maxjhandsome
/
llvm-project
forked from OSchip/llvm-project
1
0
Fork 0
Code Issues Pull Requests Packages Releases Wiki Activity
llvm-project/lld/test/COFF/pdb.test

410 lines
14 KiB
Plaintext
Raw Normal View History

PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RUN: yaml2obj < %p/Inputs/pdb1.yaml > %t1.obj
# RUN: yaml2obj < %p/Inputs/pdb2.yaml > %t2.obj
Create a TPI stream only when /debugpdb is given. Adding type records to TPI stream is too time consuming. It is reported that linking chrome_child.dll took 5 minutes. llvm-svn: 289330
2016-12-11 01:23:23 +08:00
# RUN: lld-link /debug /pdb:%t.pdb /dll /out:%t.dll /entry:main /nodefaultlib \
[COFF] Fix interaction between /DEBUG and /PDB When /DEBUG is not specified, /PDB should be ignored. When /DEBUG is specified, a PDB should be output regardless of whether or not /PDB is specified. /PDB just overrides the default name. This patch implements this behavior, and adds some tests, while also removing a dead option /DEBUGPDB which was unused in any code. Differential Revision: https://reviews.llvm.org/D33302 llvm-svn: 303352
2017-05-18 23:15:10 +08:00
# RUN: %t1.obj %t2.obj
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
Rename llvm-pdbdump -> llvm-pdbutil. This is to reflect the evolving nature of the tool as being useful for more than just dumping PDBs, as it can do many other things. Differential Revision: https://reviews.llvm.org/D34062 llvm-svn: 305106
2017-06-10 04:46:17 +08:00
# RUN: llvm-pdbutil pdb2yaml -stream-metadata -stream-directory -pdb-stream \
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RUN: -dbi-stream -ipi-stream -tpi-stream %t.pdb | FileCheck %s
Check /pdb output using llvm-pdbdump. test/COFF/rsds.test checks only RSDS directory in a DLL and didn't check the validity of the PDF file produced. (Technically the produced PDB is not valid because it is really a stub, but this test is still good to have.) llvm-svn: 281678
2016-09-16 07:14:40 +08:00
Rename llvm-pdbdump -> llvm-pdbutil. This is to reflect the evolving nature of the tool as being useful for more than just dumping PDBs, as it can do many other things. Differential Revision: https://reviews.llvm.org/D34062 llvm-svn: 305106
2017-06-10 04:46:17 +08:00
# RUN: llvm-pdbutil raw -modules -section-map -section-headers -section-contribs \
[PDB] Re-add RAW FileCheck testing to pdb.test Update the expected output to match the current llvm-pdbdump raw output. llvm-svn: 299793
2017-04-08 02:38:38 +08:00
# RUN: -tpi-records %t.pdb | FileCheck -check-prefix RAW %s
Check /pdb output using llvm-pdbdump. test/COFF/rsds.test checks only RSDS directory in a DLL and didn't check the validity of the PDF file produced. (Technically the produced PDB is not valid because it is really a stub, but this test is still good to have.) llvm-svn: 281678
2016-09-16 07:14:40 +08:00
# CHECK: MSF:
# CHECK-NEXT: SuperBlock:
Use functions in DebugInfoPDB to create dummy PDB file. I don't think we are creating valid PDB file here, but it is okay because we have never created valid PDBs before. llvm-svn: 281704
2016-09-16 12:32:33 +08:00
# CHECK-NEXT: BlockSize: 4096
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# CHECK-NEXT: FreeBlockMap: 1
Fix lld pdb test. It was assuming various things about the PDB like file size which are going to be too high maintenance to maintain in a test. llvm-svn: 297908
2017-03-16 06:58:10 +08:00
# CHECK-NEXT: NumBlocks:
# CHECK-NEXT: NumDirectoryBytes:
Use functions in DebugInfoPDB to create dummy PDB file. I don't think we are creating valid PDB file here, but it is okay because we have never created valid PDBs before. llvm-svn: 281704
2016-09-16 12:32:33 +08:00
# CHECK-NEXT: Unknown1: 0
Fix lld pdb test. It was assuming various things about the PDB like file size which are going to be too high maintenance to maintain in a test. llvm-svn: 297908
2017-03-16 06:58:10 +08:00
# CHECK-NEXT: BlockMapAddr:
# CHECK-NEXT: NumDirectoryBlocks:
# CHECK-NEXT: DirectoryBlocks:
# CHECK-NEXT: NumStreams:
# CHECK-NEXT: FileSize:
# CHECK-NEXT: StreamSizes:
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: StreamMap:
Fix lld pdb test. It was assuming various things about the PDB like file size which are going to be too high maintenance to maintain in a test. llvm-svn: 297908
2017-03-16 06:58:10 +08:00
# CHECK: PdbStream:
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: Age: 1
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# CHECK-NEXT: Guid:
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: Signature: 0
# CHECK-NEXT: Version: VC70
# CHECK-NEXT: DbiStream:
# CHECK-NEXT: VerHeader: V110
# CHECK-NEXT: Age: 1
# CHECK-NEXT: BuildNumber: 0
# CHECK-NEXT: PdbDllVersion: 0
# CHECK-NEXT: PdbDllRbld: 0
# CHECK-NEXT: Flags: 0
# CHECK-NEXT: MachineType: x86
# CHECK-NEXT: TpiStream:
# CHECK-NEXT: Version: VC80
# CHECK-NEXT: Records:
# CHECK-NEXT: - Kind: LF_ARGLIST
# CHECK-NEXT: ArgList:
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# CHECK-NEXT: ArgIndices: [ ]
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: - Kind: LF_PROCEDURE
# CHECK-NEXT: Procedure:
# CHECK-NEXT: ReturnType: 116
# CHECK-NEXT: CallConv: NearC
# CHECK-NEXT: Options: [ None ]
# CHECK-NEXT: ParameterCount: 0
# CHECK-NEXT: ArgumentList: 4096
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# CHECK-NEXT: - Kind: LF_POINTER
# CHECK-NEXT: Pointer:
# CHECK-NEXT: ReferentType: 4097
# CHECK-NEXT: Attrs: 65548
# CHECK-NEXT: - Kind: LF_ARGLIST
# CHECK-NEXT: ArgList:
# CHECK-NEXT: ArgIndices: [ 0 ]
# CHECK-NEXT: - Kind: LF_PROCEDURE
# CHECK-NEXT: Procedure:
# CHECK-NEXT: ReturnType: 116
# CHECK-NEXT: CallConv: NearC
# CHECK-NEXT: Options: [ None ]
# CHECK-NEXT: ParameterCount: 0
# CHECK-NEXT: ArgumentList: 4099
[PDB] Split item and type records when merging type streams Summary: MSVC does this when producing a PDB. Reviewers: ruiu Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D31316 llvm-svn: 298717
2017-03-25 01:26:38 +08:00
# CHECK-NEXT: IpiStream:
# CHECK-NEXT: Version: VC80
# CHECK-NEXT: Records:
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: - Kind: LF_FUNC_ID
# CHECK-NEXT: FuncId:
# CHECK-NEXT: ParentScope: 0
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# CHECK-NEXT: FunctionType: 4100
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: Name: main
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# CHECK-NEXT: - Kind: LF_FUNC_ID
# CHECK-NEXT: FuncId:
# CHECK-NEXT: ParentScope: 0
# CHECK-NEXT: FunctionType: 4097
# CHECK-NEXT: Name: foo
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: - Kind: LF_STRING_ID
# CHECK-NEXT: StringId:
# CHECK-NEXT: Id: 0
# CHECK-NEXT: String: 'D:\b'
# CHECK-NEXT: - Kind: LF_STRING_ID
# CHECK-NEXT: StringId:
# CHECK-NEXT: Id: 0
# CHECK-NEXT: String: 'C:\vs14\VC\BIN\amd64\cl.exe'
# CHECK-NEXT: - Kind: LF_STRING_ID
# CHECK-NEXT: StringId:
# CHECK-NEXT: Id: 0
# CHECK-NEXT: String: '-Z7 -c -MT -IC:\vs14\VC\INCLUDE -IC:\vs14\VC\ATLMFC\INCLUDE -I"C:\Program Files (x86)\Windows Kits\10\include\10.0.10150.0\ucrt" -I"C:\Program Files (x86)\Windows Kits\NETFXSDK\4.6\include\um" -I"C:\Program Files (x86)\Windows Kits\8.1\include\shared"'
# CHECK-NEXT: - Kind: LF_SUBSTR_LIST
Update LLD tests for codeview dumping changes llvm-svn: 298475
2017-03-22 09:43:49 +08:00
# CHECK-NEXT: StringList:
[PDB] Split item and type records when merging type streams Summary: MSVC does this when producing a PDB. Reviewers: ruiu Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D31316 llvm-svn: 298717
2017-03-25 01:26:38 +08:00
# CHECK-NEXT: StringIndices: [ 4100 ]
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: - Kind: LF_STRING_ID
# CHECK-NEXT: StringId:
[PDB] Split item and type records when merging type streams Summary: MSVC does this when producing a PDB. Reviewers: ruiu Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D31316 llvm-svn: 298717
2017-03-25 01:26:38 +08:00
# CHECK-NEXT: Id: 4101
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: String: ' -I"C:\Program Files (x86)\Windows Kits\8.1\include\um" -I"C:\Program Files (x86)\Windows Kits\8.1\include\winrt" -TC -X'
# CHECK-NEXT: - Kind: LF_STRING_ID
# CHECK-NEXT: StringId:
# CHECK-NEXT: Id: 0
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# CHECK-NEXT: String: ret42-main.c
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# CHECK-NEXT: - Kind: LF_STRING_ID
# CHECK-NEXT: StringId:
# CHECK-NEXT: Id: 0
# CHECK-NEXT: String: 'D:\b\vc140.pdb'
# CHECK-NEXT: - Kind: LF_BUILDINFO
# CHECK-NEXT: BuildInfo:
[PDB] Split item and type records when merging type streams Summary: MSVC does this when producing a PDB. Reviewers: ruiu Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D31316 llvm-svn: 298717
2017-03-25 01:26:38 +08:00
# CHECK-NEXT: ArgIndices: [ 4098, 4099, 4103, 4104, 4102 ]
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# CHECK-NEXT: - Kind: LF_STRING_ID
# CHECK-NEXT: StringId:
# CHECK-NEXT: Id: 0
# CHECK-NEXT: String: ret42-sub.c
# CHECK-NEXT: - Kind: LF_BUILDINFO
# CHECK-NEXT: BuildInfo:
[PDB] Split item and type records when merging type streams Summary: MSVC does this when producing a PDB. Reviewers: ruiu Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D31316 llvm-svn: 298717
2017-03-25 01:26:38 +08:00
# CHECK-NEXT: ArgIndices: [ 4098, 4099, 4106, 4104, 4102 ]
Check /pdb output using llvm-pdbdump. test/COFF/rsds.test checks only RSDS directory in a DLL and didn't check the validity of the PDF file produced. (Technically the produced PDB is not valid because it is really a stub, but this test is still good to have.) llvm-svn: 281678
2016-09-16 07:14:40 +08:00
[PDB] Re-add RAW FileCheck testing to pdb.test Update the expected output to match the current llvm-pdbdump raw output. llvm-svn: 299793
2017-04-08 02:38:38 +08:00
# RAW: Type Info Stream (TPI) {
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RAW-NEXT: TPI Version: 20040203
[PDB] Re-add RAW FileCheck testing to pdb.test Update the expected output to match the current llvm-pdbdump raw output. llvm-svn: 299793
2017-04-08 02:38:38 +08:00
# RAW-NEXT: Record count: 5
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RAW-NEXT: Records [
# RAW-NEXT: {
# RAW-NEXT: ArgList (0x1000) {
# RAW-NEXT: TypeLeafKind: LF_ARGLIST (0x1201)
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: NumArgs: 0
# RAW-NEXT: Arguments [
# RAW-NEXT: ]
# RAW-NEXT: }
# RAW-NEXT: }
# RAW-NEXT: {
# RAW-NEXT: Procedure (0x1001) {
# RAW-NEXT: TypeLeafKind: LF_PROCEDURE (0x1008)
# RAW-NEXT: ReturnType: int (0x74)
# RAW-NEXT: CallingConvention: NearC (0x0)
# RAW-NEXT: FunctionOptions [ (0x0)
# RAW-NEXT: ]
# RAW-NEXT: NumParameters: 0
# RAW-NEXT: ArgListType: () (0x1000)
# RAW-NEXT: }
# RAW-NEXT: }
# RAW-NEXT: {
# RAW-NEXT: Pointer (0x1002) {
# RAW-NEXT: TypeLeafKind: LF_POINTER (0x1002)
# RAW-NEXT: PointeeType: int () (0x1001)
# RAW-NEXT: PointerAttributes: 0x1000C
# RAW-NEXT: PtrType: Near64 (0xC)
# RAW-NEXT: PtrMode: Pointer (0x0)
# RAW-NEXT: IsFlat: 0
# RAW-NEXT: IsConst: 0
# RAW-NEXT: IsVolatile: 0
# RAW-NEXT: IsUnaligned: 0
# RAW-NEXT: SizeOf: 8
# RAW-NEXT: }
# RAW-NEXT: }
# RAW-NEXT: {
# RAW-NEXT: ArgList (0x1003) {
# RAW-NEXT: TypeLeafKind: LF_ARGLIST (0x1201)
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RAW-NEXT: NumArgs: 1
# RAW-NEXT: Arguments [
# RAW-NEXT: ArgType: 0x0
# RAW-NEXT: ]
# RAW-NEXT: }
# RAW-NEXT: }
# RAW-NEXT: {
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: Procedure (0x1004) {
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RAW-NEXT: TypeLeafKind: LF_PROCEDURE (0x1008)
# RAW-NEXT: ReturnType: int (0x74)
# RAW-NEXT: CallingConvention: NearC (0x0)
# RAW-NEXT: FunctionOptions [ (0x0)
# RAW-NEXT: ]
# RAW-NEXT: NumParameters: 0
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: ArgListType: (<no type>) (0x1003)
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RAW-NEXT: }
# RAW-NEXT: }
[PDB] Re-add RAW FileCheck testing to pdb.test Update the expected output to match the current llvm-pdbdump raw output. llvm-svn: 299793
2017-04-08 02:38:38 +08:00
# RAW-NEXT: TypeIndexOffsets [
[PDB] Emit index/offset pairs for TPI and IPI streams Summary: This lets PDB readers lookup type record data by type index in O(log n) time. It also enables makes `cvdump -t` work on PDBs produced by LLD. cvdump will not dump a PDB that doesn't have an index-to-offset table. The table is sorted by type index, and has an entry every 8KB. Looking up a type record by index is a binary search of this table, followed by a scan of at most 8KB. Reviewers: ruiu, zturner, inglorion Subscribers: llvm-commits Differential Revision: https://reviews.llvm.org/D31636 llvm-svn: 299958
2017-04-12 00:26:15 +08:00
# RAW-NEXT: Index: 0x1000, Offset: 0
[PDB] Re-add RAW FileCheck testing to pdb.test Update the expected output to match the current llvm-pdbdump raw output. llvm-svn: 299793
2017-04-08 02:38:38 +08:00
# RAW-NEXT: ]
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RAW-NEXT: ]
# RAW-NEXT: }
# RAW-NEXT: DBI Stream {
PDB: Add "* Linker *" module. The added module contains nothing, but it is still useful as a test to ensure that we are emitting modules that can be read back. llvm-svn: 287071
2016-11-16 09:10:46 +08:00
# RAW-NEXT: Dbi Version: 20091201
# RAW-NEXT: Age: 1
# RAW-NEXT: Incremental Linking: No
# RAW-NEXT: Has CTypes: No
# RAW-NEXT: Is Stripped: No
# RAW-NEXT: Machine Type: x86
# RAW-NEXT: Symbol Record Stream Index: 65535
# RAW-NEXT: Public Symbol Stream Index: 65535
# RAW-NEXT: Global Symbol Stream Index: 65535
# RAW-NEXT: Toolchain Version: 0.0
# RAW-NEXT: mspdb00.dll version: 0.0.0
# RAW-NEXT: Modules [
# RAW-NEXT: {
# RAW-NEXT: Name: * Linker *
[PDB] Re-add RAW FileCheck testing to pdb.test Update the expected output to match the current llvm-pdbdump raw output. llvm-svn: 299793
2017-04-08 02:38:38 +08:00
# RAW-NEXT: Debug Stream Index: 9
PDB: Add "* Linker *" module. The added module contains nothing, but it is still useful as a test to ensure that we are emitting modules that can be read back. llvm-svn: 287071
2016-11-16 09:10:46 +08:00
# RAW-NEXT: Object File Name:
# RAW-NEXT: Num Files: 0
# RAW-NEXT: Source File Name Idx: 0
# RAW-NEXT: Pdb File Name Idx: 0
# RAW-NEXT: Line Info Byte Size: 0
# RAW-NEXT: C13 Line Info Byte Size: 0
[PDB] Re-add RAW FileCheck testing to pdb.test Update the expected output to match the current llvm-pdbdump raw output. llvm-svn: 299793
2017-04-08 02:38:38 +08:00
# RAW-NEXT: Symbol Byte Size: 4
PDB: Add "* Linker *" module. The added module contains nothing, but it is still useful as a test to ensure that we are emitting modules that can be read back. llvm-svn: 287071
2016-11-16 09:10:46 +08:00
# RAW-NEXT: Type Server Index: 0
# RAW-NEXT: Has EC Info: No
# RAW-NEXT: }
# RAW-NEXT: ]
# RAW-NEXT: }
# RAW-NEXT: Section Contributions [
Emit Section Contributions. Differential Revision: https://reviews.llvm.org/D26211 llvm-svn: 286684
2016-11-12 08:00:51 +08:00
# RAW-NEXT: Contribution {
# RAW-NEXT: ISect: 0
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: Off: 1288
# RAW-NEXT: Size: 14
# RAW-NEXT: Characteristics [ (0x60500020)
# RAW-NEXT: IMAGE_SCN_ALIGN_16BYTES (0x500000)
# RAW-NEXT: IMAGE_SCN_CNT_CODE (0x20)
# RAW-NEXT: IMAGE_SCN_MEM_EXECUTE (0x20000000)
# RAW-NEXT: IMAGE_SCN_MEM_READ (0x40000000)
# RAW-NEXT: ]
# RAW-NEXT: Module {
# RAW-NEXT: Index: 0
# RAW-NEXT: Name: * Linker *
# RAW-NEXT: }
# RAW-NEXT: Data CRC: 0
# RAW-NEXT: Reloc CRC: 0
# RAW-NEXT: }
# RAW-NEXT: Contribution {
# RAW-NEXT: ISect: 0
# RAW-NEXT: Off: 1312
# RAW-NEXT: Size: 8
# RAW-NEXT: Characteristics [ (0x40300040)
# RAW-NEXT: IMAGE_SCN_ALIGN_4BYTES (0x300000)
# RAW-NEXT: IMAGE_SCN_CNT_INITIALIZED_DATA (0x40)
# RAW-NEXT: IMAGE_SCN_MEM_READ (0x40000000)
# RAW-NEXT: ]
# RAW-NEXT: Module {
# RAW-NEXT: Index: 0
# RAW-NEXT: Name: * Linker *
# RAW-NEXT: }
# RAW-NEXT: Data CRC: 0
# RAW-NEXT: Reloc CRC: 0
# RAW-NEXT: }
# RAW-NEXT: Contribution {
# RAW-NEXT: ISect: 0
# RAW-NEXT: Off: 1320
# RAW-NEXT: Size: 12
# RAW-NEXT: Characteristics [ (0x40300040)
# RAW-NEXT: IMAGE_SCN_ALIGN_4BYTES (0x300000)
# RAW-NEXT: IMAGE_SCN_CNT_INITIALIZED_DATA (0x40)
# RAW-NEXT: IMAGE_SCN_MEM_READ (0x40000000)
# RAW-NEXT: ]
# RAW-NEXT: Module {
# RAW-NEXT: Index: 0
# RAW-NEXT: Name: * Linker *
# RAW-NEXT: }
# RAW-NEXT: Data CRC: 0
# RAW-NEXT: Reloc CRC: 0
# RAW-NEXT: }
# RAW-NEXT: Contribution {
# RAW-NEXT: ISect: 0
# RAW-NEXT: Off: 1144
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RAW-NEXT: Size: 6
# RAW-NEXT: Characteristics [ (0x60500020)
# RAW-NEXT: IMAGE_SCN_ALIGN_16BYTES (0x500000)
Emit Section Contributions. Differential Revision: https://reviews.llvm.org/D26211 llvm-svn: 286684
2016-11-12 08:00:51 +08:00
# RAW-NEXT: IMAGE_SCN_CNT_CODE (0x20)
# RAW-NEXT: IMAGE_SCN_MEM_EXECUTE (0x20000000)
# RAW-NEXT: IMAGE_SCN_MEM_READ (0x40000000)
# RAW-NEXT: ]
# RAW-NEXT: Module {
# RAW-NEXT: Index: 0
PDB: Add "* Linker *" module. The added module contains nothing, but it is still useful as a test to ensure that we are emitting modules that can be read back. llvm-svn: 287071
2016-11-16 09:10:46 +08:00
# RAW-NEXT: Name: * Linker *
Emit Section Contributions. Differential Revision: https://reviews.llvm.org/D26211 llvm-svn: 286684
2016-11-12 08:00:51 +08:00
# RAW-NEXT: }
# RAW-NEXT: Data CRC: 0
# RAW-NEXT: Reloc CRC: 0
# RAW-NEXT: }
# RAW-NEXT: ]
# RAW-NEXT: Section Map [
# RAW-NEXT: Entry {
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: Flags [ (0x109)
# RAW-NEXT: AddressIs32Bit (0x8)
# RAW-NEXT: IsSelector (0x100)
# RAW-NEXT: Read (0x1)
# RAW-NEXT: ]
# RAW-NEXT: Ovl: 0
# RAW-NEXT: Group: 0
# RAW-NEXT: Frame: 1
# RAW-NEXT: SecName: 65535
# RAW-NEXT: ClassName: 65535
# RAW-NEXT: Offset: 0
# RAW-NEXT: SecByteLength: 12
# RAW-NEXT: }
# RAW-NEXT: Entry {
Emit Section Contributions. Differential Revision: https://reviews.llvm.org/D26211 llvm-svn: 286684
2016-11-12 08:00:51 +08:00
# RAW-NEXT: Flags [ (0x10D)
# RAW-NEXT: AddressIs32Bit (0x8)
Re-commit r285607: Emit Section Map stream." Removed some test fields that overspecified test conditions. llvm-svn: 285638
2016-11-01 05:09:21 +08:00
# RAW-NEXT: Execute (0x4)
# RAW-NEXT: IsSelector (0x100)
# RAW-NEXT: Read (0x1)
# RAW-NEXT: ]
# RAW-NEXT: Ovl: 0
# RAW-NEXT: Group: 0
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: Frame: 2
Re-commit r285607: Emit Section Map stream." Removed some test fields that overspecified test conditions. llvm-svn: 285638
2016-11-01 05:09:21 +08:00
# RAW-NEXT: SecName: 65535
# RAW-NEXT: ClassName: 65535
# RAW-NEXT: Offset: 0
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: SecByteLength: 22
Re-commit r285607: Emit Section Map stream." Removed some test fields that overspecified test conditions. llvm-svn: 285638
2016-11-01 05:09:21 +08:00
# RAW-NEXT: }
# RAW-NEXT: Entry {
# RAW-NEXT: Flags [ (0x109)
# RAW-NEXT: AddressIs32Bit (0x8)
# RAW-NEXT: IsSelector (0x100)
# RAW-NEXT: Read (0x1)
# RAW-NEXT: ]
# RAW-NEXT: Ovl: 0
# RAW-NEXT: Group: 0
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: Frame: 3
Re-commit r285607: Emit Section Map stream." Removed some test fields that overspecified test conditions. llvm-svn: 285638
2016-11-01 05:09:21 +08:00
# RAW-NEXT: SecName: 65535
# RAW-NEXT: ClassName: 65535
# RAW-NEXT: Offset: 0
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: SecByteLength: 8
# RAW-NEXT: }
# RAW-NEXT: Entry {
# RAW-NEXT: Flags [ (0x109)
# RAW-NEXT: AddressIs32Bit (0x8)
# RAW-NEXT: IsSelector (0x100)
# RAW-NEXT: Read (0x1)
# RAW-NEXT: ]
# RAW-NEXT: Ovl: 0
# RAW-NEXT: Group: 0
# RAW-NEXT: Frame: 4
# RAW-NEXT: SecName: 65535
# RAW-NEXT: ClassName: 65535
# RAW-NEXT: Offset: 0
Remove .rdata section size checks from pdb.test, they are not the same on Windows and Unix llvm-svn: 299795
2017-04-08 03:58:33 +08:00
# RAW-NEXT: SecByteLength:
Re-commit r285607: Emit Section Map stream." Removed some test fields that overspecified test conditions. llvm-svn: 285638
2016-11-01 05:09:21 +08:00
# RAW-NEXT: }
# RAW-NEXT: Entry {
# RAW-NEXT: Flags [ (0x208)
# RAW-NEXT: AddressIs32Bit (0x8)
# RAW-NEXT: IsAbsoluteAddress (0x200)
# RAW-NEXT: ]
# RAW-NEXT: Ovl: 0
# RAW-NEXT: Group: 0
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: Frame: 5
Re-commit r285607: Emit Section Map stream." Removed some test fields that overspecified test conditions. llvm-svn: 285638
2016-11-01 05:09:21 +08:00
# RAW-NEXT: SecName: 65535
# RAW-NEXT: ClassName: 65535
# RAW-NEXT: Offset: 0
# RAW-NEXT: SecByteLength: 4294967295
# RAW-NEXT: }
# RAW-NEXT: ]
# RAW-NEXT: Section Headers [
Re-submit r283825: Add section header stream to PDB. It was reverted because the change that depends on was reverted. Now it was submitted as r283925, so we can submit this as well. llvm-svn: 283926
2016-10-12 03:45:07 +08:00
# RAW-NEXT: {
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: Name: .pdata
# RAW-NEXT: Virtual Size: 12
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RAW-NEXT: Virtual Address: 4096
# RAW-NEXT: Size of Raw Data: 512
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: File Pointer to Raw Data: 1024
# RAW-NEXT: File Pointer to Relocations: 0
# RAW-NEXT: File Pointer to Linenumbers: 0
# RAW-NEXT: Number of Relocations: 0
# RAW-NEXT: Number of Linenumbers: 0
# RAW-NEXT: Characteristics [ (0x40000040)
# RAW-NEXT: IMAGE_SCN_CNT_INITIALIZED_DATA (0x40)
# RAW-NEXT: IMAGE_SCN_MEM_READ (0x40000000)
# RAW-NEXT: ]
# RAW-NEXT: }
# RAW-NEXT: {
# RAW-NEXT: Name: .text
# RAW-NEXT: Virtual Size: 22
# RAW-NEXT: Virtual Address: 8192
# RAW-NEXT: Size of Raw Data: 512
# RAW-NEXT: File Pointer to Raw Data: 1536
Re-submit r283825: Add section header stream to PDB. It was reverted because the change that depends on was reverted. Now it was submitted as r283925, so we can submit this as well. llvm-svn: 283926
2016-10-12 03:45:07 +08:00
# RAW-NEXT: File Pointer to Relocations: 0
# RAW-NEXT: File Pointer to Linenumbers: 0
# RAW-NEXT: Number of Relocations: 0
# RAW-NEXT: Number of Linenumbers: 0
# RAW-NEXT: Characteristics [ (0x60000020)
# RAW-NEXT: IMAGE_SCN_CNT_CODE (0x20)
# RAW-NEXT: IMAGE_SCN_MEM_EXECUTE (0x20000000)
# RAW-NEXT: IMAGE_SCN_MEM_READ (0x40000000)
# RAW-NEXT: ]
# RAW-NEXT: }
# RAW-NEXT: {
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: Name: .xdata
# RAW-NEXT: Virtual Size: 8
# RAW-NEXT: Virtual Address: 12288
# RAW-NEXT: Size of Raw Data: 512
# RAW-NEXT: File Pointer to Raw Data: 2048
# RAW-NEXT: File Pointer to Relocations: 0
# RAW-NEXT: File Pointer to Linenumbers: 0
# RAW-NEXT: Number of Relocations: 0
# RAW-NEXT: Number of Linenumbers: 0
# RAW-NEXT: Characteristics [ (0x40000040)
# RAW-NEXT: IMAGE_SCN_CNT_INITIALIZED_DATA (0x40)
# RAW-NEXT: IMAGE_SCN_MEM_READ (0x40000000)
# RAW-NEXT: ]
# RAW-NEXT: }
# RAW-NEXT: {
Re-submit r283825: Add section header stream to PDB. It was reverted because the change that depends on was reverted. Now it was submitted as r283925, so we can submit this as well. llvm-svn: 283926
2016-10-12 03:45:07 +08:00
# RAW-NEXT: Name: .rdata
Remove .rdata section size checks from pdb.test, they are not the same on Windows and Unix llvm-svn: 299795
2017-04-08 03:58:33 +08:00
# RAW-NEXT: Virtual Size:
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: Virtual Address: 16384
Add type records to TPI stream. I don't think the data I add to a TPI stream in this patch is correct, but at least it can be displayed using llvm-pdbdump. Until I add more streams to a PDB file, I'm not able to know whether the data will be accepted by MSVC tools or not. llvm-svn: 289183
2016-12-09 12:46:54 +08:00
# RAW-NEXT: Size of Raw Data: 512
PDB: Merge type info. This patch is to merge type info in multiple .debug$T sections. One mystery that needs to be solved is that it is not clear how the MSVC linker uses TPI and IPI streams. Both streams contain type info, and it is not obvious what kind of record should go which. dumppdb command in microsoft-pdb repository prints out IPI stream contents as "IDs" and TPI stream as "TYPES", but looks like the tool don't really care about which stream type recrods were read from. For now, in this patch, I emit all type records to TPI stream. It might just work with other tools. If not, we need to investigate it more. llvm-svn: 291739
2017-01-12 11:09:25 +08:00
# RAW-NEXT: File Pointer to Raw Data: 2560
Re-submit r283825: Add section header stream to PDB. It was reverted because the change that depends on was reverted. Now it was submitted as r283925, so we can submit this as well. llvm-svn: 283926
2016-10-12 03:45:07 +08:00
# RAW-NEXT: File Pointer to Relocations: 0
# RAW-NEXT: File Pointer to Linenumbers: 0
# RAW-NEXT: Number of Relocations: 0
# RAW-NEXT: Number of Linenumbers: 0
# RAW-NEXT: Characteristics [ (0x40000040)
# RAW-NEXT: IMAGE_SCN_CNT_INITIALIZED_DATA (0x40)
# RAW-NEXT: IMAGE_SCN_MEM_READ (0x40000000)
# RAW-NEXT: ]
# RAW-NEXT: }
# RAW-NEXT: ]