Removing unecessary handling of padding bits.

Why not just make half as an argument do the right thing for that case?

Formatting.

I've double-checked the Neon intrinsics reference and it indeed confirms that the only allowed value for the lane argument for vdupq_lane_f64 is 0:

Argument Preparation

Ping.

Ping.

Oh, wait, AAPCS wants half values to be passed in the *least* significant bits of a GPR, even on big-endian machines? That's certainly more convenient, but it's a weird inconsistency with the otherwise iron rule of the calling convention, which that it's exactly as if you laid all of the arguments out in memory and then popped the first four 32-bit values off. We're talking about a calling convention here that literally skips registers in order to "align" arguments.

Can we not just coerce to i16? Will LLVM not pass an i16 in the least-significant bits of a register?

Hi @rjmccall,
I agree those kind of tweaks do not look good. The issue here, though, is that argument coercion currently ignores the target's endian information when performing coercion through memory.
This happens for any type that requires memory coercion, so unfortunately using [1 x i32] does not do the trick.
Let me know if you have any other sugestions for handling this, I'd be glad to avoid the ABIArgInfo approach.

Adding example to call_mangled description comment.

Ping.

The clang-format pre-merge check keeps wanting me to update the indentation for the entire ARMSIMDIntrinsicMap and AArch64SIMDIntrinsicMap maps due to the change in their first entries.
I believe, though, that this change would not only be out of the scope of this patch, but would also bring inconsistencies with the current indentation style of the entire file.

Fixing missing format issue.

I believe rule B.2 from the AAPCS64 should take precedence over rule B.5 for HFA arguments:

B.2	If the argument type is an HFA or an HVA, then the argument is used unmodified.

Clang-format.

Clang-format.

Clang-format.

Adding check for valid range on tests.

Adding check for valid range on tests.

Applying clang-format.

Formatting patch according to clang-format and clang-tidy.

Merging content from D74617 into this revision.

Addressing review comment

This patch by itself is a NFC. It aims to enable the proper overload resolution of intrinsic calls when using the call_mangled operation introduced on D74618.
To make this more clear, I'll discard this review and merge this changes into D74619, where call_mangled gets used.

Addressing review.

Addressing review comment.

Ping.

Clang-format fixes.

Updating outdated comment.

Addressing review comments.

Removed restriction of branch instruction mnemonic for allowing $-prefixed
symbol names in operands, resolving ambriguity in favor of symbol names over
expression start.

I agree with resolving the ambiguity in favor of identifiers regardless of the mnemonic for the general-case operand parsing, avoiding inconsistencies. Specific cases where constant immediates are explicitly expected are currently handled by their own ParseMethod implementation.
If there's no objection, I'll update the patch to reflect this behavior.

Adding test for the intrinsics range checks.

Regarding the b $4 case, I found that the limitation currently lies in the expression/identifier parsing at the core AsmParser, which only detects $-prefixed Identifier or String tokens at the moment.
As this part of the code has no knowledge of the branch instruction context, I'm afraid updating it to accept Integer tokens as well might cause some inconsistencies in the instructions for which we drop the leading Dollar token. For example:

Ping.

[ARM] Updating handling of $-prefixed operands for branch instructions

In D73176#1835170, @efriedma wrote:

If the goal here is to match the GNU assembler, I don't think you've achieved that; gas treats "b $4" as a branch to the symbol "$4", not an absolute offset.