| Message ID | 20201118220731.925424-3-samitolvanen@google.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | Add support for Clang LTO | expand |
On Wed, Nov 18, 2020 at 2:07 PM Sami Tolvanen <samitolvanen@google.com> wrote: > > This change adds build system support for Clang's Link Time > Optimization (LTO). With -flto, instead of ELF object files, Clang > produces LLVM bitcode, which is compiled into native code at link > time, allowing the final binary to be optimized globally. For more > details, see: > > https://llvm.org/docs/LinkTimeOptimization.html > > The Kconfig option CONFIG_LTO_CLANG is implemented as a choice, > which defaults to LTO being disabled. To use LTO, the architecture > must select ARCH_SUPPORTS_LTO_CLANG and support: > > - compiling with Clang, > - compiling inline assembly with Clang's integrated assembler, > - and linking with LLD. > > While using full LTO results in the best runtime performance, the > compilation is not scalable in time or memory. CONFIG_THINLTO > enables ThinLTO, which allows parallel optimization and faster > incremental builds. ThinLTO is used by default if the architecture > also selects ARCH_SUPPORTS_THINLTO: > > https://clang.llvm.org/docs/ThinLTO.html > > To enable LTO, LLVM tools must be used to handle bitcode files. The > easiest way is to pass the LLVM=1 option to make: > > $ make LLVM=1 defconfig > $ scripts/config -e LTO_CLANG > $ make LLVM=1 > > Alternatively, at least the following LLVM tools must be used: > > CC=clang LD=ld.lld AR=llvm-ar NM=llvm-nm > > To prepare for LTO support with other compilers, common parts are > gated behind the CONFIG_LTO option, and LTO can be disabled for > specific files by filtering out CC_FLAGS_LTO. > > Signed-off-by: Sami Tolvanen <samitolvanen@google.com> > Reviewed-by: Kees Cook <keescook@chromium.org> > --- > Makefile | 19 +++++++- > arch/Kconfig | 75 +++++++++++++++++++++++++++++++ > include/asm-generic/vmlinux.lds.h | 11 +++-- > scripts/Makefile.build | 9 +++- > scripts/Makefile.modfinal | 9 +++- > scripts/Makefile.modpost | 21 ++++++++- > scripts/link-vmlinux.sh | 32 +++++++++---- > 7 files changed, 158 insertions(+), 18 deletions(-) > > diff --git a/Makefile b/Makefile > index 8c8feb4245a6..240560e88d69 100644 > --- a/Makefile > +++ b/Makefile > @@ -893,6 +893,21 @@ KBUILD_CFLAGS += $(CC_FLAGS_SCS) > export CC_FLAGS_SCS > endif > > +ifdef CONFIG_LTO_CLANG > +ifdef CONFIG_THINLTO > +CC_FLAGS_LTO += -flto=thin -fsplit-lto-unit > +KBUILD_LDFLAGS += --thinlto-cache-dir=$(extmod-prefix).thinlto-cache > +else > +CC_FLAGS_LTO += -flto > +endif > +CC_FLAGS_LTO += -fvisibility=default > +endif > + > +ifdef CONFIG_LTO > +KBUILD_CFLAGS += $(CC_FLAGS_LTO) > +export CC_FLAGS_LTO > +endif > + > ifdef CONFIG_DEBUG_FORCE_FUNCTION_ALIGN_32B > KBUILD_CFLAGS += -falign-functions=32 > endif > @@ -1473,7 +1488,7 @@ MRPROPER_FILES += include/config include/generated \ > *.spec > > # Directories & files removed with 'make distclean' > -DISTCLEAN_FILES += tags TAGS cscope* GPATH GTAGS GRTAGS GSYMS > +DISTCLEAN_FILES += tags TAGS cscope* GPATH GTAGS GRTAGS GSYMS .thinlto-cache > > # clean - Delete most, but leave enough to build external modules > # > @@ -1719,7 +1734,7 @@ PHONY += compile_commands.json > > clean-dirs := $(KBUILD_EXTMOD) > clean: rm-files := $(KBUILD_EXTMOD)/Module.symvers $(KBUILD_EXTMOD)/modules.nsdeps \ > - $(KBUILD_EXTMOD)/compile_commands.json > + $(KBUILD_EXTMOD)/compile_commands.json $(KBUILD_EXTMOD)/.thinlto-cache > > PHONY += help > help: > diff --git a/arch/Kconfig b/arch/Kconfig > index 56b6ccc0e32d..a41fcb3ca7c6 100644 > --- a/arch/Kconfig > +++ b/arch/Kconfig > @@ -598,6 +598,81 @@ config SHADOW_CALL_STACK > reading and writing arbitrary memory may be able to locate them > and hijack control flow by modifying the stacks. > > +config LTO > + bool > + > +config ARCH_SUPPORTS_LTO_CLANG > + bool > + help > + An architecture should select this option if it supports: > + - compiling with Clang, > + - compiling inline assembly with Clang's integrated assembler, > + - and linking with LLD. > + > +config ARCH_SUPPORTS_THINLTO > + bool > + help > + An architecture should select this option if it supports Clang's > + ThinLTO. > + > +config THINLTO > + bool "Clang ThinLTO" > + depends on LTO_CLANG && ARCH_SUPPORTS_THINLTO > + default y > + help > + This option enables Clang's ThinLTO, which allows for parallel > + optimization and faster incremental compiles. More information > + can be found from Clang's documentation: > + > + https://clang.llvm.org/docs/ThinLTO.html > + > + If you say N here, the compiler will use full LTO, which may > + produce faster code, but building the kernel will be significantly > + slower as the linker won't efficiently utilize multiple threads. > + > + If unsure, say Y. I think the order of these new configs makes it so that ThinLTO appears above LTO in menuconfig; I don't like that, and wish it came immediately after. Does `THINLTO` have to be defined _after_ the choice for LTO_NONE/LTO_CLANG, perhaps? Secondly, I don't like how ThinLTO is a config and not a choice. If I don't set ThinLTO, what am I getting? That's a rhetorical question; I know its full LTO, and I guess the help text does talk about the tradeoffs and what you would get. I guess what's curious to me is "why does it display ThinLTO? Why not FullLTO?" I can't help but wonder if a kconfig `choice` rather than a `config` would be better here, that way it's more obvious the user is making a choice between ThinLTO vs Full LTO, rather than the current patches which look like "ThinkLTO on/off." These are cosmetic concerns, feel free to ignore. Just a thought. > + > +choice > + prompt "Link Time Optimization (LTO)" > + default LTO_NONE > + help > + This option enables Link Time Optimization (LTO), which allows the > + compiler to optimize binaries globally. > + > + If unsure, select LTO_NONE. Note that LTO is very resource-intensive > + so it's disabled by default. > + > +config LTO_NONE > + bool "None" > + > +config LTO_CLANG > + bool "Clang's Link Time Optimization (EXPERIMENTAL)" > + # Clang >= 11: https://github.com/ClangBuiltLinux/linux/issues/510 > + depends on CC_IS_CLANG && CLANG_VERSION >= 110000 && LD_IS_LLD > + depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm) > + depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm) > + depends on ARCH_SUPPORTS_LTO_CLANG > + depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT > + depends on !KASAN > + depends on !GCOV_KERNEL > + depends on !MODVERSIONS > + select LTO > + help > + This option enables Clang's Link Time Optimization (LTO), which > + allows the compiler to optimize the kernel globally. If you enable > + this option, the compiler generates LLVM bitcode instead of ELF > + object files, and the actual compilation from bitcode happens at > + the LTO link step, which may take several minutes depending on the > + kernel configuration. More information can be found from LLVM's > + documentation: > + > + https://llvm.org/docs/LinkTimeOptimization.html > + > + To select this option, you also need to use LLVM tools to handle > + the bitcode by passing LLVM=1 to make. > + > +endchoice > + > config HAVE_ARCH_WITHIN_STACK_FRAMES > bool > help > diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h > index b2b3d81b1535..8988a2e445d8 100644 > --- a/include/asm-generic/vmlinux.lds.h > +++ b/include/asm-generic/vmlinux.lds.h > @@ -90,15 +90,18 @@ > * .data. We don't want to pull in .data..other sections, which Linux > * has defined. Same for text and bss. > * > + * With LTO_CLANG, the linker also splits sections by default, so we need > + * these macros to combine the sections during the final link. > + * > * RODATA_MAIN is not used because existing code already defines .rodata.x > * sections to be brought in with rodata. > */ > -#ifdef CONFIG_LD_DEAD_CODE_DATA_ELIMINATION > +#if defined(CONFIG_LD_DEAD_CODE_DATA_ELIMINATION) || defined(CONFIG_LTO_CLANG) > #define TEXT_MAIN .text .text.[0-9a-zA-Z_]* > -#define DATA_MAIN .data .data.[0-9a-zA-Z_]* .data..LPBX* > +#define DATA_MAIN .data .data.[0-9a-zA-Z_]* .data..L* .data..compoundliteral* > #define SDATA_MAIN .sdata .sdata.[0-9a-zA-Z_]* > -#define RODATA_MAIN .rodata .rodata.[0-9a-zA-Z_]* > -#define BSS_MAIN .bss .bss.[0-9a-zA-Z_]* > +#define RODATA_MAIN .rodata .rodata.[0-9a-zA-Z_]* .rodata..L* > +#define BSS_MAIN .bss .bss.[0-9a-zA-Z_]* .bss..compoundliteral* > #define SBSS_MAIN .sbss .sbss.[0-9a-zA-Z_]* > #else > #define TEXT_MAIN .text > diff --git a/scripts/Makefile.build b/scripts/Makefile.build > index 2175ddb1ee0c..ed74b2f986f7 100644 > --- a/scripts/Makefile.build > +++ b/scripts/Makefile.build > @@ -111,7 +111,7 @@ endif > # --------------------------------------------------------------------------- > > quiet_cmd_cc_s_c = CC $(quiet_modtag) $@ > - cmd_cc_s_c = $(CC) $(filter-out $(DEBUG_CFLAGS), $(c_flags)) -fverbose-asm -S -o $@ $< > + cmd_cc_s_c = $(CC) $(filter-out $(DEBUG_CFLAGS) $(CC_FLAGS_LTO), $(c_flags)) -fverbose-asm -S -o $@ $< > > $(obj)/%.s: $(src)/%.c FORCE > $(call if_changed_dep,cc_s_c) > @@ -425,8 +425,15 @@ $(obj)/lib.a: $(lib-y) FORCE > # Do not replace $(filter %.o,^) with $(real-prereqs). When a single object > # module is turned into a multi object module, $^ will contain header file > # dependencies recorded in the .*.cmd file. > +ifdef CONFIG_LTO_CLANG > +quiet_cmd_link_multi-m = AR [M] $@ > +cmd_link_multi-m = \ > + rm -f $@; \ > + $(AR) cDPrsT $@ $(filter %.o,$^) > +else > quiet_cmd_link_multi-m = LD [M] $@ > cmd_link_multi-m = $(LD) $(ld_flags) -r -o $@ $(filter %.o,$^) > +endif > > $(multi-used-m): FORCE > $(call if_changed,link_multi-m) > diff --git a/scripts/Makefile.modfinal b/scripts/Makefile.modfinal > index ae01baf96f4e..2cb9a1d88434 100644 > --- a/scripts/Makefile.modfinal > +++ b/scripts/Makefile.modfinal > @@ -6,6 +6,7 @@ > PHONY := __modfinal > __modfinal: > > +include $(objtree)/include/config/auto.conf > include $(srctree)/scripts/Kbuild.include > > # for c_flags > @@ -29,6 +30,12 @@ quiet_cmd_cc_o_c = CC [M] $@ > > ARCH_POSTLINK := $(wildcard $(srctree)/arch/$(SRCARCH)/Makefile.postlink) > > +ifdef CONFIG_LTO_CLANG > +# With CONFIG_LTO_CLANG, reuse the object file we compiled for modpost to > +# avoid a second slow LTO link > +prelink-ext := .lto > +endif > + > quiet_cmd_ld_ko_o = LD [M] $@ > cmd_ld_ko_o = \ > $(LD) -r $(KBUILD_LDFLAGS) \ > @@ -36,7 +43,7 @@ quiet_cmd_ld_ko_o = LD [M] $@ > -T scripts/module.lds -o $@ $(filter %.o, $^); \ > $(if $(ARCH_POSTLINK), $(MAKE) -f $(ARCH_POSTLINK) $@, true) > > -$(modules): %.ko: %.o %.mod.o scripts/module.lds FORCE > +$(modules): %.ko: %$(prelink-ext).o %.mod.o scripts/module.lds FORCE > +$(call if_changed,ld_ko_o) > > targets += $(modules) $(modules:.ko=.mod.o) > diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost > index f54b6ac37ac2..9ff8bfdb574d 100644 > --- a/scripts/Makefile.modpost > +++ b/scripts/Makefile.modpost > @@ -43,6 +43,9 @@ __modpost: > include include/config/auto.conf > include scripts/Kbuild.include > > +# for ld_flags > +include scripts/Makefile.lib > + > MODPOST = scripts/mod/modpost \ > $(if $(CONFIG_MODVERSIONS),-m) \ > $(if $(CONFIG_MODULE_SRCVERSION_ALL),-a) \ > @@ -102,12 +105,26 @@ $(input-symdump): > @echo >&2 'WARNING: Symbol version dump "$@" is missing.' > @echo >&2 ' Modules may not have dependencies or modversions.' > > +ifdef CONFIG_LTO_CLANG > +# With CONFIG_LTO_CLANG, .o files might be LLVM bitcode, so we need to run > +# LTO to compile them into native code before running modpost > +prelink-ext := .lto > + > +quiet_cmd_cc_lto_link_modules = LTO [M] $@ > +cmd_cc_lto_link_modules = $(LD) $(ld_flags) -r -o $@ --whole-archive $^ > + > +%.lto.o: %.o > + $(call if_changed,cc_lto_link_modules) > +endif > + > +modules := $(sort $(shell cat $(MODORDER))) > + > # Read out modules.order to pass in modpost. > # Otherwise, allmodconfig would fail with "Argument list too long". > quiet_cmd_modpost = MODPOST $@ > - cmd_modpost = sed 's/ko$$/o/' $< | $(MODPOST) -T - > + cmd_modpost = sed 's/\.ko$$/$(prelink-ext)\.o/' $< | $(MODPOST) -T - > > -$(output-symdump): $(MODORDER) $(input-symdump) FORCE > +$(output-symdump): $(MODORDER) $(input-symdump) $(modules:.ko=$(prelink-ext).o) FORCE > $(call if_changed,modpost) > > targets += $(output-symdump) > diff --git a/scripts/link-vmlinux.sh b/scripts/link-vmlinux.sh > index 6eded325c837..596507573a48 100755 > --- a/scripts/link-vmlinux.sh > +++ b/scripts/link-vmlinux.sh > @@ -56,6 +56,14 @@ modpost_link() > ${KBUILD_VMLINUX_LIBS} \ > --end-group" > > + if [ -n "${CONFIG_LTO_CLANG}" ]; then > + # This might take a while, so indicate that we're doing > + # an LTO link > + info LTO ${1} > + else > + info LD ${1} > + fi > + > ${LD} ${KBUILD_LDFLAGS} -r -o ${1} ${objects} > } > > @@ -103,13 +111,22 @@ vmlinux_link() > fi > > if [ "${SRCARCH}" != "um" ]; then > - objects="--whole-archive \ > - ${KBUILD_VMLINUX_OBJS} \ > - --no-whole-archive \ > - --start-group \ > - ${KBUILD_VMLINUX_LIBS} \ > - --end-group \ > - ${@}" > + if [ -n "${CONFIG_LTO_CLANG}" ]; then > + # Use vmlinux.o instead of performing the slow LTO > + # link again. > + objects="--whole-archive \ > + vmlinux.o \ > + --no-whole-archive \ > + ${@}" > + else > + objects="--whole-archive \ > + ${KBUILD_VMLINUX_OBJS} \ > + --no-whole-archive \ > + --start-group \ > + ${KBUILD_VMLINUX_LIBS} \ > + --end-group \ > + ${@}" > + fi > > ${LD} ${KBUILD_LDFLAGS} ${LDFLAGS_vmlinux} \ > ${strip_debug#-Wl,} \ > @@ -274,7 +291,6 @@ fi; > ${MAKE} -f "${srctree}/scripts/Makefile.build" obj=init need-builtin=1 > > #link vmlinux.o > -info LD vmlinux.o > modpost_link vmlinux.o > objtool_link vmlinux.o > > -- > 2.29.2.299.gdc1121823c-goog >
On Wed, Nov 18, 2020 at 3:49 PM Nick Desaulniers <ndesaulniers@google.com> wrote: > > On Wed, Nov 18, 2020 at 2:07 PM Sami Tolvanen <samitolvanen@google.com> wrote: > > > > This change adds build system support for Clang's Link Time > > Optimization (LTO). With -flto, instead of ELF object files, Clang > > produces LLVM bitcode, which is compiled into native code at link > > time, allowing the final binary to be optimized globally. For more > > details, see: > > > > https://llvm.org/docs/LinkTimeOptimization.html > > > > The Kconfig option CONFIG_LTO_CLANG is implemented as a choice, > > which defaults to LTO being disabled. To use LTO, the architecture > > must select ARCH_SUPPORTS_LTO_CLANG and support: > > > > - compiling with Clang, > > - compiling inline assembly with Clang's integrated assembler, > > - and linking with LLD. > > > > While using full LTO results in the best runtime performance, the > > compilation is not scalable in time or memory. CONFIG_THINLTO > > enables ThinLTO, which allows parallel optimization and faster > > incremental builds. ThinLTO is used by default if the architecture > > also selects ARCH_SUPPORTS_THINLTO: > > > > https://clang.llvm.org/docs/ThinLTO.html > > > > To enable LTO, LLVM tools must be used to handle bitcode files. The > > easiest way is to pass the LLVM=1 option to make: > > > > $ make LLVM=1 defconfig > > $ scripts/config -e LTO_CLANG > > $ make LLVM=1 > > > > Alternatively, at least the following LLVM tools must be used: > > > > CC=clang LD=ld.lld AR=llvm-ar NM=llvm-nm > > > > To prepare for LTO support with other compilers, common parts are > > gated behind the CONFIG_LTO option, and LTO can be disabled for > > specific files by filtering out CC_FLAGS_LTO. > > > > Signed-off-by: Sami Tolvanen <samitolvanen@google.com> > > Reviewed-by: Kees Cook <keescook@chromium.org> > > --- > > Makefile | 19 +++++++- > > arch/Kconfig | 75 +++++++++++++++++++++++++++++++ > > include/asm-generic/vmlinux.lds.h | 11 +++-- > > scripts/Makefile.build | 9 +++- > > scripts/Makefile.modfinal | 9 +++- > > scripts/Makefile.modpost | 21 ++++++++- > > scripts/link-vmlinux.sh | 32 +++++++++---- > > 7 files changed, 158 insertions(+), 18 deletions(-) > > > > diff --git a/Makefile b/Makefile > > index 8c8feb4245a6..240560e88d69 100644 > > --- a/Makefile > > +++ b/Makefile > > @@ -893,6 +893,21 @@ KBUILD_CFLAGS += $(CC_FLAGS_SCS) > > export CC_FLAGS_SCS > > endif > > > > +ifdef CONFIG_LTO_CLANG > > +ifdef CONFIG_THINLTO > > +CC_FLAGS_LTO += -flto=thin -fsplit-lto-unit > > +KBUILD_LDFLAGS += --thinlto-cache-dir=$(extmod-prefix).thinlto-cache > > +else > > +CC_FLAGS_LTO += -flto > > +endif > > +CC_FLAGS_LTO += -fvisibility=default > > +endif > > + > > +ifdef CONFIG_LTO > > +KBUILD_CFLAGS += $(CC_FLAGS_LTO) > > +export CC_FLAGS_LTO > > +endif > > + > > ifdef CONFIG_DEBUG_FORCE_FUNCTION_ALIGN_32B > > KBUILD_CFLAGS += -falign-functions=32 > > endif > > @@ -1473,7 +1488,7 @@ MRPROPER_FILES += include/config include/generated \ > > *.spec > > > > # Directories & files removed with 'make distclean' > > -DISTCLEAN_FILES += tags TAGS cscope* GPATH GTAGS GRTAGS GSYMS > > +DISTCLEAN_FILES += tags TAGS cscope* GPATH GTAGS GRTAGS GSYMS .thinlto-cache > > > > # clean - Delete most, but leave enough to build external modules > > # > > @@ -1719,7 +1734,7 @@ PHONY += compile_commands.json > > > > clean-dirs := $(KBUILD_EXTMOD) > > clean: rm-files := $(KBUILD_EXTMOD)/Module.symvers $(KBUILD_EXTMOD)/modules.nsdeps \ > > - $(KBUILD_EXTMOD)/compile_commands.json > > + $(KBUILD_EXTMOD)/compile_commands.json $(KBUILD_EXTMOD)/.thinlto-cache > > > > PHONY += help > > help: > > diff --git a/arch/Kconfig b/arch/Kconfig > > index 56b6ccc0e32d..a41fcb3ca7c6 100644 > > --- a/arch/Kconfig > > +++ b/arch/Kconfig > > @@ -598,6 +598,81 @@ config SHADOW_CALL_STACK > > reading and writing arbitrary memory may be able to locate them > > and hijack control flow by modifying the stacks. > > > > +config LTO > > + bool > > + > > +config ARCH_SUPPORTS_LTO_CLANG > > + bool > > + help > > + An architecture should select this option if it supports: > > + - compiling with Clang, > > + - compiling inline assembly with Clang's integrated assembler, > > + - and linking with LLD. > > + > > +config ARCH_SUPPORTS_THINLTO > > + bool > > + help > > + An architecture should select this option if it supports Clang's > > + ThinLTO. > > + > > +config THINLTO > > + bool "Clang ThinLTO" > > + depends on LTO_CLANG && ARCH_SUPPORTS_THINLTO > > + default y > > + help > > + This option enables Clang's ThinLTO, which allows for parallel > > + optimization and faster incremental compiles. More information > > + can be found from Clang's documentation: > > + > > + https://clang.llvm.org/docs/ThinLTO.html > > + > > + If you say N here, the compiler will use full LTO, which may > > + produce faster code, but building the kernel will be significantly > > + slower as the linker won't efficiently utilize multiple threads. > > + > > + If unsure, say Y. > > I think the order of these new configs makes it so that ThinLTO > appears above LTO in menuconfig; I don't like that, and wish it came > immediately after. Does `THINLTO` have to be defined _after_ the > choice for LTO_NONE/LTO_CLANG, perhaps? > > Secondly, I don't like how ThinLTO is a config and not a choice. If I > don't set ThinLTO, what am I getting? That's a rhetorical question; I > know its full LTO, and I guess the help text does talk about the > tradeoffs and what you would get. I guess what's curious to me is > "why does it display ThinLTO? Why not FullLTO?" I can't help but > wonder if a kconfig `choice` rather than a `config` would be better > here, that way it's more obvious the user is making a choice between > ThinLTO vs Full LTO, rather than the current patches which look like > "ThinkLTO on/off." Changing the ThinLTO config to a choice and moving it after the main LTO config sounds like a good idea to me. I'll see if I can change this in v8. Thanks! Sami
On Fri, Nov 20, 2020 at 08:23:11AM -0800, Sami Tolvanen wrote: > Changing the ThinLTO config to a choice and moving it after the main > LTO config sounds like a good idea to me. I'll see if I can change > this in v8. Thanks! Originally, I thought this might be a bit ugly once GCC LTO is added, but this could be just a choice like we're done for the stack initialization. Something like an "LTO" choice of NONE, CLANG_FULL, CLANG_THIN, and in the future GCC, etc.
On Fri, Nov 20, 2020 at 11:47:21AM -0800, Kees Cook wrote: > On Fri, Nov 20, 2020 at 08:23:11AM -0800, Sami Tolvanen wrote: > > Changing the ThinLTO config to a choice and moving it after the main > > LTO config sounds like a good idea to me. I'll see if I can change > > this in v8. Thanks! > > Originally, I thought this might be a bit ugly once GCC LTO is added, > but this could be just a choice like we're done for the stack > initialization. Something like an "LTO" choice of NONE, CLANG_FULL, > CLANG_THIN, and in the future GCC, etc. Having two separate choices might be a little bit cleaner though? One for the compiler (LTO_CLANG versus LTO_GCC) and one for the type (THINLTO versus FULLLTO). The type one could just have a "depends on CC_IS_CLANG" to ensure it only showed up when needed. Cheers, Nathan
On Fri, Nov 20, 2020 at 01:29:35PM -0700, Nathan Chancellor wrote: > On Fri, Nov 20, 2020 at 11:47:21AM -0800, Kees Cook wrote: > > On Fri, Nov 20, 2020 at 08:23:11AM -0800, Sami Tolvanen wrote: > > > Changing the ThinLTO config to a choice and moving it after the main > > > LTO config sounds like a good idea to me. I'll see if I can change > > > this in v8. Thanks! > > > > Originally, I thought this might be a bit ugly once GCC LTO is added, > > but this could be just a choice like we're done for the stack > > initialization. Something like an "LTO" choice of NONE, CLANG_FULL, > > CLANG_THIN, and in the future GCC, etc. > > Having two separate choices might be a little bit cleaner though? One > for the compiler (LTO_CLANG versus LTO_GCC) and one for the type > (THINLTO versus FULLLTO). The type one could just have a "depends on > CC_IS_CLANG" to ensure it only showed up when needed. Right, that's how the stack init choice works. Kconfigs that aren't supported by the compiler won't be shown. I.e. after Sami's future patch, the only choice for GCC will be CONFIG_LTO_NONE. But building under Clang, it would offer CONFIG_LTO_NONE, CONFIG_LTO_CLANG_FULL, CONFIG_LTO_CLANG_THIN, or something. (and I assume CONFIG_LTO would be def_bool y, depends on !LTO_NONE)
On Fri, Nov 20, 2020 at 12:43 PM Kees Cook <keescook@chromium.org> wrote: > > On Fri, Nov 20, 2020 at 01:29:35PM -0700, Nathan Chancellor wrote: > > On Fri, Nov 20, 2020 at 11:47:21AM -0800, Kees Cook wrote: > > > On Fri, Nov 20, 2020 at 08:23:11AM -0800, Sami Tolvanen wrote: > > > > Changing the ThinLTO config to a choice and moving it after the main > > > > LTO config sounds like a good idea to me. I'll see if I can change > > > > this in v8. Thanks! > > > > > > Originally, I thought this might be a bit ugly once GCC LTO is added, > > > but this could be just a choice like we're done for the stack > > > initialization. Something like an "LTO" choice of NONE, CLANG_FULL, > > > CLANG_THIN, and in the future GCC, etc. > > > > Having two separate choices might be a little bit cleaner though? One > > for the compiler (LTO_CLANG versus LTO_GCC) and one for the type > > (THINLTO versus FULLLTO). The type one could just have a "depends on > > CC_IS_CLANG" to ensure it only showed up when needed. > > Right, that's how the stack init choice works. Kconfigs that aren't > supported by the compiler won't be shown. I.e. after Sami's future > patch, the only choice for GCC will be CONFIG_LTO_NONE. But building > under Clang, it would offer CONFIG_LTO_NONE, CONFIG_LTO_CLANG_FULL, > CONFIG_LTO_CLANG_THIN, or something. > > (and I assume CONFIG_LTO would be def_bool y, depends on !LTO_NONE) I'm fine with adding ThinLTO as another option to the LTO choice, but it would duplicate the dependencies and a lot of the help text. I suppose we could add another config for the dependencies and have both LTO options depend on that instead. Sami
On Fri, Nov 20, 2020 at 12:58:41PM -0800, Sami Tolvanen wrote: > On Fri, Nov 20, 2020 at 12:43 PM Kees Cook <keescook@chromium.org> wrote: > > > > On Fri, Nov 20, 2020 at 01:29:35PM -0700, Nathan Chancellor wrote: > > > On Fri, Nov 20, 2020 at 11:47:21AM -0800, Kees Cook wrote: > > > > On Fri, Nov 20, 2020 at 08:23:11AM -0800, Sami Tolvanen wrote: > > > > > Changing the ThinLTO config to a choice and moving it after the main > > > > > LTO config sounds like a good idea to me. I'll see if I can change > > > > > this in v8. Thanks! > > > > > > > > Originally, I thought this might be a bit ugly once GCC LTO is added, > > > > but this could be just a choice like we're done for the stack > > > > initialization. Something like an "LTO" choice of NONE, CLANG_FULL, > > > > CLANG_THIN, and in the future GCC, etc. > > > > > > Having two separate choices might be a little bit cleaner though? One > > > for the compiler (LTO_CLANG versus LTO_GCC) and one for the type > > > (THINLTO versus FULLLTO). The type one could just have a "depends on > > > CC_IS_CLANG" to ensure it only showed up when needed. > > > > Right, that's how the stack init choice works. Kconfigs that aren't > > supported by the compiler won't be shown. I.e. after Sami's future > > patch, the only choice for GCC will be CONFIG_LTO_NONE. But building > > under Clang, it would offer CONFIG_LTO_NONE, CONFIG_LTO_CLANG_FULL, > > CONFIG_LTO_CLANG_THIN, or something. > > > > (and I assume CONFIG_LTO would be def_bool y, depends on !LTO_NONE) > > I'm fine with adding ThinLTO as another option to the LTO choice, but > it would duplicate the dependencies and a lot of the help text. I > suppose we could add another config for the dependencies and have both > LTO options depend on that instead. How about something like this? This separates the arch support, compiler support, and user choice into three separate Kconfig areas, which I think should work. diff --git a/Makefile b/Makefile index e397c4caec1b..af902718e882 100644 --- a/Makefile +++ b/Makefile @@ -897,7 +897,7 @@ export CC_FLAGS_SCS endif ifdef CONFIG_LTO_CLANG -ifdef CONFIG_THINLTO +ifdef CONFIG_LTO_CLANG_THIN CC_FLAGS_LTO += -flto=thin -fsplit-lto-unit KBUILD_LDFLAGS += --thinlto-cache-dir=$(extmod-prefix).thinlto-cache else diff --git a/arch/Kconfig b/arch/Kconfig index cdd29b5fdb56..5c22e10e4c12 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -600,6 +600,14 @@ config SHADOW_CALL_STACK config LTO bool + help + Selected if the kernel will be built using the compiler's LTO feature. + +config LTO_CLANG + bool + select LTO + help + Selected if the kernel will be built using Clang's LTO feature. config ARCH_SUPPORTS_LTO_CLANG bool @@ -609,28 +617,25 @@ config ARCH_SUPPORTS_LTO_CLANG - compiling inline assembly with Clang's integrated assembler, - and linking with LLD. -config ARCH_SUPPORTS_THINLTO +config ARCH_SUPPORTS_LTO_CLANG_THIN bool help - An architecture should select this option if it supports Clang's - ThinLTO. + An architecture should select this option if it can supports Clang's + ThinLTO mode. -config THINLTO - bool "Clang ThinLTO" - depends on LTO_CLANG && ARCH_SUPPORTS_THINLTO - default y +config HAS_LTO_CLANG + def_bool y + # Clang >= 11: https://github.com/ClangBuiltLinux/linux/issues/510 + depends on CC_IS_CLANG && CLANG_VERSION >= 110000 && LD_IS_LLD + depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm) + depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm) + depends on ARCH_SUPPORTS_LTO_CLANG + depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT + depends on !KASAN + depends on !GCOV_KERNEL help - This option enables Clang's ThinLTO, which allows for parallel - optimization and faster incremental compiles. More information - can be found from Clang's documentation: - - https://clang.llvm.org/docs/ThinLTO.html - - If you say N here, the compiler will use full LTO, which may - produce faster code, but building the kernel will be significantly - slower as the linker won't efficiently utilize multiple threads. - - If unsure, say Y. + The compiler and Kconfig options support building with Clang's + LTO. choice prompt "Link Time Optimization (LTO)" @@ -644,20 +649,14 @@ choice config LTO_NONE bool "None" + help + Build the kernel normally, without Link Time Optimization (LTO). -config LTO_CLANG - bool "Clang's Link Time Optimization (EXPERIMENTAL)" - # Clang >= 11: https://github.com/ClangBuiltLinux/linux/issues/510 - depends on CC_IS_CLANG && CLANG_VERSION >= 110000 && LD_IS_LLD - depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm) - depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm) - depends on ARCH_SUPPORTS_LTO_CLANG - depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT - depends on !KASAN - depends on !GCOV_KERNEL - select LTO +config LTO_CLANG_FULL + bool "Clang Full LTO (EXPERIMENTAL)" + select LTO_CLANG help - This option enables Clang's Link Time Optimization (LTO), which + This option enables Clang's full Link Time Optimization (LTO), which allows the compiler to optimize the kernel globally. If you enable this option, the compiler generates LLVM bitcode instead of ELF object files, and the actual compilation from bitcode happens at @@ -667,9 +666,22 @@ config LTO_CLANG https://llvm.org/docs/LinkTimeOptimization.html - To select this option, you also need to use LLVM tools to handle - the bitcode by passing LLVM=1 to make. + During link time, this option can use a large amount of RAM, and + may take much longer than the ThinLTO option. +config LTO_CLANG_THIN + bool "Clang ThinLTO (EXPERIMENTAL)" + depends on ARCH_SUPPORTS_LTO_CLANG_THIN + select LTO_CLANG + help + This option enables Clang's ThinLTO, which allows for parallel + optimization and faster incremental compiles compared to the + CONFIG_LTO_CLANG_FULL option. More information can be found + from Clang's documentation: + + https://clang.llvm.org/docs/ThinLTO.html + + If unsure, say Y. endchoice config CFI_CLANG diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig index 8bf763307544..f39df315316e 100644 --- a/arch/arm64/Kconfig +++ b/arch/arm64/Kconfig @@ -74,7 +74,7 @@ config ARM64 select ARCH_SUPPORTS_MEMORY_FAILURE select ARCH_SUPPORTS_SHADOW_CALL_STACK if CC_HAVE_SHADOW_CALL_STACK select ARCH_SUPPORTS_LTO_CLANG - select ARCH_SUPPORTS_THINLTO + select ARCH_SUPPORTS_LTO_CLANG_THIN select ARCH_SUPPORTS_ATOMIC_RMW select ARCH_SUPPORTS_INT128 if CC_HAS_INT128 && (GCC_VERSION >= 50000 || CC_IS_CLANG) select ARCH_SUPPORTS_NUMA_BALANCING diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index cb4c77a9b5ab..f99a4d3b55ae 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -93,7 +93,7 @@ config X86 select ARCH_SUPPORTS_ATOMIC_RMW select ARCH_SUPPORTS_NUMA_BALANCING if X86_64 select ARCH_SUPPORTS_LTO_CLANG if X86_64 - select ARCH_SUPPORTS_THINLTO if X86_64 + select ARCH_SUPPORTS_LTO_CLANG_THIN if X86_64 select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_QUEUED_RWLOCKS select ARCH_USE_QUEUED_SPINLOCKS diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index 3106636375c0..96505113b907 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -161,7 +161,7 @@ static unsigned long kallsyms_sym_address(int idx) return kallsyms_relative_base - 1 - kallsyms_offsets[idx]; } -#if defined(CONFIG_CFI_CLANG) && defined(CONFIG_THINLTO) +#if defined(CONFIG_CFI_CLANG) && defined(CONFIG_LTO_CLANG_THIN) /* * LLVM appends a hash to static function names when ThinLTO and CFI are * both enabled, which causes confusion and potentially breaks user space
On Fri, Nov 20, 2020 at 3:59 PM Kees Cook <keescook@chromium.org> wrote: > > On Fri, Nov 20, 2020 at 12:58:41PM -0800, Sami Tolvanen wrote: > > On Fri, Nov 20, 2020 at 12:43 PM Kees Cook <keescook@chromium.org> wrote: > > > > > > On Fri, Nov 20, 2020 at 01:29:35PM -0700, Nathan Chancellor wrote: > > > > On Fri, Nov 20, 2020 at 11:47:21AM -0800, Kees Cook wrote: > > > > > On Fri, Nov 20, 2020 at 08:23:11AM -0800, Sami Tolvanen wrote: > > > > > > Changing the ThinLTO config to a choice and moving it after the main > > > > > > LTO config sounds like a good idea to me. I'll see if I can change > > > > > > this in v8. Thanks! > > > > > > > > > > Originally, I thought this might be a bit ugly once GCC LTO is added, > > > > > but this could be just a choice like we're done for the stack > > > > > initialization. Something like an "LTO" choice of NONE, CLANG_FULL, > > > > > CLANG_THIN, and in the future GCC, etc. > > > > > > > > Having two separate choices might be a little bit cleaner though? One > > > > for the compiler (LTO_CLANG versus LTO_GCC) and one for the type > > > > (THINLTO versus FULLLTO). The type one could just have a "depends on > > > > CC_IS_CLANG" to ensure it only showed up when needed. > > > > > > Right, that's how the stack init choice works. Kconfigs that aren't > > > supported by the compiler won't be shown. I.e. after Sami's future > > > patch, the only choice for GCC will be CONFIG_LTO_NONE. But building > > > under Clang, it would offer CONFIG_LTO_NONE, CONFIG_LTO_CLANG_FULL, > > > CONFIG_LTO_CLANG_THIN, or something. > > > > > > (and I assume CONFIG_LTO would be def_bool y, depends on !LTO_NONE) > > > > I'm fine with adding ThinLTO as another option to the LTO choice, but > > it would duplicate the dependencies and a lot of the help text. I > > suppose we could add another config for the dependencies and have both > > LTO options depend on that instead. > > How about something like this? This separates the arch support, compiler > support, and user choice into three separate Kconfig areas, which I > think should work. Sure, this looks good to me, I'll use this in v8. The only minor concern I have is that ThinLTO cannot be set as the default LTO mode, but I assume anyone who selects LTO is also capable of deciding which mode is better for them. > diff --git a/Makefile b/Makefile > index e397c4caec1b..af902718e882 100644 > --- a/Makefile > +++ b/Makefile > @@ -897,7 +897,7 @@ export CC_FLAGS_SCS > endif > > ifdef CONFIG_LTO_CLANG > -ifdef CONFIG_THINLTO > +ifdef CONFIG_LTO_CLANG_THIN > CC_FLAGS_LTO += -flto=thin -fsplit-lto-unit > KBUILD_LDFLAGS += --thinlto-cache-dir=$(extmod-prefix).thinlto-cache > else > diff --git a/arch/Kconfig b/arch/Kconfig > index cdd29b5fdb56..5c22e10e4c12 100644 > --- a/arch/Kconfig > +++ b/arch/Kconfig > @@ -600,6 +600,14 @@ config SHADOW_CALL_STACK > > config LTO > bool > + help > + Selected if the kernel will be built using the compiler's LTO feature. > + > +config LTO_CLANG > + bool > + select LTO > + help > + Selected if the kernel will be built using Clang's LTO feature. > > config ARCH_SUPPORTS_LTO_CLANG > bool > @@ -609,28 +617,25 @@ config ARCH_SUPPORTS_LTO_CLANG > - compiling inline assembly with Clang's integrated assembler, > - and linking with LLD. > > -config ARCH_SUPPORTS_THINLTO > +config ARCH_SUPPORTS_LTO_CLANG_THIN > bool > help > - An architecture should select this option if it supports Clang's > - ThinLTO. > + An architecture should select this option if it can supports Clang's > + ThinLTO mode. > > -config THINLTO > - bool "Clang ThinLTO" > - depends on LTO_CLANG && ARCH_SUPPORTS_THINLTO > - default y > +config HAS_LTO_CLANG > + def_bool y > + # Clang >= 11: https://github.com/ClangBuiltLinux/linux/issues/510 > + depends on CC_IS_CLANG && CLANG_VERSION >= 110000 && LD_IS_LLD > + depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm) > + depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm) > + depends on ARCH_SUPPORTS_LTO_CLANG > + depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT > + depends on !KASAN > + depends on !GCOV_KERNEL > help > - This option enables Clang's ThinLTO, which allows for parallel > - optimization and faster incremental compiles. More information > - can be found from Clang's documentation: > - > - https://clang.llvm.org/docs/ThinLTO.html > - > - If you say N here, the compiler will use full LTO, which may > - produce faster code, but building the kernel will be significantly > - slower as the linker won't efficiently utilize multiple threads. > - > - If unsure, say Y. > + The compiler and Kconfig options support building with Clang's > + LTO. > > choice > prompt "Link Time Optimization (LTO)" > @@ -644,20 +649,14 @@ choice > > config LTO_NONE > bool "None" > + help > + Build the kernel normally, without Link Time Optimization (LTO). > > -config LTO_CLANG > - bool "Clang's Link Time Optimization (EXPERIMENTAL)" > - # Clang >= 11: https://github.com/ClangBuiltLinux/linux/issues/510 > - depends on CC_IS_CLANG && CLANG_VERSION >= 110000 && LD_IS_LLD > - depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm) > - depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm) > - depends on ARCH_SUPPORTS_LTO_CLANG > - depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT > - depends on !KASAN > - depends on !GCOV_KERNEL > - select LTO > +config LTO_CLANG_FULL > + bool "Clang Full LTO (EXPERIMENTAL)" > + select LTO_CLANG > help > - This option enables Clang's Link Time Optimization (LTO), which > + This option enables Clang's full Link Time Optimization (LTO), which > allows the compiler to optimize the kernel globally. If you enable > this option, the compiler generates LLVM bitcode instead of ELF > object files, and the actual compilation from bitcode happens at > @@ -667,9 +666,22 @@ config LTO_CLANG > > https://llvm.org/docs/LinkTimeOptimization.html > > - To select this option, you also need to use LLVM tools to handle > - the bitcode by passing LLVM=1 to make. > + During link time, this option can use a large amount of RAM, and > + may take much longer than the ThinLTO option. > > +config LTO_CLANG_THIN > + bool "Clang ThinLTO (EXPERIMENTAL)" > + depends on ARCH_SUPPORTS_LTO_CLANG_THIN > + select LTO_CLANG > + help > + This option enables Clang's ThinLTO, which allows for parallel > + optimization and faster incremental compiles compared to the > + CONFIG_LTO_CLANG_FULL option. More information can be found > + from Clang's documentation: > + > + https://clang.llvm.org/docs/ThinLTO.html > + > + If unsure, say Y. > endchoice The two LTO_CLANG_* options need to depend on HAS_LTO_CLANG, of course. Sami
On Fri, Nov 20, 2020 at 05:46:44PM -0800, Sami Tolvanen wrote: > Sure, this looks good to me, I'll use this in v8. The only minor > concern I have is that ThinLTO cannot be set as the default LTO mode, > but I assume anyone who selects LTO is also capable of deciding which > mode is better for them. It could be re-arranged similar to what you had before, but like: config LTO bool "..." depends on HAS_LTO help ... choice prompt "LTO mode" if LTO default LTO_GCC if HAS_LTO_GCC default LTO_CLANG_THIN if HAS_LTO_CLANG default LTO_CLANG_FULL help ... config LTO_CLANG_THIN ... config LTO_CLANG_FULL endchoice Then the LTO is top-level yes/no, but depends on detected capabilities, and the mode is visible if LTO is chosen, etc. I'm not really sure which is better... > > +config LTO_CLANG_THIN > > + bool "Clang ThinLTO (EXPERIMENTAL)" > > + depends on ARCH_SUPPORTS_LTO_CLANG_THIN > > + select LTO_CLANG > > + help > > + This option enables Clang's ThinLTO, which allows for parallel > > + optimization and faster incremental compiles compared to the > > + CONFIG_LTO_CLANG_FULL option. More information can be found > > + from Clang's documentation: > > + > > + https://clang.llvm.org/docs/ThinLTO.html > > + > > + If unsure, say Y. > > endchoice > > The two LTO_CLANG_* options need to depend on HAS_LTO_CLANG, of course. Whoops, yes. Thanks for catching that. :)
diff --git a/Makefile b/Makefile index 8c8feb4245a6..240560e88d69 100644 --- a/Makefile +++ b/Makefile @@ -893,6 +893,21 @@ KBUILD_CFLAGS += $(CC_FLAGS_SCS) export CC_FLAGS_SCS endif +ifdef CONFIG_LTO_CLANG +ifdef CONFIG_THINLTO +CC_FLAGS_LTO += -flto=thin -fsplit-lto-unit +KBUILD_LDFLAGS += --thinlto-cache-dir=$(extmod-prefix).thinlto-cache +else +CC_FLAGS_LTO += -flto +endif +CC_FLAGS_LTO += -fvisibility=default +endif + +ifdef CONFIG_LTO +KBUILD_CFLAGS += $(CC_FLAGS_LTO) +export CC_FLAGS_LTO +endif + ifdef CONFIG_DEBUG_FORCE_FUNCTION_ALIGN_32B KBUILD_CFLAGS += -falign-functions=32 endif @@ -1473,7 +1488,7 @@ MRPROPER_FILES += include/config include/generated \ *.spec # Directories & files removed with 'make distclean' -DISTCLEAN_FILES += tags TAGS cscope* GPATH GTAGS GRTAGS GSYMS +DISTCLEAN_FILES += tags TAGS cscope* GPATH GTAGS GRTAGS GSYMS .thinlto-cache # clean - Delete most, but leave enough to build external modules # @@ -1719,7 +1734,7 @@ PHONY += compile_commands.json clean-dirs := $(KBUILD_EXTMOD) clean: rm-files := $(KBUILD_EXTMOD)/Module.symvers $(KBUILD_EXTMOD)/modules.nsdeps \ - $(KBUILD_EXTMOD)/compile_commands.json + $(KBUILD_EXTMOD)/compile_commands.json $(KBUILD_EXTMOD)/.thinlto-cache PHONY += help help: diff --git a/arch/Kconfig b/arch/Kconfig index 56b6ccc0e32d..a41fcb3ca7c6 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -598,6 +598,81 @@ config SHADOW_CALL_STACK reading and writing arbitrary memory may be able to locate them and hijack control flow by modifying the stacks. +config LTO + bool + +config ARCH_SUPPORTS_LTO_CLANG + bool + help + An architecture should select this option if it supports: + - compiling with Clang, + - compiling inline assembly with Clang's integrated assembler, + - and linking with LLD. + +config ARCH_SUPPORTS_THINLTO + bool + help + An architecture should select this option if it supports Clang's + ThinLTO. + +config THINLTO + bool "Clang ThinLTO" + depends on LTO_CLANG && ARCH_SUPPORTS_THINLTO + default y + help + This option enables Clang's ThinLTO, which allows for parallel + optimization and faster incremental compiles. More information + can be found from Clang's documentation: + + https://clang.llvm.org/docs/ThinLTO.html + + If you say N here, the compiler will use full LTO, which may + produce faster code, but building the kernel will be significantly + slower as the linker won't efficiently utilize multiple threads. + + If unsure, say Y. + +choice + prompt "Link Time Optimization (LTO)" + default LTO_NONE + help + This option enables Link Time Optimization (LTO), which allows the + compiler to optimize binaries globally. + + If unsure, select LTO_NONE. Note that LTO is very resource-intensive + so it's disabled by default. + +config LTO_NONE + bool "None" + +config LTO_CLANG + bool "Clang's Link Time Optimization (EXPERIMENTAL)" + # Clang >= 11: https://github.com/ClangBuiltLinux/linux/issues/510 + depends on CC_IS_CLANG && CLANG_VERSION >= 110000 && LD_IS_LLD + depends on $(success,$(NM) --help | head -n 1 | grep -qi llvm) + depends on $(success,$(AR) --help | head -n 1 | grep -qi llvm) + depends on ARCH_SUPPORTS_LTO_CLANG + depends on !FTRACE_MCOUNT_USE_RECORDMCOUNT + depends on !KASAN + depends on !GCOV_KERNEL + depends on !MODVERSIONS + select LTO + help + This option enables Clang's Link Time Optimization (LTO), which + allows the compiler to optimize the kernel globally. If you enable + this option, the compiler generates LLVM bitcode instead of ELF + object files, and the actual compilation from bitcode happens at + the LTO link step, which may take several minutes depending on the + kernel configuration. More information can be found from LLVM's + documentation: + + https://llvm.org/docs/LinkTimeOptimization.html + + To select this option, you also need to use LLVM tools to handle + the bitcode by passing LLVM=1 to make. + +endchoice + config HAVE_ARCH_WITHIN_STACK_FRAMES bool help diff --git a/include/asm-generic/vmlinux.lds.h b/include/asm-generic/vmlinux.lds.h index b2b3d81b1535..8988a2e445d8 100644 --- a/include/asm-generic/vmlinux.lds.h +++ b/include/asm-generic/vmlinux.lds.h @@ -90,15 +90,18 @@ * .data. We don't want to pull in .data..other sections, which Linux * has defined. Same for text and bss. * + * With LTO_CLANG, the linker also splits sections by default, so we need + * these macros to combine the sections during the final link. + * * RODATA_MAIN is not used because existing code already defines .rodata.x * sections to be brought in with rodata. */ -#ifdef CONFIG_LD_DEAD_CODE_DATA_ELIMINATION +#if defined(CONFIG_LD_DEAD_CODE_DATA_ELIMINATION) || defined(CONFIG_LTO_CLANG) #define TEXT_MAIN .text .text.[0-9a-zA-Z_]* -#define DATA_MAIN .data .data.[0-9a-zA-Z_]* .data..LPBX* +#define DATA_MAIN .data .data.[0-9a-zA-Z_]* .data..L* .data..compoundliteral* #define SDATA_MAIN .sdata .sdata.[0-9a-zA-Z_]* -#define RODATA_MAIN .rodata .rodata.[0-9a-zA-Z_]* -#define BSS_MAIN .bss .bss.[0-9a-zA-Z_]* +#define RODATA_MAIN .rodata .rodata.[0-9a-zA-Z_]* .rodata..L* +#define BSS_MAIN .bss .bss.[0-9a-zA-Z_]* .bss..compoundliteral* #define SBSS_MAIN .sbss .sbss.[0-9a-zA-Z_]* #else #define TEXT_MAIN .text diff --git a/scripts/Makefile.build b/scripts/Makefile.build index 2175ddb1ee0c..ed74b2f986f7 100644 --- a/scripts/Makefile.build +++ b/scripts/Makefile.build @@ -111,7 +111,7 @@ endif # --------------------------------------------------------------------------- quiet_cmd_cc_s_c = CC $(quiet_modtag) $@ - cmd_cc_s_c = $(CC) $(filter-out $(DEBUG_CFLAGS), $(c_flags)) -fverbose-asm -S -o $@ $< + cmd_cc_s_c = $(CC) $(filter-out $(DEBUG_CFLAGS) $(CC_FLAGS_LTO), $(c_flags)) -fverbose-asm -S -o $@ $< $(obj)/%.s: $(src)/%.c FORCE $(call if_changed_dep,cc_s_c) @@ -425,8 +425,15 @@ $(obj)/lib.a: $(lib-y) FORCE # Do not replace $(filter %.o,^) with $(real-prereqs). When a single object # module is turned into a multi object module, $^ will contain header file # dependencies recorded in the .*.cmd file. +ifdef CONFIG_LTO_CLANG +quiet_cmd_link_multi-m = AR [M] $@ +cmd_link_multi-m = \ + rm -f $@; \ + $(AR) cDPrsT $@ $(filter %.o,$^) +else quiet_cmd_link_multi-m = LD [M] $@ cmd_link_multi-m = $(LD) $(ld_flags) -r -o $@ $(filter %.o,$^) +endif $(multi-used-m): FORCE $(call if_changed,link_multi-m) diff --git a/scripts/Makefile.modfinal b/scripts/Makefile.modfinal index ae01baf96f4e..2cb9a1d88434 100644 --- a/scripts/Makefile.modfinal +++ b/scripts/Makefile.modfinal @@ -6,6 +6,7 @@ PHONY := __modfinal __modfinal: +include $(objtree)/include/config/auto.conf include $(srctree)/scripts/Kbuild.include # for c_flags @@ -29,6 +30,12 @@ quiet_cmd_cc_o_c = CC [M] $@ ARCH_POSTLINK := $(wildcard $(srctree)/arch/$(SRCARCH)/Makefile.postlink) +ifdef CONFIG_LTO_CLANG +# With CONFIG_LTO_CLANG, reuse the object file we compiled for modpost to +# avoid a second slow LTO link +prelink-ext := .lto +endif + quiet_cmd_ld_ko_o = LD [M] $@ cmd_ld_ko_o = \ $(LD) -r $(KBUILD_LDFLAGS) \ @@ -36,7 +43,7 @@ quiet_cmd_ld_ko_o = LD [M] $@ -T scripts/module.lds -o $@ $(filter %.o, $^); \ $(if $(ARCH_POSTLINK), $(MAKE) -f $(ARCH_POSTLINK) $@, true) -$(modules): %.ko: %.o %.mod.o scripts/module.lds FORCE +$(modules): %.ko: %$(prelink-ext).o %.mod.o scripts/module.lds FORCE +$(call if_changed,ld_ko_o) targets += $(modules) $(modules:.ko=.mod.o) diff --git a/scripts/Makefile.modpost b/scripts/Makefile.modpost index f54b6ac37ac2..9ff8bfdb574d 100644 --- a/scripts/Makefile.modpost +++ b/scripts/Makefile.modpost @@ -43,6 +43,9 @@ __modpost: include include/config/auto.conf include scripts/Kbuild.include +# for ld_flags +include scripts/Makefile.lib + MODPOST = scripts/mod/modpost \ $(if $(CONFIG_MODVERSIONS),-m) \ $(if $(CONFIG_MODULE_SRCVERSION_ALL),-a) \ @@ -102,12 +105,26 @@ $(input-symdump): @echo >&2 'WARNING: Symbol version dump "$@" is missing.' @echo >&2 ' Modules may not have dependencies or modversions.' +ifdef CONFIG_LTO_CLANG +# With CONFIG_LTO_CLANG, .o files might be LLVM bitcode, so we need to run +# LTO to compile them into native code before running modpost +prelink-ext := .lto + +quiet_cmd_cc_lto_link_modules = LTO [M] $@ +cmd_cc_lto_link_modules = $(LD) $(ld_flags) -r -o $@ --whole-archive $^ + +%.lto.o: %.o + $(call if_changed,cc_lto_link_modules) +endif + +modules := $(sort $(shell cat $(MODORDER))) + # Read out modules.order to pass in modpost. # Otherwise, allmodconfig would fail with "Argument list too long". quiet_cmd_modpost = MODPOST $@ - cmd_modpost = sed 's/ko$$/o/' $< | $(MODPOST) -T - + cmd_modpost = sed 's/\.ko$$/$(prelink-ext)\.o/' $< | $(MODPOST) -T - -$(output-symdump): $(MODORDER) $(input-symdump) FORCE +$(output-symdump): $(MODORDER) $(input-symdump) $(modules:.ko=$(prelink-ext).o) FORCE $(call if_changed,modpost) targets += $(output-symdump) diff --git a/scripts/link-vmlinux.sh b/scripts/link-vmlinux.sh index 6eded325c837..596507573a48 100755 --- a/scripts/link-vmlinux.sh +++ b/scripts/link-vmlinux.sh @@ -56,6 +56,14 @@ modpost_link() ${KBUILD_VMLINUX_LIBS} \ --end-group" + if [ -n "${CONFIG_LTO_CLANG}" ]; then + # This might take a while, so indicate that we're doing + # an LTO link + info LTO ${1} + else + info LD ${1} + fi + ${LD} ${KBUILD_LDFLAGS} -r -o ${1} ${objects} } @@ -103,13 +111,22 @@ vmlinux_link() fi if [ "${SRCARCH}" != "um" ]; then - objects="--whole-archive \ - ${KBUILD_VMLINUX_OBJS} \ - --no-whole-archive \ - --start-group \ - ${KBUILD_VMLINUX_LIBS} \ - --end-group \ - ${@}" + if [ -n "${CONFIG_LTO_CLANG}" ]; then + # Use vmlinux.o instead of performing the slow LTO + # link again. + objects="--whole-archive \ + vmlinux.o \ + --no-whole-archive \ + ${@}" + else + objects="--whole-archive \ + ${KBUILD_VMLINUX_OBJS} \ + --no-whole-archive \ + --start-group \ + ${KBUILD_VMLINUX_LIBS} \ + --end-group \ + ${@}" + fi ${LD} ${KBUILD_LDFLAGS} ${LDFLAGS_vmlinux} \ ${strip_debug#-Wl,} \ @@ -274,7 +291,6 @@ fi; ${MAKE} -f "${srctree}/scripts/Makefile.build" obj=init need-builtin=1 #link vmlinux.o -info LD vmlinux.o modpost_link vmlinux.o objtool_link vmlinux.o