[PATCH 0/4] Expose the isa-string via the AT_BASE_PLATFORM aux vector

Palmer Dabbelt palmer at dabbelt.com
Fri Apr 28 12:28:09 PDT 2023


On Fri, 28 Apr 2023 11:59:31 PDT (-0700), philipp.tomsich at vrull.eu wrote:
> On Fri, 28 Apr 2023 at 20:48, Christoph Müllner
> <christoph.muellner at vrull.eu> wrote:
>>
>> On Fri, Apr 28, 2023 at 4:57 PM Palmer Dabbelt <palmer at dabbelt.com> wrote:
>> >
>> > On Mon, 24 Apr 2023 12:49:07 PDT (-0700), heiko at sntech.de wrote:
>> > > From: Heiko Stuebner <heiko.stuebner at vrull.eu>
>> > >
>> > > The hwprobing infrastructure was merged recently [0] and contains a
>> > > mechanism to probe both extensions but also microarchitecural features
>> > > on a per-core level of detail.
>> > >
>> > > While discussing the solution internally we identified some possible issues,
>> > > tried to understand the underlying issue and come up with a solution for it.
>> > > All these deliberations overlapped with hwprobing being merged, but that
>> > > shouldn't really be an issue, as both have their usability - see below.
>> > >
>> > > Also please see the "Things to consider" at the bottom!
>> > >
>> > >
>> > > Possible issues:
>> > > - very much limited to Linux
>> > > - schedulers run processes on all cores by default, so will need
>> > >   the common set of extensions in most cases
>> > > - each new extensions requires an uapi change, requiring at least
>> > >   two pieces of software to be changed
>> > > - adding another extension requires a review process (only known
>> > >   extensions can be exposed to user-space)
>> > > - vendor extensions have special needs and therefore possibly
>> > >   don’t fit well
>> > >
>> > >
>> > > Limited to Linux:
>> > > -----------------
>> > >
>> > > The syscall and its uapi is Linux-specific and other OSes probably
>> > > will not defer to our review process and requirements just to get
>> > > new bits in. Instead most likely they'll build their own systems,
>> > > leading to fragmentation.
>> > >
>> > >
>> > > Feature on all cores:
>> > > ---------------------
>> > >
>> > > Arnd previously ([1]) commented in the discussion, that there
>> > > should not be a need for optimization towards hardware with an
>> > > asymmetric set of features. We believe so as well, especially
>> > > when talking about an interface that helps processes to identify
>> > > the optimized routines they can execute.
>> > >
>> > > Of course seeing it with this finality might not take into account
>> > > the somewhat special nature of RISC-V, but nevertheless it describes
>> > > the common case for programs very well.
>> > >
>> > > For starters the scheduler in its default behaviour, will try to use any
>> > > available core, so the standard program behaviour will always need the
>> > > intersection of available extensions over all cores.
>> > >
>> > >
>> > > Limiting program execution to specific cores will likely always be a
>> > > special use case and already requires Linux-specific syscalls to
>> > > select the set of cores.
>> > >
>> > > So while it can come in handy to get per-core information down the road
>> > > via the hwprobing interface, most programs will just want to know if
>> > > they can use a extension on just any core.
>> > >
>> > >
>> > > Review process:
>> > > ---------------
>> > >
>> > > There are so many (multi-letter-)extensions already with even more in
>> > > the pipeline. To expose all of them, each will require a review process
>> > > and uapi change that will make defining all of them slow as the
>> > > kernel needs patching after which userspace needs to sync in the new
>> > > api header.
>> >
>> > The whole reason we're doing hwprobe with bitmaps is beacuse the ISA
>> > strings are not a stable interface, and thus are not suitable for
>> > building uABI around.
>>
>> The ISA string is the main description of the RISC-V ISA that a
>> system/core/hart implements. It is suitable and stable enough for all toolchain
>> components (-march string, ELF header, etc.).
>> It is also used in the DTB that the kernel uses to identify available
>> extensions.
>> So it is reasonable to argue that it is good enough for all runtime components.
>> Also, I don't see any evidence that users are affected by any stability issues
>> of the ISA strings in the interfaces where it is used at the moment.
>> Quite the opposite, users are expecting ISA string interfaces everywhere
>> because of existing interfaces.
>>
>> Besides that, also the kernel stable ABI documentation allows changes:
>>   "Userspace programs are free to use these
>>   interfaces with no restrictions, and backward compatibility for
>>   them will be guaranteed for at least 2 years." [1]
>> I did not come across any issues in the RISC-V ISA string that would violate
>> these requirements. Did you? Further, the vDSO is covered by the stable ABI
>> requirements, but not the auxiliary vector. This does not imply that an ISA
>> string interface in the aux vector does not have to be stable at all, but there
>> is certainly enough room for any ISA extension errata that may pop up in the
>> future. Other architectures can live with that risk as well.
>
> To provide a slightly different perspective, arriving at a similar conclusion...
>
> The ISA string is part of the psABI (see
> https://github.com/riscv-non-isa/riscv-elf-psabi-doc/releases/tag/v1.0)
> to identify the target architecture through Tag_RISCV_arch.  As such,
> it already provides an interface that the kernel will have to consume
> (during process startup) and has to be reasonably stable. The ELF
> auxiliary vector is closely related to and should generally follow the
> lead of the psABI definitions (which already use this string), which
> makes the ISA string a natural encoding for exposing this information
> to userspace programs.

There were so many breakages due to that tag we just turned it off.

> Cheers,
> Philipp.
>
>>
>>
>> BR
>> Christoph
>>
>> [1] https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/tree/Documentation/ABI/README
>>
>> >
>> > >
>> > >
>> > > Vendor-extensions:
>> > > ------------------
>> > >
>> > > Vendor extensions are special in their own right.
>> > > Userspace probably will want to know about them, but we as the kernel
>> > > don't want to care about them too much (except as errata), as they're
>> > > not part of the official RISC-V ISA spec.
>> > >
>> > > Getting vendor extensions from the dt to userspace via hwprobe would
>> > > require coordination efforts and as vendors have the tendency to invent
>> > > things during their development process before trying to submit changes
>> > > upstream this likely would result in conflicts with assigned ids down
>> > > the road. Which in turn then may create compatibility-issues with
>> > > userspace builds built on top of the mainline kernel or a pre-
>> > > existing vendor kernel.
>> > >
>> > > The special case also is that vendor A could in theory implement an
>> > > extension from vendor B. So this would require to actually assign
>> > > separate hwprobe keys to vendors (key for xthead extensions, key for
>> > > xventana extensions, etc). This in turn would require vendors to
>> > > come to the mainline kernel to get assigned a key (which in reality
>> > > probably won't happen), which would then make the kernel community
>> > > sort of an id authority.
>> > >
>> > >
>> > >
>> > >
>> > > To address these, the attached patch series adds a second interface
>> > > for the common case and "just" exposes the isa-string via the
>> > > AT_BASE_PLATFORM aux vector.
>> > >
>> > > In the total cost of program start, parsing the string does not create
>> > > too much overhead. The extension list in the kernel already contains
>> > > the extensions list limited to the ones available on all harts and
>> > > the string form allows us to just pipe through additional stuff too, as
>> > > can be seen in the example for T-Head extensions [2] .
>> > >
>> > > This of course does not handle the microarchitecture things that
>> > > the hwprobe syscall provides but allows a more generalized view
>> > > onto the available ISA extensions, so is not intended to replace
>> > > hwprobe, but to supplement it.
>> > >
>> > > AT_BASE_PLATFORM itself is somewhat well established, with PPC already
>> > > using it to also expose a platform-specific string that identifies
>> > > the platform in finer grain so this aux-vector field could in theory
>> > > be used by other OSes as well.
>> > >
>> > >
>> > > A random riscv64-qemu could for example provide:
>> > >     rv64imafdcvh_zicbom_zihintpause_zbb_sscofpmf_sstc_svpbmt
>> > >
>> > > where a d1-nezha provides:
>> > >     rv64imafdc_xtheadba_xtheadbb_xtheadbs_xtheadcmo_xtheadcondmov_xtheadfmemidx_xtheadint_xtheadmac_xtheadmemidx_xtheadmempair_xtheadsync
>> > >
>> > >
>> > > Things to still consider:
>> > > -------------------------
>> > >
>> > > Right now both hwprobe and this approach will only pass through
>> > > extensions the kernel actually knows about itself. This should not
>> > > necessarily be needed (but could be an optional feature for e.g. virtualization).
>> > >
>> > > Most extensions don’t introduce new user-mode state that the kernel needs to manage (e.g. new registers). Extension that do introduce new user-mode state are usually disabled by default and have to be enabled by S mode or M mode (e.g. FS[1:0] for the floating-point extension). So there should not be a reason to filter any extensions that are unknown.
>> > >
>> > > So it might make more sense to just pass through a curated list (common
>> > > set) created from the core's isa strings and remove state-handling
>> > > extensions when they are not enabled in the kernel-side (sort of
>> > > blacklisting extensions that need actual kernel support).
>> > >
>> > > However, this is a very related, but still independent discussion.
>> > >
>> > >
>> > > [0] https://lore.kernel.org/lkml/168191462224.22791.2281450562691381145.git-patchwork-notify@kernel.org/
>> > > [1] https://lore.kernel.org/all/605fb2fd-bda2-4922-92bf-e3e416d54398@app.fastmail.com/
>> > > [2] These are the T-Head extensions available in _upstream_ toolchains
>> > >
>> > > Heiko Stuebner (4):
>> > >   RISC-V: create ISA string separately - not as part of cpuinfo
>> > >   RISC-V: don't parse dt isa string to get rv32/rv64
>> > >   RISC-V: export the ISA string of the running machine in the aux vector
>> > >   RISC-V: add support for vendor-extensions via AT_BASE_PLATFORM and
>> > >     xthead
>> > >
>> > >  arch/riscv/errata/thead/errata.c     | 43 ++++++++++++
>> > >  arch/riscv/include/asm/alternative.h |  4 ++
>> > >  arch/riscv/include/asm/elf.h         | 10 +++
>> > >  arch/riscv/kernel/alternative.c      | 21 ++++++
>> > >  arch/riscv/kernel/cpu.c              | 98 +++++++++++++++++++++++++---
>> > >  5 files changed, 168 insertions(+), 8 deletions(-)



More information about the linux-riscv mailing list