public inbox for gdb-prs@sourceware.org help / color / mirror / Atom feed
From: "nuno.m.paulino at inesctec dot pt" <sourceware-bugzilla@sourceware.org> To: gdb-prs@sourceware.org Subject: [Bug backtrace/28389] New: Assertion depth > 0 failed in inline_frame_sniffer Date: Mon, 27 Sep 2021 12:40:57 +0000 [thread overview] Message-ID: <bug-28389-4717@http.sourceware.org/bugzilla/> (raw) https://sourceware.org/bugzilla/show_bug.cgi?id=28389 Bug ID: 28389 Summary: Assertion depth > 0 failed in inline_frame_sniffer Product: gdb Version: HEAD Status: UNCONFIRMED Severity: normal Priority: P2 Component: backtrace Assignee: unassigned at sourceware dot org Reporter: nuno.m.paulino at inesctec dot pt Target Milestone: --- Created attachment 13680 --> https://sourceware.org/bugzilla/attachment.cgi?id=13680&action=edit ELF File and setup required to compile (if necessary) Assertion in inline_frame_sniffer (line 245) fails when encountering any inlined function for target microblaze-xilinx-elf. I am using a GDB + QEMU setup to execute my target ELF files. GDB version: GNU gdb (GDB) 10.0.50.20200730-git (built from either Xilinx's fork, the mirror at bminor - https://github.com/bminor/binutils-gdb/releases - or even the latest commit in sourceware, i.e., a few hours ago at the time of this bug report, 27 sep 2021) QEMU: QEMU emulator version 5.1.0 (v2.6.0-35778-ge40b634b24-dirty) (note that this is the Xilinx branch of QEMU, since a specific target machine `microblaze-fdt-plnx' is required; DTB included in attachment; repo for this fork is at https://github.com/Xilinx/qemu) Attached are the source, makefile, and compiled ELF. Steps to reproduce: 1. Launch QEMU with `qemu-system-microblazeel -nographic -M microblaze-fdt-plnx -m 128 -display none -kernel dotprod.elf -dtb system.dtb -gdb tcp::1234 -S` 2. Launch mb-gdb 2.1 file dotprod.elf 2.2 target remote localhost:1234 2.3 break frame_dummy (this is only the first case in which the issue occurs, and its in initialization code common to all ELFs) 2.4 c 2.5 (when break hits, then stepi), until this assertion: Breakpoint 1, 0x000001f0 in frame_dummy () (gdb) stepi 0x000001f4 in frame_dummy () (gdb) 0x00000214 in frame_dummy () (gdb) 0x00000218 in frame_dummy () (gdb) inline-frame.c:244: internal-error: int inline_frame_sniffer(const frame_unwind*, frame_info*, void**): Assertion `depth > 0' failed. A problem internal to GDB has been detected, further debugging may prove unreliable. Quit this debugging session? (y or n) n Continuing is impossible. Note that I can set a breakpoint at "main" (or any other user function) and continue to that point. By then issuing "stepi", this assertion fails on any function that the compiler has inlined, OR any function declared as static (due to implicing inlining I presume). I've verified that by removing all "static" qualifiers, and using the " -fno-inline-functions -fno-inline-small-functions" flags, I can step through all user code normally. However, once the _exit routines are reached, the assertion fails again, since this library/init code has been compiled with inlined functions (assumption?). -- You are receiving this mail because: You are on the CC list for the bug.
next reply other threads:[~2021-09-27 12:40 UTC|newest] Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top 2021-09-27 12:40 nuno.m.paulino at inesctec dot pt [this message] 2024-03-25 16:11 ` [Bug backtrace/28389] " sean.anderson at seco dot com
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-28389-4717@http.sourceware.org/bugzilla/ \ --to=sourceware-bugzilla@sourceware.org \ --cc=gdb-prs@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).