Skip to contentSkip to conversation

BPF fix with regards to unlocking

PATCH In review Daniel Borkmann 3 participants 5 messages Feb 21, 2017

Subsystems (11)
lore
Share & tools
lore

Daniel Borkmann

author maintainer
#0 lore raw

cover BPF fix with regards to unlocking

This set fixes the issue Eric was reporting recently [1]. First patch is a prerequisite discussed with Laura that is needed for the later fix in the second one. I've tested this extensively and it does not reproduce anymore on my side after the fix. Thanks & sorry about that!

[1] https://www.spinics.net/lists/netdev/msg421877.html

Daniel Borkmann (2):
arch: add ARCH_HAS_SET_MEMORY config
bpf: fix unlocking of jited image when module ronx not set

 arch/Kconfig                  |  4 ++++
 arch/arm/Kconfig              |  1 +
 arch/arm64/Kconfig            |  1 +
 arch/arm64/net/bpf_jit_comp.c |  2 +-
 arch/s390/Kconfig             |  1 +
 arch/s390/net/bpf_jit_comp.c  |  2 +-
 arch/x86/Kconfig              |  1 +
 arch/x86/net/bpf_jit_comp.c   |  2 +-
 include/linux/filter.h        | 13 +++++++++++--
 9 files changed, 22 insertions(+), 5 deletions(-)
Signature · 1 line
1.9.3
author maintainer
#2 lore raw

PATCH 2/2 bpf: fix unlocking of jited image when module ronx not set

Eric and Willem reported that they recently saw random crashes when JIT was in use and bisected this to 74451e66d516 ("bpf: make jited programs visible in traces"). Issue was that the consolidation part added bpf_jit_binary_unlock_ro() that would unlock previously made read-only memory back to read-write. However, DEBUG_SET_MODULE_RONX cannot be used for this to test for presence of set_memory_*() functions. We need to use ARCH_HAS_SET_MEMORY instead to fix this; also add the corresponding bpf_jit_binary_lock_ro() to filter.h.

Fixes 74451e66d516 ("bpf: make jited programs visible in traces")
---
 arch/arm64/net/bpf_jit_comp.c |  2 +-
 arch/s390/net/bpf_jit_comp.c  |  2 +-
 arch/x86/net/bpf_jit_comp.c   |  2 +-
 include/linux/filter.h        | 13 +++++++++++--
 4 files changed, 14 insertions(+), 5 deletions(-)
arch/arm64/net/bpf_jit_comp.c +1 -1
index 05d1210..a785554 100644
@@ -898,7 +898,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 
 	bpf_flush_icache(header, ctx.image + ctx.idx);
 
-	set_memory_ro((unsigned long)header, header->pages);
+	bpf_jit_binary_lock_ro(header);
 	prog->bpf_func = (void *)ctx.image;
 	prog->jited = 1;
 
arch/s390/net/bpf_jit_comp.c +1 -1
index f1d0e62..b49c52a 100644
@@ -1327,7 +1327,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
 			print_fn_code(jit.prg_buf, jit.size_prg);
 	}
 	if (jit.prg_buf) {
-		set_memory_ro((unsigned long)header, header->pages);
+		bpf_jit_binary_lock_ro(header);
 		fp->bpf_func = (void *) jit.prg_buf;
 		fp->jited = 1;
 	}
arch/x86/net/bpf_jit_comp.c +1 -1
index 18a62e2..32322ce 100644
@@ -1165,7 +1165,7 @@ struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *prog)
 
 	if (image) {
 		bpf_flush_icache(header, image + proglen);
-		set_memory_ro((unsigned long)header, header->pages);
+		bpf_jit_binary_lock_ro(header);
 		prog->bpf_func = (void *)image;
 		prog->jited = 1;
 	} else {
include/linux/filter.h +11 -2
index 0c1cc91..0c167fd 100644
@@ -551,7 +551,7 @@ static inline bool bpf_prog_was_classic(const struct bpf_prog *prog)
 
 #define bpf_classic_proglen(fprog) (fprog->len * sizeof(fprog->filter[0]))
 
-#ifdef CONFIG_DEBUG_SET_MODULE_RONX
+#ifdef CONFIG_ARCH_HAS_SET_MEMORY
 static inline void bpf_prog_lock_ro(struct bpf_prog *fp)
 {
 	set_memory_ro((unsigned long)fp, fp->pages);
@@ -562,6 +562,11 @@ static inline void bpf_prog_unlock_ro(struct bpf_prog *fp)
 	set_memory_rw((unsigned long)fp, fp->pages);
 }
 
+static inline void bpf_jit_binary_lock_ro(struct bpf_binary_header *hdr)
+{
+	set_memory_ro((unsigned long)hdr, hdr->pages);
+}
+
 static inline void bpf_jit_binary_unlock_ro(struct bpf_binary_header *hdr)
 {
 	set_memory_rw((unsigned long)hdr, hdr->pages);
@@ -575,10 +580,14 @@ static inline void bpf_prog_unlock_ro(struct bpf_prog *fp)
 {
 }
 
+static inline void bpf_jit_binary_lock_ro(struct bpf_binary_header *hdr)
+{
+}
+
 static inline void bpf_jit_binary_unlock_ro(struct bpf_binary_header *hdr)
 {
 }
-#endif /* CONFIG_DEBUG_SET_MODULE_RONX */
+#endif /* CONFIG_ARCH_HAS_SET_MEMORY */
 
 static inline struct bpf_binary_header *
 bpf_jit_binary_hdr(const struct bpf_prog *fp)
Signature · 1 line
1.9.3
author maintainer
#1 lore raw

PATCH 1/2 arch: add ARCH_HAS_SET_MEMORY config

In reply to Daniel Borkmann

Currently, there's no good way to test for the presence of set_memory_ro/rw/x/nx() helpers implemented by archs such as x86, arm, arm64 and s390.

There's DEBUG_SET_MODULE_RONX and DEBUG_RODATA, however both don't really reflect that: set_memory_*() are also available even when DEBUG_SET_MODULE_RONX is turned off, and DEBUG_RODATA is set by parisc, but doesn't implement above functions. Thus, add ARCH_HAS_SET_MEMORY that is selected by mentioned archs, where generic code can test against this.

This also allows later on to move DEBUG_SET_MODULE_RONX out of the arch specific Kconfig to define it only once depending on ARCH_HAS_SET_MEMORY.

---
 arch/Kconfig       | 4 ++++
 arch/arm/Kconfig   | 1 +
 arch/arm64/Kconfig | 1 +
 arch/s390/Kconfig  | 1 +
 arch/x86/Kconfig   | 1 +
 5 files changed, 8 insertions(+)
arch/Kconfig +4 -0
index bd04eac..e8ada79 100644
@@ -222,6 +222,10 @@ config GENERIC_SMP_IDLE_THREAD
 config GENERIC_IDLE_POLL_SETUP
        bool
 
+# Select if arch has all set_memory_ro/rw/x/nx() functions in asm/cacheflush.h
+config ARCH_HAS_SET_MEMORY
+	bool
+
 # Select if arch init_task initializer is different to init/init_task.c
 config ARCH_INIT_TASK
        bool
arch/arm/Kconfig +1 -0
index 186c4c2..edae056 100644
@@ -4,6 +4,7 @@ config ARM
 	select ARCH_CLOCKSOURCE_DATA
 	select ARCH_HAS_DEVMEM_IS_ALLOWED
 	select ARCH_HAS_ELF_RANDOMIZE
+	select ARCH_HAS_SET_MEMORY
 	select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST
 	select ARCH_HAVE_CUSTOM_GPIO_H
 	select ARCH_HAS_GCOV_PROFILE_ALL
arch/arm64/Kconfig +1 -0
index 1117421..1853405 100644
@@ -12,6 +12,7 @@ config ARM64
 	select ARCH_HAS_GCOV_PROFILE_ALL
 	select ARCH_HAS_GIGANTIC_PAGE
 	select ARCH_HAS_KCOV
+	select ARCH_HAS_SET_MEMORY
 	select ARCH_HAS_SG_CHAIN
 	select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST
 	select ARCH_USE_CMPXCHG_LOCKREF
arch/s390/Kconfig +1 -0
index c6722112..094deb1 100644
@@ -72,6 +72,7 @@ config S390
 	select ARCH_HAS_GCOV_PROFILE_ALL
 	select ARCH_HAS_GIGANTIC_PAGE
 	select ARCH_HAS_KCOV
+	select ARCH_HAS_SET_MEMORY
 	select ARCH_HAS_SG_CHAIN
 	select ARCH_HAS_UBSAN_SANITIZE_ALL
 	select ARCH_HAVE_NMI_SAFE_CMPXCHG
arch/x86/Kconfig +1 -0
index e487493..434dd2a 100644
@@ -53,6 +53,7 @@ config X86
 	select ARCH_HAS_KCOV			if X86_64
 	select ARCH_HAS_MMIO_FLUSH
 	select ARCH_HAS_PMEM_API		if X86_64
+	select ARCH_HAS_SET_MEMORY
 	select ARCH_HAS_SG_CHAIN
 	select ARCH_HAS_UBSAN_SANITIZE_ALL
 	select ARCH_HAVE_NMI_SAFE_CMPXCHG
Signature · 1 line
1.9.3
Tested-by on v1 2/2
#3 lore raw

Re: [PATCH net-next 2/2] bpf: fix unlocking of jited image when module ronx not set

In reply to Daniel Borkmann
Quoted 9 lines · Daniel Borkmann

Eric and Willem reported that they recently saw random crashes when JIT was in use and bisected this to 74451e66d516 ("bpf: make jited programs visible in traces"). Issue was that the consolidation part added bpf_jit_binary_unlock_ro() that would unlock previously made read-only memory back to read-write. However, DEBUG_SET_MODULE_RONX cannot be used for this to test for presence of set_memory_*() functions. We need to use ARCH_HAS_SET_MEMORY instead to fix this; also add the corresponding bpf_jit_binary_lock_ro() to filter.h.

Quoted 5 lines
Fixes 74451e66d516 ("bpf: make jited programs visible in traces")

This fixes the issue I observed. Thanks, Daniel.

maintainer
#4 lore raw
In reply to Daniel Borkmann

From: Daniel Borkmann <[email protected]>
Date: Tue, 21 Feb 2017 16:09:32 +0100

Quoted 7 lines

This set fixes the issue Eric was reporting recently [1]. First patch is a prerequisite discussed with Laura that is needed for the later fix in the second one. I've tested this extensively and it does not reproduce anymore on my side after the fix. Thanks & sorry about that!

[1] https://www.spinics.net/lists/netdev/msg421877.html

Series applied, thanks Daniel.