[Bug ld/23195] New: [AArch64] ld-aarch64/ifunc-9 failing
https://sourceware.org/bugzilla/show_bug.cgi?id=23195 Bug ID: 23195 Summary: [AArch64] ld-aarch64/ifunc-9 failing Product: binutils Version: unspecified Status: NEW Severity: normal Priority: P2 Component: ld Assignee: unassigned at sourceware dot org Reporter: sudi at gcc dot gnu.org Target Milestone: --- Since the following commit: commit 4ec0995016801cc5d5cf13baf6e10163861e6852 Author: H.J. Lu Date: Mon May 14 03:55:37 2018 -0700 x86; Allow IFUNC pointer defined in PDE If IFUNC symbol is defined in position-dependent executable, we should change it to the normal function and set its address to its PLT entry which should be resolved by R_*_IRELATIVE at run-time. All external references should be resolved to its PLT in executable. bfd/ PR ld/23169 * elf-ifunc.c (_bfd_elf_allocate_ifunc_dyn_relocs): Don't issue an error on IFUNC pointer defined in PDE. * elf32-i386.c (elf_i386_finish_dynamic_symbol): Call _bfd_x86_elf_link_fixup_ifunc_symbol. * elf64-x86-64.c (elf_x86_64_finish_dynamic_symbol): Likewise. * elfxx-x86.c (_bfd_x86_elf_link_fixup_ifunc_symbol): New function. * elfxx-x86.h (_bfd_x86_elf_link_fixup_ifunc_symbol): New. ld/ PR ld/23169 * testsuite/ld-ifunc/ifunc-9-i386.d: New file. * testsuite/ld-ifunc/ifunc-9-x86-64.d: Likewise. * testsuite/ld-ifunc/pr23169a.c: Likewise. * testsuite/ld-ifunc/pr23169a.rd: Likewise. * testsuite/ld-ifunc/pr23169b.c: Likewise. * testsuite/ld-ifunc/pr23169b.c: Likewise. * testsuite/ld-ifunc/pr23169c.rd: Likewise. * testsuite/ld-ifunc/pr23169c.rd: Likewise. * testsuite/ld-ifunc/ifunc-9-x86.d: Removed. * testsuite/ld-ifunc/ifunc.exp: Run PR ld/23169 tests. I have observed the following failure on aarch64-none-linux-gnu succeeded with: <>, expected: <.*dynamic STT_GNU_IFUNC symbol `foo' with pointer equality in `.*.o' can not be used when making an executable; recompile with -fPIE and relink with -pie> FAIL: ld-aarch64/ifunc-9 I have also observed some of the new tests failing: FAIL: Build pr23169a FAIL: Build pr23169d FAIL: Run pr23169a FAIL: Run pr23169d -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/23195] [AArch64] ld-aarch64/ifunc-9 failing
https://sourceware.org/bugzilla/show_bug.cgi?id=23195 Sudakshina Das changed: What|Removed |Added Target||aarch64-none-linux-gnu -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug ld/23195] [AArch64] ld-aarch64/ifunc-9 failing
https://sourceware.org/bugzilla/show_bug.cgi?id=23195 --- Comment #2 from Sudakshina Das --- Thanks! -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/24258] v8.5 extension "frintts" missing as a separate feature
https://sourceware.org/bugzilla/show_bug.cgi?id=24258 --- Comment #1 from Sudakshina Das --- Hi Richard As far as I can see, frintts is not on optional feature on Armv8.5-A and thus I don't believe a command line argument is needed, unless I missed something. Sudi -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/24279] ARMv8.5 extension incorrectly named "predres" instead of "predinv"
https://sourceware.org/bugzilla/show_bug.cgi?id=24279 --- Comment #2 from Sudakshina Das --- Hi Richard We chose the name based on what instructions it is turning on - [Control Flow/Data Value/Cache Prefetch] Prediction Restriction by Context I agree that the internal naming is different. However, I believe the user option would be more suited to reflect the instructions it turns on. This name was also agreed on with our LLVM team where it is already upstream too. Sudi -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/24258] v8.5 extension "frintts" missing as a separate feature
https://sourceware.org/bugzilla/show_bug.cgi?id=24258 --- Comment #3 from Sudakshina Das --- Hi Richard Sorry for the delay in getting back on this. I agree that the specs say that any Armv8.[N+1] feature is optional on Armv8.x. However in order to implement that, we would have to make every feature optional not just frintts. We chose to keep these features independent internally by giving them new feature macros so that if and when someone implements a Armv8.[N+1] feature optionally in a Armv8.N, it can be easily updated. Thus not giving features like frintts a user facing option was a conscious decision we took while implementing. Sudi -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/24258] v8.5 extension "frintts" missing as a separate feature
https://sourceware.org/bugzilla/show_bug.cgi?id=24258 --- Comment #5 from Sudakshina Das --- If we take Armv8.5-A as an example and if we check include/opcode/aarch64.h for all the new feature bits added /* Flag Manipulation insns. */ #define AARCH64_FEATURE_FLAGMANIP 0x40ULL /* FRINT[32,64][Z,X] insns. */ #define AARCH64_FEATURE_FRINTTS 0x80ULL /* SB instruction. */ #define AARCH64_FEATURE_SB 0x100ULL /* Execution and Data Prediction Restriction instructions. */ #define AARCH64_FEATURE_PREDRES 0x200ULL /* DC CVADP. */ #define AARCH64_FEATURE_CVADP 0x400ULL /* Random Number instructions. */ #define AARCH64_FEATURE_RNG 0x800ULL /* BTI instructions. */ #define AARCH64_FEATURE_BTI 0x1000ULL /* SCXTNUM_ELx. */ #define AARCH64_FEATURE_SCXTNUM 0x2000ULL /* ID_PFR2 instructions. */ #define AARCH64_FEATURE_ID_PFR2 0x4000ULL /* SSBS mechanism enabled. */ #define AARCH64_FEATURE_SSBS0x8000ULL /* Memory Tagging Extension. */ #define AARCH64_FEATURE_MEMTAG 0x1ULL These can all be potentially added by a ARMv8.4-A implementation but do not all have command line options. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug gas/24258] v8.5 extension "frintts" missing as a separate feature
https://sourceware.org/bugzilla/show_bug.cgi?id=24258 --- Comment #7 from Sudakshina Das --- We have added extra options for every feature that has been explicitly mentioned as optional and backported in the architecture. The reason why we are not adding the rest is to prevent an explosion of commandline options. Since these remaining extensions are mandatory at Armv8.5-A we expect a user to just use -march=armv8.5-a, and if we have an explicit use case for someone having implemented one of these features under the Armv8.[N+1] rule we can consider giving it a command line option. -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/24460] bfd/elf32-arm.c: get_value_helper, uninitialized value used
https://sourceware.org/bugzilla/show_bug.cgi?id=24460 Sudakshina Das changed: What|Removed |Added CC||sudi at gcc dot gnu.org Assignee|unassigned at sourceware dot org |sudi at gcc dot gnu.org -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/24460] bfd/elf32-arm.c: get_value_helper, uninitialized value used
https://sourceware.org/bugzilla/show_bug.cgi?id=24460 Sudakshina Das changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Last reconfirmed||2019-04-17 Ever confirmed|0 |1 -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
[Bug binutils/24460] bfd/elf32-arm.c: get_value_helper, uninitialized value used
https://sourceware.org/bugzilla/show_bug.cgi?id=24460 --- Comment #1 from Sudakshina Das --- Patch proposed https://sourceware.org/ml/binutils/2019-04/msg00227.html -- You are receiving this mail because: You are on the CC list for the bug. ___ bug-binutils mailing list bug-binutils@gnu.org https://lists.gnu.org/mailman/listinfo/bug-binutils
