[PATCH v3 0/3] Change canonicalization for non-split speed [BZ 33340]

[H.J. Lu]

On Tue, Nov 18, 2025 at 11:52 AM H. Peter Anvin <hpa@zytor.com> wrote:
>
> On November 17, 2025 7:37:43 PM PST, "H.J. Lu" <hjl.tools@gmail.com> wrote:
> >On Tue, Nov 18, 2025 at 11:02 AM H. Peter Anvin <hpa@zytor.com> wrote:
> >>
> >> On November 17, 2025 6:40:29 PM PST, "H.J. Lu" <hjl.tools@gmail.com> wrote:
> >> >On Tue, Nov 18, 2025 at 8:28 AM H. Peter Anvin <hpa@zytor.com> wrote:
> >> >>
> >> >> On 2025-10-20 13:42, H. Peter Anvin wrote:
> >> >> > So I finally found some time to look at this again. The Alpha build
> >> >> > fix turned out to be quite trivial.
> >> >> >
> >> >> > This patch changes the canonicalization for the termios interface to
> >> >> > set c_cflag & CIBAUD == 0 when not using split speed, as that appears
> >> >> > to be more compatible with applications that frob the kernel interface
> >> >> > directly, due to the lack of glibc support for arbitrary speeds for
> >> >> > some 17 years.
> >> >> >
> >> >> > This also adds tests for the kernel-side canonicalization.
> >> >> >
> >> >> > [ v2: fixed comment style ]
> >> >> > [ v3: fixed build breakage on Alpha ]
> >> >> >
> >> >>
> >> >> <ping again?>
> >> >>
> >> >
> >> >All patches have been reviewed.  First 2 have been approved and
> >> >the third patch needs some updates:
> >> >
> >> >https://patchwork.sourceware.org/project/glibc/patch/20251020204213.712303-4-hpa@zytor.com/
> >> >
> >> >
> >>
> >> Yes, and I posted an update; inadvertently labeled v2.1 instead of v3.1.
> >
> >I think you can commit the first 2 patches and repost the 3rd one with
> >the correct label.
> >
>
> Well, I don't have commit privileges.

I pushed the first 2 patches for you.

-- 
H.J.