[Bug binutils/24456] bfd elf.c assertion for multiple relocations to same section
https://sourceware.org/bugzilla/show_bug.cgi?id=24456 --- Comment #5 from Nick Clifton --- Created attachment 11987 --> https://sourceware.org/bugzilla/attachment.cgi?id=11987&action=edit Test binary -- 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/24991] the assembler generates incorrect offset for jumps on arm thumb2
https://sourceware.org/bugzilla/show_bug.cgi?id=24991 Nick Clifton changed: What|Removed |Added CC||nickc at redhat dot com --- Comment #1 from Nick Clifton --- Hi Mikulas, Please could you upload a copy of the compiled executable *and* a copy of the assembler output from gcc. It is possible that this is a compiler bug - ie it could be generating a branch that is not going to the correct location - and the assembler output should help us determine if this is the case. It could also be both a compiler and an assembler bug. eg the compiler might be producing a branch that is out of range, and the assembler could be assembling it without complaining. 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 gas/24991] the assembler generates incorrect offset for jumps on arm thumb2
https://sourceware.org/bugzilla/show_bug.cgi?id=24991 --- Comment #2 from Mikulas Patocka --- Created attachment 11988 --> https://sourceware.org/bugzilla/attachment.cgi?id=11988&action=edit Compiler output -- 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/24991] the assembler generates incorrect offset for jumps on arm thumb2
https://sourceware.org/bugzilla/show_bug.cgi?id=24991 --- Comment #3 from Mikulas Patocka --- Created attachment 11989 --> https://sourceware.org/bugzilla/attachment.cgi?id=11989&action=edit Assembler output -- 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/24991] the assembler generates incorrect offset for jumps on arm thumb2
https://sourceware.org/bugzilla/show_bug.cgi?id=24991 --- Comment #4 from Mikulas Patocka --- Created attachment 11990 --> https://sourceware.org/bugzilla/attachment.cgi?id=11990&action=edit Linker output -- 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/24991] the assembler generates incorrect offset for jumps on arm thumb2
https://sourceware.org/bugzilla/show_bug.cgi?id=24991 --- Comment #5 from Mikulas Patocka --- I've attached the files. It is both the compiler and assembler bug - the compiler generates out of range jump and the assembler incorrectly assembles it as a backward jump. -- 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/24907] Objdump produces wrong branch opcode interpretation in some cases
https://sourceware.org/bugzilla/show_bug.cgi?id=24907 --- Comment #4 from cvs-commit at gcc dot gnu.org --- The master branch has been updated by Alan Modra : https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=0a4632b5637fa63ea4dc7651cdbae21e8b7005cd commit 0a4632b5637fa63ea4dc7651cdbae21e8b7005cd Author: Alan Modra Date: Wed Sep 11 17:33:00 2019 +0930 Re: Enhance the disassembler PR 24907 * objdump.c (disassemble_bytes): Adjust code to avoid overlong lines. Correct max_reloc_offset_into_insn test. -- 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/24992] New: RISC-V: partial relaxing against global pointer with sdata section alignment
https://sourceware.org/bugzilla/show_bug.cgi?id=24992
Bug ID: 24992
Summary: RISC-V: partial relaxing against global pointer with
sdata section alignment
Product: binutils
Version: 2.30
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: ld
Assignee: unassigned at sourceware dot org
Reporter: yitingwang16 at outlook dot com
Target Milestone: ---
A simple test case below to trigger the weird behavior of GP linker relaxation:
$ cat test.c
char __attribute__ ((aligned(2))) x = 'x';
char a = 1;
char b = 2;
char c = 3;
char d = 4;
char e = 5;
int main()
{
return a + b + c + d + e;
}
$ riscv64-unknown-linux-gnu-gcc -nostdlib -o test test.c
HI20/LO12 relocations that reference c, d and e are relaxed against gp:
$ riscv64-unknown-linux-gnu-objdump -D test
000100e8 :
100e8: 1141addisp,sp,-16
100ea: e422sd s0,8(sp)
100ec: 0800addis0,sp,16
100ee: 67c5lui a5,0x11
100f0: 1317c783lbu a5,305(a5) # 11131
100f4: 0007871bsext.w a4,a5
100f8: 67c5lui a5,0x11
100fa: 1327c783lbu a5,306(a5) # 11132
100fe: 2781sext.w a5,a5
10100: 9fb9addwa5,a5,a4
10102: 0007871bsext.w a4,a5
10106: 8031c783lbu a5,-2045(gp) # 11133
1010a: 2781sext.w a5,a5
1010c: 9fb9addwa5,a5,a4
1010e: 0007871bsext.w a4,a5
10112: 8041c783lbu a5,-2044(gp) # 11134
10116: 2781sext.w a5,a5
10118: 9fb9addwa5,a5,a4
1011a: 0007871bsext.w a4,a5
1011e: 8051c783lbu a5,-2043(gp) # 11135
10122: 2781sext.w a5,a5
10124: 9fb9addwa5,a5,a4
10126: 2781sext.w a5,a5
10128: 853emv a0,a5
1012a: 6422ld s0,8(sp)
1012c: 0141addisp,sp,16
1012e: 8082ret
Change test.c line#1 to:
char __attribute__ ((aligned(4))) x = 'x';
Only HI20/LO12 relocation that references e is relaxed against gp:
$ riscv64-unknown-linux-gnu-objdump -D test
000100e8 :
100e8: 1141addisp,sp,-16
100ea: e422sd s0,8(sp)
100ec: 0800addis0,sp,16
100ee: 67c5lui a5,0x11
100f0: 1357c783lbu a5,309(a5) # 11135
100f4: 0007871bsext.w a4,a5
100f8: 67c5lui a5,0x11
100fa: 1367c783lbu a5,310(a5) # 11136
100fe: 2781sext.w a5,a5
10100: 9fb9addwa5,a5,a4
10102: 0007871bsext.w a4,a5
10106: 67c5lui a5,0x11
10108: 1377c783lbu a5,311(a5) # 11137
1010c: 2781sext.w a5,a5
1010e: 9fb9addwa5,a5,a4
10110: 0007871bsext.w a4,a5
10114: 67c5lui a5,0x11
10116: 1387c783lbu a5,312(a5) # 11138
1011a: 2781sext.w a5,a5
1011c: 9fb9addwa5,a5,a4
1011e: 0007871bsext.w a4,a5
10122: 8051c783lbu a5,-2043(gp) # 11139
10126: 2781sext.w a5,a5
10128: 9fb9addwa5,a5,a4
1012a: 2781sext.w a5,a5
1012c: 853emv a0,a5
1012e: 6422ld s0,8(sp)
10130: 0141addisp,sp,16
10132: 8082ret
Change test.c line#1 to:
char __attribute__ ((aligned(8))) x = 'x';
Then there is no relaxation against gp.
--
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/24993] New: RISC-V: Address 0x00000000000xxxxx is out of bounds when "objdump -D"
https://sourceware.org/bugzilla/show_bug.cgi?id=24993
Bug ID: 24993
Summary: RISC-V: Address 0x000x is out of bounds
when "objdump -D"
Product: binutils
Version: 2.30
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: binutils
Assignee: unassigned at sourceware dot org
Reporter: yitingwang16 at outlook dot com
Target Milestone: ---
$ cat test.c
char __attribute__ ((aligned(2))) x = 'x';
char a = 1;
char b = 2;
char c = 3;
char d = 4;
char e = 5;
int main()
{
return a + b + c + d + e;
}
$ riscv64-unknown-linux-gnu-gcc -nostdlib -o test test.c
$ riscv64-unknown-linux-gnu-objdump -D test
...
Disassembly of section .sdata:
00011130 :
11130: addia4,sp,140
00011131 :
11131: additp,tp,0
00011132 :
11132: sllit1,t1,0x0
00011133 :
11133: lb s0,0(zero) # 0
00011134 :
11134: addis1,sp,640
00011135 :
11135: 05 Address 0x00011135 is out of bounds.
11139:
--
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/24992] RISC-V: partial relaxing against global pointer with sdata section alignment
https://sourceware.org/bugzilla/show_bug.cgi?id=24992 Jim Wilson changed: What|Removed |Added CC||wilson at gcc dot gnu.org --- Comment #1 from Jim Wilson --- This is essentially a dup of 24983. -- 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/24993] RISC-V: Address 0x00000000000xxxxx is out of bounds when "objdump -D"
https://sourceware.org/bugzilla/show_bug.cgi?id=24993 Jim Wilson changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED CC||wilson at gcc dot gnu.org Resolution|--- |INVALID --- Comment #1 from Jim Wilson --- the problem is that you are using objdump -D, and this is almost always the wrong thing to do. The correct option is "-d". -D will dump data sections as if they are text sections, but data sections are not text, and thus (normally) all you will get is garbage. There is also the problem that risc-v instructions require nature alignment and power of 2 sizes, and data does not. So in this case, you have an odd number of bytes in the data section, which can't be translated into a RISC-V instruction, and so you are getting an objdump error, which is correct. The error message could perhaps be a little friendlier, but the error is correct. I think there is some old RISC-V docs somewhere telling people to use objdump -D when they should not be using it, as there is an epidemic of RISC-V folks reporting bugs with objdump -D when it isn't broken, and when they shouldn't have used -D in the first place. If you are aware of docs someplace recommending people use objdump -D, then please report bugs against those docs, not against binutils. -- 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/24983] RISC-V GP linker relaxation is not performed with -nostdlib
https://sourceware.org/bugzilla/show_bug.cgi?id=24983 Jim Wilson changed: What|Removed |Added CC||yitingwang16 at outlook dot com --- Comment #2 from Jim Wilson --- *** Bug 24992 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 ld/24992] RISC-V: partial relaxing against global pointer with sdata section alignment
https://sourceware.org/bugzilla/show_bug.cgi?id=24992 Jim Wilson changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|--- |DUPLICATE --- Comment #2 from Jim Wilson --- Same problem as 24983. *** This bug has been marked as a duplicate of bug 24983 *** -- 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
