public inbox for glibc-bugs@sourceware.org help / color / mirror / Atom feed
From: "i at maskray dot me" <sourceware-bugzilla@sourceware.org> To: glibc-bugs@sourceware.org Subject: [Bug dynamic-link/27924] New: ld.so: Support DT_RELR relative relocation format Date: Thu, 27 May 2021 18:14:01 +0000 [thread overview] Message-ID: <bug-27924-131@http.sourceware.org/bugzilla/> (raw) https://sourceware.org/bugzilla/show_bug.cgi?id=27924 Bug ID: 27924 Summary: ld.so: Support DT_RELR relative relocation format Product: glibc Version: unspecified Status: UNCONFIRMED Severity: normal Priority: P2 Component: dynamic-link Assignee: unassigned at sourceware dot org Reporter: i at maskray dot me Target Milestone: --- Position independent executables and shared objects built with -Bsymbolic/-Bsymbolic-functions may have many relative relocations (instead of absolute relocations (e.g. R_X86_64_64)). (The number of R_*_RELATIVE is the dominating factor between a PDE and a PIE in terms of their size differences.) Relative relocations can be packed in a compact format to save a lot of space. See the last few messages on https://groups.google.com/g/generic-abi/c/bX460iggiKg An Elf64_Rela takes 24 bytes. 64 consecutive R_*_RELATIVE take 24*64 bytes. In RELR, this just needs 8*2 bytes. DT_RELR has support in the Linux kernel (arch/arm64/Kconfig ARCH_HAS_RELR) since 2019 https://git.kernel.org/linus/5cf896fb6be3effd9aea455b22213e27be8bdb1d SHT_RELR/DT_RELR has dumper support in llvm-readobj/llvm-readelf. ld.lld --pack-dyn-relocs=relr can encode R_*_RELATIVE in RELR. GNU ld feature request: https://sourceware.org/bugzilla/show_bug.cgi?id=27923 -- You are receiving this mail because: You are on the CC list for the bug.
next reply other threads:[~2021-05-27 18:14 UTC|newest] Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-05-27 18:14 i at maskray dot me [this message] 2021-05-27 18:40 ` [Bug dynamic-link/27924] " i at maskray dot me 2021-05-27 19:01 ` i at maskray dot me 2021-05-27 23:20 ` i at maskray dot me 2021-06-24 16:08 ` carlos at redhat dot com 2021-06-24 16:08 ` carlos at redhat dot com 2021-06-24 16:30 ` carlos at redhat dot com 2021-06-24 18:03 ` i at maskray dot me 2021-06-24 18:15 ` carlos at redhat dot com 2021-06-24 21:38 ` ccoutant at gmail dot com 2021-06-24 21:38 ` ccoutant at gmail dot com 2021-10-08 21:52 ` i at maskray dot me 2021-10-15 22:00 ` sam at gentoo dot org 2021-10-19 13:11 ` carlos at redhat dot com 2021-10-19 16:51 ` i at maskray dot me 2021-10-25 13:47 ` fweimer at redhat dot com 2021-10-25 13:50 ` fweimer at redhat dot com 2022-01-05 0:25 ` hjl.tools at gmail dot com 2022-01-05 1:11 ` i at maskray dot me 2022-01-05 1:49 ` hjl.tools at gmail dot com 2022-01-05 1:50 ` hjl.tools at gmail dot com 2022-01-05 3:27 ` i at maskray dot me 2022-01-05 3:39 ` hjl.tools at gmail dot com 2022-01-05 3:42 ` hjl.tools at gmail dot com 2022-01-05 22:54 ` hjl.tools at gmail dot com 2022-01-08 20:36 ` hjl.tools at gmail dot com 2022-01-09 17:10 ` hjl.tools at gmail dot com 2022-06-07 8:05 ` i at maskray dot me
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=bug-27924-131@http.sourceware.org/bugzilla/ \ --to=sourceware-bugzilla@sourceware.org \ --cc=glibc-bugs@sourceware.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).