d865f32fe8
This patch uses the previously build infrastructure to parse multiple FDE entries into a single unwind plan. There is one catch though: we parse only one FDE entry per unwind range. This is not fully correct because lldb coalesces adjecant address ranges, which means that something that originally looked like two separate address ranges (and two FDE entries) may get merged into one because if the linker decides to put the two ranges next to each other. In this case, we will ignore the second FDE entry. It would be more correct to try to parse another entry when the one we found turns out to be short, but I'm not doing this (yet), because: - this is how we've done things so far (although, monolithic functions are unlikely to have more than one FDE entry) - in cases where we don't have debug info or (full) symbol tables, we can end up with "symbols" which appear to span many megabytes (potentially, the whole module). If we tried to fill short FDE entries, we could end up parsing the entire eh_frame section in a single go. In a way, this would be more correct, but it would also probably be very slow. I haven't quite decided what to do about this case yet, though it's not particularly likely to happen in the "production" cases as typically the functions are split into two parts (hot/cold) instead of one part per basic block.
49 lines
2.5 KiB
Plaintext
49 lines
2.5 KiB
Plaintext
# Test unwind info for functions which have been split into two or more parts.
|
|
# In particular, check we return the correct set of unwind plans when querying
|
|
# addresses which are in the middle of discontinuous functions.
|
|
|
|
# REQUIRES: lld, target-x86_64
|
|
|
|
# RUN: llvm-mc -triple=x86_64-pc-linux -filetype=obj \
|
|
# RUN: %S/Inputs/basic-block-sections-with-dwarf.s > %t.o
|
|
# RUN: ld.lld %t.o -o %t
|
|
## NB: This minidump exists only as a receptacle for the object file built
|
|
## above. This is a workaround for the fact that "image show-unwind" does not
|
|
## work without a Process object.
|
|
# RUN: yaml2obj %S/Inputs/linux-x86_64.yaml > %t.core
|
|
# RUN: %lldb -c %t.core %t -o "image load --file %t --slide 0" -s %s -o exit | \
|
|
# RUN: FileCheck --implicit-check-not="UNWIND PLANS" %s
|
|
|
|
image show-unwind --cached true -n foo
|
|
# CHECK: UNWIND PLANS for {{.*}}`foo
|
|
#
|
|
# CHECK: eh_frame UnwindPlan:
|
|
# CHECK-NEXT: This UnwindPlan originally sourced from eh_frame CFI
|
|
# CHECK-NEXT: This UnwindPlan is sourced from the compiler: yes.
|
|
# CHECK-NEXT: This UnwindPlan is valid at all instruction locations: no.
|
|
# CHECK-NEXT: This UnwindPlan is for a trap handler function: no.
|
|
# CHECK-NEXT: Address range of this UnwindPlan: [{{.*}}.text + 6-0x0000000000000010)[{{.*}}.text + 17-0x000000000000001c)[{{.*}}.text + 44-0x0000000000000037)[{{.*}}.text + 56-0x000000000000003d)
|
|
# CHECK-NEXT: row[0]: 0: CFA=rsp +8 => rip=[CFA-8]
|
|
# CHECK-NEXT: row[1]: 1: CFA=rsp+16 => rbx=[CFA-16] rip=[CFA-8]
|
|
# CHECK-NEXT: row[2]: 11: CFA=rsp+16 => rbx=[CFA-16] rip=[CFA-8]
|
|
# CHECK-NEXT: row[3]: 15: CFA=rsp+32 => rbx=[CFA-16] rip=[CFA-8]
|
|
# CHECK-NEXT: row[4]: 38: CFA=rsp+16 => rbx=[CFA-16] rip=[CFA-8]
|
|
# CHECK-NEXT: row[5]: 42: CFA=rsp+32 => rbx=[CFA-16] rip=[CFA-8]
|
|
# CHECK-NEXT: row[6]: 50: CFA=rsp+32 => rbx=[CFA-16] rip=[CFA-8]
|
|
# CHECK-NEXT: row[7]: 54: CFA=rsp +8 => rbx=[CFA-16] rip=[CFA-8]
|
|
# CHECK-EMPTY:
|
|
|
|
image show-unwind --cached true -n bar
|
|
# CHECK: UNWIND PLANS for {{.*}}`bar
|
|
|
|
# CHECK: eh_frame UnwindPlan:
|
|
# CHECK-NEXT: This UnwindPlan originally sourced from eh_frame CFI
|
|
# CHECK-NEXT: This UnwindPlan is sourced from the compiler: yes.
|
|
# CHECK-NEXT: This UnwindPlan is valid at all instruction locations: no.
|
|
# CHECK-NEXT: This UnwindPlan is for a trap handler function: no.
|
|
# CHECK-NEXT: Address range of this UnwindPlan: [{{.*}}.text + 28-0x000000000000002c)
|
|
# CHECK-NEXT: row[0]: 0: CFA=rsp +8 => rip=[CFA-8]
|
|
# CHECK-NEXT: row[1]: 4: CFA=rsp+32 => rip=[CFA-8]
|
|
# CHECK-NEXT: row[2]: 15: CFA=rsp +8 => rip=[CFA-8]
|
|
# CHECK-EMPTY:
|