Re: Avoid assembler warnings from AArch64 constructor/destructor priorities
On 17/07/18 12:09, Kyrill Tkachov wrote: > > On 02/02/18 15:14, Kyrill Tkachov wrote: >> >> On 01/02/18 17:26, Joseph Myers wrote: >>> On Thu, 1 Feb 2018, Kyrill Tkachov wrote: >>> Hi Joseph, aarch64 maintainers, On 28/09/17 13:31, Joseph Myers wrote: > Many GCC tests fail for AArch64 with current binutils because of > assembler warnings of the form "Warning: ignoring incorrect section > type for .init_array.00100". The same issue was fixed for ARM in > r247015 by using SECTION_NOTYPE when creating those sections; this > patch applies the same fix to AArch64. > > Tested with no regressions with cross to aarch64-linux-gnu. OK to > commit? There is a user request to backport this patch to the GCC 7 and 6 branches: PR 84168. Would that be acceptable? >>> I don't object, but I'm not aware of this being a regression. >> >> I think it's been present since the beginning of the port so >> it's not a regression, just a user request for a bugfix backport. >> >> The patch applies, bootstraps and tests cleanly on both branches FWIW. >> > > Gentle ping on this (I had almost forgotten about it). > Do the aarch64 maintainers mind this patch being packported to the GCC 7 > and 6 branches? > > https://gcc.gnu.org/ml/gcc-patches/2018-02/msg00054.html > > Thanks, > Kyrill > > It looks pretty safe to me. OK unless the RMs object. R.
Re: Avoid assembler warnings from AArch64 constructor/destructor priorities
On 28/09/17 13:31, Joseph Myers wrote: > Many GCC tests fail for AArch64 with current binutils because of > assembler warnings of the form "Warning: ignoring incorrect section > type for .init_array.00100". The same issue was fixed for ARM in > r247015 by using SECTION_NOTYPE when creating those sections; this > patch applies the same fix to AArch64. > > Tested with no regressions with cross to aarch64-linux-gnu. OK to > commit? > > 2017-09-28 Joseph Myers > > * config/aarch64/aarch64.c (aarch64_elf_asm_constructor) > (aarch64_elf_asm_destructor): Pass SECTION_NOTYPE to get_section > when creating .init_array and .fini_array sections with priority > specified. > > Index: gcc/config/aarch64/aarch64.c > === > --- gcc/config/aarch64/aarch64.c (revision 253248) > +++ gcc/config/aarch64/aarch64.c (working copy) > @@ -6095,7 +6095,7 @@ aarch64_elf_asm_constructor (rtx symbol, int prior > -Wformat-truncation false positive, use a larger size. */ >char buf[23]; >snprintf (buf, sizeof (buf), ".init_array.%.5u", priority); > - s = get_section (buf, SECTION_WRITE, NULL); > + s = get_section (buf, SECTION_WRITE | SECTION_NOTYPE, NULL); >switch_to_section (s); >assemble_align (POINTER_SIZE); >assemble_aligned_integer (POINTER_BYTES, symbol); > @@ -6115,7 +6115,7 @@ aarch64_elf_asm_destructor (rtx symbol, int priori > -Wformat-truncation false positive, use a larger size. */ >char buf[23]; >snprintf (buf, sizeof (buf), ".fini_array.%.5u", priority); > - s = get_section (buf, SECTION_WRITE, NULL); > + s = get_section (buf, SECTION_WRITE | SECTION_NOTYPE, NULL); >switch_to_section (s); >assemble_align (POINTER_SIZE); >assemble_aligned_integer (POINTER_BYTES, symbol); > Ok, sorry I missed this one. R.
Re: Avoid assembler warnings from AArch64 constructor/destructor priorities
On 02/02/18 15:14, Kyrill Tkachov wrote: On 01/02/18 17:26, Joseph Myers wrote: On Thu, 1 Feb 2018, Kyrill Tkachov wrote: Hi Joseph, aarch64 maintainers, On 28/09/17 13:31, Joseph Myers wrote: Many GCC tests fail for AArch64 with current binutils because of assembler warnings of the form "Warning: ignoring incorrect section type for .init_array.00100". The same issue was fixed for ARM in r247015 by using SECTION_NOTYPE when creating those sections; this patch applies the same fix to AArch64. Tested with no regressions with cross to aarch64-linux-gnu. OK to commit? There is a user request to backport this patch to the GCC 7 and 6 branches: PR 84168. Would that be acceptable? I don't object, but I'm not aware of this being a regression. I think it's been present since the beginning of the port so it's not a regression, just a user request for a bugfix backport. The patch applies, bootstraps and tests cleanly on both branches FWIW. Gentle ping on this (I had almost forgotten about it). Do the aarch64 maintainers mind this patch being packported to the GCC 7 and 6 branches? https://gcc.gnu.org/ml/gcc-patches/2018-02/msg00054.html Thanks, Kyrill
Re: Avoid assembler warnings from AArch64 constructor/destructor priorities
On 01/02/18 17:26, Joseph Myers wrote: On Thu, 1 Feb 2018, Kyrill Tkachov wrote: Hi Joseph, aarch64 maintainers, On 28/09/17 13:31, Joseph Myers wrote: Many GCC tests fail for AArch64 with current binutils because of assembler warnings of the form "Warning: ignoring incorrect section type for .init_array.00100". The same issue was fixed for ARM in r247015 by using SECTION_NOTYPE when creating those sections; this patch applies the same fix to AArch64. Tested with no regressions with cross to aarch64-linux-gnu. OK to commit? There is a user request to backport this patch to the GCC 7 and 6 branches: PR 84168. Would that be acceptable? I don't object, but I'm not aware of this being a regression. I think it's been present since the beginning of the port so it's not a regression, just a user request for a bugfix backport. The patch applies, bootstraps and tests cleanly on both branches FWIW. Thanks, Kyrill
Re: Avoid assembler warnings from AArch64 constructor/destructor priorities
On Thu, 1 Feb 2018, Kyrill Tkachov wrote: > Hi Joseph, aarch64 maintainers, > > On 28/09/17 13:31, Joseph Myers wrote: > > Many GCC tests fail for AArch64 with current binutils because of > > assembler warnings of the form "Warning: ignoring incorrect section > > type for .init_array.00100". The same issue was fixed for ARM in > > r247015 by using SECTION_NOTYPE when creating those sections; this > > patch applies the same fix to AArch64. > > > > Tested with no regressions with cross to aarch64-linux-gnu. OK to > > commit? > > There is a user request to backport this patch to the GCC 7 and 6 > branches: PR 84168. > > Would that be acceptable? I don't object, but I'm not aware of this being a regression. -- Joseph S. Myers jos...@codesourcery.com
Re: Avoid assembler warnings from AArch64 constructor/destructor priorities
Hi Joseph, aarch64 maintainers, On 28/09/17 13:31, Joseph Myers wrote: Many GCC tests fail for AArch64 with current binutils because of assembler warnings of the form "Warning: ignoring incorrect section type for .init_array.00100". The same issue was fixed for ARM in r247015 by using SECTION_NOTYPE when creating those sections; this patch applies the same fix to AArch64. Tested with no regressions with cross to aarch64-linux-gnu. OK to commit? There is a user request to backport this patch to the GCC 7 and 6 branches: PR 84168. Would that be acceptable? If so, I'm happy to help with the testing on those branches. Thanks, Kyrill 2017-09-28 Joseph Myers* config/aarch64/aarch64.c (aarch64_elf_asm_constructor) (aarch64_elf_asm_destructor): Pass SECTION_NOTYPE to get_section when creating .init_array and .fini_array sections with priority specified. Index: gcc/config/aarch64/aarch64.c === --- gcc/config/aarch64/aarch64.c(revision 253248) +++ gcc/config/aarch64/aarch64.c(working copy) @@ -6095,7 +6095,7 @@ aarch64_elf_asm_constructor (rtx symbol, int prior -Wformat-truncation false positive, use a larger size. */ char buf[23]; snprintf (buf, sizeof (buf), ".init_array.%.5u", priority); - s = get_section (buf, SECTION_WRITE, NULL); + s = get_section (buf, SECTION_WRITE | SECTION_NOTYPE, NULL); switch_to_section (s); assemble_align (POINTER_SIZE); assemble_aligned_integer (POINTER_BYTES, symbol); @@ -6115,7 +6115,7 @@ aarch64_elf_asm_destructor (rtx symbol, int priori -Wformat-truncation false positive, use a larger size. */ char buf[23]; snprintf (buf, sizeof (buf), ".fini_array.%.5u", priority); - s = get_section (buf, SECTION_WRITE, NULL); + s = get_section (buf, SECTION_WRITE | SECTION_NOTYPE, NULL); switch_to_section (s); assemble_align (POINTER_SIZE); assemble_aligned_integer (POINTER_BYTES, symbol); -- Joseph S. Myers jos...@codesourcery.com
Re: Avoid assembler warnings from AArch64 constructor/destructor priorities
On 28/09/17 13:31, Joseph Myers wrote: > Many GCC tests fail for AArch64 with current binutils because of > assembler warnings of the form "Warning: ignoring incorrect section > type for .init_array.00100". The same issue was fixed for ARM in > r247015 by using SECTION_NOTYPE when creating those sections; this > patch applies the same fix to AArch64. > > Tested with no regressions with cross to aarch64-linux-gnu. OK to > commit? > > 2017-09-28 Joseph Myers> > * config/aarch64/aarch64.c (aarch64_elf_asm_constructor) > (aarch64_elf_asm_destructor): Pass SECTION_NOTYPE to get_section > when creating .init_array and .fini_array sections with priority > specified. > OK. R. > Index: gcc/config/aarch64/aarch64.c > === > --- gcc/config/aarch64/aarch64.c (revision 253248) > +++ gcc/config/aarch64/aarch64.c (working copy) > @@ -6095,7 +6095,7 @@ aarch64_elf_asm_constructor (rtx symbol, int prior > -Wformat-truncation false positive, use a larger size. */ >char buf[23]; >snprintf (buf, sizeof (buf), ".init_array.%.5u", priority); > - s = get_section (buf, SECTION_WRITE, NULL); > + s = get_section (buf, SECTION_WRITE | SECTION_NOTYPE, NULL); >switch_to_section (s); >assemble_align (POINTER_SIZE); >assemble_aligned_integer (POINTER_BYTES, symbol); > @@ -6115,7 +6115,7 @@ aarch64_elf_asm_destructor (rtx symbol, int priori > -Wformat-truncation false positive, use a larger size. */ >char buf[23]; >snprintf (buf, sizeof (buf), ".fini_array.%.5u", priority); > - s = get_section (buf, SECTION_WRITE, NULL); > + s = get_section (buf, SECTION_WRITE | SECTION_NOTYPE, NULL); >switch_to_section (s); >assemble_align (POINTER_SIZE); >assemble_aligned_integer (POINTER_BYTES, symbol); >
Avoid assembler warnings from AArch64 constructor/destructor priorities
Many GCC tests fail for AArch64 with current binutils because of assembler warnings of the form "Warning: ignoring incorrect section type for .init_array.00100". The same issue was fixed for ARM in r247015 by using SECTION_NOTYPE when creating those sections; this patch applies the same fix to AArch64. Tested with no regressions with cross to aarch64-linux-gnu. OK to commit? 2017-09-28 Joseph Myers* config/aarch64/aarch64.c (aarch64_elf_asm_constructor) (aarch64_elf_asm_destructor): Pass SECTION_NOTYPE to get_section when creating .init_array and .fini_array sections with priority specified. Index: gcc/config/aarch64/aarch64.c === --- gcc/config/aarch64/aarch64.c(revision 253248) +++ gcc/config/aarch64/aarch64.c(working copy) @@ -6095,7 +6095,7 @@ aarch64_elf_asm_constructor (rtx symbol, int prior -Wformat-truncation false positive, use a larger size. */ char buf[23]; snprintf (buf, sizeof (buf), ".init_array.%.5u", priority); - s = get_section (buf, SECTION_WRITE, NULL); + s = get_section (buf, SECTION_WRITE | SECTION_NOTYPE, NULL); switch_to_section (s); assemble_align (POINTER_SIZE); assemble_aligned_integer (POINTER_BYTES, symbol); @@ -6115,7 +6115,7 @@ aarch64_elf_asm_destructor (rtx symbol, int priori -Wformat-truncation false positive, use a larger size. */ char buf[23]; snprintf (buf, sizeof (buf), ".fini_array.%.5u", priority); - s = get_section (buf, SECTION_WRITE, NULL); + s = get_section (buf, SECTION_WRITE | SECTION_NOTYPE, NULL); switch_to_section (s); assemble_align (POINTER_SIZE); assemble_aligned_integer (POINTER_BYTES, symbol); -- Joseph S. Myers jos...@codesourcery.com