[Bug ld/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
https://sourceware.org/bugzilla/show_bug.cgi?id=15302 Terry Guo terry.guo at arm dot com changed: What|Removed |Added Status|REOPENED|RESOLVED Resolution|--- |FIXED --- Comment #9 from Terry Guo terry.guo at arm dot com --- Fixed by a updated patch at https://sourceware.org/ml/binutils/2013-07/msg00094.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
[Bug ld/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302 peter info at fagus dot ch changed: What|Removed |Added CC||info at fagus dot ch -- 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/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302 --- Comment #8 from cvs-commit at gcc dot gnu.org cvs-commit at gcc dot gnu.org --- CVSROOT:/cvs/src Module name:src Changes by:ni...@sourceware.org2013-06-28 12:22:42 Modified files: bfd: ChangeLog elf32-arm.c Log message: PR ld/15302 * elf32-arm.c (allocate_dynrelocs_for_symbol): Revert previous patch. Patches: http://sourceware.org/cgi-bin/cvsweb.cgi/src/bfd/ChangeLog.diff?cvsroot=srcr1=1.6088r2=1.6089 http://sourceware.org/cgi-bin/cvsweb.cgi/src/bfd/elf32-arm.c.diff?cvsroot=srcr1=1.318r2=1.319 -- 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/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302
Terry Guo terry.guo at arm dot com changed:
What|Removed |Added
Status|RESOLVED|REOPENED
Last reconfirmed||2013-06-21
Resolution|FIXED |---
Ever confirmed|0 |1
--- Comment #7 from Terry Guo terry.guo at arm dot com ---
I am sorry that I have to reopen it. The proposed patch will cause below issue.
In the below concrete case, the __init_array_start is allocated at address
0x11a40 actually. But any relocations using address of __init_array_start will
be resolved to use address 0x11a41 by linker, thus the subsequential deference
on __init_array_start is a deference on wrong address (0x11a41).
This is because the proposed patch set ST_BRANCH_TO_THUMB to the symbol
__init_array_start, then the last bit of its address is set by the linker
function elf32_arm_swap_symbol_out.
To reproduce this issue, an ARM baremetal toolchain using Newlib as C library
is needed. Then use below command to build a project with empty main function:
arm-none-eabi-gcc -mthumb -mcpu=cortex-m3 -O2 -lc -lrdimon -specs=rdimon.specs
test.c -o test.out -Wl,-verbose -Wl,-Map=test.map
Then run test.out on QEMU, a Segmentation falut will be reported.
qemu-system-arm -cpu cortex-m3 -nographic -monitor null -serial null
-semihosting -kernel ./test.out
This QEMU segmentation fault is caused by deference of __init_array_start in
function __libc_init_array.
Here are some investigations:
Disassemble the test.out with command arm-none-eabi-objdump -D test.out and
we will see:
Disassembly of section .init_array:
00011a40 __init_array_start:
11a40: 802dandeq r8, r0, sp, lsr #32
The __init_array_start is allocated at address 0x11a40 actually.
Meanwhile the command arm-none-eabi-readelf -s test.out, we will get symbol
table:
222: 00011a41 0 FUNCLOCAL DEFAULT7 __init_array_start
Thus any instructions using __init_array_start will be resolved to 0x11a41
instead of 0x11a40.
movwr5, #:lower16:__init_array_start
movtr5, #:upper16:__init_array_start
ldr r3, [r5, #4]!
blx r3
in the above case, the r5 will get wrong value 0x11a41, then r3 will get wrong
value, finally the blx will cause the fault.
Here are some background on Newlib:
In Newlib init.c we have code:
extern void (*__init_array_start []) (void) __attribute__((weak));
extern void (*__init_array_end []) (void) __attribute__((weak));
void
__libc_init_array (void)
{
size_t count;
size_t i;
count = __init_array_end - __init_array_start;
for (i = 0; i count; i++)
__init_array_start[i] ();
}
Both __init_array_start and __init_array_end are defined in linker script as:
.init_array :
{
PROVIDE_HIDDEN (__init_array_start = .);
KEEP (*(SORT(.init_array.*)))
KEEP (*(.init_array ))
PROVIDE_HIDDEN (__init_array_end = .);
}
--
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/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302 --- Comment #5 from cvs-commit at gcc dot gnu.org cvs-commit at gcc dot gnu.org --- CVSROOT:/cvs/src Module name:src Changes by:ni...@sourceware.org2013-06-13 12:36:02 Modified files: bfd: ChangeLog elf32-arm.c ld/testsuite : ChangeLog ld/testsuite/ld-arm: arm-elf.exp Added files: ld/testsuite/ld-arm: branch-lks-sym.ld thumb-b-lks-sym.d thumb-b-lks-sym.s thumb-bl-lks-sym.d thumb-bl-lks-sym.s Log message: PR ld/15302 * elf32-arm.c (allocate_dynrelocs_for_symbol): Transform ST_BRANCH_TO_ARM into ST_BRANCH_TO_THUMB if the target only supports thumb instructions. PR ld/15302 * ld-arm/branch-lks-sym.ld: New script. * ld-arm/thumb-b-lks-sym.s: New test. * ld-arm/thumb-b-lks-sym.d: Expected disassembly. * ld-arm/thumb-bl-lks-sym.s: New test. * ld-arm/thumb-bl-lks-sym.d: Expected disassembly. * ld-arm/arm-elf.exp: Run the new tests. Patches: http://sourceware.org/cgi-bin/cvsweb.cgi/src/bfd/ChangeLog.diff?cvsroot=srcr1=1.6073r2=1.6074 http://sourceware.org/cgi-bin/cvsweb.cgi/src/bfd/elf32-arm.c.diff?cvsroot=srcr1=1.317r2=1.318 http://sourceware.org/cgi-bin/cvsweb.cgi/src/ld/testsuite/ChangeLog.diff?cvsroot=srcr1=1.1726r2=1.1727 http://sourceware.org/cgi-bin/cvsweb.cgi/src/ld/testsuite/ld-arm/branch-lks-sym.ld.diff?cvsroot=srcr1=NONEr2=1.1 http://sourceware.org/cgi-bin/cvsweb.cgi/src/ld/testsuite/ld-arm/thumb-b-lks-sym.d.diff?cvsroot=srcr1=NONEr2=1.1 http://sourceware.org/cgi-bin/cvsweb.cgi/src/ld/testsuite/ld-arm/thumb-b-lks-sym.s.diff?cvsroot=srcr1=NONEr2=1.1 http://sourceware.org/cgi-bin/cvsweb.cgi/src/ld/testsuite/ld-arm/thumb-bl-lks-sym.d.diff?cvsroot=srcr1=NONEr2=1.1 http://sourceware.org/cgi-bin/cvsweb.cgi/src/ld/testsuite/ld-arm/thumb-bl-lks-sym.s.diff?cvsroot=srcr1=NONEr2=1.1 http://sourceware.org/cgi-bin/cvsweb.cgi/src/ld/testsuite/ld-arm/arm-elf.exp.diff?cvsroot=srcr1=1.103r2=1.104 -- 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/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302 Nick Clifton nickc at redhat dot com changed: What|Removed |Added Status|UNCONFIRMED |RESOLVED Resolution|--- |FIXED --- Comment #6 from Nick Clifton nickc at redhat dot com --- Hi Terry, Thanks for the patch - I liked it, so I have applied it! I made one small change - there was no need for the globals local variable as htab already contained the same pointer. 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 ld/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302 --- Comment #4 from Terry Guo terry.guo at arm dot com --- Created attachment 7076 -- http://sourceware.org/bugzilla/attachment.cgi?id=7076action=edit A proposal to fix this bug Hello Nick and Pavel, I am inspired by Nick's patch and managed to come up with this one. It did fixed this issue without causing any new failures in regress test. Can you guys please help to check it? Is this patch good enough to be sent to Binutils mailing list for review? -- 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/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302 Terry Guo terry.guo at arm dot com changed: What|Removed |Added CC||nickc at redhat dot com -- 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/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302
--- Comment #3 from Terry Guo terry.guo at arm dot com ---
(In reply to Nick Clifton from comment #1)
Created attachment 6968 [details]
Do not enable blx use on thumb-only architectures.
Hi Terry,
Please could you try out the uploaded patch and let me know if it works
for you.
Cheers
Nick
Apologize for my slow response. With your patch I got code like:
terguo01@terry-pc01:case$ ../install-native/bin/arm-none-eabi-objdump -d a.out
a.out: file format elf32-littlearm
Disassembly of section .text:
main:
0:b580 push{r7, lr}
2:af00 addr7, sp, #0
4:f000 f804 bl10 __extFunc_from_thumb
8:bd80 pop{r7, pc}
a:bf00 nop
c: movsr0, r0
...
0010 __extFunc_from_thumb:
10:4778 bxpc
12:46c0 nop; (mov r8, r8)
14:ea0003f9 b1000 __extFunc_from_thumb+0xff0
terguo01@terry-pc01:case$
It's same with Pavel's issue. Can you please continue to look into it?
--
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/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302 Pavel Král pavel.kral at omsquare dot com changed: What|Removed |Added Status|NEW |UNCONFIRMED CC||pavel.kral at omsquare dot ||com Ever Confirmed|1 |0 --- Comment #2 from Pavel Král pavel.kral at omsquare dot com 2013-0 5-20 21:57:34 UTC --- My use case is similar. I just did smoke test and there is another issue. T est code is executed from RAM (0x1000) where it try to call firmware stored at flash (in this example function at 0x3779) BLX instruction was replaced by BL, but at immediate address is generated thumb-ARM switch instruction BX PC which cause hard fault 1a94:f000 f898 bl1bc8 ___xxx_from_thumb # ok BLX repl aced by BL ... ___xxx_from_thumb(): 1bc8:4778 bxpc ; hard fault 1bca:46c0 nop; 1bcc:e51ff004 ldrpc, [pc, #-4]; 1bd0 _farAway_from_thumb+0x8 1bd0:3779 andeqr3, r0, r9, ror r7 ld just generate these __from_thumb stubs which is not what we wont at Cort ex Ms. -- Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email --- 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/15302] Branching to linker script symbol is resolved to improperly BLX instruction for cortex-m3
http://sourceware.org/bugzilla/show_bug.cgi?id=15302 --- Comment #1 from Nick Clifton nickc at redhat dot com 2013-04-05 17:04:22 UTC --- Created attachment 6968 -- http://sourceware.org/bugzilla/attachment.cgi?id=6968 Do not enable blx use on thumb-only architectures. Hi Terry, Please could you try out the uploaded patch and let me know if it works for you. Cheers Nick -- Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email --- 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
