Summary:
Fixes PR35242. A simplified reproduce:
thread_local int i; int f() { return i; }
% {g++,clang++} -fPIC -shared -ftls-model=local-dynamic -fuse-ld=lld a.cc
ld.lld: error: can't create dynamic relocation R_X86_64_DTPOFF32 against symbol: i in readonly segment; recompile object files with -fPIC or pass '-Wl,-z,notext' to allow text relocations in the output
In isStaticLinkTimeConstant(), Syn.IsPreemptible is true, so it is not
seen as a constant. The error is then issued in processRelocAux().
A symbol of the local-dynamic TLS model cannot be preempted but it can
preempt symbols of the global-dynamic TLS model in other DSOs.
So it makes some sense that the variable is not static.
This patch fixes the linking error by changing getRelExpr() on
R_386_TLS_LDO_32 and R_X86_64_DTPOFF{32,64} from R_ABS to R_DTPREL.
R_PPC64_DTPREL_* and R_MIPS_TLS_DTPREL_* need similar fixes, but they are not handled in this patch.
As a bonus, we use `if (Expr == R_ABS && !Config->Shared)` to find
ld-to-le opportunities. R_ABS is overloaded here for such STT_TLS symbols.
A dedicated R_DTPREL is clearer.
Differential Revision: https://reviews.llvm.org/D60945
llvm-svn: 358870
Summary:
This should address remaining issues discussed in PR36555.
Currently R_GOT*_FROM_END are exclusively used by x86 and x86_64 to
express relocations types relative to the GOT base. We have
_GLOBAL_OFFSET_TABLE_ (GOT base) = start(.got.plt) but end(.got) !=
start(.got.plt)
This can have problems when _GLOBAL_OFFSET_TABLE_ is used as a symbol, e.g.
glibc dl_machine_dynamic assumes _GLOBAL_OFFSET_TABLE_ is start(.got.plt),
which is not true.
extern const ElfW(Addr) _GLOBAL_OFFSET_TABLE_[] attribute_hidden;
return _GLOBAL_OFFSET_TABLE_[0]; // R_X86_64_GOTPC32
In this patch, we
* Change all GOT*_FROM_END to GOTPLT* to fix the problem.
* Add HasGotPltOffRel to denote whether .got.plt should be kept even if
the section is empty.
* Simplify GotSection::empty and GotPltSection::empty by setting
HasGotOffRel and HasGotPltOffRel according to GlobalOffsetTable early.
The change of R_386_GOTPC makes X86::writePltHeader simpler as we don't
have to compute the offset start(.got.plt) - Ebx (it is constant 0).
We still diverge from ld.bfd (at least in most cases) and gold in that
.got.plt and .got are not adjacent, but the advantage doing that is
unclear.
Reviewers: ruiu, sivachandra, espindola
Subscribers: emaste, mehdi_amini, arichardson, dexonsmith, jdoerfert, llvm-commits
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D59594
llvm-svn: 356968
Previously, we showed the following message for an unknown relocation:
foo.o: unrecognized reloc 256
This patch improves it so that the error message includes a symbol name:
foo.o: unknown relocation (256) against symbol bar
llvm-svn: 354040
R_X86_64_PC{8,16} relocations are sign-extended, so when we check
for relocation overflow, we had to use checkInt instead of checkUInt.
I confirmed that GNU linkers create the same output for the test case.
llvm-svn: 353437
This is the same as D57749, but for x64 target.
"ELF Handling For Thread-Local Storage" p41 says (https://www.akkadia.org/drepper/tls.pdf):
R_X86_64_GOTTPOFF relocation is used for IE TLS models.
Hence if linker sees this relocation we should add DF_STATIC_TLS flag.
Differential revision: https://reviews.llvm.org/D57821
llvm-svn: 353378
Summary:
R_PPC64_TLSGD and R_PPC64_TLSLD are used as markers on TLS code sequences. After GD-to-IE or GD-to-LE relaxation, the next relocation R_PPC64_REL24 should be skipped to not create a false dependency on __tls_get_addr. When linking statically, the false dependency may cause an "undefined symbol: __tls_get_addr" error.
R_PPC64_GOT_TLSGD16_HA
R_PPC64_GOT_TLSGD16_LO
R_PPC64_TLSGD R_TLSDESC_CALL
R_PPC64_REL24 __tls_get_addr
Reviewers: ruiu, sfertile, syzaara, espindola
Reviewed By: sfertile
Subscribers: emaste, nemanjai, arichardson, kbarton, jsji, llvm-commits, tamur
Tags: #llvm
Differential Revision: https://reviews.llvm.org/D57673
llvm-svn: 353262
to reflect the new license.
We understand that people may be surprised that we're moving the header
entirely to discuss the new license. We checked this carefully with the
Foundation's lawyer and we believe this is the correct approach.
Essentially, all code in the project is now made available by the LLVM
project under our new license, so you will see that the license headers
include that license only. Some of our contributors have contributed
code under our old license, and accordingly, we have retained a copy of
our old license notice in the top-level files in each project and
repository.
llvm-svn: 351636
This patch also makes getPltEntryOffset a non-member function because
it doesn't depend on any private members of the TargetInfo class.
I tried a few different ideas, and it seems this change fits in best to me.
Differential Revision: https://reviews.llvm.org/D54981
llvm-svn: 347781
The uint32_t type does not clearly convey that these fields are interpreted
in the target endianness. Converting them to byte arrays should make this
more obvious and less error-prone.
Patch by James Clarke
Differential Revision: http://reviews.llvm.org/D54207
llvm-svn: 346893
Recommitting https://reviews.llvm.org/rL344544 after fixing undefined behavior
from left-shifting a negative value. Original commit message:
This support is slightly different then the X86_64 implementation in that calls
to __morestack don't need to get rewritten to calls to __moresatck_non_split
when a split-stack caller calls a non-split-stack callee. Instead the size of
the stack frame requested by the caller is adjusted prior to the call to
__morestack. The size the stack-frame will be adjusted by is tune-able through a
new --split-stack-adjust-size option.
llvm-svn: 344622
This reverts commit https://reviews.llvm.org/rL344544, which causes failures on
a undefined behaviour sanitizer bot -->
lld/ELF/Arch/PPC64.cpp:849:35: runtime error: left shift of negative value -1
llvm-svn: 344551
This support is slightly different then the X86_64 implementation in that calls
to __morestack don't need to get rewritten to calls to __moresatck_non_split
when a split-stack caller calls a non-split-stack callee. Instead the size of
the stack frame requested by the caller is adjusted prior to the call to
__morestack. The size the stack-frame will be adjusted by is tune-able through a
new --split-stack-adjust-size option.
Differential Revision: https://reviews.llvm.org/D52099
llvm-svn: 344544
This is https://bugs.llvm.org//show_bug.cgi?id=38919.
Currently, LLD may report "unsupported relocation target while parsing debug info"
when parsing the debug information.
At the same time LLD does that for zeroed R_X86_64_NONE relocations,
which obviously has "invalid" targets.
The nature of R_*_NONE relocation assumes them should be ignored.
This patch teaches LLD to stop reporting the debug information parsing errors for them.
Differential revision: https://reviews.llvm.org/D52408
llvm-svn: 343078
Previously, if you invoke lld's `main` more than once in the same process,
the second invocation could fail or produce a wrong result due to a stale
pointer values of the previous run.
Differential Revision: https://reviews.llvm.org/D52506
llvm-svn: 343009
Summary:
R_X86_64_GOTOFF64: S + A - GOT
R_X86_64_GOTPC{32,64}: GOT + A - P (R_GOTONLY_PC_FROM_END)
R_X86_64_GOTOFF64 should use R_GOTREL_FROM_END so that in conjunction with
R_X86_64_GOTPC{32,64}, the `GOT` term is neutralized. This also matches
the handling of R_386_GOTOFF (S + A - GOT).
Reviewers: ruiu, espindola
Subscribers: emaste, arichardson, llvm-commits
Differential Revision: https://reviews.llvm.org/D48095
llvm-svn: 334672
R_X86_64_GOTOFF64 is a relocation type to set to a distance betwween
a symbol and the beginning of the .got section. Previously, we always
created a dynamic relocation for the relocation type even though it
can be resolved at link-time.
Creating a dynamic relocation for R_X86_64_GOTOFF64 caused link failure
for some programs that do have a relocation of the type in a .text
section, as text relocations are prohibited in most configurations.
Differential Revision: https://reviews.llvm.org/D48058
llvm-svn: 334534
As was mentioned in comments for D45158,
isPicRel's name does not make much sense,
because what this method does is checks if
we need to create the dynamic relocation or not.
Instead of renaming it to something different,
we can 'isPicRel' completely.
We can reuse the getDynRel method.
They are logically very close, getDynRel can just return
R_*_NONE in case no dynamic relocation should be produced
and that would simplify things and avoid functionality
correlation/duplication with 'isPicRel'.
The patch does this change.
Differential revision: https://reviews.llvm.org/D45248
llvm-svn: 329275
isPicRel is used to check if we want to create the dynamic relocations.
Not all of the dynamic relocations we create are passing through this
check, but those that are, probably better be whitelisted.
Differential revision: https://reviews.llvm.org/D45252
llvm-svn: 329203
Having 8/16 bits dynamic relocations is incorrect.
Both gold and bfd (built from latest sources) disallow
that too.
Differential revision: https://reviews.llvm.org/D45158
llvm-svn: 329059
The PLT retpoline support for X86 and X86_64 did not include the padding
when writing the header and entries. This issue was revealed when linker
scripts were used, as this disables the built-in behaviour of filling
the last page of executable segments with trap instructions. This
particular behaviour was hiding the missing padding.
Added retpoline tests with linker scripts.
Differential Revision: https://reviews.llvm.org/D44682
llvm-svn: 328777
This is the same as 327248 except Arm defining _GLOBAL_OFFSET_TABLE_ to
be the base of the .got section as some existing code is relying upon it.
For most Targets the _GLOBAL_OFFSET_TABLE_ symbol is expected to be at
the start of the .got.plt section so that _GLOBAL_OFFSET_TABLE_[0] =
reserved value that is by convention the address of the dynamic section.
Previously we had defined _GLOBAL_OFFSET_TABLE_ as either the start or end
of the .got section with the intention that the .got.plt section would
follow the .got. However this does not always hold with the current
default section ordering so _GLOBAL_OFFSET_TABLE_[0] may not be consistent
with the reserved first entry of the .got.plt.
X86, X86_64 and AArch64 will use the .got.plt. Arm, Mips and Power use .got
Fixes PR36555
Differential Revision: https://reviews.llvm.org/D44259
llvm-svn: 327823
This change broke ARM code that expects to be able to add
_GLOBAL_OFFSET_TABLE_ to the result of an R_ARM_REL32.
I will provide a reproducer on llvm-commits.
llvm-svn: 327688
the start of the .got.plt section so that _GLOBAL_OFFSET_TABLE_[0] =
reserved value that is by convention the address of the dynamic section.
Previously we had defined _GLOBAL_OFFSET_TABLE_ as either the start or end
of the .got section with the intention that the .got.plt section would
follow the .got. However this does not always hold with the current
default section ordering so _GLOBAL_OFFSET_TABLE_[0] may not be consistent
with the reserved first entry of the .got.plt.
X86, X86_64, Arm and AArch64 will use the .got.plt. Mips and Power use .got
Fixes PR36555
Differential Revision: https://reviews.llvm.org/D44259
llvm-svn: 327248
Summary:
First, we need to explain the core of the vulnerability. Note that this
is a very incomplete description, please see the Project Zero blog post
for details:
https://googleprojectzero.blogspot.com/2018/01/reading-privileged-memory-with-side.html
The basis for branch target injection is to direct speculative execution
of the processor to some "gadget" of executable code by poisoning the
prediction of indirect branches with the address of that gadget. The
gadget in turn contains an operation that provides a side channel for
reading data. Most commonly, this will look like a load of secret data
followed by a branch on the loaded value and then a load of some
predictable cache line. The attacker then uses timing of the processors
cache to determine which direction the branch took *in the speculative
execution*, and in turn what one bit of the loaded value was. Due to the
nature of these timing side channels and the branch predictor on Intel
processors, this allows an attacker to leak data only accessible to
a privileged domain (like the kernel) back into an unprivileged domain.
The goal is simple: avoid generating code which contains an indirect
branch that could have its prediction poisoned by an attacker. In many
cases, the compiler can simply use directed conditional branches and
a small search tree. LLVM already has support for lowering switches in
this way and the first step of this patch is to disable jump-table
lowering of switches and introduce a pass to rewrite explicit indirectbr
sequences into a switch over integers.
However, there is no fully general alternative to indirect calls. We
introduce a new construct we call a "retpoline" to implement indirect
calls in a non-speculatable way. It can be thought of loosely as
a trampoline for indirect calls which uses the RET instruction on x86.
Further, we arrange for a specific call->ret sequence which ensures the
processor predicts the return to go to a controlled, known location. The
retpoline then "smashes" the return address pushed onto the stack by the
call with the desired target of the original indirect call. The result
is a predicted return to the next instruction after a call (which can be
used to trap speculative execution within an infinite loop) and an
actual indirect branch to an arbitrary address.
On 64-bit x86 ABIs, this is especially easily done in the compiler by
using a guaranteed scratch register to pass the target into this device.
For 32-bit ABIs there isn't a guaranteed scratch register and so several
different retpoline variants are introduced to use a scratch register if
one is available in the calling convention and to otherwise use direct
stack push/pop sequences to pass the target address.
This "retpoline" mitigation is fully described in the following blog
post: https://support.google.com/faqs/answer/7625886
We also support a target feature that disables emission of the retpoline
thunk by the compiler to allow for custom thunks if users want them.
These are particularly useful in environments like kernels that
routinely do hot-patching on boot and want to hot-patch their thunk to
different code sequences. They can write this custom thunk and use
`-mretpoline-external-thunk` *in addition* to `-mretpoline`. In this
case, on x86-64 thu thunk names must be:
```
__llvm_external_retpoline_r11
```
or on 32-bit:
```
__llvm_external_retpoline_eax
__llvm_external_retpoline_ecx
__llvm_external_retpoline_edx
__llvm_external_retpoline_push
```
And the target of the retpoline is passed in the named register, or in
the case of the `push` suffix on the top of the stack via a `pushl`
instruction.
There is one other important source of indirect branches in x86 ELF
binaries: the PLT. These patches also include support for LLD to
generate PLT entries that perform a retpoline-style indirection.
The only other indirect branches remaining that we are aware of are from
precompiled runtimes (such as crt0.o and similar). The ones we have
found are not really attackable, and so we have not focused on them
here, but eventually these runtimes should also be replicated for
retpoline-ed configurations for completeness.
For kernels or other freestanding or fully static executables, the
compiler switch `-mretpoline` is sufficient to fully mitigate this
particular attack. For dynamic executables, you must compile *all*
libraries with `-mretpoline` and additionally link the dynamic
executable and all shared libraries with LLD and pass `-z retpolineplt`
(or use similar functionality from some other linker). We strongly
recommend also using `-z now` as non-lazy binding allows the
retpoline-mitigated PLT to be substantially smaller.
When manually apply similar transformations to `-mretpoline` to the
Linux kernel we observed very small performance hits to applications
running typical workloads, and relatively minor hits (approximately 2%)
even for extremely syscall-heavy applications. This is largely due to
the small number of indirect branches that occur in performance
sensitive paths of the kernel.
When using these patches on statically linked applications, especially
C++ applications, you should expect to see a much more dramatic
performance hit. For microbenchmarks that are switch, indirect-, or
virtual-call heavy we have seen overheads ranging from 10% to 50%.
However, real-world workloads exhibit substantially lower performance
impact. Notably, techniques such as PGO and ThinLTO dramatically reduce
the impact of hot indirect calls (by speculatively promoting them to
direct calls) and allow optimized search trees to be used to lower
switches. If you need to deploy these techniques in C++ applications, we
*strongly* recommend that you ensure all hot call targets are statically
linked (avoiding PLT indirection) and use both PGO and ThinLTO. Well
tuned servers using all of these techniques saw 5% - 10% overhead from
the use of retpoline.
We will add detailed documentation covering these components in
subsequent patches, but wanted to make the core functionality available
as soon as possible. Happy for more code review, but we'd really like to
get these patches landed and backported ASAP for obvious reasons. We're
planning to backport this to both 6.0 and 5.0 release streams and get
a 5.0 release with just this cherry picked ASAP for distros and vendors.
This patch is the work of a number of people over the past month: Eric, Reid,
Rui, and myself. I'm mailing it out as a single commit due to the time
sensitive nature of landing this and the need to backport it. Huge thanks to
everyone who helped out here, and everyone at Intel who helped out in
discussions about how to craft this. Also, credit goes to Paul Turner (at
Google, but not an LLVM contributor) for much of the underlying retpoline
design.
Reviewers: echristo, rnk, ruiu, craig.topper, DavidKreitzer
Subscribers: sanjoy, emaste, mcrosier, mgorny, mehdi_amini, hiraditya, llvm-commits
Differential Revision: https://reviews.llvm.org/D41723
llvm-svn: 323155
This is an aesthetic change to represent a placeholder for later
binary patching as "0, 0, 0, 0" instead of "0x00, 0x00, 0x00, 0x00".
The former is how we represent it in COFF, and I found it easier to
read than the latter.
llvm-svn: 321471
Now that we have only SymbolBody as the symbol class. So, "SymbolBody"
is a bit strange name now. This is a mechanical change generated by
perl -i -pe s/SymbolBody/Symbol/g $(git grep -l SymbolBody lld/ELF lld/COFF)
nd clang-format-diff.
Differential Revision: https://reviews.llvm.org/D39459
llvm-svn: 317370
Summary:
The COFF linker and the ELF linker have long had similar but separate
Error.h and Error.cpp files to implement error handling. This change
introduces new error handling code in Common/ErrorHandler.h, changes the
COFF and ELF linkers to use it, and removes the old, separate
implementations.
Reviewers: ruiu
Reviewed By: ruiu
Subscribers: smeenai, jyknight, emaste, sdardis, nemanjai, nhaehnle, mgorny, javed.absar, kbarton, fedor.sergeev, llvm-commits
Differential Revision: https://reviews.llvm.org/D39259
llvm-svn: 316624
A section was passed to getRelExpr just to create an error message.
But if there's an invalid relocation, we would eventually report it
in relocateOne. So we don't have to pass a section to getRelExpr.
llvm-svn: 315552
We were using uint32_t as the type of relocation kind. It has a
readability issue because what Type really means in `uint32_t Type`
is not obvious. It could be a section type, a symbol type or a
relocation type.
Since we do not do any arithemetic operations on relocation types
(e.g. adding one to R_X86_64_PC32 doesn't make sense), it would be
more natural if they are represented as enums. Unfortunately, that
is not doable because relocation type definitions are spread into
multiple header files.
So I decided to use typedef. This still should be better than the
plain uint32_t because the intended type is now obvious.
llvm-svn: 315525
This patch fills holes in executable sections with 0xd4 (ARM) or
0xef (MIPS). These trap instructions were suggested by Theo de Raadt.
llvm-svn: 306322
On many architectures gcc and clang will recognize _GLOBAL_OFFSET_TABLE_ - .
and produce a relocation that can be processed without needing to know the
value of _GLOBAL_OFFSET_TABLE_. This is not always the case; for example ARM
gcc produces R_ARM_BASE_PREL but clang produces the more general
R_ARM_REL32 to _GLOBAL_OFFSET_TABLE_. To evaluate this relocation
correctly _GLOBAL_OFFSET_TABLE_ must be defined to be the either the base of
the GOT or end of the GOT dependent on architecture..
If/when llvm-mc is changed to recognize _GLOBAL_OFFSET_TABLE_ - . this
change will not be necessary for new objects. However there may still be
old objects and versions of clang.
Differential Revision: https://reviews.llvm.org/D34355
llvm-svn: 306282
Target.cpp contains code for all the targets that LLD supports. It was
simple and easy, but as the number of supported targets increased,
it got messy.
This patch splits the file into per-target files under ELF/arch directory.
Differential Revision: https://reviews.llvm.org/D34222
llvm-svn: 305565
Fangrui Song