-
Notifications
You must be signed in to change notification settings - Fork 14.1k
[feature][riscv] handle target address calculation in llvm-objdump disassembly for riscv #144620
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Open
arjunUpatel
wants to merge
5
commits into
llvm:main
Choose a base branch
from
arjunUpatel:riscv-address-resolution
base: main
Could not load branches
Branch not found: {{ refName }}
Loading
Could not load tags
Nothing to show
Loading
Are you sure you want to change the base?
Some commits from the old base branch may be removed from the timeline,
and old review comments may become outdated.
Open
Changes from all commits
Commits
Show all changes
5 commits
Select commit
Hold shift + click to select a range
52ebb65
Remove changes affecting non-RISCV targets
arjunUpatel 49f276e
Update test output to match previous functionality
arjunUpatel 5e1cf12
Add support for zclsd and zilsd extensions + tests
arjunUpatel 11c50dc
Pass subtargetinfo as function argument
arjunUpatel 2241583
Run clang format
arjunUpatel File filter
Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
There are no files selected for viewing
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Binary file not shown.
Binary file not shown.
Binary file not shown.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,2 @@ | ||
if not "RISCV" in config.root.targets: | ||
config.unsupported = True |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,57 @@ | ||
# RUN: llvm-objdump -d %p/Inputs/riscv-ar | FileCheck %s | ||
|
||
# CHECK: auipc a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: ld a0, {{-?0x[0-9a-fA-F]+}}(a0) | ||
# CHECK: auipc a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: addi a0, a0, {{-?0x[0-9a-fA-F]+}} <gdata> | ||
# CHECK: auipc a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: addi a0, a0, {{-?0x[0-9a-fA-F]+}} <gdata> | ||
# CHECK: auipc a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: lw a0, {{-?0x[0-9a-fA-F]+}}(a0) <gdata> | ||
# CHECK: auipc a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: addi a0, a0, {{-?0x[0-9a-fA-F]+}} <ldata> | ||
# CHECK: auipc ra, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: jalr {{-?0x[0-9a-fA-F]+}}(ra) <func> | ||
# CHECK: auipc t1, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: jr {{-?0x[0-9a-fA-F]+}}(t1) <func> | ||
# CHECK: lui a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: addiw a0, a0, {{-?0x[0-9a-fA-F]+}} <gdata+0x12242678> | ||
# CHECK: lui a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: addiw a0, a0, {{-?0x[0-9a-fA-F]+}} <gdata+0x1438ad> | ||
# CHECK: slli a0, a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: addi a0, a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: slli a0, a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: addi a0, a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: slli a0, a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: addi a0, a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: lui a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: lui a0, {{-?0x[0-9a-fA-F]+}} | ||
# CHECK: addiw a0, a0, {{-?0x[0-9a-fA-F]+}} <_start+0xfefff> | ||
|
||
.global _start | ||
.text | ||
_start: | ||
la a0, gdata | ||
lla a0, gdata | ||
lla a0, gdata | ||
lw a0, gdata | ||
lla a0, ldata | ||
|
||
call func | ||
tail func | ||
|
||
li a0, 0x12345678 | ||
li a0, 0x1234567890abcdef | ||
li a0, 0x10000 | ||
li a0, 0xfffff | ||
|
||
.skip 0x100000 | ||
func: | ||
ret | ||
|
||
ldata: | ||
.int 0 | ||
|
||
.data | ||
gdata: | ||
.int 0 |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,30 @@ | ||
# RUN: llvm-objdump -d %p/Inputs/riscv32-ar-coverage | FileCheck %s | ||
|
||
# CHECK: 00001000 <_start>: | ||
# CHECK-NEXT: 1000: 00000517 auipc a0, 0x0 | ||
# CHECK-NEXT: 1004: 0559 addi a0, a0, 0x16 <target> | ||
# CHECK-NEXT: 1006: 00000517 auipc a0, 0x0 | ||
# CHECK-NEXT: 100a: 6910 ld a2, 0x10(a0) <target> | ||
# CHECK-NEXT: 100c: 00000517 auipc a0, 0x0 | ||
# CHECK-NEXT: 1010: 00c53523 sd a2, 0xa(a0) <target> | ||
# CHECK-NEXT: 1014: 0000 unimp | ||
|
||
# the structure of this test file is similar to that of riscv64-ar-coverage | ||
# with the major difference being that these tests are focused on instructions | ||
# for 32 bit architecture | ||
|
||
.global _start | ||
.text | ||
_start: | ||
auipc a0, 0x0 | ||
addi a0, a0, 0x16 # addi -- behavior changes with differentr architectures | ||
|
||
auipc a0, 0x0 | ||
c.ld a2, 0x10(a0) # zclsd instruction | ||
|
||
auipc a0, 0x0 | ||
sd a2, 0xa(a0) # zilsd instruction | ||
|
||
.skip 0x2 | ||
target: | ||
ret: |
106 changes: 106 additions & 0 deletions
106
llvm/test/tools/llvm-objdump/RISCV/riscv64-ar-coverage.s
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Original file line number | Diff line number | Diff line change |
---|---|---|
@@ -0,0 +1,106 @@ | ||
# RUN: llvm-objdump -d %p/Inputs/riscv64-ar-coverage | FileCheck %s | ||
|
||
# CHECK: 0000000000001000 <_start>: | ||
# CHECK-NEXT: 1000: 00001517 auipc a0, 0x1 | ||
# CHECK-NEXT: 1004: 00450513 addi a0, a0, 0x4 <target> | ||
# CHECK-NEXT: 1008: 00001517 auipc a0, 0x1 | ||
# CHECK-NEXT: 100c: 1571 addi a0, a0, -0x4 <target> | ||
# CHECK-NEXT: 100e: 6509 lui a0, 0x2 | ||
# CHECK-NEXT: 1010: 0045059b addiw a1, a0, 0x4 <target> | ||
# CHECK-NEXT: 1014: 6509 lui a0, 0x2 | ||
# CHECK-NEXT: 1016: 2511 addiw a0, a0, 0x4 <target> | ||
# CHECK-NEXT: 1018: 00102537 lui a0, 0x102 | ||
# CHECK-NEXT: 101c: c50c sw a1, 0x8(a0) <far_target> | ||
# CHECK-NEXT: 101e: 00102537 lui a0, 0x102 | ||
# CHECK-NEXT: 1022: 4508 lw a0, 0x8(a0) <far_target> | ||
# CHECK-NEXT: 1024: 6509 lui a0, 0x2 | ||
# CHECK-NEXT: 1026: 6585 lui a1, 0x1 | ||
# CHECK-NEXT: 1028: 0306 slli t1, t1, 0x1 | ||
# CHECK-NEXT: 102a: 0511 addi a0, a0, 0x4 <target> | ||
# CHECK-NEXT: 102c: 0505 addi a0, a0, 0x1 | ||
# CHECK-NEXT: 102e: 00200037 lui zero, 0x200 | ||
# CHECK-NEXT: 1032: 00a02423 sw a0, 0x8(zero) | ||
# CHECK-NEXT: 1036: 00101097 auipc ra, 0x101 | ||
# CHECK-NEXT: 103a: fd6080e7 jalr -0x2a(ra) <func> | ||
# CHECK-NEXT: 103e: 640d lui s0, 0x3 | ||
# CHECK-NEXT: 1040: 8800 sb s0, 0x0(s0) <zcb> | ||
# CHECK-NEXT: 1042: 4522 lw a0, 0x8(sp) | ||
|
||
|
||
.global _start | ||
.text | ||
|
||
# The core of the feature being added was address resolution for instruction | ||
# sequences where an register is populated by immediate values via two | ||
# separate instructions. First by an instruction that provides the upper bits | ||
# (auipc, lui ...) followed by another instruction for the lower bits (addi, | ||
# jalr, ld ...). | ||
|
||
_start: | ||
# Test block 1-3 each focus on a certain starting instruction in a sequences, | ||
# the ones that provide the upper bits. The other sequence is another | ||
# instruction the provides the lower bits. The second instruction is | ||
# arbitrarily chosen to increase code coverage | ||
|
||
# test block #1 | ||
lla a0, target # addi | ||
auipc a0, 0x1 | ||
c.addi a0, -0x4 # c.addi | ||
|
||
# test block #2 | ||
c.lui a0, 0x2 | ||
addiw a1, a0, 0x4 # addiw | ||
c.lui a0, 0x2 | ||
c.addiw a0, 0x4 # c.addiw | ||
|
||
# test block #3 | ||
lui a0, 0x102 | ||
sw a1, 0x8(a0) # sw | ||
lui a0, 0x102 | ||
c.lw a0, 0x8(a0) # lw | ||
|
||
# Test block 4 tests instruction interleaving, essentially the code's | ||
# ability to keep track of a valid sequence even if multiple other unrelated | ||
# instructions separate the two | ||
|
||
# test #4 | ||
lui a0, 0x2 | ||
lui a1, 0x1 # unrelated instruction | ||
slli t1, t1, 0x1 # unrelated instruction | ||
addi a0, a0, 0x4 | ||
addi a0, a0, 0x1 # verify register tracking terminates | ||
|
||
# Test 5 ensures that an instruction writing into the zero register does | ||
# not trigger resolution because that register's value cannot change and | ||
# the sequence is equivalent to never running the first instruction | ||
|
||
# test #5 | ||
lui x0, 0x200 | ||
sw a0, 0x8(x0) | ||
|
||
# Test 6 ensures that the newly added functionality is compatible with | ||
# code that already worked for branch instructions | ||
|
||
# test #6 | ||
call func | ||
|
||
# test #7 zcb extension | ||
lui x8, 0x3 | ||
c.sb x8, 0(x8) | ||
|
||
# test #8 stack based load/stores | ||
c.lwsp a0, 0x8(sp) | ||
|
||
# these are the labels that the instructions above are expecteed to resolve to | ||
.section .data | ||
.skip 0x4 | ||
target: | ||
.word 1 | ||
.skip 0xff8 | ||
zcb: | ||
.word 1 | ||
.skip 0xff004 | ||
far_target: | ||
.word 2 | ||
func: | ||
ret |
Oops, something went wrong.
Oops, something went wrong.
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
We try hard not to introduce precanned binaries for testing. They are opaque, difficult to update, and typically increase the repo size more than text files.
If you want to test an executable, linked by lld, there is a recent example at
cross-project-tests/tools/llvm-objdump/ARM/
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I can run clang as the first run line and pass the output to the run line that currently takes as input a precanned binary. My though process was that changes to clang may affect the binary that will be emitted and lead to test failures as clang evolves. Is this something I even worry about or should I instead be having the binary built at the time the test run instead?