From: Yinghai Lu <ying...@kernel.org> Current z_extract_offset is calculated in boot/compressed/mkpiggy.c. The problem is in mkpiggy.c we don't know the detail of decompressor. Then we just get a rough z_extract_offset according to extra_bytes. As we know extra_bytes can only promise a safety margin when decompressing. In fact this brings some risks:
- output+output_len could be much bigger than input+input_len. In this cass decompressed kernel plus relocs could overwrite decompressing method code even when they are running. - ehead of ZO could be bigger than z_extract_offset. In this case overwrite could happen when head code is running to move ZO to end of buffer. Though currently the size of head code is very small it's still a potential risk. Because no rule to limit the size of head code of ZO, it has possibility to be very big. Now move z_extract_offset calculation to header.S, and adjust z_extract_offset to make sure that above two cases never happen. Besides we have made ZO always be in the end of decompressing buffer, z_extract_offset is only used here to calculate an appropriate INIT_SIZE, no other place need it now. So no need to put it in voffset.h. Signed-off-by: Yinghai Lu <ying...@kernel.org> Signed-off-by: Baoquan He <b...@redhat.com> --- v2->v3: Tune the patch log. Remove code comment above init_size. I still think extra_bytes shoule be taken as below formula to cover the worst case of xz. #define ZO_z_extra_bytes ((ZO_z_output_len >> 12) + 65536 + 128) arch/x86/boot/Makefile | 2 +- arch/x86/boot/compressed/misc.c | 5 +---- arch/x86/boot/compressed/mkpiggy.c | 15 +-------------- arch/x86/boot/header.S | 22 +++++++++++++++++++++- 4 files changed, 24 insertions(+), 20 deletions(-) diff --git a/arch/x86/boot/Makefile b/arch/x86/boot/Makefile index bbe1a62..bd021e5 100644 --- a/arch/x86/boot/Makefile +++ b/arch/x86/boot/Makefile @@ -87,7 +87,7 @@ targets += voffset.h $(obj)/voffset.h: vmlinux FORCE $(call if_changed,voffset) -sed-zoffset := -e 's/^\([0-9a-fA-F]*\) [ABCDGRSTVW] \(startup_32\|startup_64\|efi32_stub_entry\|efi64_stub_entry\|efi_pe_entry\|input_data\|_end\|z_.*\)$$/\#define ZO_\2 0x\1/p' +sed-zoffset := -e 's/^\([0-9a-fA-F]*\) [ABCDGRSTVW] \(startup_32\|startup_64\|efi32_stub_entry\|efi64_stub_entry\|efi_pe_entry\|input_data\|_end\|_ehead\|_text\|z_.*\)$$/\#define ZO_\2 0x\1/p' quiet_cmd_zoffset = ZOFFSET $@ cmd_zoffset = $(NM) $< | sed -n $(sed-zoffset) > $@ diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c index c4477d5..e2a998f 100644 --- a/arch/x86/boot/compressed/misc.c +++ b/arch/x86/boot/compressed/misc.c @@ -83,13 +83,10 @@ * To avoid problems with the compressed data's meta information an extra 18 * bytes are needed. Leading to the formula: * - * extra_bytes = (uncompressed_size >> 12) + 32768 + 18 + decompressor_size. + * extra_bytes = (uncompressed_size >> 12) + 32768 + 18. * * Adding 8 bytes per 32K is a bit excessive but much easier to calculate. * Adding 32768 instead of 32767 just makes for round numbers. - * Adding the decompressor_size is necessary as it musht live after all - * of the data as well. Last I measured the decompressor is about 14K. - * 10K of actual data and 4K of bss. * */ diff --git a/arch/x86/boot/compressed/mkpiggy.c b/arch/x86/boot/compressed/mkpiggy.c index b980046..a613c84 100644 --- a/arch/x86/boot/compressed/mkpiggy.c +++ b/arch/x86/boot/compressed/mkpiggy.c @@ -21,8 +21,7 @@ * ----------------------------------------------------------------------- */ /* - * Compute the desired load offset from a compressed program; outputs - * a small assembly wrapper with the appropriate symbols defined. + * outputs a small assembly wrapper with the appropriate symbols defined. */ #include <stdlib.h> @@ -35,7 +34,6 @@ int main(int argc, char *argv[]) { uint32_t olen; long ilen; - unsigned long offs; unsigned long run_size; FILE *f = NULL; int retval = 1; @@ -67,15 +65,6 @@ int main(int argc, char *argv[]) ilen = ftell(f); olen = get_unaligned_le32(&olen); - /* - * Now we have the input (compressed) and output (uncompressed) - * sizes, compute the necessary decompression offset... - */ - - offs = (olen > ilen) ? olen - ilen : 0; - offs += olen >> 12; /* Add 8 bytes for each 32K block */ - offs += 64*1024 + 128; /* Add 64K + 128 bytes slack */ - offs = (offs+4095) & ~4095; /* Round to a 4K boundary */ run_size = atoi(argv[2]); printf(".section \".rodata..compressed\",\"a\",@progbits\n"); @@ -83,8 +72,6 @@ int main(int argc, char *argv[]) printf("z_input_len = %lu\n", ilen); printf(".globl z_output_len\n"); printf("z_output_len = %lu\n", (unsigned long)olen); - printf(".globl z_extract_offset\n"); - printf("z_extract_offset = 0x%lx\n", offs); printf(".globl z_run_size\n"); printf("z_run_size = %lu\n", run_size); diff --git a/arch/x86/boot/header.S b/arch/x86/boot/header.S index 6236b9e..1c057e3 100644 --- a/arch/x86/boot/header.S +++ b/arch/x86/boot/header.S @@ -440,7 +440,27 @@ setup_data: .quad 0 # 64-bit physical pointer to pref_address: .quad LOAD_PHYSICAL_ADDR # preferred load addr -#define ZO_INIT_SIZE (ZO__end - ZO_startup_32 + ZO_z_extract_offset) +/* Check arch/x86/boot/compressed/misc.c for the formula of extra_bytes*/ +#define ZO_z_extra_bytes ((ZO_z_output_len >> 12) + 65536 + 128) +#if ZO_z_output_len > ZO_z_input_len +#define ZO_z_extract_offset (ZO_z_output_len + ZO_z_extra_bytes - ZO_z_input_len) +#else +#define ZO_z_extract_offset ZO_z_extra_bytes +#endif + +/* + * extract_offset has to be bigger than ZO head section. Otherwise + * during head code running to move ZO to end of buffer, it will + * overwrite head code itself. + */ +#if (ZO__ehead - ZO_startup_32) > ZO_z_extract_offset +#define ZO_z_min_extract_offset ((ZO__ehead - ZO_startup_32 + 4095) & ~4095) +#else +#define ZO_z_min_extract_offset ((ZO_z_extract_offset + 4095) & ~4095) +#endif + +#define ZO_INIT_SIZE (ZO__end - ZO_startup_32 + ZO_z_min_extract_offset) + #define VO_INIT_SIZE (VO__end - VO__text) #if ZO_INIT_SIZE > VO_INIT_SIZE #define INIT_SIZE ZO_INIT_SIZE -- 2.5.0