| Message ID | 20250226104556.36324-1-oleksii.kurochko@gmail.com (mailing list archive) |
|---|---|
| State | Superseded |
| Headers | show |
| Series | [for,4.20,v2] CHANGELOG.md: Finalize changes in 4.20 release cycle | expand |
On 26.02.2025 11:45, Oleksii Kurochko wrote: > @@ -34,6 +40,9 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) > - On x86: > - xl suspend/resume subcommands. > - `wallclock` command line option to select time source. > + - Add Support for Paging-Write Feature. That EPT (i.e. Intel) only, which may want making explicit? > + - Zen5 support (including new hardware support to mitigate the SRSO > + speculative vulnerability). I'd also suggest to qualify Zen5 with AMD. Whether to mention this here when I think I backported all the pieces isn't entirely clear to me either. Jan
On 2/26/25 11:50 AM, Jan Beulich wrote: > On 26.02.2025 11:45, Oleksii Kurochko wrote: >> @@ -34,6 +40,9 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) >> - On x86: >> - xl suspend/resume subcommands. >> - `wallclock` command line option to select time source. >> + - Add Support for Paging-Write Feature. > That EPT (i.e. Intel) only, which may want making explicit? Agree, it would be better to write: "Add support forEPT paging-write feature". > >> + - Zen5 support (including new hardware support to mitigate the SRSO >> + speculative vulnerability). > I'd also suggest to qualify Zen5 with AMD. I thought that it is clear just from the name for a CPU microachitecture: Zen5 which I expect to be develop by AMD. Anyway, if it is really better I will add AMD before Zen5. > Whether to mention this here > when I think I backported all the pieces isn't entirely clear to me either. What is the better place then? ~ Oleksii
On 26.02.2025 14:03, Oleksii Kurochko wrote: > > On 2/26/25 11:50 AM, Jan Beulich wrote: >> On 26.02.2025 11:45, Oleksii Kurochko wrote: >>> @@ -34,6 +40,9 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) >>> - On x86: >>> - xl suspend/resume subcommands. >>> - `wallclock` command line option to select time source. >>> + - Add Support for Paging-Write Feature. >> That EPT (i.e. Intel) only, which may want making explicit? > > Agree, it would be better to write: "Add support forEPT paging-write feature". "Intel EPT" perhaps, for the same reason as using "AMD" below. >>> + - Zen5 support (including new hardware support to mitigate the SRSO >>> + speculative vulnerability). >> I'd also suggest to qualify Zen5 with AMD. > > I thought that it is clear just from the name for a CPU microachitecture: Zen5 which > I expect to be develop by AMD. Anyway, if it is really better I will add AMD before Zen5. > >> Whether to mention this here >> when I think I backported all the pieces isn't entirely clear to me either. > > What is the better place then? The question isn't where to put it, but whether to in the first place. Jan
On 2/26/25 2:13 PM, Jan Beulich wrote: >>>> + - Zen5 support (including new hardware support to mitigate the SRSO >>>> + speculative vulnerability). >>> I'd also suggest to qualify Zen5 with AMD. >> I thought that it is clear just from the name for a CPU microachitecture: Zen5 which >> I expect to be develop by AMD. Anyway, if it is really better I will add AMD before Zen5. >> >>> Whether to mention this here >>> when I think I backported all the pieces isn't entirely clear to me either. >> What is the better place then? > The question isn't where to put it, but whether to in the first place. Wouldn't it be useful to highlight that Xen now supports the new security feature for mitigating SRSO vulnerabilities on AMD Zen5? ~ Oleksii
On 26.02.2025 15:31, Oleksii Kurochko wrote: > > On 2/26/25 2:13 PM, Jan Beulich wrote: >>>>> + - Zen5 support (including new hardware support to mitigate the SRSO >>>>> + speculative vulnerability). >>>> I'd also suggest to qualify Zen5 with AMD. >>> I thought that it is clear just from the name for a CPU microachitecture: Zen5 which >>> I expect to be develop by AMD. Anyway, if it is really better I will add AMD before Zen5. >>> >>>> Whether to mention this here >>>> when I think I backported all the pieces isn't entirely clear to me either. >>> What is the better place then? >> The question isn't where to put it, but whether to in the first place. > > Wouldn't it be useful to highlight that Xen now supports the new security feature > for mitigating SRSO vulnerabilities on AMD Zen5? I don't know. Thing is what we list here is supposedly new in 4.20. Yet here we're talking about something that was already backported to older versions. I'll admit though I didn't check how much of that made it into any stable release. Jan
On 26/02/2025 10:45 am, Oleksii Kurochko wrote: > Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com> > --- > Changes in v2: > - Drop "Support device passthrough when dom0 is PVH on Xen" from > CHANGELOD.md becuase it isn't really ready: > https://lore.kernel.org/xen-devel/31db7d34-3338-4d88-8721-f2cd4b68f3b9@gmail.com/T/#m725b559864e5ed6163b59a088b437aa10c36ff16 > --- > CHANGELOG.md | 9 +++++++++ > 1 file changed, 9 insertions(+) > > diff --git a/CHANGELOG.md b/CHANGELOG.md > index 1979166820..5f5a40855a 100644 > --- a/CHANGELOG.md > +++ b/CHANGELOG.md > @@ -18,6 +18,11 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) > - Fixed blkif protocol specification for sector sizes different than 512b. > - The dombuilder in libxenguest no longer un-gzips secondary modules, instead > leaving this to the guest kernel to do in guest context. > + - Reduce xenstore library dependencies. What is this in reference to? I don't think all of Juergen's series has been merged yet. > + - On Arm: > + - Several FF-A support improvements: add indirect messages support, transmit > + RXTX buffer to the SPMC, fix version negotication and partition information > + retrieval. > - On x86: > - Prefer ACPI reboot over UEFI ResetSystem() run time service call. > - Prefer CMOS over EFI_GET_TIME as time source. > @@ -25,6 +30,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) > interrupts instead of logical destination mode. > > ### Added > + - Enable CONFIG_UBSAN (Arm, x86, RISC-V) for GitLab CI. +PPC (just backported that). Also, best to say ARM64, because ARM32 is pending the list.h fix which we deemed too invasive. > - On Arm: > - Experimental support for Armv8-R. > - Support for NXP S32G3 Processors Family and NXP LINFlexD UART driver. > @@ -34,6 +40,9 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) > - On x86: > - xl suspend/resume subcommands. > - `wallclock` command line option to select time source. > + - Add Support for Paging-Write Feature. (Just so all my feedback is in one place), "Intel EPT". The average person reading these notes isn't enough of an x86 expert to equate EPT with Intel. > + - Zen5 support (including new hardware support to mitigate the SRSO > + speculative vulnerability). AMD Zen5. Again, the target audience aren't all experts. Although, I'd phrase that as "support, including" without brackets. ~Andrew
On 26/02/2025 2:33 pm, Jan Beulich wrote: > On 26.02.2025 15:31, Oleksii Kurochko wrote: >> On 2/26/25 2:13 PM, Jan Beulich wrote: >>>>>> + - Zen5 support (including new hardware support to mitigate the SRSO >>>>>> + speculative vulnerability). >>>>> I'd also suggest to qualify Zen5 with AMD. >>>> I thought that it is clear just from the name for a CPU microachitecture: Zen5 which >>>> I expect to be develop by AMD. Anyway, if it is really better I will add AMD before Zen5. >>>> >>>>> Whether to mention this here >>>>> when I think I backported all the pieces isn't entirely clear to me either. >>>> What is the better place then? >>> The question isn't where to put it, but whether to in the first place. >> Wouldn't it be useful to highlight that Xen now supports the new security feature >> for mitigating SRSO vulnerabilities on AMD Zen5? > I don't know. Thing is what we list here is supposedly new in 4.20. Yet > here we're talking about something that was already backported to older > versions. I'll admit though I didn't check how much of that made it into > any stable release. This was my suggested wording. Yes we've backported it, but it was also new feature work done during the 4.20 window. ~Andrew
On 26.02.25 16:12, Andrew Cooper wrote: > On 26/02/2025 10:45 am, Oleksii Kurochko wrote: >> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com> >> --- >> Changes in v2: >> - Drop "Support device passthrough when dom0 is PVH on Xen" from >> CHANGELOD.md becuase it isn't really ready: >> https://lore.kernel.org/xen-devel/31db7d34-3338-4d88-8721-f2cd4b68f3b9@gmail.com/T/#m725b559864e5ed6163b59a088b437aa10c36ff16 >> --- >> CHANGELOG.md | 9 +++++++++ >> 1 file changed, 9 insertions(+) >> >> diff --git a/CHANGELOG.md b/CHANGELOG.md >> index 1979166820..5f5a40855a 100644 >> --- a/CHANGELOG.md >> +++ b/CHANGELOG.md >> @@ -18,6 +18,11 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) >> - Fixed blkif protocol specification for sector sizes different than 512b. >> - The dombuilder in libxenguest no longer un-gzips secondary modules, instead >> leaving this to the guest kernel to do in guest context. >> + - Reduce xenstore library dependencies. > > What is this in reference to? I don't think all of Juergen's series has > been merged yet. Not all of the series has been merged, but some library dependencies have been dropped already (e.g. to libxenguest). This is especially affecting the build of xenstore-stubdom positively. Juergen
On 26/02/2025 3:17 pm, Juergen Gross wrote: > On 26.02.25 16:12, Andrew Cooper wrote: >> On 26/02/2025 10:45 am, Oleksii Kurochko wrote: >>> Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com> >>> --- >>> Changes in v2: >>> - Drop "Support device passthrough when dom0 is PVH on Xen" from >>> CHANGELOD.md becuase it isn't really ready: >>> >>> https://lore.kernel.org/xen-devel/31db7d34-3338-4d88-8721-f2cd4b68f3b9@gmail.com/T/#m725b559864e5ed6163b59a088b437aa10c36ff16 >>> --- >>> CHANGELOG.md | 9 +++++++++ >>> 1 file changed, 9 insertions(+) >>> >>> diff --git a/CHANGELOG.md b/CHANGELOG.md >>> index 1979166820..5f5a40855a 100644 >>> --- a/CHANGELOG.md >>> +++ b/CHANGELOG.md >>> @@ -18,6 +18,11 @@ The format is based on [Keep a >>> Changelog](https://keepachangelog.com/en/1.0.0/) >>> - Fixed blkif protocol specification for sector sizes different >>> than 512b. >>> - The dombuilder in libxenguest no longer un-gzips secondary >>> modules, instead >>> leaving this to the guest kernel to do in guest context. >>> + - Reduce xenstore library dependencies. >> >> What is this in reference to? I don't think all of Juergen's series has >> been merged yet. > > Not all of the series has been merged, but some library dependencies have > been dropped already (e.g. to libxenguest). This is especially affecting > the build of xenstore-stubdom positively. Oh, that's good to hear. ~Andrew
On 2/26/25 4:20 PM, Andrew Cooper wrote: > On 26/02/2025 3:17 pm, Juergen Gross wrote: >> On 26.02.25 16:12, Andrew Cooper wrote: >>> On 26/02/2025 10:45 am, Oleksii Kurochko wrote: >>>> Signed-off-by: Oleksii Kurochko<oleksii.kurochko@gmail.com> >>>> --- >>>> Changes in v2: >>>> - Drop "Support device passthrough when dom0 is PVH on Xen" from >>>> CHANGELOD.md becuase it isn't really ready: >>>> >>>> https://lore.kernel.org/xen-devel/31db7d34-3338-4d88-8721-f2cd4b68f3b9@gmail.com/T/#m725b559864e5ed6163b59a088b437aa10c36ff16 >>>> --- >>>> CHANGELOG.md | 9 +++++++++ >>>> 1 file changed, 9 insertions(+) >>>> >>>> diff --git a/CHANGELOG.md b/CHANGELOG.md >>>> index 1979166820..5f5a40855a 100644 >>>> --- a/CHANGELOG.md >>>> +++ b/CHANGELOG.md >>>> @@ -18,6 +18,11 @@ The format is based on [Keep a >>>> Changelog](https://keepachangelog.com/en/1.0.0/) >>>> - Fixed blkif protocol specification for sector sizes different >>>> than 512b. >>>> - The dombuilder in libxenguest no longer un-gzips secondary >>>> modules, instead >>>> leaving this to the guest kernel to do in guest context. >>>> + - Reduce xenstore library dependencies. >>> What is this in reference to? I don't think all of Juergen's series has >>> been merged yet. >> Not all of the series has been merged, but some library dependencies have >> been dropped already (e.g. to libxenguest). This is especially affecting >> the build of xenstore-stubdom positively. Yes, it is connected to stubdom: https://lore.kernel.org/xen-devel/20241010155459.22389-1-jgross@suse.com/ ~ Oleksii > Oh, that's good to hear. > > ~Andrew
On 2/26/25 4:27 PM, Oleksii Kurochko wrote: > > > On 2/26/25 4:20 PM, Andrew Cooper wrote: >> On 26/02/2025 3:17 pm, Juergen Gross wrote: >>> On 26.02.25 16:12, Andrew Cooper wrote: >>>> On 26/02/2025 10:45 am, Oleksii Kurochko wrote: >>>>> Signed-off-by: Oleksii Kurochko<oleksii.kurochko@gmail.com> >>>>> --- >>>>> Changes in v2: >>>>> - Drop "Support device passthrough when dom0 is PVH on Xen" from >>>>> CHANGELOD.md becuase it isn't really ready: >>>>> >>>>> https://lore.kernel.org/xen-devel/31db7d34-3338-4d88-8721-f2cd4b68f3b9@gmail.com/T/#m725b559864e5ed6163b59a088b437aa10c36ff16 >>>>> --- >>>>> CHANGELOG.md | 9 +++++++++ >>>>> 1 file changed, 9 insertions(+) >>>>> >>>>> diff --git a/CHANGELOG.md b/CHANGELOG.md >>>>> index 1979166820..5f5a40855a 100644 >>>>> --- a/CHANGELOG.md >>>>> +++ b/CHANGELOG.md >>>>> @@ -18,6 +18,11 @@ The format is based on [Keep a >>>>> Changelog](https://keepachangelog.com/en/1.0.0/) >>>>> - Fixed blkif protocol specification for sector sizes different >>>>> than 512b. >>>>> - The dombuilder in libxenguest no longer un-gzips secondary >>>>> modules, instead >>>>> leaving this to the guest kernel to do in guest context. >>>>> + - Reduce xenstore library dependencies. >>>> What is this in reference to? I don't think all of Juergen's series has >>>> been merged yet. >>> Not all of the series has been merged, but some library dependencies have >>> been dropped already (e.g. to libxenguest). This is especially affecting >>> the build of xenstore-stubdom positively. > Yes, it is connected to stubdom: > https://lore.kernel.org/xen-devel/20241010155459.22389-1-jgross@suse.com/ > > ~ Oleksii Do we need some rewording for the item in CHANGELOG.md? ~ Oleksii
diff --git a/CHANGELOG.md b/CHANGELOG.md index 1979166820..5f5a40855a 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -18,6 +18,11 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) - Fixed blkif protocol specification for sector sizes different than 512b. - The dombuilder in libxenguest no longer un-gzips secondary modules, instead leaving this to the guest kernel to do in guest context. + - Reduce xenstore library dependencies. + - On Arm: + - Several FF-A support improvements: add indirect messages support, transmit + RXTX buffer to the SPMC, fix version negotication and partition information + retrieval. - On x86: - Prefer ACPI reboot over UEFI ResetSystem() run time service call. - Prefer CMOS over EFI_GET_TIME as time source. @@ -25,6 +30,7 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) interrupts instead of logical destination mode. ### Added + - Enable CONFIG_UBSAN (Arm, x86, RISC-V) for GitLab CI. - On Arm: - Experimental support for Armv8-R. - Support for NXP S32G3 Processors Family and NXP LINFlexD UART driver. @@ -34,6 +40,9 @@ The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/) - On x86: - xl suspend/resume subcommands. - `wallclock` command line option to select time source. + - Add Support for Paging-Write Feature. + - Zen5 support (including new hardware support to mitigate the SRSO + speculative vulnerability). ### Removed - On x86:
Signed-off-by: Oleksii Kurochko <oleksii.kurochko@gmail.com> --- Changes in v2: - Drop "Support device passthrough when dom0 is PVH on Xen" from CHANGELOD.md becuase it isn't really ready: https://lore.kernel.org/xen-devel/31db7d34-3338-4d88-8721-f2cd4b68f3b9@gmail.com/T/#m725b559864e5ed6163b59a088b437aa10c36ff16 --- CHANGELOG.md | 9 +++++++++ 1 file changed, 9 insertions(+)