Commit Graph

18 Commits

Author SHA1 Message Date
Peter Collingbourne 66f1c9a858 Reland r330223, "COFF: Merge .idata, .didat and .edata into .rdata by default.", which was reverted in r330228.
In this reland I removed an unnecessary use of /debug in the test
delayimports32.test and used the /pdbaltpath flag in the test
pdb-publics-import.test, both of which avoid embedding absolute PDB
paths in executables which could affect later RVAs.

Original commit message:
> COFF: Merge .idata, .didat and .edata into .rdata by default.
>
> This saves a little space and matches what link.exe does.
>
> Tested using the chromium Windows trybots:
> https://chromium-review.googlesource.com/c/chromium/src/+/1014784

Differential Revision: https://reviews.llvm.org/D45737

llvm-svn: 330233
2018-04-17 23:28:52 +00:00
Peter Collingbourne 1254f3e77c Revert r330223, "COFF: Merge .idata, .didat and .edata into .rdata by default."
Seems to have uncovered some sort of non-determinism on the bots.

llvm-svn: 330228
2018-04-17 22:16:39 +00:00
Peter Collingbourne 09e0e2e656 COFF: Merge .idata, .didat and .edata into .rdata by default.
This saves a little space and matches what link.exe does.

Tested using the chromium Windows trybots:
https://chromium-review.googlesource.com/c/chromium/src/+/1014784

Differential Revision: https://reviews.llvm.org/D45737

llvm-svn: 330223
2018-04-17 21:44:31 +00:00
Hans Wennborg 9a9fc78744 COFF: Layout sections in the same order as link.exe
One place where this seems to matter is to make sure the .rsrc section comes
after .text. The Win32 UpdateResource() function can change the contents of
.rsrc. It will move the sections that come after, but if .text gets moved, the
entry point header will not get updated and the executable breaks. This was
found by a test in Chromium.

Differential Revision: https://reviews.llvm.org/D45260

llvm-svn: 329221
2018-04-04 19:15:55 +00:00
Peter Collingbourne 5e80bdebd2 COFF: Stop emitting a non-standard COFF symbol table into PEs.
Now that our support for PDB emission is reasonably good, there is
no longer a need to emit a COFF symbol table.

Also fix a bug where we would fail to emit a string table for long
section names if /debug was not specified.

Differential Revision: https://reviews.llvm.org/D40189

llvm-svn: 318548
2017-11-17 19:51:20 +00:00
Jon Chesterfield e0ca2ff070 [LLD] Mark a number of x86 only tests to require x86
Noticed while testing for an out of tree target. There are probably more tests that should be so marked.
I'm not sure who owns these tests so I've added a few names I recognise from the recent history.

With advice from probinson, ruiu, rafael and dramatically improved by davidb. Thank you all!

Differential Revision: https://reviews.llvm.org/D34685

llvm-svn: 308335
2017-07-18 18:40:50 +00:00
Saleem Abdulrasool 8fcff934ba COFF: add beginnings of debug directory creation
The IMAGE_FILE_HEADER structure contains a (RVA, size) to an array of
COFF_DEBUG_DIRECTORY records. Each one of these records contains an RVA to a OMF
Debug Directory. These OMF debug directories are derived into newer types such
as PDB70, PDB20, etc. This constructs a PDB70 structure which will allow us to
associate a GUID with a build to actually tie debug information.

llvm-svn: 280012
2016-08-29 21:20:46 +00:00
Rui Ueyama 6872455c6d COFF: Make test commands shorter. NFC.
llvm-svn: 244227
2015-08-06 16:47:46 +00:00
Rui Ueyama 251b0e268b COFF: Remove the old COFF linker and make link an alias to link2.
It's time to remove old COFF linker because the new one is now complete.

llvm-svn: 244226
2015-08-06 16:19:35 +00:00
David Majnemer bf1e5c0ee9 COFF: Reenable the lld test disabled in r243758.
llvm-svn: 243762
2015-07-31 17:41:11 +00:00
Reid Kleckner 60dde1fbda Disable objdump disassembly test after r243755
Right now PE image section addresses are RVAs and symbol addresses are
VAs.  We should probably fix this by changing section addresses to match
symbol addresses. Fixing this might take a few hours, so temporarily
disable the objdump part of this test.

llvm-svn: 243758
2015-07-31 16:41:23 +00:00
Rui Ueyama 7e387a68de COFF: Fix 32-bit delay-import address table.
The address table entry is 32-bit wide on 32-bit and 64-bit on 64-bit.
Previously, it was 64-bit even on 32-bit.

llvm-svn: 243372
2015-07-28 02:54:18 +00:00
Rui Ueyama 5c437cd1e9 COFF: Fix image base address for 32-bit.
0x140000000 or 0x180000000 are not correct image base addresses
for 32-bit. They are actually much smaller.

llvm-svn: 243228
2015-07-25 21:42:33 +00:00
Rui Ueyama 1c341a54de COFF: Do not align import thunks on 16-byte boundaries on x86.
Looks like MSVC linker aligns them only on x64.

llvm-svn: 243194
2015-07-25 01:16:04 +00:00
Rui Ueyama 2296dc137c COFF: Fix base relocation type for x86.
llvm-svn: 243178
2015-07-24 23:24:45 +00:00
Rui Ueyama 759c8aa9a0 COFF: Fix offset in x86 delay-load thunks.
llvm-svn: 242353
2015-07-15 23:01:36 +00:00
Rui Ueyama ef0e647581 COFF: Implement x86 delay-load thunks.
llvm-svn: 242343
2015-07-15 22:26:57 +00:00
Rui Ueyama 6d24908fe7 COFF: Fix x86 delay-load helper function name.
If /delayload option is given, we have to resolve __delayLoadHelper2
since the function is the dynamic loader to delay-load DLLs.
The function name is mangled in x86 as ___delayLoadHelper2@8.

llvm-svn: 242078
2015-07-13 22:31:45 +00:00