-
Notifications
You must be signed in to change notification settings - Fork 10.8k
Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
[ELF] Add --compress-section to compress matched non-SHF_ALLOC sections
--compress-sections <section-glib>=[none|zlib|zstd] is similar to --compress-debug-sections but applies to broader sections without the SHF_ALLOC flag. lld will report an error if a SHF_ALLOC section is matched. An interesting use case is to compress `.strtab`/`.symtab`, which consume a significant portion of the file size (15.1% for a release build of Clang). An older revision is available at https://reviews.llvm.org/D154641 . This patch focuses on non-allocated sections for safety. Moving `maybeCompress` as D154641 does not handle STT_SECTION symbols for `-r --compress-debug-sections=zlib` (see `relocatable-section-symbol.s` from #66804). Since different output sections may use different compression algorithms, we need CompressedData::type to generalize config->compressDebugSections. GNU ld feature request: https://sourceware.org/bugzilla/show_bug.cgi?id=27452 Link: https://discourse.llvm.org/t/rfc-compress-arbitrary-sections-with-ld-lld-compress-sections/71674 Pull Request: #84855
- Loading branch information
Showing
11 changed files
with
259 additions
and
19 deletions.
There are no files selected for viewing
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,31 @@ | ||
# REQUIRES: x86, zlib | ||
|
||
# RUN: rm -rf %t && mkdir %t && cd %t | ||
# RUN: llvm-mc -filetype=obj -triple=x86_64 %s -o a.o | ||
# RUN: ld.lld -pie a.o --compress-sections .strtab=zlib --compress-sections .symtab=zlib -o out | ||
# RUN: llvm-readelf -Ss -x .strtab out 2>&1 | FileCheck %s | ||
|
||
# CHECK: nonalloc0 PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 00 0 0 1 | ||
# CHECK: .symtab SYMTAB 0000000000000000 [[#%x,]] [[#%x,]] 18 C 12 3 1 | ||
# CHECK-NEXT: .shstrtab STRTAB 0000000000000000 [[#%x,]] [[#%x,]] 00 0 0 1 | ||
# CHECK-NEXT: .strtab STRTAB 0000000000000000 [[#%x,]] [[#%x,]] 00 C 0 0 1 | ||
|
||
## TODO Add compressed SHT_STRTAB/SHT_SYMTAB support to llvm-readelf | ||
# CHECK: warning: {{.*}}: unable to get the string table for the SHT_SYMTAB section: SHT_STRTAB string table section | ||
|
||
# CHECK: Hex dump of section '.strtab': | ||
# CHECK-NEXT: 01000000 00000000 1a000000 00000000 | ||
# CHECK-NEXT: 01000000 00000000 {{.*}} | ||
|
||
# RUN: not ld.lld -shared a.o --compress-sections .dynstr=zlib 2>&1 | FileCheck %s --check-prefix=ERR-ALLOC | ||
# ERR-ALLOC: error: --compress-sections: section '.dynstr' with the SHF_ALLOC flag cannot be compressed | ||
|
||
.globl _start, g0, g1 | ||
_start: | ||
l0: | ||
g0: | ||
g1: | ||
|
||
.section nonalloc0,"" | ||
.quad .text+1 | ||
.quad .text+2 |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,91 @@ | ||
# REQUIRES: x86, zlib, zstd | ||
|
||
# RUN: rm -rf %t && mkdir %t && cd %t | ||
# RUN: llvm-mc -filetype=obj -triple=x86_64 %s -o a.o | ||
# RUN: ld.lld -pie a.o -o out --compress-sections '*0=zlib' --compress-sections '*0=none' --compress-sections 'nomatch=none' | ||
# RUN: llvm-readelf -SrsX out | FileCheck %s --check-prefix=CHECK1 | ||
|
||
# CHECK1: Name Type Address Off Size ES Flg Lk Inf Al | ||
# CHECK1: foo0 PROGBITS [[#%x,FOO0:]] [[#%x,]] [[#%x,]] 00 A 0 0 8 | ||
# CHECK1-NEXT: foo1 PROGBITS [[#%x,FOO1:]] [[#%x,]] [[#%x,]] 00 A 0 0 8 | ||
# CHECK1-NEXT: .text PROGBITS [[#%x,TEXT:]] [[#%x,]] [[#%x,]] 00 AX 0 0 4 | ||
# CHECK1: nonalloc0 PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 00 0 0 8 | ||
# CHECK1-NEXT: nonalloc1 PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 00 0 0 8 | ||
# CHECK1-NEXT: .debug_str PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 01 MS 0 0 1 | ||
|
||
# CHECK1: 0000000000000010 0 NOTYPE LOCAL DEFAULT [[#]] (nonalloc0) sym0 | ||
# CHECK1: 0000000000000008 0 NOTYPE LOCAL DEFAULT [[#]] (nonalloc1) sym1 | ||
|
||
# RUN: ld.lld -pie a.o --compress-sections '*c0=zlib' --compress-sections .debug_str=zstd -o out2 | ||
# RUN: llvm-readelf -SrsX -x nonalloc0 -x .debug_str out2 | FileCheck %s --check-prefix=CHECK2 | ||
|
||
# CHECK2: Name Type Address Off Size ES Flg Lk Inf Al | ||
# CHECK2: foo0 PROGBITS [[#%x,FOO0:]] [[#%x,]] [[#%x,]] 00 A 0 0 8 | ||
# CHECK2-NEXT: foo1 PROGBITS [[#%x,FOO1:]] [[#%x,]] [[#%x,]] 00 A 0 0 8 | ||
# CHECK2-NEXT: .text PROGBITS [[#%x,TEXT:]] [[#%x,]] [[#%x,]] 00 AX 0 0 4 | ||
# CHECK2: nonalloc0 PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 00 C 0 0 1 | ||
# CHECK2-NEXT: nonalloc1 PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 00 0 0 8 | ||
# CHECK2-NEXT: .debug_str PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 01 MSC 0 0 1 | ||
|
||
# CHECK2: 0000000000000010 0 NOTYPE LOCAL DEFAULT [[#]] (nonalloc0) sym0 | ||
# CHECK2: 0000000000000008 0 NOTYPE LOCAL DEFAULT [[#]] (nonalloc1) sym1 | ||
|
||
# CHECK2: Hex dump of section 'nonalloc0': | ||
## zlib with ch_size=0x10 | ||
# CHECK2-NEXT: 01000000 00000000 10000000 00000000 | ||
# CHECK2-NEXT: 01000000 00000000 {{.*}} | ||
# CHECK2: Hex dump of section '.debug_str': | ||
## zstd with ch_size=0x38 | ||
# CHECK2-NEXT: 02000000 00000000 38000000 00000000 | ||
# CHECK2-NEXT: 01000000 00000000 {{.*}} | ||
|
||
## --compress-debug-sections=none takes precedence. | ||
# RUN: ld.lld a.o --compress-debug-sections=none --compress-sections .debug_str=zstd -o out3 | ||
# RUN: llvm-readelf -S out3 | FileCheck %s --check-prefix=CHECK3 | ||
|
||
# CHECK3: .debug_str PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 01 MS 0 0 1 | ||
|
||
# RUN: not ld.lld a.o --compress-sections '*0=zlib' 2>&1 | \ | ||
# RUN: FileCheck %s --check-prefix=ERR-ALLOC --implicit-check-not=error: | ||
# ERR-ALLOC: error: --compress-sections: section 'foo0' with the SHF_ALLOC flag cannot be compressed | ||
|
||
# RUN: not ld.lld --compress-sections=foo a.o 2>&1 | \ | ||
# RUN: FileCheck %s --check-prefix=ERR1 --implicit-check-not=error: | ||
# ERR1: error: --compress-sections: parse error, not 'section-glob=[none|zlib|zstd]' | ||
|
||
# RUN: not ld.lld --compress-sections 'a[=zlib' a.o 2>&1 | \ | ||
# RUN: FileCheck %s --check-prefix=ERR2 --implicit-check-not=error: | ||
# ERR2: error: --compress-sections: invalid glob pattern, unmatched '[' | ||
|
||
# RUN: not ld.lld a.o --compress-sections='.debug*=zlib-gabi' --compress-sections='.debug*=' 2>&1 | \ | ||
# RUN: FileCheck -check-prefix=ERR3 %s | ||
# ERR3: unknown --compress-sections value: zlib-gabi | ||
# ERR3-NEXT: --compress-sections: parse error, not 'section-glob=[none|zlib|zstd]' | ||
|
||
.globl _start | ||
_start: | ||
ret | ||
|
||
.section foo0,"a" | ||
.balign 8 | ||
.quad .text-. | ||
.quad .text-. | ||
.section foo1,"a" | ||
.balign 8 | ||
.quad .text-. | ||
.quad .text-. | ||
.section nonalloc0,"" | ||
.balign 8 | ||
.quad .text+1 | ||
.quad .text+2 | ||
sym0: | ||
.section nonalloc1,"" | ||
.balign 8 | ||
.quad 42 | ||
sym1: | ||
|
||
.section .debug_str,"MS",@progbits,1 | ||
.Linfo_string0: | ||
.asciz "AAAAAAAAAAAAAAAAAAAAAAAAAAA" | ||
.Linfo_string1: | ||
.asciz "BBBBBBBBBBBBBBBBBBBBBBBBBBB" |
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,62 @@ | ||
# REQUIRES: x86, zlib | ||
|
||
# RUN: rm -rf %t && split-file %s %t && cd %t | ||
# RUN: llvm-mc -filetype=obj -triple=x86_64 a.s -o a.o | ||
# RUN: ld.lld -T a.lds a.o --compress-sections nonalloc=zlib --compress-sections str=zlib -o out | ||
# RUN: llvm-readelf -SsXz -p str out | FileCheck %s | ||
|
||
# CHECK: Name Type Address Off Size ES Flg Lk Inf Al | ||
# CHECK: nonalloc PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 00 C 0 0 1 | ||
# CHECK-NEXT: str PROGBITS 0000000000000000 [[#%x,]] [[#%x,]] 01 MSC 0 0 1 | ||
|
||
# CHECK: 0000000000000000 0 NOTYPE GLOBAL DEFAULT [[#]] (nonalloc) nonalloc_start | ||
# CHECK: 0000000000000023 0 NOTYPE GLOBAL DEFAULT [[#]] (nonalloc) nonalloc_end | ||
# CHECK: String dump of section 'str': | ||
# CHECK-NEXT: [ 0] AAA | ||
# CHECK-NEXT: [ 4] BBB | ||
|
||
## TODO The uncompressed size of 'nonalloc' is dependent on linker script | ||
## commands, which is not handled. We should report an error. | ||
# RUN: ld.lld -T b.lds a.o --compress-sections nonalloc=zlib | ||
|
||
#--- a.s | ||
.globl _start | ||
_start: | ||
ret | ||
|
||
.section nonalloc0,"" | ||
.balign 8 | ||
.quad .text | ||
.quad .text | ||
.section nonalloc1,"" | ||
.balign 8 | ||
.quad 42 | ||
|
||
.section str,"MS",@progbits,1 | ||
.asciz "AAA" | ||
.asciz "BBB" | ||
|
||
#--- a.lds | ||
SECTIONS { | ||
.text : { *(.text) } | ||
c = SIZEOF(.text); | ||
b = c+1; | ||
a = b+1; | ||
nonalloc : { | ||
nonalloc_start = .; | ||
## In general, using data commands is error-prone. This case is correct, though. | ||
*(nonalloc*) QUAD(SIZEOF(.text)) | ||
. += a; | ||
nonalloc_end = .; | ||
} | ||
str : { *(str) } | ||
} | ||
|
||
#--- b.lds | ||
SECTIONS { | ||
nonalloc : { *(nonalloc*) . += a; } | ||
.text : { *(.text) } | ||
a = b+1; | ||
b = c+1; | ||
c = SIZEOF(.text); | ||
} |