LZ4 final literal copy could be overlapped when doing
in-place decompression, so it's unsafe to just use memcpy()
on an optimized memcpy approach but memmove() instead.
Upstream LZ4 has updated this years ago [1] (and the impact
is non-sensible [2]), this commit just synchronizes LZ4
upstream code to the kernel side as well.
It can be observed as EROFS in-place decompression failure
on specific files when X86_FEATURE_ERMS is unsupported,
memcpy() optimization of commit 59daa706fbec ("x86, mem:
Optimize memcpy by avoiding memory false dependece") will
be enabled then.
Currently most modern x86-CPUs support ERMS, these CPUs
just use "rep movsb" approach so no problem at all. However,
it can still be verified with forcely disabling ERMS feature...
arch/x86/lib/memcpy_64.S:
ALTERNATIVE_2 "jmp memcpy_orig", "", X86_FEATURE_REP_GOOD, \
- "jmp memcpy_erms", X86_FEATURE_ERMS
+ "jmp memcpy_orig", X86_FEATURE_ERMS
We didn't observed any strange on arm64/arm platform before,
but it really needs an update for sure considering this
unsafe overlapping memcpy() behavior for LZ4 in-place
decompression.
[1] https://github.com/lz4/lz4/commit/33cb8518ac385835cc17be9a770b27b40cd0e15b
[2] https://github.com/lz4/lz4/pull/717#issuecomment-497818921
Cc: Yann Collet <[email protected]>
Cc: Miao Xie <[email protected]>
Cc: Chao Yu <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
---
Hi Andrew,
Could you kindly consider picking this patch up, although
the impact is EROFS but it touchs in-kernel lz4 library
anyway...
Thanks,
Gao Xiang
lib/lz4/lz4_decompress.c | 2 +-
lib/lz4/lz4defs.h | 1 +
2 files changed, 2 insertions(+), 1 deletion(-)
diff --git a/lib/lz4/lz4_decompress.c b/lib/lz4/lz4_decompress.c
index 00cb0d0b73e1..fa88e13ef00b 100644
--- a/lib/lz4/lz4_decompress.c
+++ b/lib/lz4/lz4_decompress.c
@@ -263,7 +263,7 @@ static FORCE_INLINE int LZ4_decompress_generic(
}
}
- LZ4_memcpy(op, ip, length);
+ LZ4_memmove(op, ip, length);
ip += length;
op += length;
diff --git a/lib/lz4/lz4defs.h b/lib/lz4/lz4defs.h
index c91dd96ef629..673bd206aa98 100644
--- a/lib/lz4/lz4defs.h
+++ b/lib/lz4/lz4defs.h
@@ -146,6 +146,7 @@ static FORCE_INLINE void LZ4_writeLE16(void *memPtr, U16 value)
* environments. This is needed when decompressing the Linux Kernel, for example.
*/
#define LZ4_memcpy(dst, src, size) __builtin_memcpy(dst, src, size)
+#define LZ4_memmove(dst, src, size) __builtin_memmove(dst, src, size)
static FORCE_INLINE void LZ4_copy8(void *dst, const void *src)
{
--
2.18.4
LZ4 final literal copy could be overlapped when doing
in-place decompression, so it's unsafe to just use memcpy()
on an optimized memcpy approach but memmove() instead.
Upstream LZ4 has updated this years ago [1] (and the impact
is non-sensible [2] plus only a few bytes remain), this commit
just synchronizes LZ4 upstream code to the kernel side as well.
It can be observed as EROFS in-place decompression failure
on specific files when X86_FEATURE_ERMS is unsupported,
memcpy() optimization of commit 59daa706fbec ("x86, mem:
Optimize memcpy by avoiding memory false dependece") will
be enabled then.
Currently most modern x86-CPUs support ERMS, these CPUs just
use "rep movsb" approach so no problem at all. However, it can
still be verified with forcely disabling ERMS feature...
arch/x86/lib/memcpy_64.S:
ALTERNATIVE_2 "jmp memcpy_orig", "", X86_FEATURE_REP_GOOD, \
- "jmp memcpy_erms", X86_FEATURE_ERMS
+ "jmp memcpy_orig", X86_FEATURE_ERMS
We didn't observe any strange on arm64/arm/x86 platform before
since most memcpy() would behave in an increasing address order
("copy upwards" [3]) and it's the correct order of in-place
decompression but it really needs an update to memmove() for sure
considering it's an undefined behavior according to the standard
and some unique optimization already exists in the kernel.
[1] https://github.com/lz4/lz4/commit/33cb8518ac385835cc17be9a770b27b40cd0e15b
[2] https://github.com/lz4/lz4/pull/717#issuecomment-497818921
[3] https://sourceware.org/bugzilla/show_bug.cgi?id=12518
Cc: Yann Collet <[email protected]>
Cc: Nick Terrell <[email protected]>
Cc: Miao Xie <[email protected]>
Cc: Chao Yu <[email protected]>
Cc: Li Guifu <[email protected]>
Cc: Guo Xuenan <[email protected]>
Signed-off-by: Gao Xiang <[email protected]>
---
changes since v1:
- refine commit message;
- Cc more people.
Hi Andrew,
Could you kindly consider picking this patch up, although
the impact is EROFS but it touchs in-kernel lz4 library
anyway...
Thanks,
Gao Xiang
lib/lz4/lz4_decompress.c | 6 +++++-
lib/lz4/lz4defs.h | 1 +
2 files changed, 6 insertions(+), 1 deletion(-)
diff --git a/lib/lz4/lz4_decompress.c b/lib/lz4/lz4_decompress.c
index 00cb0d0b73e1..8a7724a6ce2f 100644
--- a/lib/lz4/lz4_decompress.c
+++ b/lib/lz4/lz4_decompress.c
@@ -263,7 +263,11 @@ static FORCE_INLINE int LZ4_decompress_generic(
}
}
- LZ4_memcpy(op, ip, length);
+ /*
+ * supports overlapping memory regions; only matters
+ * for in-place decompression scenarios
+ */
+ LZ4_memmove(op, ip, length);
ip += length;
op += length;
diff --git a/lib/lz4/lz4defs.h b/lib/lz4/lz4defs.h
index c91dd96ef629..673bd206aa98 100644
--- a/lib/lz4/lz4defs.h
+++ b/lib/lz4/lz4defs.h
@@ -146,6 +146,7 @@ static FORCE_INLINE void LZ4_writeLE16(void *memPtr, U16 value)
* environments. This is needed when decompressing the Linux Kernel, for example.
*/
#define LZ4_memcpy(dst, src, size) __builtin_memcpy(dst, src, size)
+#define LZ4_memmove(dst, src, size) __builtin_memmove(dst, src, size)
static FORCE_INLINE void LZ4_copy8(void *dst, const void *src)
{
--
2.18.4
> On Nov 21, 2020, at 7:07 PM, Gao Xiang <[email protected]> wrote:
>
> LZ4 final literal copy could be overlapped when doing
> in-place decompression, so it's unsafe to just use memcpy()
> on an optimized memcpy approach but memmove() instead.
>
> Upstream LZ4 has updated this years ago [1] (and the impact
> is non-sensible [2] plus only a few bytes remain), this commit
> just synchronizes LZ4 upstream code to the kernel side as well.
>
> It can be observed as EROFS in-place decompression failure
> on specific files when X86_FEATURE_ERMS is unsupported,
> memcpy() optimization of commit 59daa706fbec ("x86, mem:
> Optimize memcpy by avoiding memory false dependece") will
> be enabled then.
>
> Currently most modern x86-CPUs support ERMS, these CPUs just
> use "rep movsb" approach so no problem at all. However, it can
> still be verified with forcely disabling ERMS feature...
>
> arch/x86/lib/memcpy_64.S:
> ALTERNATIVE_2 "jmp memcpy_orig", "", X86_FEATURE_REP_GOOD, \
> - "jmp memcpy_erms", X86_FEATURE_ERMS
> + "jmp memcpy_orig", X86_FEATURE_ERMS
>
> We didn't observe any strange on arm64/arm/x86 platform before
> since most memcpy() would behave in an increasing address order
> ("copy upwards" [3]) and it's the correct order of in-place
> decompression but it really needs an update to memmove() for sure
> considering it's an undefined behavior according to the standard
> and some unique optimization already exists in the kernel.
>
> [1] https://github.com/lz4/lz4/commit/33cb8518ac385835cc17be9a770b27b40cd0e15b
> [2] https://github.com/lz4/lz4/pull/717#issuecomment-497818921
> [3] https://sourceware.org/bugzilla/show_bug.cgi?id=12518
> Cc: Yann Collet <[email protected]>
> Cc: Nick Terrell <[email protected]>
> Cc: Miao Xie <[email protected]>
> Cc: Chao Yu <[email protected]>
> Cc: Li Guifu <[email protected]>
> Cc: Guo Xuenan <[email protected]>
> Signed-off-by: Gao Xiang <[email protected]>
> ---
> changes since v1:
> - refine commit message;
> - Cc more people.
>
> Hi Andrew,
>
> Could you kindly consider picking this patch up, although
> the impact is EROFS but it touchs in-kernel lz4 library
> anyway...
>
> Thanks,
> Gao Xiang
>
> lib/lz4/lz4_decompress.c | 6 +++++-
> lib/lz4/lz4defs.h | 1 +
> 2 files changed, 6 insertions(+), 1 deletion(-)
>
> diff --git a/lib/lz4/lz4_decompress.c b/lib/lz4/lz4_decompress.c
> index 00cb0d0b73e1..8a7724a6ce2f 100644
> --- a/lib/lz4/lz4_decompress.c
> +++ b/lib/lz4/lz4_decompress.c
> @@ -263,7 +263,11 @@ static FORCE_INLINE int LZ4_decompress_generic(
> }
> }
>
> - LZ4_memcpy(op, ip, length);
> + /*
> + * supports overlapping memory regions; only matters
> + * for in-place decompression scenarios
> + */
> + LZ4_memmove(op, ip, length);
> ip += length;
> op += length;
>
> diff --git a/lib/lz4/lz4defs.h b/lib/lz4/lz4defs.h
> index c91dd96ef629..673bd206aa98 100644
> --- a/lib/lz4/lz4defs.h
> +++ b/lib/lz4/lz4defs.h
> @@ -146,6 +146,7 @@ static FORCE_INLINE void LZ4_writeLE16(void *memPtr, U16 value)
> * environments. This is needed when decompressing the Linux Kernel, for example.
> */
> #define LZ4_memcpy(dst, src, size) __builtin_memcpy(dst, src, size)
> +#define LZ4_memmove(dst, src, size) __builtin_memmove(dst, src, size)
>
> static FORCE_INLINE void LZ4_copy8(void *dst, const void *src)
> {
> --
> 2.18.4
>
Looks good to me! You can add:
Reviewed-by: Nick Terrell <[email protected]>
On Mon, Nov 30, 2020 at 06:19:34PM +0000, Nick Terrell wrote:
>
<too long snip>
>
> Looks good to me! You can add:
>
> Reviewed-by: Nick Terrell <[email protected]>
Thanks for reviewing this!
Thanks,
Gao Xiang
>