Damien Miller
2024-Oct-27 04:45 UTC
FYI: fix for big-endian systems pushed to V_9_9 branch
Hi, This is mostly a note for downstream distributors of OpenSSH. I've just pushed fixes to the V_9_9 stable branch for a bug in the mlkem768x25519-sha256 key exchange algorithm that was added in this release that causes connection failures when connecting between big-endian and little-endian hosts. The problem is on the big-endian side. No change is required for the more common little-endian architectures (e.g. x86, ARM). If you distribute OpenSSH to big-endian systems and have packaged OpenSSH 9.9 already, then I recommend you include these fixes as the next release of OpenSSH will make this key exchange algorithm the default. Thanks, Damien -------------- next part -------------- An embedded message was scrubbed... From: git+noreply at mindrot.org Subject: [openssh-commits] [openssh] branch V_9_9 updated (19bcb2d9 -> 33c5f384) Date: Sun, 27 Oct 2024 15:37:01 +1100 Size: 4551 URL: <http://lists.mindrot.org/pipermail/openssh-unix-dev/attachments/20241027/f6444af8/attachment.eml>
On Sun, Oct 27, 2024 at 03:45:33PM +1100, Damien Miller wrote:> This is mostly a note for downstream distributors of OpenSSH. I've > just pushed fixes to the V_9_9 stable branch for a bug in the > mlkem768x25519-sha256 key exchange algorithm that was added in this > release that causes connection failures when connecting between > big-endian and little-endian hosts. > > The problem is on the big-endian side. No change is required for > the more common little-endian architectures (e.g. x86, ARM). > > If you distribute OpenSSH to big-endian systems and have packaged > OpenSSH 9.9 already, then I recommend you include these fixes as the > next release of OpenSSH will make this key exchange algorithm the > default.Thanks, applied in Debian. (I skipped the third patch, since I believe we have endian.h everywhere.) -- Colin Watson (he/him) [cjwatson at debian.org]
Dmitry Belyavskiy
2024-Oct-28 17:24 UTC
FYI: fix for big-endian systems pushed to V_9_9 branch
Thanks, applied to Fedora rawhide On Sun, Oct 27, 2024 at 5:48?AM Damien Miller <djm at mindrot.org> wrote:> Hi, > > This is mostly a note for downstream distributors of OpenSSH. I've > just pushed fixes to the V_9_9 stable branch for a bug in the > mlkem768x25519-sha256 key exchange algorithm that was added in this > release that causes connection failures when connecting between > big-endian and little-endian hosts. > > The problem is on the big-endian side. No change is required for > the more common little-endian architectures (e.g. x86, ARM). > > If you distribute OpenSSH to big-endian systems and have packaged > OpenSSH 9.9 already, then I recommend you include these fixes as the > next release of OpenSSH will make this key exchange algorithm the > default. > > Thanks, > Damien > > > ---------- Forwarded message ---------- > From: git+noreply at mindrot.org > To: openssh-commits at mindrot.org > Cc: > Bcc: > Date: Sun, 27 Oct 2024 15:37:01 +1100 > Subject: [openssh-commits] [openssh] branch V_9_9 updated (19bcb2d9 -> > 33c5f384) > > This is an automated email from the git hooks/post-receive script. > > djm pushed a change to branch V_9_9 > in repository openssh. > > from 19bcb2d9 upstream: fix previous change to ssh_config Match, which > broken on > new 11f34819 upstream: fix ML-KEM768x25519 KEX on big-endian systems; > spotted by > new fe8d28a7 upstream: explicitly include endian.h > new 33c5f384 htole64() etc for systems without endian.h > > The 3 revisions listed above as "new" are entirely new to this > repository and will be described in separate emails. The revisions > listed as "add" were already present in the repository and have only > been added to this reference. > > > Detailed log of new commits: > > commit 33c5f384ae03a5d1a0bd46ca0fac3c62e4eaf784 > Author: Damien Miller <djm at mindrot.org> > Date: Sun Oct 27 13:28:11 2024 +1100 > > htole64() etc for systems without endian.h > > commit fe8d28a7ebbaa35cfc04a21263627f05c237e460 > Author: djm at openbsd.org <djm at openbsd.org> > Date: Sun Oct 27 02:06:59 2024 +0000 > > upstream: explicitly include endian.h > > OpenBSD-Commit-ID: 13511fdef7535bdbc35b644c90090013da43a318 > > commit 11f348196b3fb51c3d8d1f4f36db9d73f03149ed > Author: djm at openbsd.org <djm at openbsd.org> > Date: Sun Oct 27 02:06:01 2024 +0000 > > upstream: fix ML-KEM768x25519 KEX on big-endian systems; spotted by > > jsg@ feedback/ok deraadt@ > > OpenBSD-Commit-ID: 26d81a430811672bc762687166986cad40d28cc0 > > Summary of changes: > configure.ac | 1 - > defines.h | 26 ++++++++++++++++++++++++++ > kexmlkem768x25519.c | 5 ++++- > libcrux_mlkem768_sha3.h | 8 +++++--- > mlkem768.sh | 17 ++++++++++++----- > 5 files changed, 47 insertions(+), 10 deletions(-) > > -- > To stop receiving notification emails like this one, please contact > djm at mindrot.org. > _______________________________________________ > openssh-commits mailing list > openssh-commits at mindrot.org > https://lists.mindrot.org/mailman/listinfo/openssh-commits > _______________________________________________ > openssh-unix-dev mailing list > openssh-unix-dev at mindrot.org > https://lists.mindrot.org/mailman/listinfo/openssh-unix-dev >-- Dmitry Belyavskiy
Tom G. Christensen
2024-Oct-29 20:34 UTC
FYI: fix for big-endian systems pushed to V_9_9 branch
On 27/10/2024 05:45, Damien Miller wrote:> If you distribute OpenSSH to big-endian systems and have packaged > OpenSSH 9.9 already, then I recommend you include these fixes as the > next release of OpenSSH will make this key exchange algorithm the > default. >I tried to update my Solaris builds but ran into a build error. In file included from kexmlkem768x25519.c:50: libcrux_mlkem768_sha3.h: In function 'core_num__u64_9__to_le_bytes': libcrux_mlkem768_sha3.h:164: error: '__uint64_t' undeclared (first use in this function) libcrux_mlkem768_sha3.h:164: error: (Each undeclared identifier is reported only once libcrux_mlkem768_sha3.h:164: error: for each function it appears in.) libcrux_mlkem768_sha3.h: In function 'core_num__u64_9__from_le_bytes': libcrux_mlkem768_sha3.h:170: error: '__uint64_t' undeclared (first use in this function) gmake: *** [kexmlkem768x25519.o] Error 1 AFAICT __uint64_t is not available on at least Solaris 10 and older. -tgc
Reasonably Related Threads
- FYI: fix for big-endian systems pushed to V_9_9 branch
- FYI: fix for big-endian systems pushed to V_9_9 branch
- FYI: fix for big-endian systems pushed to V_9_9 branch
- [Bug 3739] New: Match parsing requires space before '='
- reminder about our stable branches