From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: by sourceware.org (Postfix, from userid 2205) id 72ECB3858C2D; Thu, 10 Feb 2022 08:02:48 +0000 (GMT) DKIM-Filter: OpenDKIM Filter v2.11.0 sourceware.org 72ECB3858C2D MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Type: text/plain; charset="utf-8" From: Tom de Vries To: gcc-cvs@gcc.gnu.org Subject: [gcc r12-7168] nvptx: Tweak constraints on copysign instructions X-Act-Checkin: gcc X-Git-Author: Roger Sayle X-Git-Refname: refs/heads/master X-Git-Oldrev: 9bacd7af2e3bba9ddad17e7de4e2d299419d819d X-Git-Newrev: 6d98e83b2c919bd9fba2c61333d613bafc37357f Message-Id: <20220210080248.72ECB3858C2D@sourceware.org> Date: Thu, 10 Feb 2022 08:02:48 +0000 (GMT) X-BeenThere: gcc-cvs@gcc.gnu.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Gcc-cvs mailing list List-Unsubscribe: , List-Archive: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 10 Feb 2022 08:02:48 -0000 https://gcc.gnu.org/g:6d98e83b2c919bd9fba2c61333d613bafc37357f commit r12-7168-g6d98e83b2c919bd9fba2c61333d613bafc37357f Author: Roger Sayle Date: Tue Feb 8 20:56:55 2022 +0100 nvptx: Tweak constraints on copysign instructions Many thanks to Thomas Schwinge for confirming my hypothesis that the register usage regression, PR target/104345, is solely due to libgcc's _muldc3 function. In addition to the isinf functionality in the previously proposed nvptx patch at https://gcc.gnu.org/pipermail/gcc-patches/2022-January/588453.html which significantly reduces the number of instructions in _muldc3, the patch below further reduces both the number of instructions and the number of explicitly declared registers, by permitting floating point constant immediate operands in nvptx's copysign instruction. Fingers-crossed, the combination with all of the previous proposed nvptx patches improves things. Ultimately, increasing register usage from 50 to 51 registers, reducing the number of concurrent threads by ~2%, can easily be countered if we're now executing significantly fewer instructions in each kernel, for a net performance win. This patch has been tested on nvptx-none hosted on x86_64-pc-linux-gnu with a "make" and "make -k check" with no new failures. gcc/ChangeLog: * config/nvptx/nvptx.md (copysign3): Allow immediate floating point constants as operands 1 and/or 2. Diff: --- gcc/config/nvptx/nvptx.md | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/gcc/config/nvptx/nvptx.md b/gcc/config/nvptx/nvptx.md index ad642e78ae3..bb0c0b3b9a5 100644 --- a/gcc/config/nvptx/nvptx.md +++ b/gcc/config/nvptx/nvptx.md @@ -1209,8 +1209,8 @@ (define_insn "copysign3" [(set (match_operand:SDFM 0 "nvptx_register_operand" "=R") - (unspec:SDFM [(match_operand:SDFM 1 "nvptx_register_operand" "R") - (match_operand:SDFM 2 "nvptx_register_operand" "R")] + (unspec:SDFM [(match_operand:SDFM 1 "nvptx_nonmemory_operand" "RF") + (match_operand:SDFM 2 "nvptx_nonmemory_operand" "RF")] UNSPEC_COPYSIGN))] "" "%.\\tcopysign%t0\\t%0, %2, %1;")