[Bug ld/22970] Add support for R_AARCH64_TLSLE_LDST8_TPREL_LO12 relocation.
https://sourceware.org/bugzilla/show_bug.cgi?id=22970 Nick Clifton changed: What|Removed |Added CC||nickc at redhat dot com --- Comment #1 from Nick Clifton --- Created attachment 10907 --> https://sourceware.org/bugzilla/attachment.cgi?id=10907&action=edit Proposed patch Hi Peter, Please could you try out this patch and let me know if it works for you ? I am especially concerned that I might have missed something when evaluating the TLSLE_LDST8_LO12_NC reloc in the linker, so if you can, please check that linked binaries do actually work as expected... :-) Cheers Nick -- 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 gold/22994] New: GOLD drops STV_PROTECTED status from weak function symbols
https://sourceware.org/bugzilla/show_bug.cgi?id=22994 Bug ID: 22994 Summary: GOLD drops STV_PROTECTED status from weak function symbols Product: binutils Version: 2.31 (HEAD) Status: NEW Severity: normal Priority: P2 Component: gold Assignee: ccoutant at gmail dot com Reporter: nickc at redhat dot com CC: ian at airs dot com Target Milestone: --- Target: x86_64 Created attachment 10909 --> https://sourceware.org/bugzilla/attachment.cgi?id=10909&action=edit Object files and shell script to demonstrate the bug When gold is used to link the LLVM IR files in the uploaded test it drops the STV_PROTECTED status from the f1 function symbol. Earlier versions of gold (2.29 or previous) did not do this. Nor does the 2.30 bfd linker. To reproduce install llvm and then run the build.sh script. Using ld.bfd in the script produces: 16: 1000 6 FUNCWEAK PROTECTED6 f1 8: 3f10 0 SECTION LOCAL DEFAULT8 11: 1010 2 FUNCLOCAL DEFAULT6 f1.2 14: 3f10 0 OBJECT LOCAL DEFAULT8 _DYNAMIC 26: 1000 6 FUNCWEAK PROTECTED6 f1 whereas ld.gold produces: 18: 05b0 6 FUNCWEAK DEFAULT8 f1 2: 05c0 2 FUNCLOCAL DEFAULT8 f1.2 24: 05b0 6 FUNCWEAK DEFAULT8 f1 -- 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/22948] Modify ld search path to match libxxx.lib
https://sourceware.org/bugzilla/show_bug.cgi?id=22948 --- Comment #2 from Ryan Scott --- That patch does fix the issue I was having, but in order to make it work, I had to copy those changes to the pep.em file (not just pe.em). -- 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 gold/22994] GOLD drops STV_PROTECTED status from weak function symbols
https://sourceware.org/bugzilla/show_bug.cgi?id=22994 Cary Coutant changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |DUPLICATE --- Comment #1 from Cary Coutant --- Duplicate of PR 22868. *** This bug has been marked as a duplicate of bug 22868 *** -- 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 gold/22868] Gold does not select proper symbol visibility when used with LLVM gold-plugin
https://sourceware.org/bugzilla/show_bug.cgi?id=22868 --- Comment #4 from Michael Matz --- FWIW, I'm using this patch in our binutils package: -- Fixes two testsuite fails in the gold plugin tests of LLVM. Aka binutils/PR22868 Index: binutils-2.30/gold/resolve.cc === --- binutils-2.30.orig/gold/resolve.cc 2018-01-13 14:31:16.0 +0100 +++ binutils-2.30/gold/resolve.cc 2018-03-06 16:58:42.0 +0100 @@ -265,10 +265,13 @@ Symbol_table::resolve(Sized_symbol return; // Likewise for an absolute symbol defined twice with the same value. + // plugin-symbols are always absolute with same value here, so ignore those if (!is_ordinary && st_shndx == elfcpp::SHN_ABS && !to_is_ordinary && to_shndx == elfcpp::SHN_ABS + && object->pluginobj() == NULL + && to->object()->pluginobj() == NULL && to->value() == sym.get_st_value()) return; -- 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 gold/22868] Gold does not select proper symbol visibility when used with LLVM gold-plugin
https://sourceware.org/bugzilla/show_bug.cgi?id=22868 Cary Coutant changed: What|Removed |Added CC||nickc at redhat dot com --- Comment #3 from Cary Coutant --- *** Bug 22994 has been marked as a duplicate of this bug. *** -- 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 gold/22868] Gold does not select proper symbol visibility when used with LLVM gold-plugin
https://sourceware.org/bugzilla/show_bug.cgi?id=22868 Cary Coutant changed: What|Removed |Added Status|UNCONFIRMED |ASSIGNED Last reconfirmed||2018-03-22 Ever confirmed|0 |1 --- Comment #5 from Cary Coutant --- I'm not able to reproduce this problem in my test framework. Whether or not we override the symbol when scanning the IR, we still pick up the visibility when processing the replacement files. Could it be the case that your replacement objects do not have the STV_PROTECTED visibility on the symbol? -- 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 gold/22883] Gold mis-evaluates R_X86_64_PLT32 reloc
https://sourceware.org/bugzilla/show_bug.cgi?id=22883 Cary Coutant changed: What|Removed |Added Status|NEW |RESOLVED Resolution|--- |INVALID --- Comment #1 from Cary Coutant --- I don't see how the Gnu linker is making this work. All your PLT32 relocations are missing the "-4" bias for a pc-relative reference. As far as I can see, gold is doing the right thing, and the assembler is emitting bad relocations. Relocation section '.rela.text' at offset 0x1cdc contains 30 entries: Offset Info Type Symbol's Value Symbol's Name + Addend 0017 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 005d 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 0086 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 00b0 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 00da 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 0104 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 012e 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 0158 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 0182 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 01ac 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 01d6 000a0004 R_X86_64_PLT32 0026 mx_ownsCopy_8u + 0 01f8 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 02d2 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 02f7 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 031b 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 033f 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 0363 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 0387 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 03ab 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 03cf 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 03f3 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 0417 00160004 R_X86_64_PLT32 0208 mx_ownsZero_8u + 0 0438 00220004 R_X86_64_PLT32 0448 mx_ownsSet_8u + 0 053c 00240004 R_X86_64_PLT32 054c mx_ownsSet_16u + 0 066e 00270004 R_X86_64_PLT32 067e mx_ownsSet_32s + 0 0757 00270004 R_X86_64_PLT32 067e mx_ownsSet_32s + 0 07c0 002c0004 R_X86_64_PLT32 07d0 mx_ownsSet_64s + 0 08c7 002e0004 R_X86_64_PLT32 08dc mx_ownsSet_64sc + 0 0a0c 002c0004 R_X86_64_PLT32 07d0 mx_ownsSet_64s + 0 0a5b 002e0004 R_X86_64_PLT32 08dc mx_ownsSet_64sc + 0 -- 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 gold/22883] Gold mis-evaluates R_X86_64_PLT32 reloc
https://sourceware.org/bugzilla/show_bug.cgi?id=22883 --- Comment #2 from Cary Coutant --- With a top-of-trunk assembler, I get the "- 4". $ cat test.s .globl bar .text foo: callq bar@PLT ret $ as -o test.o test.s $ readelf -rW test.o Relocation section '.rela.text' at offset 0x110 contains 1 entries: Offset Info Type Symbol's Value Symbol's Name + Addend 0001 00050004 R_X86_64_PLT32 bar - 4 -- 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/22970] Add support for R_AARCH64_TLSLE_LDST8_TPREL_LO12 relocation.
https://sourceware.org/bugzilla/show_bug.cgi?id=22970 --- Comment #2 from Peter Smith --- Hello Nick, Thanks for the patch. I'm travelling at the moment; I will try this out when I get back to the office on Monday. Peter -- 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
