* [Buildroot] [PATCH v2] toolchain-external: adjust logic for musl dynamic linker symlink
@ 2017-06-24 21:14 Thomas Petazzoni
2017-06-24 21:49 ` Arnout Vandecappelle
0 siblings, 1 reply; 3+ messages in thread
From: Thomas Petazzoni @ 2017-06-24 21:14 UTC (permalink / raw)
To: buildroot
This commit changes the strategy to know what symlink should be
created for the musl dynamic linker. For now, we have in Buildroot the
logic to decide what the symlink name should be. However, despite
numerous fixes, there are still some cases that are missing, such as
mips64el-n32, where the symlink is ld-musl-mipsn32el.so.1.
So, instead of trying to make Buildroot aware of all the possible
names, simply get it from the external toolchain.
Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
---
Changes since v1:
- As suggested by Yann E. Morin, build a sample program, and get its
program interpreter with readelf, in order to find the program
interpreter.
Note: I have tried to refactor this with the check_musl function in
toolchain/helpers.mk, with the idea of introducing a function
get_program_interpreter that would, given CC/READELF would return the
program interpreter path. However, for the life of me, I couldn't get
a make function invoking multiple shell commands to return a string.
---
toolchain/toolchain-external/pkg-toolchain-external.mk | 17 +++--------------
1 file changed, 3 insertions(+), 14 deletions(-)
diff --git a/toolchain/toolchain-external/pkg-toolchain-external.mk b/toolchain/toolchain-external/pkg-toolchain-external.mk
index 8269345..938f73d 100644
--- a/toolchain/toolchain-external/pkg-toolchain-external.mk
+++ b/toolchain/toolchain-external/pkg-toolchain-external.mk
@@ -476,21 +476,10 @@ endef
# of the dynamic library loader. We just need to create a symbolic
# link to libc.so with the appropriate name.
ifeq ($(BR2_TOOLCHAIN_EXTERNAL_MUSL),y)
-ifeq ($(BR2_i386),y)
-MUSL_ARCH = i386
-else ifeq ($(BR2_ARM_EABIHF),y)
-MUSL_ARCH = armhf
-else ifeq ($(BR2_mips):$(BR2_SOFT_FLOAT),y:y)
-MUSL_ARCH = mips-sf
-else ifeq ($(BR2_mipsel):$(BR2_SOFT_FLOAT),y:y)
-MUSL_ARCH = mipsel-sf
-else ifeq ($(BR2_sh),y)
-MUSL_ARCH = sh
-else
-MUSL_ARCH = $(ARCH)
-endif
define TOOLCHAIN_EXTERNAL_MUSL_LD_LINK
- ln -sf libc.so $(TARGET_DIR)/lib/ld-musl-$(MUSL_ARCH).so.1
+ $(Q)echo 'void main(void) {}' | $(TOOLCHAIN_EXTERNAL_CC) -x c -o $(BUILD_DIR)/.br-toolchain-test.tmp - >/dev/null 2>&1; \
+ PROGRAM_INTERPRETER=`$(TOOLCHAIN_EXTERNAL_READELF) -l $(BUILD_DIR)/.br-toolchain-test.tmp 2> /dev/null | sed -r -e '/.*Requesting program interpreter: (.+)\]$$/!d; s//\1/'` ; \
+ ln -sf libc.so $(TARGET_DIR)/$${PROGRAM_INTERPRETER}
endef
endif
--
2.9.4
^ permalink raw reply related [flat|nested] 3+ messages in thread
* [Buildroot] [PATCH v2] toolchain-external: adjust logic for musl dynamic linker symlink
2017-06-24 21:14 [Buildroot] [PATCH v2] toolchain-external: adjust logic for musl dynamic linker symlink Thomas Petazzoni
@ 2017-06-24 21:49 ` Arnout Vandecappelle
2017-06-25 17:15 ` Yann E. MORIN
0 siblings, 1 reply; 3+ messages in thread
From: Arnout Vandecappelle @ 2017-06-24 21:49 UTC (permalink / raw)
To: buildroot
On 24-06-17 23:14, Thomas Petazzoni wrote:
> This commit changes the strategy to know what symlink should be
> created for the musl dynamic linker. For now, we have in Buildroot the
> logic to decide what the symlink name should be. However, despite
> numerous fixes, there are still some cases that are missing, such as
> mips64el-n32, where the symlink is ld-musl-mipsn32el.so.1.
>
> So, instead of trying to make Buildroot aware of all the possible
> names, simply get it from the external toolchain.
>
> Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
With some nits.
> ---
> Changes since v1:
> - As suggested by Yann E. Morin, build a sample program, and get its
> program interpreter with readelf, in order to find the program
> interpreter.
>
> Note: I have tried to refactor this with the check_musl function in
> toolchain/helpers.mk, with the idea of introducing a function
> get_program_interpreter that would, given CC/READELF would return the
> program interpreter path. However, for the life of me, I couldn't get
> a make function invoking multiple shell commands to return a string.
Huh?
get_program_interpreter = $(shell \
echo 'void main(void) {}' | $(TOOLCHAIN_EXTERNAL_CC) -x c -o
$(BUILD_DIR)/.br-toolchain-test.tmp - >/dev/null 2>&1; \
$(TOOLCHAIN_EXTERNAL_READELF) -l $(BUILD_DIR)/.br-toolchain-test.tmp 2>
/dev/null | sed -r -e '/.*Requesting program interpreter: (.+)\]$$/!d; s//\1/')
$(warning $(get_program_interpreter))
toolchain/helpers.mk:496: /lib/ld-musl-i386.so.1
> ---
> toolchain/toolchain-external/pkg-toolchain-external.mk | 17 +++--------------
> 1 file changed, 3 insertions(+), 14 deletions(-)
>
> diff --git a/toolchain/toolchain-external/pkg-toolchain-external.mk b/toolchain/toolchain-external/pkg-toolchain-external.mk
> index 8269345..938f73d 100644
> --- a/toolchain/toolchain-external/pkg-toolchain-external.mk
> +++ b/toolchain/toolchain-external/pkg-toolchain-external.mk
> @@ -476,21 +476,10 @@ endef
> # of the dynamic library loader. We just need to create a symbolic
> # link to libc.so with the appropriate name.
> ifeq ($(BR2_TOOLCHAIN_EXTERNAL_MUSL),y)
> -ifeq ($(BR2_i386),y)
> -MUSL_ARCH = i386
> -else ifeq ($(BR2_ARM_EABIHF),y)
> -MUSL_ARCH = armhf
> -else ifeq ($(BR2_mips):$(BR2_SOFT_FLOAT),y:y)
> -MUSL_ARCH = mips-sf
> -else ifeq ($(BR2_mipsel):$(BR2_SOFT_FLOAT),y:y)
> -MUSL_ARCH = mipsel-sf
> -else ifeq ($(BR2_sh),y)
> -MUSL_ARCH = sh
> -else
> -MUSL_ARCH = $(ARCH)
> -endif
> define TOOLCHAIN_EXTERNAL_MUSL_LD_LINK
> - ln -sf libc.so $(TARGET_DIR)/lib/ld-musl-$(MUSL_ARCH).so.1
> + $(Q)echo 'void main(void) {}' | $(TOOLCHAIN_EXTERNAL_CC) -x c -o $(BUILD_DIR)/.br-toolchain-test.tmp - >/dev/null 2>&1; \
> + PROGRAM_INTERPRETER=`$(TOOLCHAIN_EXTERNAL_READELF) -l $(BUILD_DIR)/.br-toolchain-test.tmp 2> /dev/null | sed -r -e '/.*Requesting program interpreter: (.+)\]$$/!d; s//\1/'` ; \
Perhaps split this ridiculously long line after the | ? Perhaps also split the
two sed expressions with two -e.
I personally prefer sed -n -e '/.../s//\1/p' but I don't care much.
Regards,
Arnout
> + ln -sf libc.so $(TARGET_DIR)/$${PROGRAM_INTERPRETER}
> endef
> endif
>
>
--
Arnout Vandecappelle arnout at mind be
Senior Embedded Software Architect +32-16-286500
Essensium/Mind http://www.mind.be
G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven
LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
GPG fingerprint: 7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Buildroot] [PATCH v2] toolchain-external: adjust logic for musl dynamic linker symlink
2017-06-24 21:49 ` Arnout Vandecappelle
@ 2017-06-25 17:15 ` Yann E. MORIN
0 siblings, 0 replies; 3+ messages in thread
From: Yann E. MORIN @ 2017-06-25 17:15 UTC (permalink / raw)
To: buildroot
Thomas, All,
On 2017-06-24 23:49 +0200, Arnout Vandecappelle spake thusly:
> On 24-06-17 23:14, Thomas Petazzoni wrote:
> > This commit changes the strategy to know what symlink should be
> > created for the musl dynamic linker. For now, we have in Buildroot the
> > logic to decide what the symlink name should be. However, despite
> > numerous fixes, there are still some cases that are missing, such as
> > mips64el-n32, where the symlink is ld-musl-mipsn32el.so.1.
> >
> > So, instead of trying to make Buildroot aware of all the possible
> > names, simply get it from the external toolchain.
> >
> > Signed-off-by: Thomas Petazzoni <thomas.petazzoni@free-electrons.com>
>
> Reviewed-by: Arnout Vandecappelle (Essensium/Mind) <arnout@mind.be>
Not from me, see below (the worst offender being the missing LC_ALL).
> With some nits.
>
> > ---
> > Changes since v1:
> > - As suggested by Yann E. Morin, build a sample program, and get its
> > program interpreter with readelf, in order to find the program
> > interpreter.
> >
> > Note: I have tried to refactor this with the check_musl function in
> > toolchain/helpers.mk, with the idea of introducing a function
> > get_program_interpreter that would, given CC/READELF would return the
> > program interpreter path. However, for the life of me, I couldn't get
> > a make function invoking multiple shell commands to return a string.
>
> Huh?
>
> get_program_interpreter = $(shell \
> echo 'void main(void) {}' | $(TOOLCHAIN_EXTERNAL_CC) -x c -o
> $(BUILD_DIR)/.br-toolchain-test.tmp - >/dev/null 2>&1; \
> $(TOOLCHAIN_EXTERNAL_READELF) -l $(BUILD_DIR)/.br-toolchain-test.tmp 2>
> /dev/null | sed -r -e '/.*Requesting program interpreter: (.+)\]$$/!d; s//\1/')
>
> $(warning $(get_program_interpreter))
>
> toolchain/helpers.mk:496: /lib/ld-musl-i386.so.1
Yup, works for me too...
Although I prefer such long macros be defined rather than assigned.
> > ---
> > toolchain/toolchain-external/pkg-toolchain-external.mk | 17 +++--------------
> > 1 file changed, 3 insertions(+), 14 deletions(-)
> >
> > diff --git a/toolchain/toolchain-external/pkg-toolchain-external.mk b/toolchain/toolchain-external/pkg-toolchain-external.mk
> > index 8269345..938f73d 100644
> > --- a/toolchain/toolchain-external/pkg-toolchain-external.mk
> > +++ b/toolchain/toolchain-external/pkg-toolchain-external.mk
> > @@ -476,21 +476,10 @@ endef
> > # of the dynamic library loader. We just need to create a symbolic
> > # link to libc.so with the appropriate name.
> > ifeq ($(BR2_TOOLCHAIN_EXTERNAL_MUSL),y)
> > -ifeq ($(BR2_i386),y)
> > -MUSL_ARCH = i386
> > -else ifeq ($(BR2_ARM_EABIHF),y)
> > -MUSL_ARCH = armhf
> > -else ifeq ($(BR2_mips):$(BR2_SOFT_FLOAT),y:y)
> > -MUSL_ARCH = mips-sf
> > -else ifeq ($(BR2_mipsel):$(BR2_SOFT_FLOAT),y:y)
> > -MUSL_ARCH = mipsel-sf
> > -else ifeq ($(BR2_sh),y)
> > -MUSL_ARCH = sh
> > -else
> > -MUSL_ARCH = $(ARCH)
> > -endif
> > define TOOLCHAIN_EXTERNAL_MUSL_LD_LINK
> > - ln -sf libc.so $(TARGET_DIR)/lib/ld-musl-$(MUSL_ARCH).so.1
> > + $(Q)echo 'void main(void) {}' | $(TOOLCHAIN_EXTERNAL_CC) -x c -o $(BUILD_DIR)/.br-toolchain-test.tmp - >/dev/null 2>&1; \
> > + PROGRAM_INTERPRETER=`$(TOOLCHAIN_EXTERNAL_READELF) -l $(BUILD_DIR)/.br-toolchain-test.tmp 2> /dev/null | sed -r -e '/.*Requesting program interpreter: (.+)\]$$/!d; s//\1/'` ; \
>
> Perhaps split this ridiculously long line after the | ? Perhaps also split the
> two sed expressions with two -e.
>
> I personally prefer sed -n -e '/.../s//\1/p' but I don't care much.
Also:
- force LC_ALL=C when calling readelf, to avoid any translation,
- remove the generated file.
Regards,
Yann E. MORIN.
> Regards,
> Arnout
>
> > + ln -sf libc.so $(TARGET_DIR)/$${PROGRAM_INTERPRETER}
> > endef
> > endif
> >
> >
>
> --
> Arnout Vandecappelle arnout at mind be
> Senior Embedded Software Architect +32-16-286500
> Essensium/Mind http://www.mind.be
> G.Geenslaan 9, 3001 Leuven, Belgium BE 872 984 063 RPR Leuven
> LinkedIn profile: http://www.linkedin.com/in/arnoutvandecappelle
> GPG fingerprint: 7493 020B C7E3 8618 8DEC 222C 82EB F404 F9AC 0DDF
--
.-----------------.--------------------.------------------.--------------------.
| Yann E. MORIN | Real-Time Embedded | /"\ ASCII RIBBON | Erics' conspiracy: |
| +33 662 376 056 | Software Designer | \ / CAMPAIGN | ___ |
| +33 223 225 172 `------------.-------: X AGAINST | \e/ There is no |
| http://ymorin.is-a-geek.org/ | _/*\_ | / \ HTML MAIL | v conspiracy. |
'------------------------------^-------^------------------^--------------------'
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2017-06-25 17:15 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-06-24 21:14 [Buildroot] [PATCH v2] toolchain-external: adjust logic for musl dynamic linker symlink Thomas Petazzoni
2017-06-24 21:49 ` Arnout Vandecappelle
2017-06-25 17:15 ` Yann E. MORIN
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox