_COUNT, neoverse_n2
+#endif /* WORKAROUND_CVE_2022_23960 */
+
 /* --------------------------------------------------
  * Errata Workaround for Neoverse N2 Erratum 2002655.
  * This applies to revision r0p0 of Neoverse N2. it is still open.
@@ -333,6 +338,15 @@
 	b	cpu_rev_var_ls
 endfunc check_errata_2280757
 
+func check_errata_cve_2022_23960
+#if WORKAROUND_CVE_2022_23960
+	mov	x0, #ERRATA_APPLIES
+#else
+	mov	x0, #ERRATA_MISSING
+#endif
+	ret
+endfunc check_errata_cve_2022_23960
+
 	/* -------------------------------------------
 	 * The CPU Ops reset function for Neoverse N2.
 	 * -------------------------------------------
@@ -428,6 +442,15 @@
 	bl	errata_n2_2002655_wa
 #endif
 
+#if IMAGE_BL31 && WORKAROUND_CVE_2022_23960
+	/*
+	 * The Neoverse-N2 generic vectors are overridden to apply errata
+         * mitigation on exception entry from lower ELs.
+	 */
+	adr	x0, wa_cve_vbar_neoverse_n2
+	msr	vbar_el3, x0
+#endif /* IMAGE_BL31 && WORKAROUND_CVE_2022_23960 */
+
 	isb
 	ret	x19
 endfunc neoverse_n2_reset_func
@@ -469,6 +492,7 @@
 	report_errata ERRATA_N2_2138958, neoverse_n2, 2138958
 	report_errata ERRATA_N2_2242400, neoverse_n2, 2242400
 	report_errata ERRATA_N2_2280757, neoverse_n2, 2280757
+	report_errata WORKAROUND_CVE_2022_23960, neoverse_n2, cve_2022_23960
 
 	ldp	x8, x30, [sp], #16
 	ret
diff --git a/lib/cpus/aarch64/neoverse_v1.S b/lib/cpus/aarch64/neoverse_v1.S
index 62a7a30..6adb3a8 100644
--- a/lib/cpus/aarch64/neoverse_v1.S
+++ b/lib/cpus/aarch64/neoverse_v1.S
@@ -1,5 +1,5 @@
 /*
- * Copyright (c) 2019-2021, Arm Limited. All rights reserved.
+ * Copyright (c) 2019-2022, Arm Limited. All rights reserved.
  *
  * SPDX-License-Identifier: BSD-3-Clause
  */
@@ -10,6 +10,7 @@
 #include <neoverse_v1.h>
 #include <cpu_macros.S>
 #include <plat_macros.S>
+#include "wa_cve_2022_23960_bhb_vector.S"
 
 /* Hardware handled coherency */
 #if HW_ASSISTED_COHERENCY == 0
@@ -21,6 +22,10 @@
 #error "Neoverse-V1 supports only AArch64. Compile with CTX_INCLUDE_AARCH32_REGS=0"
 #endif
 
+#if WORKAROUND_CVE_2022_23960
+	wa_cve_2022_23960_bhb_vector_table NEOVERSE_V1_BHB_LOOP_COUNT, neoverse_v1
+#endif /* WORKAROUND_CVE_2022_23960 */
+
 	/* --------------------------------------------------
 	 * Errata Workaround for Neoverse V1 Errata #1774420.
 	 * This applies to revisions r0p0 and r1p0, fixed in r1p1.
@@ -325,6 +330,15 @@
 	b	cpu_rev_var_range
 endfunc check_errata_2216392
 
+func check_errata_cve_2022_23960
+#if WORKAROUND_CVE_2022_23960
+	mov	x0, #ERRATA_APPLIES
+#else
+	mov	x0, #ERRATA_MISSING
+#endif
+	ret
+endfunc check_errata_cve_2022_23960
+
 	/* ---------------------------------------------
 	 * HW will do the cache maintenance while powering down
 	 * ---------------------------------------------
@@ -364,6 +378,7 @@
 	report_errata ERRATA_V1_2139242, neoverse_v1, 2139242
 	report_errata ERRATA_V1_2108267, neoverse_v1, 2108267
 	report_errata ERRATA_V1_2216392, neoverse_v1, 2216392
+	report_errata WORKAROUND_CVE_2022_23960, neoverse_v1, cve_2022_23960
 
 	ldp	x8, x30, [sp], #16
 	ret
@@ -422,6 +437,16 @@
 	bl	errata_neoverse_v1_2216392_wa
 #endif
 
+#if IMAGE_BL31 && WORKAROUND_CVE_2022_23960
+	/*
+	 * The Neoverse-V1 generic vectors are overridden to apply errata
+         * mitigation on exception entry from lower ELs.
+	 */
+	adr	x0, wa_cve_vbar_neoverse_v1
+	msr	vbar_el3, x0
+#endif /* IMAGE_BL31 && WORKAROUND_CVE_2022_23960 */
+
+	isb
 	ret	x19
 endfunc neoverse_v1_reset_func
 
diff --git a/lib/cpus/aarch64/wa_cve_2022_23960_bhb.S b/lib/cpus/aarch64/wa_cve_2022_23960_bhb.S
new file mode 100644
index 0000000..e0e41cc
--- /dev/null
+++ b/lib/cpus/aarch64/wa_cve_2022_23960_bhb.S
@@ -0,0 +1,31 @@
+/*
+ * Copyright (c) 2022, ARM Limited and Contributors. All rights reserved.
+ *
+ * SPDX-License-Identifier: BSD-3-Clause
+ */
+
+#include <context.h>
+
+#if WORKAROUND_CVE_2022_23960
+	/*
+	 * This macro applies the mitigation for CVE-2022-23960.
+         * The macro saves x2-x3 to the CPU context.
+         * SP should point to the CPU context.
+	 */
+	.macro	apply_cve_2022_23960_bhb_wa _bhb_loop_count
+	stp	x2, x3, [sp, #CTX_GPREGS_OFFSET + CTX_GPREG_X2]
+
+	/* CVE-BHB-NUM loop count */
+	mov	x2, \_bhb_loop_count
+
+1:
+	/* b pc+4 part of the workaround */
+	b	2f
+2:
+	subs	x2, x2, #1
+	bne	1b
+	dsb	sy
+	isb
+	ldp	x2, x3, [sp, #CTX_GPREGS_OFFSET + CTX_GPREG_X2]
+	.endm
+#endif /* WORKAROUND_CVE_2022_23960 */
diff --git a/lib/cpus/aarch64/wa_cve_2022_23960_bhb_vector.S b/lib/cpus/aarch64/wa_cve_2022_23960_bhb_vector.S
new file mode 100644
index 0000000..220fa11
--- /dev/null
+++ b/lib/cpus/aarch64/wa_cve_2022_23960_bhb_vector.S
@@ -0,0 +1,108 @@
+/*
+ * Copyright (c) 2022, Arm Limited. All rights reserved.
+ *
+ * SPDX-License-Identifier: BSD-3-Clause
+ */
+
+#include <arch.h>
+#include <asm_macros.S>
+#include <services/arm_arch_svc.h>
+#include "wa_cve_2022_23960_bhb.S"
+
+	/*
+	 * This macro is used to isolate the vector table for relevant CPUs
+	 * used in the mitigation for CVE_2022_23960.
+	 */
+	.macro wa_cve_2022_23960_bhb_vector_table _bhb_loop_count, _cpu
+
+	.globl	wa_cve_vbar_\_cpu
+
+vector_base wa_cve_vbar_\_cpu
+	/* ---------------------------------------------------------------------
+	 * Current EL with SP_EL0 : 0x0 - 0x200
+	 * ---------------------------------------------------------------------
+	 */
+vector_entry bhb_sync_exception_sp_el0_\_cpu
+	b	sync_exception_sp_el0
+end_vector_entry bhb_sync_exception_sp_el0_\_cpu
+
+vector_entry bhb_irq_sp_el0_\_cpu
+	b	irq_sp_el0
+end_vector_entry bhb_irq_sp_el0_\_cpu
+
+vector_entry bhb_fiq_sp_el0_\_cpu
+	b	fiq_sp_el0
+end_vector_entry bhb_fiq_sp_el0_\_cpu
+
+vector_entry bhb_serror_sp_el0_\_cpu
+	b	serror_sp_el0
+end_vector_entry bhb_serror_sp_el0_\_cpu
+
+	/* ---------------------------------------------------------------------
+	 * Current EL with SP_ELx: 0x200 - 0x400
+	 * ---------------------------------------------------------------------
+	 */
+vector_entry bhb_sync_exception_sp_elx_\_cpu
+	b	sync_exception_sp_elx
+end_vector_entry bhb_sync_exception_sp_elx_\_cpu
+
+vector_entry bhb_irq_sp_elx_\_cpu
+	b	irq_sp_elx
+end_vector_entry bhb_irq_sp_elx_\_cpu
+
+vector_entry bhb_fiq_sp_elx_\_cpu
+	b	fiq_sp_elx
+end_vector_entry bhb_fiq_sp_elx_\_cpu
+
+vector_entry bhb_serror_sp_elx_\_cpu
+	b	serror_sp_elx
+end_vector_entry bhb_serror_sp_elx_\_cpu
+
+	/* ---------------------------------------------------------------------
+	 * Lower EL using AArch64 : 0x400 - 0x600
+	 * ---------------------------------------------------------------------
+	 */
+vector_entry bhb_sync_exception_aarch64_\_cpu
+	apply_cve_2022_23960_bhb_wa \_bhb_loop_count
+	b	sync_exception_aarch64
+end_vector_entry bhb_sync_exception_aarch64_\_cpu
+
+vector_entry bhb_irq_aarch64_\_cpu
+	apply_cve_2022_23960_bhb_wa \_bhb_loop_count
+	b	irq_aarch64
+end_vector_entry bhb_irq_aarch64_\_cpu
+
+vector_entry bhb_fiq_aarch64_\_cpu
+	apply_cve_2022_23960_bhb_wa \_bhb_loop_count
+	b	fiq_aarch64
+end_vector_entry bhb_fiq_aarch64_\_cpu
+
+vector_entry bhb_serror_aarch64_\_cpu
+	apply_cve_2022_23960_bhb_wa \_bhb_loop_count
+	b	serror_aarch64
+end_vector_entry bhb_serror_aarch64_\_cpu
+
+	/* ---------------------------------------------------------------------
+	 * Lower EL using AArch32 : 0x600 - 0x800
+	 * ---------------------------------------------------------------------
+	 */
+vector_entry bhb_sync_exception_aarch32_\_cpu
+	apply_cve_2022_23960_bhb_wa \_bhb_loop_count
+	b	sync_exception_aarch32
+end_vector_entry bhb_sync_exception_aarch32_\_cpu
+
+vector_entry bhb_irq_aarch32_\_cpu
+	apply_cve_2022_23960_bhb_wa \_bhb_loop_count
+	b	irq_aarch32
+end_vector_entry bhb_irq_aarch32_\_cpu
+
+vector_entry bhb_fiq_aarch32_\_cpu
+	apply_cve_2022_23960_bhb_wa \_bhb_loop_count
+	b	fiq_aarch32
+end_vector_entry bhb_fiq_aarch32_\_cpu
+
+vector_entry bhb_serror_aarch32_\_cpu
+	apply_cve_2022_23960_bhb_wa \_bhb_loop_count
+	b	serror_aarch32
+end_vector_entry bhb_serror_aarch32_\_cpu
+	.endm
diff --git a/lib/cpus/cpu-ops.mk b/lib/cpus/cpu-ops.mk
index e812c07..c7630fb 100644
--- a/lib/cpus/cpu-ops.mk
+++ b/lib/cpus/cpu-ops.mk
@@ -24,6 +24,7 @@
 WORKAROUND_CVE_2017_5715		?=1
 WORKAROUND_CVE_2018_3639		?=1
 DYNAMIC_WORKAROUND_CVE_2018_3639	?=0
+WORKAROUND_CVE_2022_23960		?=1
 
 # Flags to indicate internal or external Last level cache
 # By default internal
@@ -56,6 +57,10 @@
 $(eval $(call assert_boolean,DYNAMIC_WORKAROUND_CVE_2018_3639))
 $(eval $(call add_define,DYNAMIC_WORKAROUND_CVE_2018_3639))
 
+# Process WORKAROUND_CVE_2022_23960 flag
+$(eval $(call assert_boolean,WORKAROUND_CVE_2022_23960))
+$(eval $(call add_define,WORKAROUND_CVE_2022_23960))
+
 $(eval $(call assert_boolean,NEOVERSE_Nx_EXTERNAL_LLC))
 $(eval $(call add_define,NEOVERSE_Nx_EXTERNAL_LLC))