| Message ID | 1362484056-9778-4-git-send-email-kyungsik.lee@lge.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
On Tue, 5 Mar 2013 20:47:34 +0900 Kyungsik Lee <kyungsik.lee@lge.com> wrote: > This patch integrates the LZ4 decompression code to the arm pre-boot code. > And it depends on two patchs below > > lib: Add support for LZ4-compressed kernel > decompressor: Add LZ4 decompressor module > > ... > > - Apply CFLAGS, -Os to decompress.o to improve decompress > performance during boot-up process > > ... > > --- a/arch/arm/boot/compressed/Makefile > +++ b/arch/arm/boot/compressed/Makefile > @@ -24,6 +24,9 @@ endif > AFLAGS_head.o += -DTEXT_OFFSET=$(TEXT_OFFSET) > HEAD = head.o > OBJS += misc.o decompress.o > +ifeq ($(CONFIG_KERNEL_LZ4),y) > +CFLAGS_decompress.o := -Os > +endif Surprised. You found that -Os produces faster code than -O2? Details, please? -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Tue, Apr 30, 2013 at 01:56:02PM -0700, Andrew Morton wrote: > On Tue, 5 Mar 2013 20:47:34 +0900 Kyungsik Lee <kyungsik.lee@lge.com> wrote: > > > This patch integrates the LZ4 decompression code to the arm pre-boot code. > > And it depends on two patchs below > > > > lib: Add support for LZ4-compressed kernel > > decompressor: Add LZ4 decompressor module > > > > ... > > > > - Apply CFLAGS, -Os to decompress.o to improve decompress > > performance during boot-up process > > > > ... > > > > --- a/arch/arm/boot/compressed/Makefile > > +++ b/arch/arm/boot/compressed/Makefile > > @@ -24,6 +24,9 @@ endif > > AFLAGS_head.o += -DTEXT_OFFSET=$(TEXT_OFFSET) > > HEAD = head.o > > OBJS += misc.o decompress.o > > +ifeq ($(CONFIG_KERNEL_LZ4),y) > > +CFLAGS_decompress.o := -Os > > +endif > > Surprised. You found that -Os produces faster code than -O2? Details, > please? -Os is enforced here, based on the test result of decompression time below, slightly faster than -O2. But further tests with UA show that using -O2 will be the right choice especially in the case of the unaligned access enabled. Decompression Time(counts) UA -Os 6717 3447 -O2 6720 2728 Note: ARM v7, Kernel 3.4 counter freq. = 32768 HZ UA(Unaligned Access) gcc version 4.6.2 (Ubuntu/Linaro 4.6.2-14ubuntu2~ppa1) Thanks, Kyungsik -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig index 67874b8..0f9b363 100644 --- a/arch/arm/Kconfig +++ b/arch/arm/Kconfig @@ -38,6 +38,7 @@ config ARM select HAVE_IDE if PCI || ISA || PCMCIA select HAVE_IRQ_WORK select HAVE_KERNEL_GZIP + select HAVE_KERNEL_LZ4 select HAVE_KERNEL_LZMA select HAVE_KERNEL_LZO select HAVE_KERNEL_XZ diff --git a/arch/arm/boot/compressed/.gitignore b/arch/arm/boot/compressed/.gitignore index f79a08e..47279aa 100644 --- a/arch/arm/boot/compressed/.gitignore +++ b/arch/arm/boot/compressed/.gitignore @@ -6,6 +6,7 @@ piggy.gzip piggy.lzo piggy.lzma piggy.xzkern +piggy.lz4 vmlinux vmlinux.lds diff --git a/arch/arm/boot/compressed/Makefile b/arch/arm/boot/compressed/Makefile index 5cad8a6..2249d52 100644 --- a/arch/arm/boot/compressed/Makefile +++ b/arch/arm/boot/compressed/Makefile @@ -24,6 +24,9 @@ endif AFLAGS_head.o += -DTEXT_OFFSET=$(TEXT_OFFSET) HEAD = head.o OBJS += misc.o decompress.o +ifeq ($(CONFIG_KERNEL_LZ4),y) +CFLAGS_decompress.o := -Os +endif FONTC = $(srctree)/drivers/video/console/font_acorn_8x8.c # string library code (-Os is enforced to keep it much smaller) @@ -88,6 +91,7 @@ suffix_$(CONFIG_KERNEL_GZIP) = gzip suffix_$(CONFIG_KERNEL_LZO) = lzo suffix_$(CONFIG_KERNEL_LZMA) = lzma suffix_$(CONFIG_KERNEL_XZ) = xzkern +suffix_$(CONFIG_KERNEL_LZ4) = lz4 # Borrowed libfdt files for the ATAG compatibility mode @@ -112,7 +116,7 @@ targets := vmlinux vmlinux.lds \ font.o font.c head.o misc.o $(OBJS) # Make sure files are removed during clean -extra-y += piggy.gzip piggy.lzo piggy.lzma piggy.xzkern \ +extra-y += piggy.gzip piggy.lzo piggy.lzma piggy.xzkern piggy.lz4 \ lib1funcs.S ashldi3.S $(libfdt) $(libfdt_hdrs) ifeq ($(CONFIG_FUNCTION_TRACER),y) diff --git a/arch/arm/boot/compressed/decompress.c b/arch/arm/boot/compressed/decompress.c index 9deb56a..a95f071 100644 --- a/arch/arm/boot/compressed/decompress.c +++ b/arch/arm/boot/compressed/decompress.c @@ -53,6 +53,10 @@ extern char * strstr(const char * s1, const char *s2); #include "../../../../lib/decompress_unxz.c" #endif +#ifdef CONFIG_KERNEL_LZ4 +#include "../../../../lib/decompress_unlz4.c" +#endif + int do_decompress(u8 *input, int len, u8 *output, void (*error)(char *x)) { return decompress(input, len, NULL, NULL, output, NULL, error); diff --git a/arch/arm/boot/compressed/piggy.lz4.S b/arch/arm/boot/compressed/piggy.lz4.S new file mode 100644 index 0000000..3d9a575 --- /dev/null +++ b/arch/arm/boot/compressed/piggy.lz4.S @@ -0,0 +1,6 @@ + .section .piggydata,#alloc + .globl input_data +input_data: + .incbin "arch/arm/boot/compressed/piggy.lz4" + .globl input_data_end +input_data_end:
This patch integrates the LZ4 decompression code to the arm pre-boot code. And it depends on two patchs below lib: Add support for LZ4-compressed kernel decompressor: Add LZ4 decompressor module Signed-off-by: Kyungsik Lee <kyungsik.lee@lge.com> v2: - Apply CFLAGS, -Os to decompress.o to improve decompress performance during boot-up process --- arch/arm/Kconfig | 1 + arch/arm/boot/compressed/.gitignore | 1 + arch/arm/boot/compressed/Makefile | 6 +++++- arch/arm/boot/compressed/decompress.c | 4 ++++ arch/arm/boot/compressed/piggy.lz4.S | 6 ++++++ 5 files changed, 17 insertions(+), 1 deletion(-) create mode 100644 arch/arm/boot/compressed/piggy.lz4.S