[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

scottconstable wrote: @lvwr @maurer @rcvalle A gentle reminder to please review this PR. https://github.com/llvm/llvm-project/pull/121070 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commit

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

https://github.com/scottconstable updated https://github.com/llvm/llvm-project/pull/121070 >From 54296a3991e186629889611eff9e7cdd2c1cadca Mon Sep 17 00:00:00 2001 From: Scott D Constable Date: Mon, 23 Dec 2024 13:48:48 -0800 Subject: [PATCH] Implement a new kcfi_arity feature that encodes an in

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

scottconstable wrote: @sirmc @samitolvanen @Darksonn @lvwr @maurer @rcvalle A gentle reminder to please review this PR. https://github.com/llvm/llvm-project/pull/121070 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

scottconstable wrote: > > maurer: I think their point is that even if you are not changing the hash > > scheme, you are proposing breaking compatibility of the identifier with > > existing code. Since we don't want to do this many times, if we are > > breaking compatibility with existing code,

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

scottconstable wrote: > If hashing is changed, consider replacing xxhash64 with xxh3+_64bits @MaskRay This PR does not change the hashing scheme at all. https://github.com/llvm/llvm-project/pull/121070 ___ cfe-commits mailing list cfe-commits@lists.ll

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

@@ -254,6 +254,7 @@ FEATURE(is_trivially_constructible, LangOpts.CPlusPlus) FEATURE(is_trivially_copyable, LangOpts.CPlusPlus) FEATURE(is_union, LangOpts.CPlusPlus) FEATURE(kcfi, LangOpts.Sanitize.has(SanitizerKind::KCFI)) +FEATURE(kcfi_x86_arity, LangOpts.Sanitize.has(Sanitize

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

scottconstable wrote: @sirmc @samitolvanen @Darksonn @lvwr @maurer @rcvalle @MaskRay A gentle reminder to please review this PR. https://github.com/llvm/llvm-project/pull/121070 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llv

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

@@ -181,8 +181,26 @@ void X86AsmPrinter::emitKCFITypeId(const MachineFunction &MF) { // Embed the type hash in the X86::MOV32ri instruction to avoid special // casing object file parsers. EmitKCFITypePadding(MF); + + Register MovReg = X86::EAX; + const auto &Triple = M

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

https://github.com/scottconstable updated https://github.com/llvm/llvm-project/pull/121070 >From 73b6f6332cf699b7fa1bcf7cae90ef828cfa9158 Mon Sep 17 00:00:00 2001 From: Scott D Constable Date: Mon, 23 Dec 2024 13:48:48 -0800 Subject: [PATCH] Implement a new kcfi_x86_arity feature that encodes a

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

scottconstable wrote: > And you cannot use ESI when it's used to pass arguments. @phoebewang I think that my https://github.com/llvm/llvm-project/pull/121070#discussion_r1898074522 also applies to this concern. https://github.com/llvm/llvm-project/pull/121070 _

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

@@ -181,8 +181,26 @@ void X86AsmPrinter::emitKCFITypeId(const MachineFunction &MF) { // Embed the type hash in the X86::MOV32ri instruction to avoid special // casing object file parsers. EmitKCFITypePadding(MF); + + Register MovReg = X86::EAX; + const auto &Triple = M

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

@@ -254,6 +254,7 @@ FEATURE(is_trivially_constructible, LangOpts.CPlusPlus) FEATURE(is_trivially_copyable, LangOpts.CPlusPlus) FEATURE(is_union, LangOpts.CPlusPlus) FEATURE(kcfi, LangOpts.Sanitize.has(SanitizerKind::KCFI)) +FEATURE(kcfi_x86_arity, LangOpts.Sanitize.has(Sanitize

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > I haven't received a reply for my concerns I commented at [#117121 > (comment)](https://github.com/llvm/llvm-project/pull/117121#issuecomment-2502346476) > and [#117121 > (comment)](https://github.com/llvm/llvm-project/pull/117121#issuecomment-2516251353), > and they s

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

@@ -181,8 +181,26 @@ void X86AsmPrinter::emitKCFITypeId(const MachineFunction &MF) { // Embed the type hash in the X86::MOV32ri instruction to avoid special // casing object file parsers. EmitKCFITypePadding(MF); + + Register MovReg = X86::EAX; + const auto &Triple = M

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: @phoebewang @sirmc @samitolvanen @Darksonn @lvwr @ojeda @maurer @rcvalle @MaskRay I have created https://github.com/llvm/llvm-project/pull/121070 to implement the alternate proposal summarized in this comment above: https://github.com/llvm/llvm-project/pull/117121#issuec

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

https://github.com/scottconstable updated https://github.com/llvm/llvm-project/pull/121070 >From 70f1743e23ff8e218799e94b03a5bf033715666a Mon Sep 17 00:00:00 2001 From: Scott D Constable Date: Mon, 23 Dec 2024 13:48:48 -0800 Subject: [PATCH] Implement a new kcfi_x86_arity feature that encodes a

[clang] [llvm] [X86] Extend kCFI with a 3-bit arity indicator (PR #121070)

https://github.com/scottconstable created https://github.com/llvm/llvm-project/pull/121070 Kernel Control Flow Integrity (kCFI) is a feature that hardens indirect calls by comparing a 32-bit hash of the function pointer's type against a hash of the target function's type. If the hashes do not

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable updated https://github.com/llvm/llvm-project/pull/117121 >From b787d6d7a9c0904c5e47e32556103b8a5220a7d1 Mon Sep 17 00:00:00 2001 From: Scott D Constable Date: Tue, 19 Nov 2024 15:51:05 -0800 Subject: [PATCH] Enhance KCFI type IDs with a 3-bit arity indicator.

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: @rcvalle @maurer Do you have any feedback on the alternative proposal, https://github.com/llvm/llvm-project/pull/117121#issuecomment-2518240999? https://github.com/llvm/llvm-project/pull/117121 ___ cfe-commits mailing list cfe-co

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > If you are changing the hash scheme, please migrate from xxHash64 to xxh3. We > want to remove the legacy xxHash64 from LLVM. It seems from the discussion above that this change would require a new flag. Suppose, for instance, that a user builds a kernel with a version

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > I think Scott's second point may be the relevant one - this may be similar > strength or slightly weaker, but having an indicator stating which registers > are live is potentially needed to allow FineIBT to poison them during > speculative execution. > > Given how much

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: I also do not want to lose sight of one of the other obvious advantages that was mentioned in the PR description: > One additional benefit of this patch is that it can benefit other CFI > approaches that build on kCFI, such as FineIBT. For example, this proposed > enhanc

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: Hi @maurer, I can honestly say that I appreciate your attention to detail. While your observation that "attackers often have direct control of intentionally passed arguments" is true in general, I do not see evidence of this among indirect calls in the Linux kernel. 77% o

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > This is not a Rust concern, but re-reading the initial post, it _looks_ like > your own statistics suggest that consuming 3 bits for arity costs more than > it buys you. As stated, (didn't check your math, just going off what you > said) prior to your change, we expect

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable edited https://github.com/llvm/llvm-project/pull/117121 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable edited https://github.com/llvm/llvm-project/pull/117121 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: This PR is intended to improve security for X86 kernels, as the title suggests. I think that we can and should incorporate feedback from the other participants, several of whom have recommended that this new behavior should be explicitly enabled with a flag, rather than i

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > Sorry, what I meant was we should be looking only at the number of return > (with void being zero, and everything else being one) and parameters from the > function declaration/definition and shouldn't be looking at the parameters' > type layout (i.e., size), and how an

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > > Flag guarding this feature seems like it would also be good for any > > existing C users - for example, if trying to build a kernel module intended > > to load against a kernel image built with an older `clang`, you need to > > select the same type ID projection that

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

@@ -208,10 +209,34 @@ void llvm::setKCFIType(Module &M, Function &F, StringRef MangledType) { std::string Type = MangledType.str(); if (M.getModuleFlag("cfi-normalize-integers")) Type += ".normalized"; + + uint32_t OutHash = static_cast(llvm::xxHash64(Type)); + auto

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > Although the default calling convention uses 6 registers, others like RegCall > uses more. Do you want to check calling convention as well? AFAIK the use case for KCFI is very narrow: the x86-64 Linux kernel. And I don't believe that the kernel uses (or even allows?) an

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

@@ -208,10 +209,34 @@ void llvm::setKCFIType(Module &M, Function &F, StringRef MangledType) { std::string Type = MangledType.str(); if (M.getModuleFlag("cfi-normalize-integers")) Type += ".normalized"; + + uint32_t OutHash = static_cast(llvm::xxHash64(Type)); + auto

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

@@ -208,10 +209,34 @@ void llvm::setKCFIType(Module &M, Function &F, StringRef MangledType) { std::string Type = MangledType.str(); if (M.getModuleFlag("cfi-normalize-integers")) Type += ".normalized"; + + uint32_t OutHash = static_cast(llvm::xxHash64(Type)); + auto

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > > > > @phoebewang and @lvwr I also noticed that there is this code in LLVM: > > > > https://github.com/llvm/llvm-project/blob/9ba6672b9f0e82a1f6d4100dc832c84447ea545c/llvm/lib/Transforms/Utils/ModuleUtils.cpp#L202-L214 > > > > > > > > . As far as I can tell, this code is

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

@@ -208,10 +209,34 @@ void llvm::setKCFIType(Module &M, Function &F, StringRef MangledType) { std::string Type = MangledType.str(); if (M.getModuleFlag("cfi-normalize-integers")) Type += ".normalized"; + + uint32_t OutHash = static_cast(llvm::xxHash64(Type)); + auto

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable updated https://github.com/llvm/llvm-project/pull/117121 >From b787d6d7a9c0904c5e47e32556103b8a5220a7d1 Mon Sep 17 00:00:00 2001 From: Scott D Constable Date: Tue, 19 Nov 2024 15:51:05 -0800 Subject: [PATCH] Enhance KCFI type IDs with a 3-bit arity indicator.

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > > Second, this scheme reduces the expected number of hash collisions within > > each arity, compared against the expected number of collisions (0.01383765) > > for the 32-bit hashing scheme that includes all arities. The table below > > shows the expected number of coll

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable edited https://github.com/llvm/llvm-project/pull/117121 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable updated https://github.com/llvm/llvm-project/pull/117121 >From 4f21a0c817d221398bb93e25452518d265794265 Mon Sep 17 00:00:00 2001 From: Scott D Constable Date: Tue, 19 Nov 2024 15:51:05 -0800 Subject: [PATCH] Enhance KCFI type IDs with a 3-bit arity indicator.

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

@@ -208,10 +209,34 @@ void llvm::setKCFIType(Module &M, Function &F, StringRef MangledType) { std::string Type = MangledType.str(); if (M.getModuleFlag("cfi-normalize-integers")) Type += ".normalized"; + + uint32_t OutHash = static_cast(llvm::xxHash64(Type)); + auto

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable edited https://github.com/llvm/llvm-project/pull/117121 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable edited https://github.com/llvm/llvm-project/pull/117121 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

@@ -208,10 +209,34 @@ void llvm::setKCFIType(Module &M, Function &F, StringRef MangledType) { std::string Type = MangledType.str(); if (M.getModuleFlag("cfi-normalize-integers")) Type += ".normalized"; + + uint32_t OutHash = static_cast(llvm::xxHash64(Type)); + auto

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable updated https://github.com/llvm/llvm-project/pull/117121 >From ccb50df487043cde9414943f08988bc9e6ee Mon Sep 17 00:00:00 2001 From: Scott D Constable Date: Tue, 19 Nov 2024 15:51:05 -0800 Subject: [PATCH] Enhance KCFI type IDs with a 3-bit arity indicator.

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: > > @phoebewang and @lvwr I also noticed that there is this code in LLVM: > > https://github.com/llvm/llvm-project/blob/9ba6672b9f0e82a1f6d4100dc832c84447ea545c/llvm/lib/Transforms/Utils/ModuleUtils.cpp#L202-L214 > > > > . As far as I can tell, this code is not triggered wh

[clang] [llvm] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable updated https://github.com/llvm/llvm-project/pull/117121 >From d5ec228a1d3e059e2a64ac1a150bec6f65d573e7 Mon Sep 17 00:00:00 2001 From: Scott D Constable Date: Tue, 19 Nov 2024 15:51:05 -0800 Subject: [PATCH] Enhance KCFI type IDs with a 3-bit arity indicator.

[clang] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable edited https://github.com/llvm/llvm-project/pull/117121 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: @phoebewang and @lvwr I also noticed that there is this code in LLVM: https://github.com/llvm/llvm-project/blob/9ba6672b9f0e82a1f6d4100dc832c84447ea545c/llvm/lib/Transforms/Utils/ModuleUtils.cpp#L202-L214. As far as I can tell, this code is not triggered when I build the L

[clang] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable edited https://github.com/llvm/llvm-project/pull/117121 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

scottconstable wrote: @phoebewang @lvwr Please review this PR. https://github.com/llvm/llvm-project/pull/117121 ___ cfe-commits mailing list cfe-commits@lists.llvm.org https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

[clang] [X86] Enhance kCFI type IDs with a 3-bit arity indicator. (PR #117121)

https://github.com/scottconstable created https://github.com/llvm/llvm-project/pull/117121 Kernel Control Flow Integrity (kCFI) is a feature that hardens indirect calls by comparing a 32-bit hash of the function pointer's type against a hash of the target function's type. If the hashes do not

Fix for Paths with Whitespace

Dear Reviewer, The current doxygen.cfg.in will produce paths that cannot be understood by Doxygen/Graphviz, when any generated paths contain white space. This patch simply adds quotations to all generated paths so that white space cannot be misinterpreted by Doxygen/Graphviz. ~Scott Constable d