mikros/rust
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
93ea5bb709
rust/tests/assembly/stack-protector/stack-protector-target-support.rs

287 lines
11 KiB
Rust
Raw Normal View History

add rustc option for using LLVM stack smash protection LLVM has built-in heuristics for adding stack canaries to functions. These heuristics can be selected with LLVM function attributes. This patch adds a rustc option `-Z stack-protector={none,basic,strong,all}` which controls the use of these attributes. This gives rustc the same stack smash protection support as clang offers through options `-fno-stack-protector`, `-fstack-protector`, `-fstack-protector-strong`, and `-fstack-protector-all`. The protection this can offer is demonstrated in test/ui/abi/stack-protector.rs. This fills a gap in the current list of rustc exploit mitigations (https://doc.rust-lang.org/rustc/exploit-mitigations.html), originally discussed in #15179. Stack smash protection adds runtime overhead and is therefore still off by default, but now users have the option to trade performance for security as they see fit. An example use case is adding Rust code in an existing C/C++ code base compiled with stack smash protection. Without the ability to add stack smash protection to the Rust code, the code base artifacts could be exploitable in ways not possible if the code base remained pure C/C++. Stack smash protection support is present in LLVM for almost all the current tier 1/tier 2 targets: see test/assembly/stack-protector/stack-protector-target-support.rs. The one exception is nvptx64-nvidia-cuda. This patch follows clang's example, and adds a warning message printed if stack smash protection is used with this target (see test/ui/stack-protector/warn-stack-protector-unsupported.rs). Support for tier 3 targets has not been checked. Since the heuristics are applied at the LLVM level, the heuristics are expected to add stack smash protection to a fraction of functions comparable to C/C++. Some experiments demonstrating how Rust code is affected by the different heuristics can be found in test/assembly/stack-protector/stack-protector-heuristics-effect.rs. There is potential for better heuristics using Rust-specific safety information. For example it might be reasonable to skip stack smash protection in functions which transitively only use safe Rust code, or which uses only a subset of functions the user declares safe (such as anything under `std.*`). Such alternative heuristics could be added at a later point. LLVM also offers a "safestack" sanitizer as an alternative way to guard against stack smashing (see #26612). This could possibly also be included as a stack-protection heuristic. An alternative is to add it as a sanitizer (#39699). This is what clang does: safestack is exposed with option `-fsanitize=safe-stack`. The options are only supported by the LLVM backend, but as with other codegen options it is visible in the main codegen option help menu. The heuristic names "basic", "strong", and "all" are hopefully sufficiently generic to be usable in other backends as well. Reviewed-by: Nikita Popov <nikic@php.net> Extra commits during review: - [address-review] make the stack-protector option unstable - [address-review] reduce detail level of stack-protector option help text - [address-review] correct grammar in comment - [address-review] use compiler flag to avoid merging functions in test - [address-review] specify min LLVM version in fortanix stack-protector test Only for Fortanix test, since this target specifically requests the `--x86-experimental-lvi-inline-asm-hardening` flag. - [address-review] specify required LLVM components in stack-protector tests - move stack protector option enum closer to other similar option enums - rustc_interface/tests: sort debug option list in tracking hash test - add an explicit `none` stack-protector option Revert "set LLVM requirements for all stack protector support test revisions" This reverts commit a49b74f92a4e7d701d6f6cf63d207a8aff2e0f68.
2021-04-06 14:37:49 -05:00
// Test that stack smash protection code is emitted for all tier1 and tier2
// targets, with the exception of nvptx64-nvidia-cuda
//
[AUTO_GENERATED] Migrate compiletest to use `ui_test`-style `//@` directives
2024-02-22 06:10:29 -06:00
//@ revisions: r1 r2 r3 r4 r5 r6 r7 r8 r9 r10 r11 r12 r13 r14 r15 r16 r17 r18 r19 r20 r21 r22 r23
//@ revisions: r24 r25 r26 r27 r28 r29 r30 r31 r32 r33 r35 r36 r37 r38 r39 r40 r41 r42 r43 r44
//@ revisions: r45 r46 r47 r48 r49 r50 r51 r52 r53 r54 r55 r56 r57 r58 r59 r60 r61 r62 r63 r64 r65
//@ revisions: r66 r67 r68 r69 r70 r71 r72 r73 r74 r75 r76 r77 r78 r79 r80 r81 r82 r83 r84 r85
//@ assembly-output: emit-asm
//@ [r1] compile-flags: --target aarch64-unknown-linux-gnu
//@ [r1] needs-llvm-components: aarch64
//@ [r2] compile-flags: --target i686-pc-windows-gnu
//@ [r2] needs-llvm-components: x86
//@ [r3] compile-flags: --target i686-pc-windows-msvc
//@ [r3] needs-llvm-components: x86
//@ [r4] compile-flags: --target i686-unknown-linux-gnu
//@ [r4] needs-llvm-components: x86
//@ [r5] compile-flags: --target x86_64-apple-darwin
//@ [r5] needs-llvm-components: x86
//@ [r6] compile-flags: --target x86_64-pc-windows-gnu
//@ [r6] needs-llvm-components: x86
//@ [r7] compile-flags: --target x86_64-pc-windows-msvc
//@ [r7] needs-llvm-components: x86
//@ [r8] compile-flags: --target x86_64-unknown-linux-gnu
//@ [r8] needs-llvm-components: x86
//@ [r9] compile-flags: --target aarch64-apple-darwin
//@ [r9] needs-llvm-components: aarch64
//@ [r10] compile-flags: --target aarch64-apple-ios
//@ [r10] needs-llvm-components: aarch64
//@ [r11] compile-flags: --target aarch64-unknown-fuchsia
//@ [r11] needs-llvm-components: aarch64
//@ [r12] compile-flags: --target aarch64-linux-android
//@ [r12] needs-llvm-components: aarch64
//@ [r13] compile-flags: --target aarch64-pc-windows-msvc
//@ [r13] needs-llvm-components: aarch64
//@ [r14] compile-flags: --target aarch64-unknown-linux-musl
//@ [r14] needs-llvm-components: aarch64
//@ [r15] compile-flags: --target aarch64-unknown-none
//@ [r15] needs-llvm-components: aarch64
//@ [r16] compile-flags: --target aarch64-unknown-none-softfloat
//@ [r16] needs-llvm-components: aarch64
//@ [r17] compile-flags: --target arm-linux-androideabi
//@ [r17] needs-llvm-components: arm
//@ [r18] compile-flags: --target arm-unknown-linux-gnueabi
//@ [r18] needs-llvm-components: arm
//@ [r19] compile-flags: --target arm-unknown-linux-gnueabihf
//@ [r19] needs-llvm-components: arm
//@ [r20] compile-flags: --target arm-unknown-linux-musleabi
//@ [r20] needs-llvm-components: arm
//@ [r21] compile-flags: --target arm-unknown-linux-musleabihf
//@ [r21] needs-llvm-components: arm
//@ [r22] compile-flags: --target armebv7r-none-eabi
//@ [r22] needs-llvm-components: arm
//@ [r23] compile-flags: --target armebv7r-none-eabihf
//@ [r23] needs-llvm-components: arm
//@ [r24] compile-flags: --target armv5te-unknown-linux-gnueabi
//@ [r24] needs-llvm-components: arm
//@ [r25] compile-flags: --target armv5te-unknown-linux-musleabi
//@ [r25] needs-llvm-components: arm
//@ [r26] compile-flags: --target armv7-linux-androideabi
//@ [r26] needs-llvm-components: arm
//@ [r27] compile-flags: --target armv7a-none-eabi
//@ [r27] needs-llvm-components: arm
//@ [r28] compile-flags: --target armv7r-none-eabi
//@ [r28] needs-llvm-components: arm
//@ [r29] compile-flags: --target armv7r-none-eabihf
//@ [r29] needs-llvm-components: arm
//@ [r30] compile-flags: --target armv7-unknown-linux-gnueabi
//@ [r30] needs-llvm-components: arm
//@ [r31] compile-flags: --target armv7-unknown-linux-gnueabihf
//@ [r31] needs-llvm-components: arm
//@ [r32] compile-flags: --target armv7-unknown-linux-musleabi
//@ [r32] needs-llvm-components: arm
//@ [r33] compile-flags: --target armv7-unknown-linux-musleabihf
//@ [r33] needs-llvm-components: arm
Remove asmjs from tests
2023-10-29 01:11:03 -05:00
[AUTO_GENERATED] Migrate compiletest to use `ui_test`-style `//@` directives
2024-02-22 06:10:29 -06:00
//@ [r35] compile-flags: --target i586-pc-windows-msvc
//@ [r35] needs-llvm-components: x86
//@ [r36] compile-flags: --target i586-unknown-linux-gnu
//@ [r36] needs-llvm-components: x86
//@ [r37] compile-flags: --target i586-unknown-linux-musl
//@ [r37] needs-llvm-components: x86
//@ [r38] compile-flags: --target i686-linux-android
//@ [r38] needs-llvm-components: x86
//@ [r39] compile-flags: --target i686-unknown-freebsd
//@ [r39] needs-llvm-components: x86
//@ [r40] compile-flags: --target i686-unknown-linux-musl
//@ [r40] needs-llvm-components: x86
//@ [r41] compile-flags: --target mips-unknown-linux-gnu
//@ [r41] needs-llvm-components: mips
//@ [r42] compile-flags: --target mips-unknown-linux-musl
//@ [r42] needs-llvm-components: mips
//@ [r43] compile-flags: --target mips64-unknown-linux-gnuabi64
//@ [r43] needs-llvm-components: mips
//@ [r44] compile-flags: --target mips64-unknown-linux-muslabi64
//@ [r44] needs-llvm-components: mips
//@ [r45] compile-flags: --target mips64el-unknown-linux-gnuabi64
//@ [r45] needs-llvm-components: mips
//@ [r46] compile-flags: --target mips64el-unknown-linux-muslabi64
//@ [r46] needs-llvm-components: mips
//@ [r47] compile-flags: --target mipsel-unknown-linux-gnu
//@ [r47] needs-llvm-components: mips
//@ [r48] compile-flags: --target mipsel-unknown-linux-musl
//@ [r48] needs-llvm-components: mips
//@ [r49] compile-flags: --target nvptx64-nvidia-cuda
//@ [r49] needs-llvm-components: nvptx
//@ [r50] compile-flags: --target powerpc-unknown-linux-gnu
//@ [r50] needs-llvm-components: powerpc
//@ [r51] compile-flags: --target powerpc64-unknown-linux-gnu
//@ [r51] needs-llvm-components: powerpc
//@ [r52] compile-flags: --target powerpc64le-unknown-linux-gnu
//@ [r52] needs-llvm-components: powerpc
//@ [r53] compile-flags: --target riscv32i-unknown-none-elf
//@ [r53] needs-llvm-components: riscv
//@ [r54] compile-flags: --target riscv32imac-unknown-none-elf
//@ [r54] needs-llvm-components: riscv
//@ [r55] compile-flags:--target riscv32imc-unknown-none-elf
//@ [r55] needs-llvm-components: riscv
//@ [r56] compile-flags:--target riscv64gc-unknown-linux-gnu
//@ [r56] needs-llvm-components: riscv
//@ [r57] compile-flags:--target riscv64gc-unknown-none-elf
//@ [r57] needs-llvm-components: riscv
//@ [r58] compile-flags:--target riscv64imac-unknown-none-elf
//@ [r58] needs-llvm-components: riscv
//@ [r59] compile-flags:--target s390x-unknown-linux-gnu
//@ [r59] needs-llvm-components: systemz
//@ [r60] compile-flags:--target sparc64-unknown-linux-gnu
//@ [r60] needs-llvm-components: sparc
//@ [r61] compile-flags:--target sparcv9-sun-solaris
//@ [r61] needs-llvm-components: sparc
//@ [r62] compile-flags:--target thumbv6m-none-eabi
//@ [r62] needs-llvm-components: arm
//@ [r63] compile-flags:--target thumbv7em-none-eabi
//@ [r63] needs-llvm-components: arm
//@ [r64] compile-flags:--target thumbv7em-none-eabihf
//@ [r64] needs-llvm-components: arm
//@ [r65] compile-flags:--target thumbv7m-none-eabi
//@ [r65] needs-llvm-components: arm
//@ [r66] compile-flags:--target thumbv7neon-linux-androideabi
//@ [r66] needs-llvm-components: arm
//@ [r67] compile-flags:--target thumbv7neon-unknown-linux-gnueabihf
//@ [r67] needs-llvm-components: arm
//@ [r68] compile-flags:--target thumbv8m.base-none-eabi
//@ [r68] needs-llvm-components: arm
//@ [r69] compile-flags:--target thumbv8m.main-none-eabi
//@ [r69] needs-llvm-components: arm
//@ [r70] compile-flags:--target thumbv8m.main-none-eabihf
//@ [r70] needs-llvm-components: arm
//@ [r71] compile-flags:--target wasm32-unknown-emscripten
//@ [r71] needs-llvm-components: webassembly
//@ [r72] compile-flags:--target wasm32-unknown-unknown
//@ [r72] needs-llvm-components: webassembly
Add a new `wasm32-wasip1` target to rustc This commit adds a new target called `wasm32-wasip1` to rustc. This new target is explained in these two MCPs: * https://github.com/rust-lang/compiler-team/issues/607 * https://github.com/rust-lang/compiler-team/issues/695 In short, the previous `wasm32-wasi` target is going to be renamed to `wasm32-wasip1` to better live alongside the [new `wasm32-wasip2` target](https://github.com/rust-lang/rust/pull/119616). This new target is added alongside the `wasm32-wasi` target and has the exact same definition as the previous target. This PR is effectively a rename of `wasm32-wasi` to `wasm32-wasip1`. Note, however, that as explained in rust-lang/compiler-team#695 the previous `wasm32-wasi` target is not being removed at this time. This change will reach stable Rust before even a warning about the rename will be printed. At this time this change is just the start where a new target is introduced and users can start migrating if they support only Nightly for example.
2024-01-29 10:19:42 -06:00
//@ [r73] compile-flags:--target wasm32-wasip1
[AUTO_GENERATED] Migrate compiletest to use `ui_test`-style `//@` directives
2024-02-22 06:10:29 -06:00
//@ [r73] needs-llvm-components: webassembly
Rename `wasm32-wasi-preview1-threads` to `wasm32-wasip1-threads` This commit renames the current `wasm32-wasi-preview1-threads` target to `wasm32-wasip1-threads`. The need for this rename is a bit unfortunate as the previous name was chosen in an attempt to be future-compatible with other WASI targets. Originally this target was proposed to be `wasm32-wasi-threads`, and that's what was originally implemented in wasi-sdk as well. After discussion though and with the plans for the upcoming component-model target (now named `wasm32-wasip2`) the "preview1" naming was chosen for the threads-based target. The WASI subgroup later decided that it was time to drop the "preview" terminology and recommends "pX" instead, hence previous PRs to add `wasm32-wasip2` and rename `wasm32-wasi` to `wasm32-wasip1`. So, with all that history, the "proper name" for this target is different than its current name, so one way or another a rename is required. This PR proposes renaming this target cold-turkey, unlike `wasm32-wasi` which is having a long transition period to change its name. The threads-based target is predicted to see only a fraction of the traffic of `wasm32-wasi` due to the unstable nature of the WASI threads proposal itself. While I was here I updated the in-tree documentation in the target spec file itself as most of the documentation was copied from the original WASI target and wasn't as applicable to this target. Also, as an aside, I can at least try to apologize for all the naming confusion here, but this is hopefully the last WASI-related rename.
2024-03-07 20:01:45 -06:00
//@ [r74] compile-flags:--target wasm32-wasip1-threads
[AUTO_GENERATED] Migrate compiletest to use `ui_test`-style `//@` directives
2024-02-22 06:10:29 -06:00
//@ [r74] needs-llvm-components: webassembly
//@ [r75] compile-flags:--target x86_64-apple-ios
//@ [r75] needs-llvm-components: x86
//@ [r76] compile-flags:--target x86_64-fortanix-unknown-sgx
//@ [r76] needs-llvm-components: x86
//@ [r77] compile-flags:--target x86_64-unknown-fuchsia
//@ [r77] needs-llvm-components: x86
//@ [r78] compile-flags:--target x86_64-linux-android
//@ [r78] needs-llvm-components: x86
//@ [r79] compile-flags:--target x86_64-pc-solaris
//@ [r79] needs-llvm-components: x86
//@ [r80] compile-flags:--target x86_64-unknown-freebsd
//@ [r80] needs-llvm-components: x86
//@ [r81] compile-flags:--target x86_64-unknown-illumos
//@ [r81] needs-llvm-components: x86
//@ [r82] compile-flags:--target x86_64-unknown-linux-gnux32
//@ [r82] needs-llvm-components: x86
//@ [r83] compile-flags:--target x86_64-unknown-linux-musl
//@ [r83] needs-llvm-components: x86
//@ [r84] compile-flags:--target x86_64-unknown-netbsd
//@ [r84] needs-llvm-components: x86
//@ [r85] compile-flags: --target x86_64-unknown-redox
//@ [r85] needs-llvm-components: x86
//@ compile-flags: -Z stack-protector=all
//@ compile-flags: -C opt-level=2
add rustc option for using LLVM stack smash protection LLVM has built-in heuristics for adding stack canaries to functions. These heuristics can be selected with LLVM function attributes. This patch adds a rustc option `-Z stack-protector={none,basic,strong,all}` which controls the use of these attributes. This gives rustc the same stack smash protection support as clang offers through options `-fno-stack-protector`, `-fstack-protector`, `-fstack-protector-strong`, and `-fstack-protector-all`. The protection this can offer is demonstrated in test/ui/abi/stack-protector.rs. This fills a gap in the current list of rustc exploit mitigations (https://doc.rust-lang.org/rustc/exploit-mitigations.html), originally discussed in #15179. Stack smash protection adds runtime overhead and is therefore still off by default, but now users have the option to trade performance for security as they see fit. An example use case is adding Rust code in an existing C/C++ code base compiled with stack smash protection. Without the ability to add stack smash protection to the Rust code, the code base artifacts could be exploitable in ways not possible if the code base remained pure C/C++. Stack smash protection support is present in LLVM for almost all the current tier 1/tier 2 targets: see test/assembly/stack-protector/stack-protector-target-support.rs. The one exception is nvptx64-nvidia-cuda. This patch follows clang's example, and adds a warning message printed if stack smash protection is used with this target (see test/ui/stack-protector/warn-stack-protector-unsupported.rs). Support for tier 3 targets has not been checked. Since the heuristics are applied at the LLVM level, the heuristics are expected to add stack smash protection to a fraction of functions comparable to C/C++. Some experiments demonstrating how Rust code is affected by the different heuristics can be found in test/assembly/stack-protector/stack-protector-heuristics-effect.rs. There is potential for better heuristics using Rust-specific safety information. For example it might be reasonable to skip stack smash protection in functions which transitively only use safe Rust code, or which uses only a subset of functions the user declares safe (such as anything under `std.*`). Such alternative heuristics could be added at a later point. LLVM also offers a "safestack" sanitizer as an alternative way to guard against stack smashing (see #26612). This could possibly also be included as a stack-protection heuristic. An alternative is to add it as a sanitizer (#39699). This is what clang does: safestack is exposed with option `-fsanitize=safe-stack`. The options are only supported by the LLVM backend, but as with other codegen options it is visible in the main codegen option help menu. The heuristic names "basic", "strong", and "all" are hopefully sufficiently generic to be usable in other backends as well. Reviewed-by: Nikita Popov <nikic@php.net> Extra commits during review: - [address-review] make the stack-protector option unstable - [address-review] reduce detail level of stack-protector option help text - [address-review] correct grammar in comment - [address-review] use compiler flag to avoid merging functions in test - [address-review] specify min LLVM version in fortanix stack-protector test Only for Fortanix test, since this target specifically requests the `--x86-experimental-lvi-inline-asm-hardening` flag. - [address-review] specify required LLVM components in stack-protector tests - move stack protector option enum closer to other similar option enums - rustc_interface/tests: sort debug option list in tracking hash test - add an explicit `none` stack-protector option Revert "set LLVM requirements for all stack protector support test revisions" This reverts commit a49b74f92a4e7d701d6f6cf63d207a8aff2e0f68.
2021-04-06 14:37:49 -05:00
#![crate_type = "lib"]
#![feature(no_core, lang_items)]
#![crate_type = "lib"]
#![no_core]
#[lang = "sized"]
trait Sized {}
#[lang = "copy"]
trait Copy {}
#[no_mangle]
pub fn foo() {
// CHECK: foo{{:|()}}
// MSVC does the stack checking within a stack-check function:
// r3: calll @__security_check_cookie
// r7: callq __security_check_cookie
// r13: bl __security_check_cookie
// r35: calll @__security_check_cookie
// cuda doesn't support stack-smash protection
// r49-NOT: __security_check_cookie
// r49-NOT: __stack_chk_fail
// Other targets do stack checking within the function, and call a failure function on error
// r1: __stack_chk_fail
// r2: __stack_chk_fail
// r4: __stack_chk_fail
// r5: __stack_chk_fail
// r6: __stack_chk_fail
// r8: __stack_chk_fail
// r9: __stack_chk_fail
// r10: __stack_chk_fail
// r11: __stack_chk_fail
// r12: __stack_chk_fail
// r14: __stack_chk_fail
// r15: __stack_chk_fail
// r16: __stack_chk_fail
// r17: __stack_chk_fail
// r18: __stack_chk_fail
// r19: __stack_chk_fail
// r20: __stack_chk_fail
// r21: __stack_chk_fail
// r22: __stack_chk_fail
// r23: __stack_chk_fail
// r24: __stack_chk_fail
// r25: __stack_chk_fail
// r26: __stack_chk_fail
// r27: __stack_chk_fail
// r28: __stack_chk_fail
// r29: __stack_chk_fail
// r30: __stack_chk_fail
// r31: __stack_chk_fail
// r32: __stack_chk_fail
// r33: __stack_chk_fail
// r34: __stack_chk_fail
// r36: __stack_chk_fail
// r37: __stack_chk_fail
// r38: __stack_chk_fail
// r39: __stack_chk_fail
// r40: __stack_chk_fail
// r41: __stack_chk_fail
// r42: __stack_chk_fail
// r43: __stack_chk_fail
// r44: __stack_chk_fail
// r45: __stack_chk_fail
// r46: __stack_chk_fail
// r47: __stack_chk_fail
// r48: __stack_chk_fail
// r50: __stack_chk_fail
// r51: __stack_chk_fail
// r52: __stack_chk_fail
// r53: __stack_chk_fail
// r54: __stack_chk_fail
// r55: __stack_chk_fail
// r56: __stack_chk_fail
// r57: __stack_chk_fail
// r58: __stack_chk_fail
// r59: __stack_chk_fail
// r60: __stack_chk_fail
// r61: __stack_chk_fail
// r62: __stack_chk_fail
// r63: __stack_chk_fail
// r64: __stack_chk_fail
// r65: __stack_chk_fail
// r66: __stack_chk_fail
// r67: __stack_chk_fail
// r68: __stack_chk_fail
// r69: __stack_chk_fail
// r70: __stack_chk_fail
// r71: __stack_chk_fail
// r72: __stack_chk_fail
// r73: __stack_chk_fail
// r74: __stack_chk_fail
// r75: __stack_chk_fail
// r76: __stack_chk_fail
// r77: __stack_chk_fail
// r78: __stack_chk_fail
// r79: __stack_chk_fail
// r80: __stack_chk_fail
// r81: __stack_chk_fail
// r82: __stack_chk_fail
// r83: __stack_chk_fail
// r84: __stack_chk_fail
Fix tests/assembly/stack-protector/stack-protector-target-support.rs Previously the test seems to not actually test redox.
2024-02-20 18:48:37 -06:00
// r85: __stack_chk_fail
add rustc option for using LLVM stack smash protection LLVM has built-in heuristics for adding stack canaries to functions. These heuristics can be selected with LLVM function attributes. This patch adds a rustc option `-Z stack-protector={none,basic,strong,all}` which controls the use of these attributes. This gives rustc the same stack smash protection support as clang offers through options `-fno-stack-protector`, `-fstack-protector`, `-fstack-protector-strong`, and `-fstack-protector-all`. The protection this can offer is demonstrated in test/ui/abi/stack-protector.rs. This fills a gap in the current list of rustc exploit mitigations (https://doc.rust-lang.org/rustc/exploit-mitigations.html), originally discussed in #15179. Stack smash protection adds runtime overhead and is therefore still off by default, but now users have the option to trade performance for security as they see fit. An example use case is adding Rust code in an existing C/C++ code base compiled with stack smash protection. Without the ability to add stack smash protection to the Rust code, the code base artifacts could be exploitable in ways not possible if the code base remained pure C/C++. Stack smash protection support is present in LLVM for almost all the current tier 1/tier 2 targets: see test/assembly/stack-protector/stack-protector-target-support.rs. The one exception is nvptx64-nvidia-cuda. This patch follows clang's example, and adds a warning message printed if stack smash protection is used with this target (see test/ui/stack-protector/warn-stack-protector-unsupported.rs). Support for tier 3 targets has not been checked. Since the heuristics are applied at the LLVM level, the heuristics are expected to add stack smash protection to a fraction of functions comparable to C/C++. Some experiments demonstrating how Rust code is affected by the different heuristics can be found in test/assembly/stack-protector/stack-protector-heuristics-effect.rs. There is potential for better heuristics using Rust-specific safety information. For example it might be reasonable to skip stack smash protection in functions which transitively only use safe Rust code, or which uses only a subset of functions the user declares safe (such as anything under `std.*`). Such alternative heuristics could be added at a later point. LLVM also offers a "safestack" sanitizer as an alternative way to guard against stack smashing (see #26612). This could possibly also be included as a stack-protection heuristic. An alternative is to add it as a sanitizer (#39699). This is what clang does: safestack is exposed with option `-fsanitize=safe-stack`. The options are only supported by the LLVM backend, but as with other codegen options it is visible in the main codegen option help menu. The heuristic names "basic", "strong", and "all" are hopefully sufficiently generic to be usable in other backends as well. Reviewed-by: Nikita Popov <nikic@php.net> Extra commits during review: - [address-review] make the stack-protector option unstable - [address-review] reduce detail level of stack-protector option help text - [address-review] correct grammar in comment - [address-review] use compiler flag to avoid merging functions in test - [address-review] specify min LLVM version in fortanix stack-protector test Only for Fortanix test, since this target specifically requests the `--x86-experimental-lvi-inline-asm-hardening` flag. - [address-review] specify required LLVM components in stack-protector tests - move stack protector option enum closer to other similar option enums - rustc_interface/tests: sort debug option list in tracking hash test - add an explicit `none` stack-protector option Revert "set LLVM requirements for all stack protector support test revisions" This reverts commit a49b74f92a4e7d701d6f6cf63d207a8aff2e0f68.
2021-04-06 14:37:49 -05:00
}
Reference in New Issue Copy Permalink