| Message ID | 20191116052815.nop3xkmd4umqsdsb@google.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | configure: Use lld --image-base for --disable-pie user mode binaries | expand |
On 2019-11-15, Fangrui Song wrote: >For lld, --image-base is the preferred way to set the base address. >lld does not actually implement -Ttext-segment, but treats it as an alias for >-Ttext. -Ttext-segment=0x60000000 combined with --no-rosegment can >create a 1.6GB executable. > >Fix the problem by using --image-base for lld. GNU ld and gold will >still get -Ttext-segment. Also delete the ld --verbose fallback introduced >in 2013, which is no longer relevant or correct (the default linker >script has changed). > >Signed-off-by: Fangrui Song <i@maskray.me> >--- > configure | 33 ++++++++++++--------------------- > 1 file changed, 12 insertions(+), 21 deletions(-) > >diff --git a/configure b/configure >index 6099be1d84..2d45af0d09 100755 >--- a/configure >+++ b/configure >@@ -6336,43 +6336,34 @@ fi > > # Probe for the need for relocating the user-only binary. > if ( [ "$linux_user" = yes ] || [ "$bsd_user" = yes ] ) && [ "$pie" = no ]; then >- textseg_addr= >+ image_base= > case "$cpu" in > arm | i386 | ppc* | s390* | sparc* | x86_64 | x32) >- # ??? Rationale for choosing this address >- textseg_addr=0x60000000 >+ # An arbitrary address that makes it unlikely to collide with user >+ # programs. >+ image_base=0x60000000 > ;; > mips) > # A 256M aligned address, high in the address space, with enough > # room for the code_gen_buffer above it before the stack. >- textseg_addr=0x60000000 >+ image_base=0x60000000 > ;; > esac >- if [ -n "$textseg_addr" ]; then >+ if [ -n "$image_base" ]; then > cat > $TMPC <<EOF > int main(void) { return 0; } > EOF >- textseg_ldflags="-Wl,-Ttext-segment=$textseg_addr" >- if ! compile_prog "" "$textseg_ldflags"; then >- # In case ld does not support -Ttext-segment, edit the default linker >- # script via sed to set the .text start addr. This is needed on FreeBSD >- # at least. >- if ! $ld --verbose >/dev/null 2>&1; then >+ image_base_ldflags="-Wl,--image-base=$image_base" >+ if ! compile_prog "" "$image_base_ldflags"; then >+ image_base_ldflags="-Wl,-Ttext-segment=$image_base" >+ if ! compile_prog "" "$image_base_ldflags"; then > error_exit \ > "We need to link the QEMU user mode binaries at a" \ > "specific text address. Unfortunately your linker" \ >- "doesn't support either the -Ttext-segment option or" \ >- "printing the default linker script with --verbose." \ >+ "supports neither --image-base nor -Ttext-segment. " \ > "If you don't want the user mode binaries, pass the" \ > "--disable-user option to configure." > fi >- >- $ld --verbose | sed \ >- -e '1,/==================================================/d' \ >- -e '/==================================================/,$d' \ >- -e "s/[.] = [0-9a-fx]* [+] SIZEOF_HEADERS/. = $textseg_addr + SIZEOF_HEADERS/" \ >- -e "s/__executable_start = [0-9a-fx]*/__executable_start = $textseg_addr/" > config-host.ld >- textseg_ldflags="-Wl,-T../config-host.ld" > fi > fi > fi >@@ -7945,7 +7936,7 @@ if test "$gprof" = "yes" ; then > fi > > if test "$target_linux_user" = "yes" || test "$target_bsd_user" = "yes" ; then >- ldflags="$ldflags $textseg_ldflags" >+ ldflags="$ldflags $image_base_ldflags" > fi > > # Newer kernels on s390 check for an S390_PGSTE program header and >-- >2.24.0 > Ping :)
On 2019-11-20, Fangrui Song wrote: >On 2019-11-15, Fangrui Song wrote: >>For lld, --image-base is the preferred way to set the base address. >>lld does not actually implement -Ttext-segment, but treats it as an alias for >>-Ttext. -Ttext-segment=0x60000000 combined with --no-rosegment can >>create a 1.6GB executable. >> >>Fix the problem by using --image-base for lld. GNU ld and gold will >>still get -Ttext-segment. Also delete the ld --verbose fallback introduced >>in 2013, which is no longer relevant or correct (the default linker >>script has changed). >> >>Signed-off-by: Fangrui Song <i@maskray.me> >>--- >> configure | 33 ++++++++++++--------------------- >> 1 file changed, 12 insertions(+), 21 deletions(-) >> >>diff --git a/configure b/configure >>index 6099be1d84..2d45af0d09 100755 >>--- a/configure >>+++ b/configure >>@@ -6336,43 +6336,34 @@ fi >> >> # Probe for the need for relocating the user-only binary. >> if ( [ "$linux_user" = yes ] || [ "$bsd_user" = yes ] ) && [ "$pie" = no ]; then >>- textseg_addr= >>+ image_base= >> case "$cpu" in >> arm | i386 | ppc* | s390* | sparc* | x86_64 | x32) >>- # ??? Rationale for choosing this address >>- textseg_addr=0x60000000 >>+ # An arbitrary address that makes it unlikely to collide with user >>+ # programs. >>+ image_base=0x60000000 >> ;; >> mips) >> # A 256M aligned address, high in the address space, with enough >> # room for the code_gen_buffer above it before the stack. >>- textseg_addr=0x60000000 >>+ image_base=0x60000000 >> ;; >> esac >>- if [ -n "$textseg_addr" ]; then >>+ if [ -n "$image_base" ]; then >> cat > $TMPC <<EOF >> int main(void) { return 0; } >> EOF >>- textseg_ldflags="-Wl,-Ttext-segment=$textseg_addr" >>- if ! compile_prog "" "$textseg_ldflags"; then >>- # In case ld does not support -Ttext-segment, edit the default linker >>- # script via sed to set the .text start addr. This is needed on FreeBSD >>- # at least. >>- if ! $ld --verbose >/dev/null 2>&1; then >>+ image_base_ldflags="-Wl,--image-base=$image_base" >>+ if ! compile_prog "" "$image_base_ldflags"; then >>+ image_base_ldflags="-Wl,-Ttext-segment=$image_base" >>+ if ! compile_prog "" "$image_base_ldflags"; then >> error_exit \ >> "We need to link the QEMU user mode binaries at a" \ >> "specific text address. Unfortunately your linker" \ >>- "doesn't support either the -Ttext-segment option or" \ >>- "printing the default linker script with --verbose." \ >>+ "supports neither --image-base nor -Ttext-segment. " \ >> "If you don't want the user mode binaries, pass the" \ >> "--disable-user option to configure." >> fi >>- >>- $ld --verbose | sed \ >>- -e '1,/==================================================/d' \ >>- -e '/==================================================/,$d' \ >>- -e "s/[.] = [0-9a-fx]* [+] SIZEOF_HEADERS/. = $textseg_addr + SIZEOF_HEADERS/" \ >>- -e "s/__executable_start = [0-9a-fx]*/__executable_start = $textseg_addr/" > config-host.ld >>- textseg_ldflags="-Wl,-T../config-host.ld" >> fi >> fi >> fi >>@@ -7945,7 +7936,7 @@ if test "$gprof" = "yes" ; then >> fi >> >> if test "$target_linux_user" = "yes" || test "$target_bsd_user" = "yes" ; then >>- ldflags="$ldflags $textseg_ldflags" >>+ ldflags="$ldflags $image_base_ldflags" >> fi >> >> # Newer kernels on s390 check for an S390_PGSTE program header and >>-- >>2.24.0 >> > >Ping :) Ping :)
Fangrui Song <i@maskray.me> writes: > For lld, --image-base is the preferred way to set the base address. > lld does not actually implement -Ttext-segment, but treats it as an alias for > -Ttext. -Ttext-segment=0x60000000 combined with --no-rosegment can > create a 1.6GB executable. > > Fix the problem by using --image-base for lld. GNU ld and gold will > still get -Ttext-segment. Also delete the ld --verbose fallback introduced > in 2013, which is no longer relevant or correct (the default linker > script has changed). > > Signed-off-by: Fangrui Song <i@maskray.me> This patch no longer applies cleanly to configure so I couldn't test it. > --- > configure | 33 ++++++++++++--------------------- > 1 file changed, 12 insertions(+), 21 deletions(-) > > diff --git a/configure b/configure > index 6099be1d84..2d45af0d09 100755 > --- a/configure > +++ b/configure > @@ -6336,43 +6336,34 @@ fi > > # Probe for the need for relocating the user-only binary. > if ( [ "$linux_user" = yes ] || [ "$bsd_user" = yes ] ) && [ "$pie" = no ]; then > - textseg_addr= > + image_base= > case "$cpu" in > arm | i386 | ppc* | s390* | sparc* | x86_64 | x32) > - # ??? Rationale for choosing this address > - textseg_addr=0x60000000 > + # An arbitrary address that makes it unlikely to collide with user > + # programs. > + image_base=0x60000000 The comment probably belongs up above when we define the empty variable unless it really is specifically about these targets. Renaming textseg_addr seems like unnecessary churn for this patch. > ;; > mips) > # A 256M aligned address, high in the address space, with enough > # room for the code_gen_buffer above it before the stack. > - textseg_addr=0x60000000 > + image_base=0x60000000 > ;; > esac > - if [ -n "$textseg_addr" ]; then > + if [ -n "$image_base" ]; then > cat > $TMPC <<EOF > int main(void) { return 0; } > EOF > - textseg_ldflags="-Wl,-Ttext-segment=$textseg_addr" > - if ! compile_prog "" "$textseg_ldflags"; then > - # In case ld does not support -Ttext-segment, edit the default linker > - # script via sed to set the .text start addr. This is needed on FreeBSD > - # at least. > - if ! $ld --verbose >/dev/null 2>&1; then > + image_base_ldflags="-Wl,--image-base=$image_base" > + if ! compile_prog "" "$image_base_ldflags"; then > + image_base_ldflags="-Wl,-Ttext-segment=$image_base" > + if ! compile_prog "" "$image_base_ldflags"; then > error_exit \ > "We need to link the QEMU user mode binaries at a" \ > "specific text address. Unfortunately your linker" \ > - "doesn't support either the -Ttext-segment option or" \ > - "printing the default linker script with --verbose." \ > + "supports neither --image-base nor -Ttext-segment. " \ > "If you don't want the user mode binaries, pass the" \ > "--disable-user option to configure." > fi > - > - $ld --verbose | sed \ > - -e '1,/==================================================/d' \ > - -e '/==================================================/,$d' \ > - -e "s/[.] = [0-9a-fx]* [+] SIZEOF_HEADERS/. = $textseg_addr + SIZEOF_HEADERS/" \ > - -e "s/__executable_start = [0-9a-fx]*/__executable_start = $textseg_addr/" > config-host.ld > - textseg_ldflags="-Wl,-T../config-host.ld" > fi > fi > fi > @@ -7945,7 +7936,7 @@ if test "$gprof" = "yes" ; then > fi > > if test "$target_linux_user" = "yes" || test "$target_bsd_user" = "yes" ; then > - ldflags="$ldflags $textseg_ldflags" > + ldflags="$ldflags $image_base_ldflags" > fi > > # Newer kernels on s390 check for an S390_PGSTE program header and
On 11/27/19 6:36 PM, Fangrui Song wrote: > On 2019-11-20, Fangrui Song wrote: >> On 2019-11-15, Fangrui Song wrote: >>> For lld, --image-base is the preferred way to set the base address. >>> lld does not actually implement -Ttext-segment, but treats it as an alias for >>> -Ttext. -Ttext-segment=0x60000000 combined with --no-rosegment can >>> create a 1.6GB executable. >>> >>> Fix the problem by using --image-base for lld. GNU ld and gold will >>> still get -Ttext-segment. Also delete the ld --verbose fallback introduced >>> in 2013, which is no longer relevant or correct (the default linker >>> script has changed). >>> >>> Signed-off-by: Fangrui Song <i@maskray.me> >>> --- >>> configure | 33 ++++++++++++--------------------- >>> 1 file changed, 12 insertions(+), 21 deletions(-) >>> >>> diff --git a/configure b/configure >>> index 6099be1d84..2d45af0d09 100755 >>> --- a/configure >>> +++ b/configure >>> @@ -6336,43 +6336,34 @@ fi >>> >>> # Probe for the need for relocating the user-only binary. >>> if ( [ "$linux_user" = yes ] || [ "$bsd_user" = yes ] ) && [ "$pie" = no ]; >>> then >>> - textseg_addr= >>> + image_base= >>> case "$cpu" in >>> arm | i386 | ppc* | s390* | sparc* | x86_64 | x32) >>> - # ??? Rationale for choosing this address >>> - textseg_addr=0x60000000 >>> + # An arbitrary address that makes it unlikely to collide with user >>> + # programs. Please don't replace this ??? with an arbitrary rationale, which clearly doesn't apply to all of these hosts. >>> + image_base=0x60000000 >>> ;; >>> mips) >>> # A 256M aligned address, high in the address space, with enough >>> # room for the code_gen_buffer above it before the stack. This is the only one with a proper rationale. That said, I'm not sure that the proper way to handle this issue with lld is to drop this code entirely. The best way to handle the underlying issue -- address conflict between interpreter and guest binary -- is PIE, for which this code is skipped. After that, we go to some pain to choose a guest_base address that allows the guest binary to load around the interpreter's reserved addresses. So what's left that this messing about with link addresses buys us? r~
Thanks for reviewing this patch! On 2019-12-01, Richard Henderson wrote: >On 11/27/19 6:36 PM, Fangrui Song wrote: >> On 2019-11-20, Fangrui Song wrote: >>> On 2019-11-15, Fangrui Song wrote: >>>> For lld, --image-base is the preferred way to set the base address. >>>> lld does not actually implement -Ttext-segment, but treats it as an alias for >>>> -Ttext. -Ttext-segment=0x60000000 combined with --no-rosegment can >>>> create a 1.6GB executable. >>>> >>>> Fix the problem by using --image-base for lld. GNU ld and gold will >>>> still get -Ttext-segment. Also delete the ld --verbose fallback introduced >>>> in 2013, which is no longer relevant or correct (the default linker >>>> script has changed). >>>> >>>> Signed-off-by: Fangrui Song <i@maskray.me> >>>> --- >>>> configure | 33 ++++++++++++--------------------- >>>> 1 file changed, 12 insertions(+), 21 deletions(-) >>>> >>>> diff --git a/configure b/configure >>>> index 6099be1d84..2d45af0d09 100755 >>>> --- a/configure >>>> +++ b/configure >>>> @@ -6336,43 +6336,34 @@ fi >>>> >>>> # Probe for the need for relocating the user-only binary. >>>> if ( [ "$linux_user" = yes ] || [ "$bsd_user" = yes ] ) && [ "$pie" = no ]; >>>> then >>>> - textseg_addr= >>>> + image_base= >>>> case "$cpu" in >>>> arm | i386 | ppc* | s390* | sparc* | x86_64 | x32) >>>> - # ??? Rationale for choosing this address >>>> - textseg_addr=0x60000000 >>>> + # An arbitrary address that makes it unlikely to collide with user >>>> + # programs. > >Please don't replace this ??? with an arbitrary rationale, which clearly >doesn't apply to all of these hosts. In https://lists.nongnu.org/archive/html/qemu-devel/2019-11/msg04646.html it was suggested to move the comment around a bit. I am not puzzled where and what I should say in the comment. Can you (or other maintainers) kindly edit the comment for me? I do not know enough about qemu to provide a good rationale here. >>>> + image_base=0x60000000 >>>> ;; >>>> mips) >>>> # A 256M aligned address, high in the address space, with enough >>>> # room for the code_gen_buffer above it before the stack. > >This is the only one with a proper rationale. > >That said, I'm not sure that the proper way to handle this issue with lld is to >drop this code entirely. The patch changes a feature that lld does not support: -Ttext-segment, to use --image-base instead. Due to the prevalence of -z separate-code in GNU ld, -Ttext-segment is no longer appropriate. I suggested that GNU linkers implement the feature https://sourceware.org/bugzilla/show_bug.cgi?id=25207 . What gets deleted is the sed script. As I explained in the commit message, it is no longer relevant. It probably applies to an old GNU ld that FreeBSD used. FreeBSD has switched to lld now. >The best way to handle the underlying issue -- address conflict between >interpreter and guest binary -- is PIE, for which this code is skipped. > >After that, we go to some pain to choose a guest_base address that allows the >guest binary to load around the interpreter's reserved addresses. > >So what's left that this messing about with link addresses buys us? I agree that --enable-pie will be a better solution, but dropping the support now will break at least FreeBSD. Its kernel supports running an ET_DYN executable but it does not perform address randomization. --disable-pie also appears to be used by ChromeOS developers who reported https://bugs.llvm.org/show_bug.cgi?id=43997 . I can communicate to them that migrating to --enable-pie is the way going forward.
On 2019-12-01, Fangrui Song wrote: > >Thanks for reviewing this patch! > >On 2019-12-01, Richard Henderson wrote: >>On 11/27/19 6:36 PM, Fangrui Song wrote: >>>On 2019-11-20, Fangrui Song wrote: >>>>On 2019-11-15, Fangrui Song wrote: >>>>>For lld, --image-base is the preferred way to set the base address. >>>>>lld does not actually implement -Ttext-segment, but treats it as an alias for >>>>>-Ttext. -Ttext-segment=0x60000000 combined with --no-rosegment can >>>>>create a 1.6GB executable. >>>>> >>>>>Fix the problem by using --image-base for lld. GNU ld and gold will >>>>>still get -Ttext-segment. Also delete the ld --verbose fallback introduced >>>>>in 2013, which is no longer relevant or correct (the default linker >>>>>script has changed). >>>>> >>>>>Signed-off-by: Fangrui Song <i@maskray.me> >>>>>--- >>>>>configure | 33 ++++++++++++--------------------- >>>>>1 file changed, 12 insertions(+), 21 deletions(-) >>>>> >>>>>diff --git a/configure b/configure >>>>>index 6099be1d84..2d45af0d09 100755 >>>>>--- a/configure >>>>>+++ b/configure >>>>>@@ -6336,43 +6336,34 @@ fi >>>>> >>>>># Probe for the need for relocating the user-only binary. >>>>>if ( [ "$linux_user" = yes ] || [ "$bsd_user" = yes ] ) && [ "$pie" = no ]; >>>>>then >>>>>- textseg_addr= >>>>>+ image_base= >>>>> case "$cpu" in >>>>> arm | i386 | ppc* | s390* | sparc* | x86_64 | x32) >>>>>- # ??? Rationale for choosing this address >>>>>- textseg_addr=0x60000000 >>>>>+ # An arbitrary address that makes it unlikely to collide with user >>>>>+ # programs. >> >>Please don't replace this ??? with an arbitrary rationale, which clearly >>doesn't apply to all of these hosts. > >In >https://lists.nongnu.org/archive/html/qemu-devel/2019-11/msg04646.html >it was suggested to move the comment around a bit. >I am not puzzled where and what I should say in the comment. >Can you (or other maintainers) kindly edit the comment for me? >I do not know enough about qemu to provide a good rationale here. > >>>>>+ image_base=0x60000000 >>>>> ;; >>>>> mips) >>>>> # A 256M aligned address, high in the address space, with enough >>>>> # room for the code_gen_buffer above it before the stack. >> >>This is the only one with a proper rationale. >> >>That said, I'm not sure that the proper way to handle this issue with lld is to >>drop this code entirely. > >The patch changes a feature that lld does not support: -Ttext-segment, >to use --image-base instead. > >Due to the prevalence of -z separate-code in GNU ld, -Ttext-segment is >no longer appropriate. I suggested that GNU linkers implement the >feature https://sourceware.org/bugzilla/show_bug.cgi?id=25207 . > >What gets deleted is the sed script. As I explained in the commit >message, it is no longer relevant. It probably applies to an old GNU ld >that FreeBSD used. FreeBSD has switched to lld now. > >>The best way to handle the underlying issue -- address conflict between >>interpreter and guest binary -- is PIE, for which this code is skipped. >> >>After that, we go to some pain to choose a guest_base address that allows the >>guest binary to load around the interpreter's reserved addresses. >> >>So what's left that this messing about with link addresses buys us? > >I agree that --enable-pie will be a better solution, but dropping the >support now will break at least FreeBSD. Its kernel supports running an >ET_DYN executable but it does not perform address randomization. >--disable-pie also appears to be used by ChromeOS developers who >reported https://bugs.llvm.org/show_bug.cgi?id=43997 . I can communicate >to them that migrating to --enable-pie is the way going forward. Ping? There is another very good reason that we need to keep --disable-pie for a while. Many users build statically linked qemu user mode binaries. -static-pie requires newer toolchain: * GCC 8 or clang 9 * binutils>=2.29 * glibc>=2.27 or musl Dropping support for GCC<8 does not sound a good idea in the year of 2019. (It may be in a few years.)
On 2019-12-09, Fangrui Song wrote: >On 2019-12-01, Fangrui Song wrote: >> >>Thanks for reviewing this patch! >> >>On 2019-12-01, Richard Henderson wrote: >>>On 11/27/19 6:36 PM, Fangrui Song wrote: >>>>On 2019-11-20, Fangrui Song wrote: >>>>>On 2019-11-15, Fangrui Song wrote: >>>>>>For lld, --image-base is the preferred way to set the base address. >>>>>>lld does not actually implement -Ttext-segment, but treats it as an alias for >>>>>>-Ttext. -Ttext-segment=0x60000000 combined with --no-rosegment can >>>>>>create a 1.6GB executable. >>>>>> >>>>>>Fix the problem by using --image-base for lld. GNU ld and gold will >>>>>>still get -Ttext-segment. Also delete the ld --verbose fallback introduced >>>>>>in 2013, which is no longer relevant or correct (the default linker >>>>>>script has changed). >>>>>> >>>>>>Signed-off-by: Fangrui Song <i@maskray.me> >>>>>>--- >>>>>>configure | 33 ++++++++++++--------------------- >>>>>>1 file changed, 12 insertions(+), 21 deletions(-) >>>>>> >>>>>>diff --git a/configure b/configure >>>>>>index 6099be1d84..2d45af0d09 100755 >>>>>>--- a/configure >>>>>>+++ b/configure >>>>>>@@ -6336,43 +6336,34 @@ fi >>>>>> >>>>>># Probe for the need for relocating the user-only binary. >>>>>>if ( [ "$linux_user" = yes ] || [ "$bsd_user" = yes ] ) && [ "$pie" = no ]; >>>>>>then >>>>>>- textseg_addr= >>>>>>+ image_base= >>>>>> case "$cpu" in >>>>>> arm | i386 | ppc* | s390* | sparc* | x86_64 | x32) >>>>>>- # ??? Rationale for choosing this address >>>>>>- textseg_addr=0x60000000 >>>>>>+ # An arbitrary address that makes it unlikely to collide with user >>>>>>+ # programs. >>> >>>Please don't replace this ??? with an arbitrary rationale, which clearly >>>doesn't apply to all of these hosts. >> >>In >>https://lists.nongnu.org/archive/html/qemu-devel/2019-11/msg04646.html >>it was suggested to move the comment around a bit. >>I am not puzzled where and what I should say in the comment. >>Can you (or other maintainers) kindly edit the comment for me? >>I do not know enough about qemu to provide a good rationale here. >> >>>>>>+ image_base=0x60000000 >>>>>> ;; >>>>>> mips) >>>>>> # A 256M aligned address, high in the address space, with enough >>>>>> # room for the code_gen_buffer above it before the stack. >>> >>>This is the only one with a proper rationale. >>> >>>That said, I'm not sure that the proper way to handle this issue with lld is to >>>drop this code entirely. >> >>The patch changes a feature that lld does not support: -Ttext-segment, >>to use --image-base instead. >> >>Due to the prevalence of -z separate-code in GNU ld, -Ttext-segment is >>no longer appropriate. I suggested that GNU linkers implement the >>feature https://sourceware.org/bugzilla/show_bug.cgi?id=25207 . >> >>What gets deleted is the sed script. As I explained in the commit >>message, it is no longer relevant. It probably applies to an old GNU ld >>that FreeBSD used. FreeBSD has switched to lld now. >> >>>The best way to handle the underlying issue -- address conflict between >>>interpreter and guest binary -- is PIE, for which this code is skipped. >>> >>>After that, we go to some pain to choose a guest_base address that allows the >>>guest binary to load around the interpreter's reserved addresses. >>> >>>So what's left that this messing about with link addresses buys us? >> >>I agree that --enable-pie will be a better solution, but dropping the >>support now will break at least FreeBSD. Its kernel supports running an >>ET_DYN executable but it does not perform address randomization. >>--disable-pie also appears to be used by ChromeOS developers who >>reported https://bugs.llvm.org/show_bug.cgi?id=43997 . I can communicate >>to them that migrating to --enable-pie is the way going forward. > >Ping? There is another very good reason that we need to keep >--disable-pie for a while. Many users build statically linked qemu user >mode binaries. -static-pie requires newer toolchain: > >* GCC 8 or clang 9 >* binutils>=2.29 >* glibc>=2.27 or musl > >Dropping support for GCC<8 does not sound a good idea in the year of 2019. (It may be in a few years.) Ping...
diff --git a/configure b/configure index 6099be1d84..2d45af0d09 100755 --- a/configure +++ b/configure @@ -6336,43 +6336,34 @@ fi # Probe for the need for relocating the user-only binary. if ( [ "$linux_user" = yes ] || [ "$bsd_user" = yes ] ) && [ "$pie" = no ]; then - textseg_addr= + image_base= case "$cpu" in arm | i386 | ppc* | s390* | sparc* | x86_64 | x32) - # ??? Rationale for choosing this address - textseg_addr=0x60000000 + # An arbitrary address that makes it unlikely to collide with user + # programs. + image_base=0x60000000 ;; mips) # A 256M aligned address, high in the address space, with enough # room for the code_gen_buffer above it before the stack. - textseg_addr=0x60000000 + image_base=0x60000000 ;; esac - if [ -n "$textseg_addr" ]; then + if [ -n "$image_base" ]; then cat > $TMPC <<EOF int main(void) { return 0; } EOF - textseg_ldflags="-Wl,-Ttext-segment=$textseg_addr" - if ! compile_prog "" "$textseg_ldflags"; then - # In case ld does not support -Ttext-segment, edit the default linker - # script via sed to set the .text start addr. This is needed on FreeBSD - # at least. - if ! $ld --verbose >/dev/null 2>&1; then + image_base_ldflags="-Wl,--image-base=$image_base" + if ! compile_prog "" "$image_base_ldflags"; then + image_base_ldflags="-Wl,-Ttext-segment=$image_base" + if ! compile_prog "" "$image_base_ldflags"; then error_exit \ "We need to link the QEMU user mode binaries at a" \ "specific text address. Unfortunately your linker" \ - "doesn't support either the -Ttext-segment option or" \ - "printing the default linker script with --verbose." \ + "supports neither --image-base nor -Ttext-segment. " \ "If you don't want the user mode binaries, pass the" \ "--disable-user option to configure." fi - - $ld --verbose | sed \ - -e '1,/==================================================/d' \ - -e '/==================================================/,$d' \ - -e "s/[.] = [0-9a-fx]* [+] SIZEOF_HEADERS/. = $textseg_addr + SIZEOF_HEADERS/" \ - -e "s/__executable_start = [0-9a-fx]*/__executable_start = $textseg_addr/" > config-host.ld - textseg_ldflags="-Wl,-T../config-host.ld" fi fi fi @@ -7945,7 +7936,7 @@ if test "$gprof" = "yes" ; then fi if test "$target_linux_user" = "yes" || test "$target_bsd_user" = "yes" ; then - ldflags="$ldflags $textseg_ldflags" + ldflags="$ldflags $image_base_ldflags" fi # Newer kernels on s390 check for an S390_PGSTE program header and
For lld, --image-base is the preferred way to set the base address. lld does not actually implement -Ttext-segment, but treats it as an alias for -Ttext. -Ttext-segment=0x60000000 combined with --no-rosegment can create a 1.6GB executable. Fix the problem by using --image-base for lld. GNU ld and gold will still get -Ttext-segment. Also delete the ld --verbose fallback introduced in 2013, which is no longer relevant or correct (the default linker script has changed). Signed-off-by: Fangrui Song <i@maskray.me> --- configure | 33 ++++++++++++--------------------- 1 file changed, 12 insertions(+), 21 deletions(-)