{"id":135950,"date":"2025-06-06T15:05:24","date_gmt":"2025-06-06T13:05:25","guid":{"rendered":"https:\/\/prohoster.info\/blog\/novosti-interneta\/predstavlen-bekend-tpde-llvm-rabotayushhij-v-10-20-raz-bystree-llvm-v-rezhime-bez-optimizaczii"},"modified":"2025-06-06T15:05:24","modified_gmt":"2025-06-06T13:05:25","slug":"predstavlen-bekend-tpde-llvm-rabotayushhij-v-10-20-raz-bystree-llvm-v-rezhime-bez-optimizaczii","status":"publish","type":"post","link":"https:\/\/prohoster.info\/en\/blog\/news\/predstavlen-bekend-tpde-llvm-rabotayushhij-v-10-20-raz-bystree-llvm-v-rezhime-bez-optimizaczii","title":{"rendered":"The TPDE-LLVM backend has been introduced, operating 10-20 times faster than LLVM in non-optimized mode.","gt_translate_keys":[{"key":"rendered","format":"text"}]},"content":{"rendered":"
Researchers from the Technical University of Munich have published the TPDE toolkit and its backend compiler for LLVM \u2014 TPDE-LLVM, which enables machine code generation for x86-64 and AArch64 architectures based on the LLVM-IR intermediate representation. During testing, TPDE-LLVM proved to be 10-20 times faster than the LLVM -O0 backend (a code generator without optimizations) while maintaining the same level of performance for the resulting machine code and increasing size by 10-30%. The project's developments are published under the Apache 2.0 license. <\/p>\n
TPDE-LLVM is initially aimed at providing compilation with minimal latency and a quality level comparable to the no-optimization build mode («-O0»). The project offers a utility for standalone execution of tpde-llc, a library for integration into applications (for example, to implement JIT compiler functionality), and patches for integration with Clang and Flang.
<\/center> <\/p>\n
The project can be used as a basic compiling component for JIT or for creating unoptimized builds. TPDE-LLVM is focused solely on compilation speed and does not attempt to compete with the optimizing backends of LLVM, which operate significantly slower than TPDE-LLVM but allow for generating faster and more compact machine code (approximately twice as fast and half the size). <\/p>\n
TPDE-LLVM employs three stages for code generation: <\/p>\n
\n
Cleaning and preparing the LLVM intermediate representation;\n
Analyzing information about loops and variable usage;\n
Generating machine code. <\/ul>\n
<\/center> <\/p>\n
It is noted that optimizations added by the TPDE authors to LLVM branches 19 and 20 have already improved the stock LLVM backend's performance by 18% on the x86-64 platform and by 13% on the ARM64 platform. According to the TPDE authors, in the future, it may be possible to accelerate the LLVM backend by another 10-20% with minor effort, but further performance improvements will require significant changes. Even with substantial reworking, it is unlikely that the existing LLVM backend can be made ten times faster. <\/p>\n
At the current stage of development, the goals of the TPDE project include providing support for the LLVM intermediate representation generated by the Clang frontend in the optimization modes «-O0» and «-O1». Processing of intermediate representation created in «-O2» mode is not yet guaranteed due to the lack of TPDE support for vector operations. Partial support is available for intermediate code from Flang and the Rust compiler. \n Source: opennet.ru<\/a> <\/p>","protected":false,"gt_translate_keys":[{"key":"rendered","format":"html"}]},"excerpt":{"rendered":"
![\"The](\"\/wp-content\/uploads\/2025\/06\/68d39bfd1329b05275aa44050fcf5ed4.png\")
<\/center> <\/p>\n![\"The](\"\/wp-content\/uploads\/2025\/06\/944e734dee08a32799457c0ce5438e57.png\")
<\/center> <\/p>\n