Backports to stable

Backports to stable

[Marek Marczykowski-Górecki]

[-- Attachment #1.1: Type: text/plain, Size: 2306 bytes --]

Hi,

I'd like to propose backporting GCC7/8 fixes to all stable branches. Below
is a list up to stable-4.6, but some of the patches are already on
select branches (developed during that release cycle, or already
backported). If you want, I can prepare separate lists for each branch,
but in practice `git cherry-pick` do the right thing already.

    e0a97098e2 x86: fix section type mismatch in mm.c
    fa7789ef18 tools/libxc: fix strncpy size
    796dea37fb tools: include sys/sysmacros.h on Linux
    850e89b3ef tools/blktap2: fix possible '\0' truncation
    3a633c2614 tools/blktap2: fix hypothetical buffer overflow
    f49fa658b5 tools:misc:xenlockprof: fix possible format string overflow
    27751d8924 tools/misc: fix hypothetical buffer overflow in xen-lowmemd
    ebdba150bf xenalyze: fix misleading indentation.
    87761277d7 xenalyze: remove cr3_compare_total
    c2a17869d5 libfsimage: replace deprecated readdir_r() with readdir()
    938c8f53b1 tools/xenpmd: fix possible '\0' truncation
    e75c9dc85f xenpmd: make 32 bit gcc 8.1 non-debug build work
    2d78f78a14 tools: fix several "format-truncation" warnings with GCC 7
    b9daff9d81 libxl: replace deprecated readdir_r() with readdir()
    b021beb21a tools: libxl: NULL terminate partially constructed hex string
    7f601f7c34 tools/gdbsx: fix -Wstringop-truncation warning

    # This one doesn't apply cleanly, because acpi stuff moved
    # tools/firmware/hvmloader/acpi -> tools/acpi
    858dbaaeda libacpi: fixes for iasl >= 20180427

    # Not sure if a good idea in a stable branch; if not acceptable, I can extract
    # gcc-7/gcc-8 related commits and add them to tools/firmware/etherboot/patches/
    461b216534 ipxe: update to newer commit

Another thing worth backporting is xs_watch thread stack size fix +
related patches:

    b4f98dc0d8 tools: fix typo in tools/Rules.mk
    448c03b3cb tools/xenstore: try to get minimum thread stack size for watch thread
    c9bd8a7365 tools/xenstore: add libdl dependency to libxenstore
    1a37319492 tools/xenstore: fix linking libxenstore with ldl

-- 
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

[-- Attachment #2: Type: text/plain, Size: 157 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: Backports to stable

[Jan Beulich]

>>> On 07.10.18 at 03:04, <marmarek@invisiblethingslab.com> wrote:
> I'd like to propose backporting GCC7/8 fixes to all stable branches. Below
> is a list up to stable-4.6, but some of the patches are already on
> select branches (developed during that release cycle, or already
> backported).

I continue to be opposed to backporting anything that is not needed
for dealing security issues to branches which are in security-support-
only mode, i.e. anything older than 4.8 at this point in time.

Furthermore I notice that 4.6 has just moved out of security support,
at the end of last week.

>     e0a97098e2 x86: fix section type mismatch in mm.c

This describes itself as a Clang fix - has it become relevant for
gcc now too? Anyway - this has been in even the original 4.7.0,
so as per above not a candidate for any actively maintained
branch.

>     # This one doesn't apply cleanly, because acpi stuff moved
>     # tools/firmware/hvmloader/acpi -> tools/acpi
>     858dbaaeda libacpi: fixes for iasl >= 20180427

Iirc I've applied this back to 4.8 already, and quite some time ago.

Jan



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: Backports to stable

[Marek Marczykowski]

[-- Attachment #1.1: Type: text/plain, Size: 1809 bytes --]

On Mon, Oct 08, 2018 at 03:29:06AM -0600, Jan Beulich wrote:
> >>> On 07.10.18 at 03:04, <marmarek@invisiblethingslab.com> wrote:
> > I'd like to propose backporting GCC7/8 fixes to all stable branches. Below
> > is a list up to stable-4.6, but some of the patches are already on
> > select branches (developed during that release cycle, or already
> > backported).
> 
> I continue to be opposed to backporting anything that is not needed
> for dealing security issues to branches which are in security-support-
> only mode, i.e. anything older than 4.8 at this point in time.

Ok, noted. It's hard to compile those still security-supported versions
on recent systems. But if one need such combination (old Xen + new other
things), then can also apply those patches locally. I just wanted to
reduce work duplication.

> Furthermore I notice that 4.6 has just moved out of security support,
> at the end of last week.

Hmm, 18+18 months from October 13, 2015 is at the end of this week.

> >     e0a97098e2 x86: fix section type mismatch in mm.c
> 
> This describes itself as a Clang fix - has it become relevant for
> gcc now too?

Yes, for gcc 8.1 at least.

> Anyway - this has been in even the original 4.7.0,
> so as per above not a candidate for any actively maintained
> branch.
> 
> >     # This one doesn't apply cleanly, because acpi stuff moved
> >     # tools/firmware/hvmloader/acpi -> tools/acpi
> >     858dbaaeda libacpi: fixes for iasl >= 20180427
> 
> Iirc I've applied this back to 4.8 already, and quite some time ago.

Yes, this one is missing only in 4.7 and 4.6.

-- 
Best Regards,
Marek Marczykowski-Górecki
Invisible Things Lab
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?

[-- Attachment #1.2: signature.asc --]
[-- Type: application/pgp-signature, Size: 488 bytes --]

[-- Attachment #2: Type: text/plain, Size: 157 bytes --]

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel

Re: Backports to stable

[Jan Beulich]

>>> On 08.10.18 at 12:40, <marmarek@invisiblethingslab.com> wrote:
> On Mon, Oct 08, 2018 at 03:29:06AM -0600, Jan Beulich wrote:
>> >>> On 07.10.18 at 03:04, <marmarek@invisiblethingslab.com> wrote:
>> > I'd like to propose backporting GCC7/8 fixes to all stable branches. Below
>> > is a list up to stable-4.6, but some of the patches are already on
>> > select branches (developed during that release cycle, or already
>> > backported).
>> 
>> I continue to be opposed to backporting anything that is not needed
>> for dealing security issues to branches which are in security-support-
>> only mode, i.e. anything older than 4.8 at this point in time.
> 
> Ok, noted. It's hard to compile those still security-supported versions
> on recent systems. But if one need such combination (old Xen + new other
> things), then can also apply those patches locally. I just wanted to
> reduce work duplication.
> 
>> Furthermore I notice that 4.6 has just moved out of security support,
>> at the end of last week.
> 
> Hmm, 18+18 months from October 13, 2015 is at the end of this week.

Quite possible - I'm not judging by announcement date, but by the
time stamp of xen/Makefile (which is always the last thing to get
updated, for the version adjustment).

Jan



_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel