target-arm queue:
* remove a line of redundant code
* convert various TCG helper fns to use 'fpst' alias
* Use float_status in helper_fcvtx_f64_to_f32
* Use float_status in helper_vfp_fcvt{ds,sd}
* Implement FEAT_XS
* hw/intc/arm_gicv3_its: Zero initialize local DTEntry etc structs
* tests/functional: update sbsa-ref firmware used in test
# -----BEGIN PGP SIGNATURE-----
#
# iQJNBAABCAA3FiEE4aXFk81BneKOgxXPPCUl7RQ2DN4FAmdhsj0ZHHBldGVyLm1h
# eWRlbGxAbGluYXJvLm9yZwAKCRA8JSXtFDYM3vldEACY/BJeDSfE3X6fi2qiVF+p
# hs2NaeiM6HS3+/1oPlPz7sW87cGV58hPW0vOCjZYlfw+Afv2wuL9zb2C7IV+PJOn
# Vy1L/tBezYvgBwi1+bRvLpCge3UKdbRNa6aowGmLnvcnQuZ6lOOcmQ8Jq/a5W1TM
# xOrvPYp0FbcfGXLcDLIluCozupsq4aJsh0gWayzr9zm2tWnzMAhb/GQLuCmLLn35
# pUiAI209xU393AOfdCpAmCGjDCqcqbjHpz0AqrIPtOwaDO3hRlJIMw1eGk2dS0BD
# R0vZG+WFSQMV972reVoTT83W7NmIeeqhZgeKDv2R347EHsbCKt5PIUpWX2zYvk+H
# rfCltOxFVWvJ0e33b9opk2/GfgckkTGw+6ZYMIdMm8UxABrdEaHcWCV0qMyLk2JH
# 4EKYlmKeR8yqkPRDbGkRVANEIDxRBmL96kN6il0wSM742y7UMwMzP3C344Jg0tf/
# AhGrwjsuRW8oeEZMgk9Z0i/J6q3CNxRQSVGGQtYsEt8fs5OXLltLrXEX+aZNF5ua
# ry5SfLWlwIR+0AO4oNaJqJYNQArqhzDUgsY4ryzrueZnaaMShobMn0AP2H3+/l4X
# W3wlIqOQ97ivk0Snc9WpDQyhPPOZuj4LN4IkTkodHu7+eoMrkvojf/BVc5kku2VL
# dE224ctbEbKsbydwubSVfQ==
# =Z/7q
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 17 Dec 2024 12:17:49 EST
# gpg: using RSA key E1A5C593CD419DE28E8315CF3C2525ED14360CDE
# gpg: issuer "peter.maydell@linaro.org"
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>" [full]
# gpg: aka "Peter Maydell <pmaydell@gmail.com>" [full]
# gpg: aka "Peter Maydell <pmaydell@chiark.greenend.org.uk>" [full]
# gpg: aka "Peter Maydell <peter@archaic.org.uk>" [unknown]
# Primary key fingerprint: E1A5 C593 CD41 9DE2 8E83 15CF 3C25 25ED 1436 0CDE
* tag 'pull-target-arm-20241217' of https://git.linaro.org/people/pmaydell/qemu-arm:
tests/functional: update sbsa-ref firmware used in test
hw/intc/arm_gicv3_its: Zero initialize local DTEntry etc structs
tests/tcg/aarch64: add system test for FEAT_XS
target/arm: Enable FEAT_XS for the max cpu
target/arm: Add decodetree entry for DSB nXS variant
target/arm: Add ARM_CP_ADD_TLBI_NXS type flag to TLBI insns
target/arm: Add ARM_CP_ADD_TLBI_NXS type flag for NXS insns
target/arm: Implement fine-grained-trap handling for FEAT_XS
target/arm: Use float_status in helper_vfp_fcvt{ds,sd}
target/arm: Use float_status in helper_fcvtx_f64_to_f32
target/arm: Convert neon_helper.c to use env alias
target/arm: Convert vec_helper.c to use env alias
target/arm: Convert sme_helper.c to fpst alias
target/arm: Convert sve_helper.c to fpst alias
target/arm: Convert neon_helper.c to fpst alias
target/arm: Convert vec_helper.c to fpst alias
target/arm: Convert helper-a64.c to fpst alias
target/arm: Convert vfp_helper.c to fpst alias
target/arm: remove redundant code
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
The macOS builds in our CI (and possibly other very recent distros)
are currently broken since the update to libnfs version 6 there.
That version apparently comes with a big API breakage. v5.0.3 was
the final release of the old API (see the libnfs commit here:
https://github.com/sahlberg/libnfs/commit/4379837 ).
Disallow version 6.x for now to get the broken CI job working
again. Once somebody had enough time to adapt our code in
block/nfs.c, we can revert this change again.
Message-ID: <20241218065157.209020-1-thuth@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Since we don't have ssh support in the functional test framework yet,
simply use the serial console for this test instead. It's also
sufficient to only boot into an initrd here, no need to fire up a
full-blown guest, so the test now finishes much faster.
While we're at it, also unplug the CPU now and check that it is gone
in the guest.
Message-ID: <20241217142020.155776-1-thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Convert the intel_iommu test to the new functional framework.
This test needs some changes since we neither support the old 'LinuxTest'
class in the functional framework yet, nor a way to use SSH for running
commands in the guest. So we now directly download a Fedora kernel and
initrd and set up the serial console for executing the commands and for
looking for the results. Instead of configuring the cloud image via
cloud-init, we now simply mount the file system manually from an initrd
rescue shell.
While the old test was exercising the network with a "dnf install"
command (which is not the best option for the CI since this depends
on third party servers), the new code is now setting up a little
HTTP server in the guest and transfers a file from the guest to the
host instead.
The test should now run much faster and more reliable (since we
don't depend on the third party servers for "dnf install" anymore),
so we can also drop the @skipUnless decorator now.
Message-ID: <20241217121550.141072-3-thuth@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
It's just a wrapper around get_info_usernet_hostfwd_port from the
qemu module that is also calling the right monitor command for
retrieving the information from QEMU.
Message-ID: <20241217121550.141072-2-thuth@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
We see periodic errors caching assets due to a combination of transient
networking and server problems. With the previous patch to skip running
a test when it has missing assets, we can now treat most cache download
errors as non-fatal.
Only HTTP 404 is retained as fatal, since it is a strong indicator of
a fully broken test rather than a transient error.
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-32-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
This helper wrappers archive.archive_extract, forcing the use of the
scratch directory, to ensure any extracted files are cleaned at test
termination. If a specific member is requested, then the path to the
extracted file is also returned.
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-23-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
There are many types of archives that the tests deal with. Provide
a generalized 'archive_extract' that can detect the format and
delegate to the appropriate helper for extraction. This ensures
that all archive extraction code follows the same design pattern.
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-22-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
More uncompress related code will be added shortly, so having a
separate file makes more sense.
The utils.py imports the functions from archive.py, so that
existing callers don't need to be modified. This avoids
redundant code churn until later in the series when all
calls will be adapted for other reasons.
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-18-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
More archive related code will be added shortly, so having a
separate file makes more sense.
The utils.py imports the functions from archive.py, so that
existing callers don't need to be modified. This avoids
redundant code churn until later in the series when all
calls will be adapted for other reasons.
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-17-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Add helper methods that construct paths for
* log files - to be preserved at the end of a test
* scratch files - to be purged at the end of a test
* build files - anything relative to the build root
* data files - anything relative to the functional test source root
* socket files - a short temporary dir to avoid UNIX socket limits
These are to be used instead of direct access to the self.workdir,
or self.logdir variables, or any other place where paths are built
manually.
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-11-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reduce repeated boilerplate with some helper decorators:
@skipIfNotPlatform("x86_64", "aarch64")
=> Skip unless the build host platform matches
@skipIfMissingCommands("mkisofs", "losetup")
=> Skips unless all listed commands are found in $PATH
@skipIfMissingImports("numpy", "cv2")
=> Skips unless all listed modules can be imported
@skipFlakyTest("https://gitlab.com/qemu-project/qemu/-/issues/NNN")
=> Skips unless env var requests flaky tests with the
reason documented in the referenced gitlab bug
@skipBigData
=> Skips unless env var permits tests creating big data files
@skipUntrustedTest
=> Skips unless env var permits tests which are potentially
dangerous to the host
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-8-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Update the URLs for the binaries we use for the firmware in the
sbsa-ref functional tests.
The firmware is built using Debian 'bookworm' cross toolchain (gcc
12.2.0).
Used versions:
- Trusted Firmware v2.12.0
- Tianocore EDK2 stable202411
- Tianocore EDK2 Platforms code commit 4b3530d
This allows us to move away from "some git commit on trunk"
to a stable release for both TF-A and EDK2.
Signed-off-by: Marcin Juszkiewicz <marcin.juszkiewicz@linaro.org>
Message-id: 20241125125448.185504-1-marcin.juszkiewicz@linaro.org
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
In the GICv3 ITS model, we have a common coding pattern which has a
local C struct like "DTEntry dte", which is a C representation of an
in-guest-memory data structure, and we call a function such as
get_dte() to read guest memory and fill in the C struct. These
functions to read in the struct sometimes have cases where they will
leave early and not fill in the whole struct (for instance get_dte()
will set "dte->valid = false" and nothing else for the case where it
is passed an entry_addr implying that there is no L2 table entry for
the DTE). This then causes potential use of uninitialized memory
later, for instance when we call a trace event which prints all the
fields of the struct. Sufficiently advanced compilers may produce
-Wmaybe-uninitialized warnings about this, especially if LTO is
enabled.
Rather than trying to carefully separate out these trace events into
"only the 'valid' field is initialized" and "all fields can be
printed", zero-init all the structs when we define them. None of
these structs are large (the biggest is 24 bytes) and having
consistent behaviour is less likely to be buggy.
Cc: qemu-stable@nongnu.org
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/2718
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-id: 20241213182337.3343068-1-peter.maydell@linaro.org
Add system test to make sure FEAT_XS is enabled for max cpu emulation
and that QEMU doesn't crash when encountering an NXS instruction
variant.
Signed-off-by: Manos Pitsidianakis <manos.pitsidianakis@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 20241211144440.2700268-7-peter.maydell@linaro.org
[PMM: In ISAR field test, mask with 0xf, not 0xff; use < rather
than an equality test to follow the standard ID register field
check guidelines]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
All of the TLBI insns with an NXS variant put that variant at the
same encoding but with a CRn field that is one greater than for the
original TLBI insn. To avoid having to define every TLBI insn
effectively twice, once in the normal way and once in a set of cpreg
arrays that are only registered when FEAT_XS is present, we define a
new ARM_CP_ADD_TLB_NXS type flag for cpregs. When this flag is set
in a cpreg struct and FEAT_XS is present,
define_one_arm_cp_reg_with_opaque() will automatically add a second
cpreg to the hash table for the TLBI NXS insn with:
* the crn+1 encoding
* an FGT field that indicates that it should honour HCR_EL2.FGTnXS
* a name with the "NXS" suffix
(If there are future TLBI NXS insns that don't use this same
encoding convention, it is also possible to define them manually.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20241211144440.2700268-3-peter.maydell@linaro.org
FEAT_XS introduces a set of new TLBI maintenance instructions with an
"nXS" qualifier. These behave like the stardard ones except that
they do not wait for memory accesses with the XS attribute to
complete. They have an interaction with the fine-grained-trap
handling: the FGT bits that a hypervisor can use to trap TLBI
maintenance instructions normally trap also the nXS variants, but the
hypervisor can elect to not trap the nXS variants by setting
HCRX_EL2.FGTnXS to 1.
Add support to our FGT mechanism for these TLBI bits. For each
TLBI-trapping FGT bit we define, for example:
* FGT_TLBIVAE1 -- the same value we do at present for the
normal variant of the insn
* FGT_TLBIVAE1NXS -- for the nXS qualified insn; the value of
this enum has an NXS bit ORed into it
In access_check_cp_reg() we can then ignore the trap bit for an
access where ri->fgt has the NXS bit set and HCRX_EL2.FGTnXS is 1.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20241211144440.2700268-2-peter.maydell@linaro.org
Stefan Hajnoczi