Fix RangeOps::Multiply to handle dependent ranges

[Copilot]

Description

RangeOps::Multiply was returning keUnknown when either range had dependent bounds, preventing range propagation for loop induction variables. This caused a ~12% regression in CSieve on ARM64.

Changes

Modified RangeOps::Multiply in src/coreclr/jit/rangecheck.h:

• Preserve existing behavior for fully constant ranges
• For ranges with dependent bounds: check if constant bound combinations overflow, then return dependent range (not keUnknown)
• Aligns with Add/Subtract patterns for dependent bound handling

Enables JIT to maintain range information for patterns like:

for (int k = i + i; k <= Size; k += i)
{
    array[k] = value;  // Can now optimize bounds checks
}

Validation

Run CSieve benchmark with EgorBot to confirm regression fix.

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• https://api.github.com/graphql
  • Triggering command: /usr/bin/gh gh issue view 124349 --json title,body,comments --jq {title, body, comments: [.comments[].body]} (http block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[dotnet-policy-service]

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch
See info in area-owners.md if you want to be subscribed.

[Copilot]

Pull request overview

Updates CoreCLR JIT range analysis to avoid immediately giving up on RangeOps::Multiply when inputs are not fully constant, aiming to preserve range propagation for induction-variable patterns and prevent ARM64 CSieve perf regressions.

Changes:

• Adjusted RangeOps::Multiply to return a range that preserves keDependent on non-constant inputs instead of always returning keUnknown.
• Kept existing constant-range multiply behavior (including overflow checks) unchanged.

[EgorBo]

PTAL @jakobbotsch @dotnet/jit-contrib fix for a perf regression. Restores back the previous logic to propagate keDependent for MUL.