[PATCH 00/11] powerpc/fsl: NXP PowerPC Spectre variant 2 workarounds

STALE2781d

14 messages, 3 authors, 2018-12-22 · open the first message on its own page

[PATCH 00/11] powerpc/fsl: NXP PowerPC Spectre variant 2 workarounds

From: Diana Craciun <hidden>
Date: 2018-12-12 14:25:16

Implement Spectre variant 2 workarounds for NXP PowerPC Book3E
processors.

Diana Craciun (11):
  Add infrastructure to fixup branch predictor flush
  Add macro to flush the branch predictor
  Fix spectre_v2 mitigations reporting
  Emulate SPRN_BUCSR register
  Add nospectre_v2 command line argument
  Flush the branch predictor at each kernel entry (64bit)
  Flush the branch predictor at each kernel entry (32 bit)
  Flush branch predictor when entering KVM
  Enable runtime patching if nospectre_v2 boot arg is used
  Update Spectre v2 reporting
  Add FSL_PPC_BOOK3E as supported arch for nospectre_v2 boot arg

 Documentation/admin-guide/kernel-parameters.txt |  2 +-
 arch/powerpc/include/asm/feature-fixups.h       | 12 +++++++++++
 arch/powerpc/include/asm/ppc_asm.h              | 10 +++++++++
 arch/powerpc/include/asm/setup.h                |  7 +++++++
 arch/powerpc/kernel/entry_64.S                  |  5 +++++
 arch/powerpc/kernel/exceptions-64e.S            | 26 ++++++++++++++++++++++-
 arch/powerpc/kernel/head_booke.h                |  6 ++++++
 arch/powerpc/kernel/head_fsl_booke.S            | 15 +++++++++++++
 arch/powerpc/kernel/security.c                  | 28 +++++++++++++++++++++++--
 arch/powerpc/kernel/setup-common.c              |  1 +
 arch/powerpc/kernel/vmlinux.lds.S               |  8 +++++++
 arch/powerpc/kvm/bookehv_interrupts.S           |  4 ++++
 arch/powerpc/kvm/e500_emulate.c                 |  5 +++++
 arch/powerpc/lib/feature-fixups.c               | 21 +++++++++++++++++++
 arch/powerpc/mm/tlb_low_64e.S                   |  7 +++++++
 15 files changed, 153 insertions(+), 4 deletions(-)

-- 
2.5.5

[PATCH 05/11] powerpc/fsl: Add nospectre_v2 command line argument

From: Diana Craciun <hidden>
Date: 2018-12-12 14:05:57

When the command line argument is present, the Spectre variant 2
mitigations are disabled.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/include/asm/setup.h |  5 +++++
 arch/powerpc/kernel/security.c   | 21 +++++++++++++++++++++
 2 files changed, 26 insertions(+)
diff --git a/arch/powerpc/include/asm/setup.h b/arch/powerpc/include/asm/setup.h
index c941c8c..65676e2 100644
--- a/arch/powerpc/include/asm/setup.h
+++ b/arch/powerpc/include/asm/setup.h
@@ -67,6 +67,11 @@ void do_barrier_nospec_fixups_range(bool enable, void *start, void *end);
 static inline void do_barrier_nospec_fixups_range(bool enable, void *start, void *end) { };
 #endif
 
+#ifdef CONFIG_PPC_FSL_BOOK3E
+void setup_spectre_v2(void);
+#else
+static inline void setup_spectre_v2(void) {};
+#endif
 void do_btb_flush_fixups(void);
 
 #endif /* !__ASSEMBLY__ */
diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c
index 1b395b8..4393a38 100644
--- a/arch/powerpc/kernel/security.c
+++ b/arch/powerpc/kernel/security.c
@@ -26,6 +26,10 @@ static enum count_cache_flush_type count_cache_flush_type = COUNT_CACHE_FLUSH_NO
 
 bool barrier_nospec_enabled;
 static bool no_nospec;
+static bool btb_flush_enabled;
+#ifdef CONFIG_PPC_FSL_BOOK3E
+static bool no_spectrev2;
+#endif
 
 static void enable_barrier_nospec(bool enable)
 {
@@ -101,6 +105,23 @@ static __init int barrier_nospec_debugfs_init(void)
 device_initcall(barrier_nospec_debugfs_init);
 #endif /* CONFIG_DEBUG_FS */
 
+#ifdef CONFIG_PPC_FSL_BOOK3E
+static int __init handle_nospectre_v2(char *p)
+{
+	no_spectrev2 = true;
+
+	return 0;
+}
+early_param("nospectre_v2", handle_nospectre_v2);
+void setup_spectre_v2(void)
+{
+	if (no_spectrev2)
+		do_btb_flush_fixups();
+	else
+		btb_flush_enabled = true;
+}
+#endif /* CONFIG_PPC_FSL_BOOK3E */
+
 #ifdef CONFIG_PPC_BOOK3S_64
 ssize_t cpu_show_meltdown(struct device *dev, struct device_attribute *attr, char *buf)
 {
-- 
2.5.5

[PATCH 06/11] powerpc/fsl: Flush the branch predictor at each kernel entry (64bit)

From: Diana Craciun <hidden>
Date: 2018-12-12 14:07:57

In order to protect against speculation attacks on
indirect branches, the branch predictor is flushed at
kernel entry to protect for the following situations:
- userspace process attacking another userspace process
- userspace process attacking the kernel
Basically when the privillege level change (i.e. the
kernel is entered), the branch predictor state is flushed.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/kernel/entry_64.S       |  5 +++++
 arch/powerpc/kernel/exceptions-64e.S | 26 +++++++++++++++++++++++++-
 arch/powerpc/mm/tlb_low_64e.S        |  7 +++++++
 3 files changed, 37 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
index 7b1693a..7c2032e 100644
--- a/arch/powerpc/kernel/entry_64.S
+++ b/arch/powerpc/kernel/entry_64.S
@@ -80,6 +80,11 @@ END_FTR_SECTION_IFSET(CPU_FTR_TM)
 	std	r0,GPR0(r1)
 	std	r10,GPR1(r1)
 	beq	2f			/* if from kernel mode */
+#ifdef CONFIG_PPC_FSL_BOOK3E
+START_BTB_FLUSH_SECTION
+	BTB_FLUSH(r10)
+END_BTB_FLUSH_SECTION
+#endif
 	ACCOUNT_CPU_USER_ENTRY(r13, r10, r11)
 2:	std	r2,GPR2(r1)
 	std	r3,GPR3(r1)
diff --git a/arch/powerpc/kernel/exceptions-64e.S b/arch/powerpc/kernel/exceptions-64e.S
index 6d6e144..afb6387 100644
--- a/arch/powerpc/kernel/exceptions-64e.S
+++ b/arch/powerpc/kernel/exceptions-64e.S
@@ -296,7 +296,8 @@ ret_from_mc_except:
 	andi.	r10,r11,MSR_PR;		/* save stack pointer */	    \
 	beq	1f;			/* branch around if supervisor */   \
 	ld	r1,PACAKSAVE(r13);	/* get kernel stack coming from usr */\
-1:	cmpdi	cr1,r1,0;		/* check if SP makes sense */	    \
+1:	type##_BTB_FLUSH		\
+	cmpdi	cr1,r1,0;		/* check if SP makes sense */	    \
 	bge-	cr1,exc_##n##_bad_stack;/* bad stack (TODO: out of line) */ \
 	mfspr	r10,SPRN_##type##_SRR0;	/* read SRR0 before touching stack */
 
@@ -328,6 +329,29 @@ ret_from_mc_except:
 #define SPRN_MC_SRR0	SPRN_MCSRR0
 #define SPRN_MC_SRR1	SPRN_MCSRR1
 
+#ifdef CONFIG_PPC_FSL_BOOK3E
+#define GEN_BTB_FLUSH			\
+	START_BTB_FLUSH_SECTION		\
+		beq 1f;			\
+		BTB_FLUSH(r10)			\
+		1:		\
+	END_BTB_FLUSH_SECTION
+
+#define CRIT_BTB_FLUSH			\
+	START_BTB_FLUSH_SECTION		\
+		BTB_FLUSH(r10)		\
+	END_BTB_FLUSH_SECTION
+
+#define DBG_BTB_FLUSH CRIT_BTB_FLUSH
+#define MC_BTB_FLUSH CRIT_BTB_FLUSH
+#define GDBELL_BTB_FLUSH GEN_BTB_FLUSH
+#else
+#define GEN_BTB_FLUSH
+#define CRIT_BTB_FLUSH
+#define DBG_BTB_FLUSH
+#define GDBELL_BTB_FLUSH
+#endif
+
 #define NORMAL_EXCEPTION_PROLOG(n, intnum, addition)			    \
 	EXCEPTION_PROLOG(n, intnum, GEN, addition##_GEN(n))
 
diff --git a/arch/powerpc/mm/tlb_low_64e.S b/arch/powerpc/mm/tlb_low_64e.S
index 7fd20c5..9ed9006 100644
--- a/arch/powerpc/mm/tlb_low_64e.S
+++ b/arch/powerpc/mm/tlb_low_64e.S
@@ -70,6 +70,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_EMB_HV)
 	std	r15,EX_TLB_R15(r12)
 	std	r10,EX_TLB_CR(r12)
 #ifdef CONFIG_PPC_FSL_BOOK3E
+START_BTB_FLUSH_SECTION
+	mfspr r11, SPRN_SRR1
+	andi. r10,r11,MSR_PR
+	beq 1f
+	BTB_FLUSH(r10)
+1:
+END_BTB_FLUSH_SECTION
 	std	r7,EX_TLB_R7(r12)
 #endif
 	TLB_MISS_PROLOG_STATS
-- 
2.5.5

[PATCH 08/11] powerpc/fsl: Flush branch predictor when entering KVM

From: Diana Craciun <hidden>
Date: 2018-12-12 14:09:57

Switching from the guest to host is another place
where the speculative accesses can be exploited.
Flush the branch predictor when entering KVM.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/kvm/bookehv_interrupts.S | 4 ++++
 1 file changed, 4 insertions(+)
diff --git a/arch/powerpc/kvm/bookehv_interrupts.S b/arch/powerpc/kvm/bookehv_interrupts.S
index 051af7d..4e5081e 100644
--- a/arch/powerpc/kvm/bookehv_interrupts.S
+++ b/arch/powerpc/kvm/bookehv_interrupts.S
@@ -75,6 +75,10 @@
 	PPC_LL	r1, VCPU_HOST_STACK(r4)
 	PPC_LL	r2, HOST_R2(r1)
 
+START_BTB_FLUSH_SECTION
+	BTB_FLUSH(r10)
+END_BTB_FLUSH_SECTION
+
 	mfspr	r10, SPRN_PID
 	lwz	r8, VCPU_HOST_PID(r4)
 	PPC_LL	r11, VCPU_SHARED(r4)
-- 
2.5.5

[PATCH 03/11] powerpc/fsl: Fix spectre_v2 mitigations reporting

From: Diana Craciun <hidden>
Date: 2018-12-12 14:11:58

Currently for CONFIG_PPC_FSL_BOOK3E
cat /sys/devices/system/cpu/vulnerabilities/spectre_v2 reports:
"Mitigation: Software count cache flush" which is wrong. Fix it
to report vulnerable for now.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/kernel/security.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c
index f6f469f..1b395b8 100644
--- a/arch/powerpc/kernel/security.c
+++ b/arch/powerpc/kernel/security.c
@@ -22,7 +22,7 @@ enum count_cache_flush_type {
 	COUNT_CACHE_FLUSH_SW	= 0x2,
 	COUNT_CACHE_FLUSH_HW	= 0x4,
 };
-static enum count_cache_flush_type count_cache_flush_type;
+static enum count_cache_flush_type count_cache_flush_type = COUNT_CACHE_FLUSH_NONE;
 
 bool barrier_nospec_enabled;
 static bool no_nospec;
-- 
2.5.5

[PATCH 01/11] powerpc/fsl: Add infrastructure to fixup branch predictor flush

From: Diana Craciun <hidden>
Date: 2018-12-12 14:14:00

In order to protect against speculation attacks (Spectre
variant 2) on NXP PowerPC platforms, the branch predictor
should be flushed when the privillege level is changed.
This patch is adding the infrastructure to fixup at runtime
the code sections that are performing the branch predictor flush
depending on a boot arg parameter which is added later in a
separate patch.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/include/asm/feature-fixups.h | 12 ++++++++++++
 arch/powerpc/include/asm/setup.h          |  2 ++
 arch/powerpc/kernel/vmlinux.lds.S         |  8 ++++++++
 arch/powerpc/lib/feature-fixups.c         | 21 +++++++++++++++++++++
 4 files changed, 43 insertions(+)
diff --git a/arch/powerpc/include/asm/feature-fixups.h b/arch/powerpc/include/asm/feature-fixups.h
index 33b6f9c..40a6c926 100644
--- a/arch/powerpc/include/asm/feature-fixups.h
+++ b/arch/powerpc/include/asm/feature-fixups.h
@@ -221,6 +221,17 @@ label##3:					       	\
 	FTR_ENTRY_OFFSET 953b-954b;			\
 	.popsection;
 
+#define START_BTB_FLUSH_SECTION			\
+955:							\
+
+#define END_BTB_FLUSH_SECTION			\
+956:							\
+	.pushsection __btb_flush_fixup,"a";	\
+	.align 2;							\
+957:						\
+	FTR_ENTRY_OFFSET 955b-957b;			\
+	FTR_ENTRY_OFFSET 956b-957b;			\
+	.popsection;
 
 #ifndef __ASSEMBLY__
 #include <linux/types.h>
@@ -230,6 +241,7 @@ extern long __start___stf_entry_barrier_fixup, __stop___stf_entry_barrier_fixup;
 extern long __start___stf_exit_barrier_fixup, __stop___stf_exit_barrier_fixup;
 extern long __start___rfi_flush_fixup, __stop___rfi_flush_fixup;
 extern long __start___barrier_nospec_fixup, __stop___barrier_nospec_fixup;
+extern long __start__btb_flush_fixup, __stop__btb_flush_fixup;
 
 void apply_feature_fixups(void);
 void setup_feature_keys(void);
diff --git a/arch/powerpc/include/asm/setup.h b/arch/powerpc/include/asm/setup.h
index 1fffbba..c941c8c 100644
--- a/arch/powerpc/include/asm/setup.h
+++ b/arch/powerpc/include/asm/setup.h
@@ -67,6 +67,8 @@ void do_barrier_nospec_fixups_range(bool enable, void *start, void *end);
 static inline void do_barrier_nospec_fixups_range(bool enable, void *start, void *end) { };
 #endif
 
+void do_btb_flush_fixups(void);
+
 #endif /* !__ASSEMBLY__ */
 
 #endif	/* _ASM_POWERPC_SETUP_H */
diff --git a/arch/powerpc/kernel/vmlinux.lds.S b/arch/powerpc/kernel/vmlinux.lds.S
index 434581b..254b757 100644
--- a/arch/powerpc/kernel/vmlinux.lds.S
+++ b/arch/powerpc/kernel/vmlinux.lds.S
@@ -170,6 +170,14 @@ SECTIONS
 	}
 #endif /* CONFIG_PPC_BARRIER_NOSPEC */
 
+#ifdef CONFIG_PPC_FSL_BOOK3E
+	. = ALIGN(8);
+	__spec_btb_flush_fixup : AT(ADDR(__spec_btb_flush_fixup) - LOAD_OFFSET) {
+		__start__btb_flush_fixup = .;
+		*(__btb_flush_fixup)
+		__stop__btb_flush_fixup = .;
+	}
+#endif
 	EXCEPTION_TABLE(0)
 
 	NOTES :kernel :notes
diff --git a/arch/powerpc/lib/feature-fixups.c b/arch/powerpc/lib/feature-fixups.c
index e613b02..02a213c 100644
--- a/arch/powerpc/lib/feature-fixups.c
+++ b/arch/powerpc/lib/feature-fixups.c
@@ -347,6 +347,27 @@ void do_barrier_nospec_fixups_range(bool enable, void *fixup_start, void *fixup_
 
 	printk(KERN_DEBUG "barrier-nospec: patched %d locations\n", i);
 }
+static void patch_btb_flush_section(long *curr)
+{
+	unsigned int *start, *end;
+
+	start = (void *)curr + *curr;
+	end = (void *)curr + *(curr + 1);
+	for (; start < end; start++) {
+		pr_devel("patching dest %lx\n", (unsigned long)start);
+		patch_instruction(start, PPC_INST_NOP);
+	}
+}
+void do_btb_flush_fixups(void)
+{
+	long *start, *end;
+
+	start = PTRRELOC(&__start__btb_flush_fixup);
+	end = PTRRELOC(&__stop__btb_flush_fixup);
+
+	for (; start < end; start += 2)
+		patch_btb_flush_section(start);
+}
 #endif /* CONFIG_PPC_FSL_BOOK3E */
 
 void do_lwsync_fixups(unsigned long value, void *fixup_start, void *fixup_end)
-- 
2.5.5

Re: [01/11] powerpc/fsl: Add infrastructure to fixup branch predictor flush

From: Michael Ellerman <hidden>
Date: 2018-12-22 10:25:38

On Wed, 2018-12-12 at 14:03:00 UTC, Diana Craciun wrote:
In order to protect against speculation attacks (Spectre
variant 2) on NXP PowerPC platforms, the branch predictor
should be flushed when the privillege level is changed.
This patch is adding the infrastructure to fixup at runtime
the code sections that are performing the branch predictor flush
depending on a boot arg parameter which is added later in a
separate patch.

Signed-off-by: Diana Craciun <redacted>
Series applied to powerpc next, thanks.

https://git.kernel.org/powerpc/c/76a5eaa38b15dda92cd6964248c39b

cheers

[PATCH 11/11] powerpc/fsl: Add FSL_PPC_BOOK3E as supported arch for nospectre_v2 boot arg

From: Diana Craciun <hidden>
Date: 2018-12-12 14:16:27

Signed-off-by: Diana Craciun <redacted>
---
 Documentation/admin-guide/kernel-parameters.txt | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
index aefd358..cf6b4c5 100644
--- a/Documentation/admin-guide/kernel-parameters.txt
+++ b/Documentation/admin-guide/kernel-parameters.txt
@@ -2827,7 +2827,7 @@
 			check bypass). With this option data leaks are possible
 			in the system.
 
-	nospectre_v2	[X86] Disable all mitigations for the Spectre variant 2
+	nospectre_v2	[X86,PPC_FSL_BOOK3E] Disable all mitigations for the Spectre variant 2
 			(indirect branch prediction) vulnerability. System may
 			allow data leaks with this option, which is equivalent
 			to spectre_v2=off.
-- 
2.5.5

[PATCH 09/11] powerpc/fsl: Enable runtime patching if nospectre_v2 boot arg is used

From: Diana Craciun <hidden>
Date: 2018-12-12 14:18:27

If the user choses not to use the mitigations, replace
the code sequence with nops.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/kernel/setup-common.c | 1 +
 1 file changed, 1 insertion(+)
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index 93ee370..f27eeda 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -974,6 +974,7 @@ void __init setup_arch(char **cmdline_p)
 		ppc_md.setup_arch();
 
 	setup_barrier_nospec();
+	setup_spectre_v2();
 
 	paging_init();
 
-- 
2.5.5

[PATCH 10/11] powerpc/fsl: Update Spectre v2 reporting

From: Diana Craciun <hidden>
Date: 2018-12-12 14:20:23

Report branch predictor state flush as a mitigation for
Spectre variant 2.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/kernel/security.c | 5 ++++-
 1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/arch/powerpc/kernel/security.c b/arch/powerpc/kernel/security.c
index 4393a38..861fab3 100644
--- a/arch/powerpc/kernel/security.c
+++ b/arch/powerpc/kernel/security.c
@@ -212,8 +212,11 @@ ssize_t cpu_show_spectre_v2(struct device *dev, struct device_attribute *attr, c
 
 		if (count_cache_flush_type == COUNT_CACHE_FLUSH_HW)
 			seq_buf_printf(&s, "(hardware accelerated)");
-	} else
+	} else if (btb_flush_enabled) {
+		seq_buf_printf(&s, "Mitigation: Branch predictor state flush");
+	} else {
 		seq_buf_printf(&s, "Vulnerable");
+	}
 
 	seq_buf_printf(&s, "\n");
 
-- 
2.5.5

[PATCH 02/11] powerpc/fsl: Add macro to flush the branch predictor

From: Diana Craciun <hidden>
Date: 2018-12-12 14:23:02

The BUCSR register can be used to invalidate the entries in the
branch prediction mechanisms.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/include/asm/ppc_asm.h | 10 ++++++++++
 1 file changed, 10 insertions(+)
diff --git a/arch/powerpc/include/asm/ppc_asm.h b/arch/powerpc/include/asm/ppc_asm.h
index b5d0236..5c901bf 100644
--- a/arch/powerpc/include/asm/ppc_asm.h
+++ b/arch/powerpc/include/asm/ppc_asm.h
@@ -821,4 +821,14 @@ END_FTR_SECTION_IFCLR(CPU_FTR_601)
 	stringify_in_c(.long (_target) - . ;)	\
 	stringify_in_c(.previous)
 
+#ifdef CONFIG_PPC_FSL_BOOK3E
+#define BTB_FLUSH(reg)			\
+	lis reg,BUCSR_INIT@h;		\
+	ori reg,reg,BUCSR_INIT@l;	\
+	mtspr SPRN_BUCSR,reg;		\
+	isync;
+#else
+#define BTB_FLUSH(reg)
+#endif /* CONFIG_PPC_FSL_BOOK3E */
+
 #endif /* _ASM_POWERPC_PPC_ASM_H */
-- 
2.5.5

[PATCH 04/11] powerpc/fsl: Emulate SPRN_BUCSR register

From: Diana Craciun <hidden>
Date: 2018-12-12 14:27:48

In order to flush the branch predictor the guest kernel
performs writes to the BUCSR register which is hypervisor
privilleged. However, the branch predictor is flushed at
each KVM entry, so the branch predictor has been already
flushed, so just return as soon as possible to guest.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/kvm/e500_emulate.c | 5 +++++
 1 file changed, 5 insertions(+)
diff --git a/arch/powerpc/kvm/e500_emulate.c b/arch/powerpc/kvm/e500_emulate.c
index 3f8189e..d0eb670 100644
--- a/arch/powerpc/kvm/e500_emulate.c
+++ b/arch/powerpc/kvm/e500_emulate.c
@@ -276,6 +276,11 @@ int kvmppc_core_emulate_mtspr_e500(struct kvm_vcpu *vcpu, int sprn, ulong spr_va
 		 */
 		vcpu->arch.pwrmgtcr0 = spr_val;
 		break;
+		/* if we are here, it means that we have already flushed the
+		 * branch predictor, so just return to guest
+		 */
+	case SPRN_BUCSR:
+		break;
 
 	/* extra exceptions */
 #ifdef CONFIG_SPE_POSSIBLE
-- 
2.5.5

[PATCH 07/11] powerpc/fsl: Flush the branch predictor at each kernel entry (32 bit)

From: Diana Craciun <hidden>
Date: 2018-12-12 14:29:43

In order to protect against speculation attacks on
indirect branches, the branch predictor is flushed at
kernel entry to protect for the following situations:
- userspace process attacking another userspace process
- userspace process attacking the kernel
Basically when the privillege level change (i.e.the kernel
is entered), the branch predictor state is flushed.

Signed-off-by: Diana Craciun <redacted>
---
 arch/powerpc/kernel/head_booke.h     |  6 ++++++
 arch/powerpc/kernel/head_fsl_booke.S | 15 +++++++++++++++
 2 files changed, 21 insertions(+)
diff --git a/arch/powerpc/kernel/head_booke.h b/arch/powerpc/kernel/head_booke.h
index d0862a1..15ac510 100644
--- a/arch/powerpc/kernel/head_booke.h
+++ b/arch/powerpc/kernel/head_booke.h
@@ -43,6 +43,9 @@
 	andi.	r11, r11, MSR_PR;	/* check whether user or kernel    */\
 	mr	r11, r1;						     \
 	beq	1f;							     \
+START_BTB_FLUSH_SECTION					\
+	BTB_FLUSH(r11)						\
+END_BTB_FLUSH_SECTION					\
 	/* if from user, start at top of this thread's kernel stack */       \
 	lwz	r11, THREAD_INFO-THREAD(r10);				     \
 	ALLOC_STACK_FRAME(r11, THREAD_SIZE);				     \
@@ -128,6 +131,9 @@
 	stw	r9,_CCR(r8);		/* save CR on stack		   */\
 	mfspr	r11,exc_level_srr1;	/* check whether user or kernel    */\
 	DO_KVM	BOOKE_INTERRUPT_##intno exc_level_srr1;		             \
+START_BTB_FLUSH_SECTION								\
+	BTB_FLUSH(r10)									\
+END_BTB_FLUSH_SECTION								\
 	andi.	r11,r11,MSR_PR;						     \
 	mfspr	r11,SPRN_SPRG_THREAD;	/* if from user, start at top of   */\
 	lwz	r11,THREAD_INFO-THREAD(r11); /* this thread's kernel stack */\
diff --git a/arch/powerpc/kernel/head_fsl_booke.S b/arch/powerpc/kernel/head_fsl_booke.S
index e2750b8..2386ce2 100644
--- a/arch/powerpc/kernel/head_fsl_booke.S
+++ b/arch/powerpc/kernel/head_fsl_booke.S
@@ -453,6 +453,13 @@ END_FTR_SECTION_IFSET(CPU_FTR_EMB_HV)
 	mfcr	r13
 	stw	r13, THREAD_NORMSAVE(3)(r10)
 	DO_KVM	BOOKE_INTERRUPT_DTLB_MISS SPRN_SRR1
+START_BTB_FLUSH_SECTION
+	mfspr r11, SPRN_SRR1
+	andi. r10,r11,MSR_PR
+	beq 1f
+	BTB_FLUSH(r10)
+1:
+END_BTB_FLUSH_SECTION
 	mfspr	r10, SPRN_DEAR		/* Get faulting address */
 
 	/* If we are faulting a kernel address, we have to use the
@@ -547,6 +554,14 @@ END_FTR_SECTION_IFSET(CPU_FTR_EMB_HV)
 	mfcr	r13
 	stw	r13, THREAD_NORMSAVE(3)(r10)
 	DO_KVM	BOOKE_INTERRUPT_ITLB_MISS SPRN_SRR1
+START_BTB_FLUSH_SECTION
+	mfspr r11, SPRN_SRR1
+	andi. r10,r11,MSR_PR
+	beq 1f
+	BTB_FLUSH(r10)
+1:
+END_BTB_FLUSH_SECTION
+
 	mfspr	r10, SPRN_SRR0		/* Get faulting address */
 
 	/* If we are faulting a kernel address, we have to use the
-- 
2.5.5

Re: [PATCH 07/11] powerpc/fsl: Flush the branch predictor at each kernel entry (32 bit)

From: kbuild test robot <hidden>
Date: 2018-12-19 09:48:33

Hi Diana,

Thank you for the patch! Perhaps something to improve:

[auto build test WARNING on powerpc/next]
[also build test WARNING on v4.20-rc7 next-20181218]
[if your patch is applied to the wrong git tree, please drop us a note to help improve the system]

url:    https://github.com/0day-ci/linux/commits/Diana-Craciun/powerpc-fsl-NXP-PowerPC-Spectre-variant-2-workarounds/20181213-015503
base:   https://git.kernel.org/pub/scm/linux/kernel/git/powerpc/linux.git next
config: powerpc-icon_defconfig (attached as .config)
compiler: powerpc-linux-gnu-gcc (Debian 7.2.0-11) 7.2.0
reproduce:
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # save the attached .config to linux build tree
        GCC_VERSION=7.2.0 make.cross ARCH=powerpc 

All warnings (new ones prefixed by >>):
quoted
powerpc-linux-gnu-ld: warning: orphan section `__btb_flush_fixup' from `arch/powerpc/kernel/head_44x.o' being placed in section `__btb_flush_fixup'.
powerpc-linux-gnu-ld: warning: orphan section `__btb_flush_fixup' from `arch/powerpc/kernel/head_44x.o' being placed in section `__btb_flush_fixup'.
powerpc-linux-gnu-ld: warning: orphan section `__btb_flush_fixup' from `arch/powerpc/kernel/head_44x.o' being placed in section `__btb_flush_fixup'.
---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help