2012-05-05 16:18:41 +08:00
|
|
|
#!/bin/sh
|
|
|
|
#
|
|
|
|
# link vmlinux
|
|
|
|
#
|
|
|
|
# vmlinux is linked from the objects selected by $(KBUILD_VMLINUX_INIT) and
|
|
|
|
# $(KBUILD_VMLINUX_MAIN). Most are built-in.o files from top-level directories
|
|
|
|
# in the kernel tree, others are specified in arch/$(ARCH)/Makefile.
|
|
|
|
# Ordering when linking is important, and $(KBUILD_VMLINUX_INIT) must be first.
|
|
|
|
#
|
|
|
|
# vmlinux
|
|
|
|
# ^
|
|
|
|
# |
|
|
|
|
# +-< $(KBUILD_VMLINUX_INIT)
|
|
|
|
# | +--< init/version.o + more
|
|
|
|
# |
|
|
|
|
# +--< $(KBUILD_VMLINUX_MAIN)
|
|
|
|
# | +--< drivers/built-in.o mm/built-in.o + more
|
|
|
|
# |
|
|
|
|
# +-< ${kallsymso} (see description in KALLSYMS section)
|
|
|
|
#
|
|
|
|
# vmlinux version (uname -v) cannot be updated during normal
|
|
|
|
# descending-into-subdirs phase since we do not yet know if we need to
|
|
|
|
# update vmlinux.
|
|
|
|
# Therefore this step is delayed until just before final link of vmlinux.
|
|
|
|
#
|
|
|
|
# System.map is generated to document addresses of all kernel symbols
|
|
|
|
|
|
|
|
# Error out on error
|
|
|
|
set -e
|
|
|
|
|
|
|
|
# Nice output in kbuild format
|
|
|
|
# Will be supressed by "make -s"
|
|
|
|
info()
|
|
|
|
{
|
|
|
|
if [ "${quiet}" != "silent_" ]; then
|
|
|
|
printf " %-7s %s\n" ${1} ${2}
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
kbuild: allow architectures to use thin archives instead of ld -r
ld -r is an incremental link used to create built-in.o files in build
subdirectories. It produces relocatable object files containing all
its input files, and these are are then pulled together and relocated
in the final link. Aside from the bloat, this constrains the final
link relocations, which has bitten large powerpc builds with
unresolvable relocations in the final link.
Alan Modra has recommended the kernel use thin archives for linking.
This is an alternative and means that the linker has more information
available to it when it links the kernel.
This patch enables a config option architectures can select, which
causes all built-in.o files to be built as thin archives. built-in.o
files in subdirectories do not get symbol table or index attached,
which improves speed and size. The final link pass creates a
built-in.o archive in the root output directory which includes the
symbol table and index. The linker then uses takes this file to link.
The --whole-archive linker option is required, because the linker now
has visibility to every individual object file, and it will otherwise
just completely avoid including those without external references
(consider a file with EXPORT_SYMBOL or initcall or hardware exceptions
as its only entry points). The traditional built works "by luck" as
built-in.o files are large enough that they're going to get external
references. However this optimisation is unpredictable for the kernel
(due to above external references), ineffective at culling unused, and
costly because the .o files have to be searched for references.
Superior alternatives for link-time culling should be used instead.
Build characteristics for inclink vs thinarc, on a small powerpc64le
pseries VM with a modest .config:
inclink thinarc
sizes
vmlinux 15 618 680 15 625 028
sum of all built-in.o 56 091 808 1 054 334
sum excluding root built-in.o 151 430
find -name built-in.o | xargs rm ; time make vmlinux
real 22.772s 21.143s
user 13.280s 13.430s
sys 4.310s 2.750s
- Final kernel pulled in only about 6K more, which shows how
ineffective the object file culling is.
- Build performance looks improved due to less pagecache activity.
On IO constrained systems it could be a bigger win.
- Build size saving is significant.
Side note, the toochain understands archives, so there's some tricks,
$ ar t built-in.o # list all files you linked with
$ size built-in.o # and their sizes
$ objdump -d built-in.o # disassembly (unrelocated) with filenames
Implementation by sfr, minor tweaks by npiggin.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michal Marek <mmarek@suse.com>
2016-08-24 20:29:19 +08:00
|
|
|
# Thin archive build here makes a final archive with
|
|
|
|
# symbol table and indexes from vmlinux objects, which can be
|
|
|
|
# used as input to linker.
|
|
|
|
#
|
|
|
|
# Traditional incremental style of link does not require this step
|
|
|
|
#
|
|
|
|
# built-in.o output file
|
|
|
|
#
|
|
|
|
archive_builtin()
|
|
|
|
{
|
|
|
|
if [ -n "${CONFIG_THIN_ARCHIVES}" ]; then
|
|
|
|
info AR built-in.o
|
|
|
|
rm -f built-in.o;
|
|
|
|
${AR} rcsT${KBUILD_ARFLAGS} built-in.o \
|
|
|
|
${KBUILD_VMLINUX_INIT} \
|
|
|
|
${KBUILD_VMLINUX_MAIN}
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
2012-05-05 16:18:41 +08:00
|
|
|
# Link of vmlinux.o used for section mismatch analysis
|
|
|
|
# ${1} output file
|
|
|
|
modpost_link()
|
|
|
|
{
|
kbuild: allow architectures to use thin archives instead of ld -r
ld -r is an incremental link used to create built-in.o files in build
subdirectories. It produces relocatable object files containing all
its input files, and these are are then pulled together and relocated
in the final link. Aside from the bloat, this constrains the final
link relocations, which has bitten large powerpc builds with
unresolvable relocations in the final link.
Alan Modra has recommended the kernel use thin archives for linking.
This is an alternative and means that the linker has more information
available to it when it links the kernel.
This patch enables a config option architectures can select, which
causes all built-in.o files to be built as thin archives. built-in.o
files in subdirectories do not get symbol table or index attached,
which improves speed and size. The final link pass creates a
built-in.o archive in the root output directory which includes the
symbol table and index. The linker then uses takes this file to link.
The --whole-archive linker option is required, because the linker now
has visibility to every individual object file, and it will otherwise
just completely avoid including those without external references
(consider a file with EXPORT_SYMBOL or initcall or hardware exceptions
as its only entry points). The traditional built works "by luck" as
built-in.o files are large enough that they're going to get external
references. However this optimisation is unpredictable for the kernel
(due to above external references), ineffective at culling unused, and
costly because the .o files have to be searched for references.
Superior alternatives for link-time culling should be used instead.
Build characteristics for inclink vs thinarc, on a small powerpc64le
pseries VM with a modest .config:
inclink thinarc
sizes
vmlinux 15 618 680 15 625 028
sum of all built-in.o 56 091 808 1 054 334
sum excluding root built-in.o 151 430
find -name built-in.o | xargs rm ; time make vmlinux
real 22.772s 21.143s
user 13.280s 13.430s
sys 4.310s 2.750s
- Final kernel pulled in only about 6K more, which shows how
ineffective the object file culling is.
- Build performance looks improved due to less pagecache activity.
On IO constrained systems it could be a bigger win.
- Build size saving is significant.
Side note, the toochain understands archives, so there's some tricks,
$ ar t built-in.o # list all files you linked with
$ size built-in.o # and their sizes
$ objdump -d built-in.o # disassembly (unrelocated) with filenames
Implementation by sfr, minor tweaks by npiggin.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michal Marek <mmarek@suse.com>
2016-08-24 20:29:19 +08:00
|
|
|
local objects
|
|
|
|
|
|
|
|
if [ -n "${CONFIG_THIN_ARCHIVES}" ]; then
|
|
|
|
objects="--whole-archive built-in.o"
|
|
|
|
else
|
|
|
|
objects="${KBUILD_VMLINUX_INIT} \
|
|
|
|
--start-group \
|
|
|
|
${KBUILD_VMLINUX_MAIN} \
|
|
|
|
--end-group"
|
|
|
|
fi
|
|
|
|
${LD} ${LDFLAGS} -r -o ${1} ${objects}
|
2012-05-05 16:18:41 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
# Link of vmlinux
|
|
|
|
# ${1} - optional extra .o files
|
|
|
|
# ${2} - output file
|
|
|
|
vmlinux_link()
|
|
|
|
{
|
|
|
|
local lds="${objtree}/${KBUILD_LDS}"
|
kbuild: allow architectures to use thin archives instead of ld -r
ld -r is an incremental link used to create built-in.o files in build
subdirectories. It produces relocatable object files containing all
its input files, and these are are then pulled together and relocated
in the final link. Aside from the bloat, this constrains the final
link relocations, which has bitten large powerpc builds with
unresolvable relocations in the final link.
Alan Modra has recommended the kernel use thin archives for linking.
This is an alternative and means that the linker has more information
available to it when it links the kernel.
This patch enables a config option architectures can select, which
causes all built-in.o files to be built as thin archives. built-in.o
files in subdirectories do not get symbol table or index attached,
which improves speed and size. The final link pass creates a
built-in.o archive in the root output directory which includes the
symbol table and index. The linker then uses takes this file to link.
The --whole-archive linker option is required, because the linker now
has visibility to every individual object file, and it will otherwise
just completely avoid including those without external references
(consider a file with EXPORT_SYMBOL or initcall or hardware exceptions
as its only entry points). The traditional built works "by luck" as
built-in.o files are large enough that they're going to get external
references. However this optimisation is unpredictable for the kernel
(due to above external references), ineffective at culling unused, and
costly because the .o files have to be searched for references.
Superior alternatives for link-time culling should be used instead.
Build characteristics for inclink vs thinarc, on a small powerpc64le
pseries VM with a modest .config:
inclink thinarc
sizes
vmlinux 15 618 680 15 625 028
sum of all built-in.o 56 091 808 1 054 334
sum excluding root built-in.o 151 430
find -name built-in.o | xargs rm ; time make vmlinux
real 22.772s 21.143s
user 13.280s 13.430s
sys 4.310s 2.750s
- Final kernel pulled in only about 6K more, which shows how
ineffective the object file culling is.
- Build performance looks improved due to less pagecache activity.
On IO constrained systems it could be a bigger win.
- Build size saving is significant.
Side note, the toochain understands archives, so there's some tricks,
$ ar t built-in.o # list all files you linked with
$ size built-in.o # and their sizes
$ objdump -d built-in.o # disassembly (unrelocated) with filenames
Implementation by sfr, minor tweaks by npiggin.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michal Marek <mmarek@suse.com>
2016-08-24 20:29:19 +08:00
|
|
|
local objects
|
2012-05-05 16:18:41 +08:00
|
|
|
|
|
|
|
if [ "${SRCARCH}" != "um" ]; then
|
kbuild: allow architectures to use thin archives instead of ld -r
ld -r is an incremental link used to create built-in.o files in build
subdirectories. It produces relocatable object files containing all
its input files, and these are are then pulled together and relocated
in the final link. Aside from the bloat, this constrains the final
link relocations, which has bitten large powerpc builds with
unresolvable relocations in the final link.
Alan Modra has recommended the kernel use thin archives for linking.
This is an alternative and means that the linker has more information
available to it when it links the kernel.
This patch enables a config option architectures can select, which
causes all built-in.o files to be built as thin archives. built-in.o
files in subdirectories do not get symbol table or index attached,
which improves speed and size. The final link pass creates a
built-in.o archive in the root output directory which includes the
symbol table and index. The linker then uses takes this file to link.
The --whole-archive linker option is required, because the linker now
has visibility to every individual object file, and it will otherwise
just completely avoid including those without external references
(consider a file with EXPORT_SYMBOL or initcall or hardware exceptions
as its only entry points). The traditional built works "by luck" as
built-in.o files are large enough that they're going to get external
references. However this optimisation is unpredictable for the kernel
(due to above external references), ineffective at culling unused, and
costly because the .o files have to be searched for references.
Superior alternatives for link-time culling should be used instead.
Build characteristics for inclink vs thinarc, on a small powerpc64le
pseries VM with a modest .config:
inclink thinarc
sizes
vmlinux 15 618 680 15 625 028
sum of all built-in.o 56 091 808 1 054 334
sum excluding root built-in.o 151 430
find -name built-in.o | xargs rm ; time make vmlinux
real 22.772s 21.143s
user 13.280s 13.430s
sys 4.310s 2.750s
- Final kernel pulled in only about 6K more, which shows how
ineffective the object file culling is.
- Build performance looks improved due to less pagecache activity.
On IO constrained systems it could be a bigger win.
- Build size saving is significant.
Side note, the toochain understands archives, so there's some tricks,
$ ar t built-in.o # list all files you linked with
$ size built-in.o # and their sizes
$ objdump -d built-in.o # disassembly (unrelocated) with filenames
Implementation by sfr, minor tweaks by npiggin.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michal Marek <mmarek@suse.com>
2016-08-24 20:29:19 +08:00
|
|
|
if [ -n "${CONFIG_THIN_ARCHIVES}" ]; then
|
|
|
|
objects="--whole-archive built-in.o ${1}"
|
|
|
|
else
|
|
|
|
objects="${KBUILD_VMLINUX_INIT} \
|
|
|
|
--start-group \
|
|
|
|
${KBUILD_VMLINUX_MAIN} \
|
|
|
|
--end-group \
|
|
|
|
${1}"
|
|
|
|
fi
|
|
|
|
|
|
|
|
${LD} ${LDFLAGS} ${LDFLAGS_vmlinux} -o ${2} \
|
|
|
|
-T ${lds} ${objects}
|
2012-05-05 16:18:41 +08:00
|
|
|
else
|
kbuild: allow architectures to use thin archives instead of ld -r
ld -r is an incremental link used to create built-in.o files in build
subdirectories. It produces relocatable object files containing all
its input files, and these are are then pulled together and relocated
in the final link. Aside from the bloat, this constrains the final
link relocations, which has bitten large powerpc builds with
unresolvable relocations in the final link.
Alan Modra has recommended the kernel use thin archives for linking.
This is an alternative and means that the linker has more information
available to it when it links the kernel.
This patch enables a config option architectures can select, which
causes all built-in.o files to be built as thin archives. built-in.o
files in subdirectories do not get symbol table or index attached,
which improves speed and size. The final link pass creates a
built-in.o archive in the root output directory which includes the
symbol table and index. The linker then uses takes this file to link.
The --whole-archive linker option is required, because the linker now
has visibility to every individual object file, and it will otherwise
just completely avoid including those without external references
(consider a file with EXPORT_SYMBOL or initcall or hardware exceptions
as its only entry points). The traditional built works "by luck" as
built-in.o files are large enough that they're going to get external
references. However this optimisation is unpredictable for the kernel
(due to above external references), ineffective at culling unused, and
costly because the .o files have to be searched for references.
Superior alternatives for link-time culling should be used instead.
Build characteristics for inclink vs thinarc, on a small powerpc64le
pseries VM with a modest .config:
inclink thinarc
sizes
vmlinux 15 618 680 15 625 028
sum of all built-in.o 56 091 808 1 054 334
sum excluding root built-in.o 151 430
find -name built-in.o | xargs rm ; time make vmlinux
real 22.772s 21.143s
user 13.280s 13.430s
sys 4.310s 2.750s
- Final kernel pulled in only about 6K more, which shows how
ineffective the object file culling is.
- Build performance looks improved due to less pagecache activity.
On IO constrained systems it could be a bigger win.
- Build size saving is significant.
Side note, the toochain understands archives, so there's some tricks,
$ ar t built-in.o # list all files you linked with
$ size built-in.o # and their sizes
$ objdump -d built-in.o # disassembly (unrelocated) with filenames
Implementation by sfr, minor tweaks by npiggin.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michal Marek <mmarek@suse.com>
2016-08-24 20:29:19 +08:00
|
|
|
if [ -n "${CONFIG_THIN_ARCHIVES}" ]; then
|
|
|
|
objects="-Wl,--whole-archive built-in.o ${1}"
|
|
|
|
else
|
|
|
|
objects="${KBUILD_VMLINUX_INIT} \
|
|
|
|
-Wl,--start-group \
|
|
|
|
${KBUILD_VMLINUX_MAIN} \
|
|
|
|
-Wl,--end-group \
|
|
|
|
${1}"
|
|
|
|
fi
|
|
|
|
|
|
|
|
${CC} ${CFLAGS_vmlinux} -o ${2} \
|
|
|
|
-Wl,-T,${lds} \
|
|
|
|
${objects} \
|
|
|
|
-lutil -lrt -lpthread
|
2012-05-05 16:18:41 +08:00
|
|
|
rm -f linux
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
|
|
|
|
|
|
|
|
# Create ${2} .o file with all symbols from the ${1} object file
|
|
|
|
kallsyms()
|
|
|
|
{
|
|
|
|
info KSYM ${2}
|
|
|
|
local kallsymopt;
|
|
|
|
|
2013-03-15 12:34:17 +08:00
|
|
|
if [ -n "${CONFIG_HAVE_UNDERSCORE_SYMBOL_PREFIX}" ]; then
|
|
|
|
kallsymopt="${kallsymopt} --symbol-prefix=_"
|
2012-09-07 05:11:25 +08:00
|
|
|
fi
|
|
|
|
|
2012-05-05 16:18:41 +08:00
|
|
|
if [ -n "${CONFIG_KALLSYMS_ALL}" ]; then
|
2012-09-07 05:11:25 +08:00
|
|
|
kallsymopt="${kallsymopt} --all-symbols"
|
2012-05-05 16:18:41 +08:00
|
|
|
fi
|
|
|
|
|
2016-03-16 05:58:12 +08:00
|
|
|
if [ -n "${CONFIG_KALLSYMS_ABSOLUTE_PERCPU}" ]; then
|
2014-03-17 11:35:46 +08:00
|
|
|
kallsymopt="${kallsymopt} --absolute-percpu"
|
|
|
|
fi
|
|
|
|
|
kallsyms: add support for relative offsets in kallsyms address table
Similar to how relative extables are implemented, it is possible to emit
the kallsyms table in such a way that it contains offsets relative to
some anchor point in the kernel image rather than absolute addresses.
On 64-bit architectures, it cuts the size of the kallsyms address table
in half, since offsets between kernel symbols can typically be expressed
in 32 bits. This saves several hundreds of kilobytes of permanent
.rodata on average. In addition, the kallsyms address table is no
longer subject to dynamic relocation when CONFIG_RELOCATABLE is in
effect, so the relocation work done after decompression now doesn't have
to do relocation updates for all these values. This saves up to 24
bytes (i.e., the size of a ELF64 RELA relocation table entry) per value,
which easily adds up to a couple of megabytes of uncompressed __init
data on ppc64 or arm64. Even if these relocation entries typically
compress well, the combined size reduction of 2.8 MB uncompressed for a
ppc64_defconfig build (of which 2.4 MB is __init data) results in a ~500
KB space saving in the compressed image.
Since it is useful for some architectures (like x86) to retain the
ability to emit absolute values as well, this patch also adds support
for capturing both absolute and relative values when
KALLSYMS_ABSOLUTE_PERCPU is in effect, by emitting absolute per-cpu
addresses as positive 32-bit values, and addresses relative to the
lowest encountered relative symbol as negative values, which are
subtracted from the runtime address of this base symbol to produce the
actual address.
Support for the above is enabled by default for all architectures except
IA-64 and Tile-GX, whose symbols are too far apart to capture in this
manner.
Signed-off-by: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Tested-by: Guenter Roeck <linux@roeck-us.net>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Kees Cook <keescook@chromium.org>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Michael Ellerman <mpe@ellerman.id.au>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Michal Marek <mmarek@suse.cz>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2016-03-16 05:58:19 +08:00
|
|
|
if [ -n "${CONFIG_KALLSYMS_BASE_RELATIVE}" ]; then
|
|
|
|
kallsymopt="${kallsymopt} --base-relative"
|
|
|
|
fi
|
|
|
|
|
2012-05-09 01:53:46 +08:00
|
|
|
local aflags="${KBUILD_AFLAGS} ${KBUILD_AFLAGS_KERNEL} \
|
|
|
|
${NOSTDINC_FLAGS} ${LINUXINCLUDE} ${KBUILD_CPPFLAGS}"
|
2012-05-05 16:18:41 +08:00
|
|
|
|
2016-02-05 18:25:05 +08:00
|
|
|
local afile="`basename ${2} .o`.S"
|
|
|
|
|
|
|
|
${NM} -n ${1} | scripts/kallsyms ${kallsymopt} > ${afile}
|
|
|
|
${CC} ${aflags} -c -o ${2} ${afile}
|
2012-05-05 16:18:41 +08:00
|
|
|
}
|
|
|
|
|
|
|
|
# Create map file with all symbols from ${1}
|
|
|
|
# See mksymap for additional details
|
|
|
|
mksysmap()
|
|
|
|
{
|
|
|
|
${CONFIG_SHELL} "${srctree}/scripts/mksysmap" ${1} ${2}
|
|
|
|
}
|
|
|
|
|
2012-05-29 01:32:28 +08:00
|
|
|
sortextable()
|
|
|
|
{
|
|
|
|
${objtree}/scripts/sortextable ${1}
|
|
|
|
}
|
|
|
|
|
2012-05-05 16:18:41 +08:00
|
|
|
# Delete output files in case of error
|
|
|
|
cleanup()
|
|
|
|
{
|
|
|
|
rm -f .old_version
|
|
|
|
rm -f .tmp_System.map
|
|
|
|
rm -f .tmp_kallsyms*
|
|
|
|
rm -f .tmp_version
|
|
|
|
rm -f .tmp_vmlinux*
|
kbuild: allow architectures to use thin archives instead of ld -r
ld -r is an incremental link used to create built-in.o files in build
subdirectories. It produces relocatable object files containing all
its input files, and these are are then pulled together and relocated
in the final link. Aside from the bloat, this constrains the final
link relocations, which has bitten large powerpc builds with
unresolvable relocations in the final link.
Alan Modra has recommended the kernel use thin archives for linking.
This is an alternative and means that the linker has more information
available to it when it links the kernel.
This patch enables a config option architectures can select, which
causes all built-in.o files to be built as thin archives. built-in.o
files in subdirectories do not get symbol table or index attached,
which improves speed and size. The final link pass creates a
built-in.o archive in the root output directory which includes the
symbol table and index. The linker then uses takes this file to link.
The --whole-archive linker option is required, because the linker now
has visibility to every individual object file, and it will otherwise
just completely avoid including those without external references
(consider a file with EXPORT_SYMBOL or initcall or hardware exceptions
as its only entry points). The traditional built works "by luck" as
built-in.o files are large enough that they're going to get external
references. However this optimisation is unpredictable for the kernel
(due to above external references), ineffective at culling unused, and
costly because the .o files have to be searched for references.
Superior alternatives for link-time culling should be used instead.
Build characteristics for inclink vs thinarc, on a small powerpc64le
pseries VM with a modest .config:
inclink thinarc
sizes
vmlinux 15 618 680 15 625 028
sum of all built-in.o 56 091 808 1 054 334
sum excluding root built-in.o 151 430
find -name built-in.o | xargs rm ; time make vmlinux
real 22.772s 21.143s
user 13.280s 13.430s
sys 4.310s 2.750s
- Final kernel pulled in only about 6K more, which shows how
ineffective the object file culling is.
- Build performance looks improved due to less pagecache activity.
On IO constrained systems it could be a bigger win.
- Build size saving is significant.
Side note, the toochain understands archives, so there's some tricks,
$ ar t built-in.o # list all files you linked with
$ size built-in.o # and their sizes
$ objdump -d built-in.o # disassembly (unrelocated) with filenames
Implementation by sfr, minor tweaks by npiggin.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michal Marek <mmarek@suse.com>
2016-08-24 20:29:19 +08:00
|
|
|
rm -f built-in.o
|
2012-05-05 16:18:41 +08:00
|
|
|
rm -f System.map
|
|
|
|
rm -f vmlinux
|
|
|
|
rm -f vmlinux.o
|
|
|
|
}
|
|
|
|
|
2015-05-07 08:36:04 +08:00
|
|
|
on_exit()
|
|
|
|
{
|
|
|
|
if [ $? -ne 0 ]; then
|
|
|
|
cleanup
|
|
|
|
fi
|
|
|
|
}
|
|
|
|
trap on_exit EXIT
|
|
|
|
|
|
|
|
on_signals()
|
|
|
|
{
|
|
|
|
exit 1
|
|
|
|
}
|
|
|
|
trap on_signals HUP INT QUIT TERM
|
|
|
|
|
2012-05-05 16:18:41 +08:00
|
|
|
#
|
|
|
|
#
|
|
|
|
# Use "make V=1" to debug this script
|
|
|
|
case "${KBUILD_VERBOSE}" in
|
|
|
|
*1*)
|
|
|
|
set -x
|
|
|
|
;;
|
|
|
|
esac
|
|
|
|
|
|
|
|
if [ "$1" = "clean" ]; then
|
|
|
|
cleanup
|
|
|
|
exit 0
|
|
|
|
fi
|
|
|
|
|
|
|
|
# We need access to CONFIG_ symbols
|
2013-02-25 20:47:53 +08:00
|
|
|
case "${KCONFIG_CONFIG}" in
|
|
|
|
*/*)
|
|
|
|
. "${KCONFIG_CONFIG}"
|
|
|
|
;;
|
|
|
|
*)
|
|
|
|
# Force using a file from the current directory
|
|
|
|
. "./${KCONFIG_CONFIG}"
|
|
|
|
esac
|
2012-05-05 16:18:41 +08:00
|
|
|
|
kbuild: allow architectures to use thin archives instead of ld -r
ld -r is an incremental link used to create built-in.o files in build
subdirectories. It produces relocatable object files containing all
its input files, and these are are then pulled together and relocated
in the final link. Aside from the bloat, this constrains the final
link relocations, which has bitten large powerpc builds with
unresolvable relocations in the final link.
Alan Modra has recommended the kernel use thin archives for linking.
This is an alternative and means that the linker has more information
available to it when it links the kernel.
This patch enables a config option architectures can select, which
causes all built-in.o files to be built as thin archives. built-in.o
files in subdirectories do not get symbol table or index attached,
which improves speed and size. The final link pass creates a
built-in.o archive in the root output directory which includes the
symbol table and index. The linker then uses takes this file to link.
The --whole-archive linker option is required, because the linker now
has visibility to every individual object file, and it will otherwise
just completely avoid including those without external references
(consider a file with EXPORT_SYMBOL or initcall or hardware exceptions
as its only entry points). The traditional built works "by luck" as
built-in.o files are large enough that they're going to get external
references. However this optimisation is unpredictable for the kernel
(due to above external references), ineffective at culling unused, and
costly because the .o files have to be searched for references.
Superior alternatives for link-time culling should be used instead.
Build characteristics for inclink vs thinarc, on a small powerpc64le
pseries VM with a modest .config:
inclink thinarc
sizes
vmlinux 15 618 680 15 625 028
sum of all built-in.o 56 091 808 1 054 334
sum excluding root built-in.o 151 430
find -name built-in.o | xargs rm ; time make vmlinux
real 22.772s 21.143s
user 13.280s 13.430s
sys 4.310s 2.750s
- Final kernel pulled in only about 6K more, which shows how
ineffective the object file culling is.
- Build performance looks improved due to less pagecache activity.
On IO constrained systems it could be a bigger win.
- Build size saving is significant.
Side note, the toochain understands archives, so there's some tricks,
$ ar t built-in.o # list all files you linked with
$ size built-in.o # and their sizes
$ objdump -d built-in.o # disassembly (unrelocated) with filenames
Implementation by sfr, minor tweaks by npiggin.
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michal Marek <mmarek@suse.com>
2016-08-24 20:29:19 +08:00
|
|
|
archive_builtin
|
|
|
|
|
2012-05-05 16:18:41 +08:00
|
|
|
#link vmlinux.o
|
|
|
|
info LD vmlinux.o
|
|
|
|
modpost_link vmlinux.o
|
|
|
|
|
|
|
|
# modpost vmlinux.o to check for section mismatches
|
|
|
|
${MAKE} -f "${srctree}/scripts/Makefile.modpost" vmlinux.o
|
|
|
|
|
|
|
|
# Update version
|
|
|
|
info GEN .version
|
|
|
|
if [ ! -r .version ]; then
|
|
|
|
rm -f .version;
|
|
|
|
echo 1 >.version;
|
|
|
|
else
|
|
|
|
mv .version .old_version;
|
|
|
|
expr 0$(cat .old_version) + 1 >.version;
|
|
|
|
fi;
|
|
|
|
|
|
|
|
# final build of init/
|
2016-05-24 06:09:38 +08:00
|
|
|
${MAKE} -f "${srctree}/scripts/Makefile.build" obj=init GCC_PLUGINS_CFLAGS="${GCC_PLUGINS_CFLAGS}"
|
2012-05-05 16:18:41 +08:00
|
|
|
|
|
|
|
kallsymso=""
|
|
|
|
kallsyms_vmlinux=""
|
|
|
|
if [ -n "${CONFIG_KALLSYMS}" ]; then
|
|
|
|
|
|
|
|
# kallsyms support
|
|
|
|
# Generate section listing all symbols and add it into vmlinux
|
|
|
|
# It's a three step process:
|
|
|
|
# 1) Link .tmp_vmlinux1 so it has all symbols and sections,
|
|
|
|
# but __kallsyms is empty.
|
|
|
|
# Running kallsyms on that gives us .tmp_kallsyms1.o with
|
|
|
|
# the right size
|
|
|
|
# 2) Link .tmp_vmlinux2 so it now has a __kallsyms section of
|
|
|
|
# the right size, but due to the added section, some
|
|
|
|
# addresses have shifted.
|
|
|
|
# From here, we generate a correct .tmp_kallsyms2.o
|
2016-11-24 00:41:37 +08:00
|
|
|
# 3) That link may have expanded the kernel image enough that
|
|
|
|
# more linker branch stubs / trampolines had to be added, which
|
|
|
|
# introduces new names, which further expands kallsyms. Do another
|
|
|
|
# pass if that is the case. In theory it's possible this results
|
|
|
|
# in even more stubs, but unlikely.
|
|
|
|
# KALLSYMS_EXTRA_PASS=1 may also used to debug or work around
|
|
|
|
# other bugs.
|
|
|
|
# 4) The correct ${kallsymso} is linked into the final vmlinux.
|
2012-05-05 16:18:41 +08:00
|
|
|
#
|
|
|
|
# a) Verify that the System.map from vmlinux matches the map from
|
|
|
|
# ${kallsymso}.
|
|
|
|
|
|
|
|
kallsymso=.tmp_kallsyms2.o
|
|
|
|
kallsyms_vmlinux=.tmp_vmlinux2
|
|
|
|
|
|
|
|
# step 1
|
|
|
|
vmlinux_link "" .tmp_vmlinux1
|
|
|
|
kallsyms .tmp_vmlinux1 .tmp_kallsyms1.o
|
|
|
|
|
|
|
|
# step 2
|
|
|
|
vmlinux_link .tmp_kallsyms1.o .tmp_vmlinux2
|
|
|
|
kallsyms .tmp_vmlinux2 .tmp_kallsyms2.o
|
|
|
|
|
2016-11-24 00:41:37 +08:00
|
|
|
# step 3
|
|
|
|
size1=$(stat -c "%s" .tmp_kallsyms1.o)
|
|
|
|
size2=$(stat -c "%s" .tmp_kallsyms2.o)
|
|
|
|
|
|
|
|
if [ $size1 -ne $size2 ] || [ -n "${KALLSYMS_EXTRA_PASS}" ]; then
|
2012-05-05 16:18:41 +08:00
|
|
|
kallsymso=.tmp_kallsyms3.o
|
|
|
|
kallsyms_vmlinux=.tmp_vmlinux3
|
|
|
|
|
|
|
|
vmlinux_link .tmp_kallsyms2.o .tmp_vmlinux3
|
|
|
|
|
|
|
|
kallsyms .tmp_vmlinux3 .tmp_kallsyms3.o
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
|
|
|
info LD vmlinux
|
|
|
|
vmlinux_link "${kallsymso}" vmlinux
|
|
|
|
|
2012-05-29 01:32:28 +08:00
|
|
|
if [ -n "${CONFIG_BUILDTIME_EXTABLE_SORT}" ]; then
|
|
|
|
info SORTEX vmlinux
|
|
|
|
sortextable vmlinux
|
|
|
|
fi
|
|
|
|
|
2012-05-05 16:18:41 +08:00
|
|
|
info SYSMAP System.map
|
|
|
|
mksysmap vmlinux System.map
|
|
|
|
|
|
|
|
# step a (see comment above)
|
|
|
|
if [ -n "${CONFIG_KALLSYMS}" ]; then
|
|
|
|
mksysmap ${kallsyms_vmlinux} .tmp_System.map
|
|
|
|
|
|
|
|
if ! cmp -s System.map .tmp_System.map; then
|
2012-07-08 05:04:40 +08:00
|
|
|
echo >&2 Inconsistent kallsyms data
|
2012-08-10 17:55:11 +08:00
|
|
|
echo >&2 Try "make KALLSYMS_EXTRA_PASS=1" as a workaround
|
2012-05-05 16:18:41 +08:00
|
|
|
exit 1
|
|
|
|
fi
|
|
|
|
fi
|
|
|
|
|
|
|
|
# We made a new kernel - delete old version file
|
|
|
|
rm -f .old_version
|