From nobody Fri May 8 02:20:08 2026 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from vger.kernel.org (vger.kernel.org [23.128.96.18]) by smtp.lore.kernel.org (Postfix) with ESMTP id B6539C433F5 for ; Thu, 12 May 2022 19:05:51 +0000 (UTC) Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1357871AbiELTFu (ORCPT ); Thu, 12 May 2022 15:05:50 -0400 Received: from lindbergh.monkeyblade.net ([23.128.96.19]:52584 "EHLO lindbergh.monkeyblade.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1357866AbiELTFr (ORCPT ); Thu, 12 May 2022 15:05:47 -0400 Received: from dfw.source.kernel.org (dfw.source.kernel.org [IPv6:2604:1380:4641:c500::1]) by lindbergh.monkeyblade.net (Postfix) with ESMTPS id 05F921128 for ; Thu, 12 May 2022 12:05:45 -0700 (PDT) Received: from smtp.kernel.org (relay.kernel.org [52.25.139.140]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dfw.source.kernel.org (Postfix) with ESMTPS id 974D761B8F for ; Thu, 12 May 2022 19:05:44 +0000 (UTC) Received: by smtp.kernel.org (Postfix) with ESMTPSA id D3C92C385B8; Thu, 12 May 2022 19:05:43 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=kernel.org; s=k20201202; t=1652382344; bh=TpojvJtg3UFM+acn3Qz4YIne1JNKro/1rJmbYwbmKkM=; h=From:To:Cc:Subject:Date:From; b=V4Rab15FjMqCZZrNlCXop5Q3VDdkrRbBNSB4AQMNq4wQPhqCguBpGnBk7I+KP7Fvc ACTw5NG97gXB3eBpG6ye1X6+lZ5PHFam8Wkyfa9PjjQ5Zox4WIVvK/2qDQunTZBnyL HgXSpJ0lwCTK6FAqSTBaHdwLVdYy6gEe4NrBzkcGO0NE9OpbgeDjojIXZuPJRL6XXG NpYr/BcUaXdbI+dr+ZGI3IWmVXTg2tONS9QLaorApO8h8ynZuFEXIbiz0+bsTh4Orj vD+RST1Put049TUH7LRCZx7+skndpmucaXYfXVy+COb5sEFDzmZXoHM7ctmv6Sv8XA oeY9Iz2uPcLIg== From: Josh Poimboeuf To: linux-kernel@vger.kernel.org Cc: Kaiwan N Billimoria , Peter Zijlstra Subject: [PATCH] scripts/faddr2line: Fix overlapping text section failures Date: Thu, 12 May 2022 12:05:27 -0700 Message-Id: <29ff99f86e3da965b6e46c1cc2d72ce6528c17c3.1652382321.git.jpoimboe@kernel.org> X-Mailer: git-send-email 2.34.1 MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Precedence: bulk List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Type: text/plain; charset="utf-8" There have been some recent reports of faddr2line failures: $ scripts/faddr2line sound/soundcore.ko sound_devnode+0x5/0x35 bad symbol size: base: 0x0000000000000000 end: 0x0000000000000000 $ ./scripts/faddr2line vmlinux.o enter_from_user_mode+0x24 bad symbol size: base: 0x0000000000005fe0 end: 0x0000000000005fe0 The problem is that faddr2line is based on 'nm', which has a major limitation: it doesn't know how to distinguish between different text sections. So if an offset exists in multiple text sections in the object, it may fail. Rewrite faddr2line to be section-aware, by basing it on readelf. Fixes: 67326666e2d4 ("scripts: add script for translating stack dump functi= on offsets") Reported-by: Kaiwan N Billimoria Reported-by: Peter Zijlstra Signed-off-by: Josh Poimboeuf --- scripts/faddr2line | 150 +++++++++++++++++++++++++++++---------------- 1 file changed, 97 insertions(+), 53 deletions(-) diff --git a/scripts/faddr2line b/scripts/faddr2line index 6c6439f69a72..0e6268d59883 100755 --- a/scripts/faddr2line +++ b/scripts/faddr2line @@ -44,17 +44,6 @@ set -o errexit set -o nounset =20 -READELF=3D"${CROSS_COMPILE:-}readelf" -ADDR2LINE=3D"${CROSS_COMPILE:-}addr2line" -SIZE=3D"${CROSS_COMPILE:-}size" -NM=3D"${CROSS_COMPILE:-}nm" - -command -v awk >/dev/null 2>&1 || die "awk isn't installed" -command -v ${READELF} >/dev/null 2>&1 || die "readelf isn't installed" -command -v ${ADDR2LINE} >/dev/null 2>&1 || die "addr2line isn't installed" -command -v ${SIZE} >/dev/null 2>&1 || die "size isn't installed" -command -v ${NM} >/dev/null 2>&1 || die "nm isn't installed" - usage() { echo "usage: faddr2line [--list] ..." >&2 exit 1 @@ -69,6 +58,14 @@ die() { exit 1 } =20 +READELF=3D"${CROSS_COMPILE:-}readelf" +ADDR2LINE=3D"${CROSS_COMPILE:-}addr2line" +AWK=3D"awk" + +command -v ${AWK} >/dev/null 2>&1 || die "${AWK} isn't installed" +command -v ${READELF} >/dev/null 2>&1 || die "${READELF} isn't installed" +command -v ${ADDR2LINE} >/dev/null 2>&1 || die "${ADDR2LINE} isn't install= ed" + # Try to figure out the source directory prefix so we can remove it from t= he # addr2line output. HACK ALERT: This assumes that start_kernel() is in # init/main.c! This only works for vmlinux. Otherwise it falls back to @@ -76,7 +73,7 @@ die() { find_dir_prefix() { local objfile=3D$1 =20 - local start_kernel_addr=3D$(${READELF} -sW $objfile | awk '$8 =3D=3D "sta= rt_kernel" {printf "0x%s", $2}') + local start_kernel_addr=3D$(${READELF} --symbols --wide $objfile | ${AWK}= '$8 =3D=3D "start_kernel" {printf "0x%s", $2}') [[ -z $start_kernel_addr ]] && return =20 local file_line=3D$(${ADDR2LINE} -e $objfile $start_kernel_addr) @@ -97,86 +94,133 @@ __faddr2line() { local dir_prefix=3D$3 local print_warnings=3D$4 =20 - local func=3D${func_addr%+*} + local sym_name=3D${func_addr%+*} local offset=3D${func_addr#*+} offset=3D${offset%/*} - local size=3D - [[ $func_addr =3D~ "/" ]] && size=3D${func_addr#*/} + local user_size=3D + [[ $func_addr =3D~ "/" ]] && user_size=3D${func_addr#*/} =20 - if [[ -z $func ]] || [[ -z $offset ]] || [[ $func =3D $func_addr ]]; then + if [[ -z $sym_name ]] || [[ -z $offset ]] || [[ $sym_name =3D $func_addr = ]]; then warn "bad func+offset $func_addr" DONE=3D1 return fi =20 # Go through each of the object's symbols which match the func name. - # In rare cases there might be duplicates. - file_end=3D$(${SIZE} -Ax $objfile | awk '$1 =3D=3D ".text" {print $2}') - while read symbol; do - local fields=3D($symbol) - local sym_base=3D0x${fields[0]} - local sym_type=3D${fields[1]} - local sym_end=3D${fields[3]} - - # calculate the size - local sym_size=3D$(($sym_end - $sym_base)) + # In rare cases there might be duplicates, in which case we print all + # matches. + while read line; do + local fields=3D($line) + local sym_addr=3D0x${fields[1]} + local sym_elf_size=3D${fields[2]} + local sym_sec=3D${fields[6]} + + # Get the section size: + local sec_size=3D$(${READELF} --section-headers --wide $objfile | + sed 's/\[ /\[/' | + ${AWK} -v sec=3D$sym_sec '$1 =3D=3D "[" sec "]" { print "0x" $6; exit }= ') + + if [[ -z $sec_size ]]; then + warn "bad section size: section: $sym_sec" + DONE=3D1 + return + fi + + # Calculate the symbol size. + # + # Unfortunately we can't use the ELF size, because kallsyms + # also includes the padding bytes in its size calculation. For + # kallsyms, the size calculation is the distance between the + # symbol and the next symbol in a sorted list. + local sym_size + local cur_sym_addr + local found=3D0 + while read line; do + local fields=3D($line) + cur_sym_addr=3D0x${fields[1]} + local cur_sym_elf_size=3D${fields[2]} + local cur_sym_name=3D${fields[7]:-} + + if [[ $cur_sym_addr =3D $sym_addr ]] && + [[ $cur_sym_elf_size =3D $sym_elf_size ]] && + [[ $cur_sym_name =3D $sym_name ]]; then + found=3D1 + continue + fi + + if [[ $found =3D 1 ]]; then + sym_size=3D$(($cur_sym_addr - $sym_addr)) + [[ $sym_size -lt $sym_elf_size ]] && continue; + found=3D2 + break + fi + done < <(${READELF} --symbols --wide $objfile | ${AWK} -v sec=3D$sym_sec= '$7 =3D=3D sec' | sort --key=3D2) + + if [[ $found =3D 0 ]]; then + warn "can't find symbol: sym_name: $sym_name sym_sec: $sym_sec sym_addr= : $sym_addr sym_elf_size: $sym_elf_size" + DONE=3D1 + return + fi + + # If nothing was found after the symbol, assume it's the last + # symbol in the section. + [[ $found =3D 1 ]] && sym_size=3D$(($sec_size - $sym_addr)) + if [[ -z $sym_size ]] || [[ $sym_size -le 0 ]]; then - warn "bad symbol size: base: $sym_base end: $sym_end" + warn "bad symbol size: sym_addr: $sym_addr cur_sym_addr: $cur_sym_addr" DONE=3D1 return fi + sym_size=3D0x$(printf %x $sym_size) =20 - # calculate the address - local addr=3D$(($sym_base + $offset)) + # Calculate the section address from user-supplied offset: + local addr=3D$(($sym_addr + $offset)) if [[ -z $addr ]] || [[ $addr =3D 0 ]]; then - warn "bad address: $sym_base + $offset" + warn "bad address: $sym_addr + $offset" DONE=3D1 return fi addr=3D0x$(printf %x $addr) =20 - # weed out non-function symbols - if [[ $sym_type !=3D t ]] && [[ $sym_type !=3D T ]]; then - [[ $print_warnings =3D 1 ]] && - echo "skipping $func address at $addr due to non-function symbol of ty= pe '$sym_type'" - continue - fi - - # if the user provided a size, make sure it matches the symbol's size - if [[ -n $size ]] && [[ $size -ne $sym_size ]]; then + # If the user provided a size, make sure it matches the symbol's size: + if [[ -n $user_size ]] && [[ $user_size -ne $sym_size ]]; then [[ $print_warnings =3D 1 ]] && - echo "skipping $func address at $addr due to size mismatch ($size !=3D= $sym_size)" + echo "skipping $sym_name address at $addr due to size mismatch ($user_= size !=3D $sym_size)" continue; fi =20 - # make sure the provided offset is within the symbol's range + # Make sure the provided offset is within the symbol's range: if [[ $offset -gt $sym_size ]]; then [[ $print_warnings =3D 1 ]] && - echo "skipping $func address at $addr due to size mismatch ($offset > = $sym_size)" + echo "skipping $sym_name address at $addr due to size mismatch ($offse= t > $sym_size)" continue fi =20 - # separate multiple entries with a blank line + # In case of duplicates or multiple addresses specified on the + # cmdline, separate multiple entries with a blank line: [[ $FIRST =3D 0 ]] && echo FIRST=3D0 =20 - # pass real address to addr2line - echo "$func+$offset/$sym_size:" - local file_lines=3D$(${ADDR2LINE} -fpie $objfile $addr | sed "s; $dir_pr= efix\(\./\)*; ;") - [[ -z $file_lines ]] && return + echo "$sym_name+$offset/$sym_size:" =20 + # Pass section address to addr2line and strip absolute paths + # from the output: + local output=3D$(${ADDR2LINE} -fpie $objfile $addr | sed "s; $dir_prefix= \(\./\)*; ;") + [[ -z $output ]] && continue + + # Default output (non --list): if [[ $LIST =3D 0 ]]; then - echo "$file_lines" | while read -r line + echo "$output" | while read -r line do echo $line done DONE=3D1; - return + continue fi =20 - # show each line with context - echo "$file_lines" | while read -r line + # For --list, show each line with its corresponding source code: + echo "$output" | while read -r line do echo echo $line @@ -184,12 +228,12 @@ __faddr2line() { n1=3D$[$n-5] n2=3D$[$n+5] f=3D$(echo $line | sed 's/.*at \(.\+\):.*/\1/g') - awk 'NR>=3Dstrtonum("'$n1'") && NR<=3Dstrtonum("'$n2'") { if (NR=3D=3D'= $n') printf(">%d<", NR); else printf(" %d ", NR); printf("\t%s\n", $0)}' $f + ${AWK} 'NR>=3Dstrtonum("'$n1'") && NR<=3Dstrtonum("'$n2'") { if (NR=3D= =3D'$n') printf(">%d<", NR); else printf(" %d ", NR); printf("\t%s\n", $0)}= ' $f done =20 DONE=3D1 =20 - done < <(${NM} -n $objfile | awk -v fn=3D$func -v end=3D$file_end '$3 =3D= =3D fn { found=3D1; line=3D$0; start=3D$1; next } found =3D=3D 1 { found=3D= 0; print line, "0x"$1 } END {if (found =3D=3D 1) print line, end; }') + done < <(${READELF} --symbols --wide $objfile | ${AWK} -v fn=3D$sym_name = '$4 =3D=3D "FUNC" && $8 =3D=3D fn') } =20 [[ $# -lt 2 ]] && usage --=20 2.34.1