unofficial mirror of libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: "H.J. Lu via Libc-alpha" <libc-alpha@sourceware.org>
To: "Carlos O'Donell" <carlos@redhat.com>
Cc: libc-alpha@sourceware.org
Subject: Re: [PATCH 1/2] Map ABI/VERSION of rtld to ABI/VERSION of ld [BZ #28132]
Date: Thu, 6 Oct 2022 12:52:57 -0700	[thread overview]
Message-ID: <CAMe9rOqxcM+Bp_Y5SESt3iSeMVKrkWqcajF0tiyE6HXohKoyqg@mail.gmail.com> (raw)
In-Reply-To: <CAMe9rOoKtOkqqnNSh3qHCPTh5u2TNjV7TsN3ojP0eb=KVL-qDQ@mail.gmail.com>

On Thu, Oct 6, 2022 at 11:46 AM H.J. Lu <hjl.tools@gmail.com> wrote:
>
> On Thu, Oct 6, 2022 at 9:39 AM Carlos O'Donell <carlos@redhat.com> wrote:
> >
> > On Sun, Aug 01, 2021 at 09:29:39PM -0700, H.J. Lu via Libc-alpha wrote:
> > > Since the module name of ld is rtld, map ABI/VERSION of rtld to ABI/VERSION
> > > of ld.  This fixes BZ #28132.
> >
> > Reviewing old patches that are still outstanding as part of the queue
> > review in patchwork. Hopefully we catch up to the point where I'm not
> > reviewing year old patches in the queue. However, some of these patches
> > are interesting and valuable so I'm reviving them to review potential
> > solutions.
> >
> > In scripts/gen-libc-modules.awk we have this code:
> >  21   if (name == "ld")
> >  22     name = "rtld"
> >
> > This means we are already handling this processing in other places. A
> > clean solution needs to consider:
> >
> > scripts/abi-version.awk (no changes, should just use rtld)
> > scripts/gen-libc-modules.awk (ld vs rtld)
> > shlib-versions (ld=...)
> >
> > It feels like we need to refactor from the top just use rtld everywhere
> > we accidentally used ld as our identifier.
>
> Since LD_SO is defined in lib-names.h, change ld.so to rtld.so requires
> scripts/lib-names.awk change.   It is a much bigger and messy change.
>

We change all internal usages to rtld by

diff --git a/elf/dl-compat.c b/elf/dl-compat.c
index 05c986a8be..6691ad1cc9 100644
--- a/elf/dl-compat.c
+++ b/elf/dl-compat.c
@@ -22,7 +22,7 @@
 /* The GLIBC_2.35 symbol version is present naturally for later ports.
    Use OTHER_SHLIB_COMPAT because the module is called rtld, but the
    ABI version uses ld.  */
-#if OTHER_SHLIB_COMPAT (ld, GLIBC_2_0, GLIBC_2_35)
+#if OTHER_SHLIB_COMPAT (rtld, GLIBC_2_0, GLIBC_2_35)
 void
 attribute_compat_text_section
 __attribute_used__
@@ -30,6 +30,6 @@ __rtld_version_placeholder_1 (void)
 {
 }

-compat_symbol (ld, __rtld_version_placeholder_1,
+compat_symbol (rtld, __rtld_version_placeholder_1,
                __rtld_version_placeholder, GLIBC_2_34);
 #endif
diff --git a/scripts/abi-versions.awk b/scripts/abi-versions.awk
index c369793459..5d229b2977 100644
--- a/scripts/abi-versions.awk
+++ b/scripts/abi-versions.awk
@@ -25,6 +25,8 @@ $2 == "=" {
   gsub(/[^A-Za-z0-9_ ]/, "_");
   oldid = $1; newid = $3;

+  if (libid == "ld")
+    libid = "rtld";
   printf "#define ABI_%s_%s\tABI_%s_%s\n", libid, oldid, libid, newid;
   printf "#define VERSION_%s_%s\t%s\n", libid, oldid, new;

@@ -36,6 +38,8 @@ $2 == "=" {
   gsub(/[^A-Za-z0-9_ ]/, "_");
   versid = $1;

+  if (libid == "ld")
+    libid = "rtld";
   printf "#define ABI_%s_%s\t%d\t/* support %s */\n", libid, versid, ++n, vers;
   printf "#define VERSION_%s_%s\t%s\n", libid, versid, vers;
   next;



-- 
H.J.

      reply	other threads:[~2022-10-06 19:53 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-08-02  4:29 [PATCH 1/2] Map ABI/VERSION of rtld to ABI/VERSION of ld [BZ #28132] H.J. Lu via Libc-alpha
2021-08-02  4:29 ` [PATCH 2/2] ld.so: Hide _r_debug to support DT_DEBUG [BZ #28130] H.J. Lu via Libc-alpha
2021-08-02 13:29   ` Andreas Schwab
2021-08-02 20:43     ` H.J. Lu via Libc-alpha
2021-08-02 21:19     ` Florian Weimer via Libc-alpha
2021-08-02 21:46       ` H.J. Lu via Libc-alpha
2021-08-02 23:06         ` H.J. Lu via Libc-alpha
2022-10-06 16:38 ` [PATCH 1/2] Map ABI/VERSION of rtld to ABI/VERSION of ld [BZ #28132] Carlos O'Donell via Libc-alpha
2022-10-06 18:46   ` H.J. Lu via Libc-alpha
2022-10-06 19:52     ` H.J. Lu via Libc-alpha [this message]

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

  List information: https://www.gnu.org/software/libc/involved.html

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=CAMe9rOqxcM+Bp_Y5SESt3iSeMVKrkWqcajF0tiyE6HXohKoyqg@mail.gmail.com \
    --to=libc-alpha@sourceware.org \
    --cc=carlos@redhat.com \
    --cc=hjl.tools@gmail.com \
    /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: link
Be 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).