One of the stages in IFS image loading process involves loading individual
chunks (test patterns) from test image file to secure memory.
Driver issues a WRMSR(MSR_AUTHENTICATE_AND_COPY_CHUNK) operation to do
this. This operation can take up to 5 msec, and if an interrupt occurs
in between, the AUTH_AND_COPY_CHUNK u-code implementation aborts the
operation.
Interrupt sources such as NMI or SMI are handled by retrying. Regular
interrupts may occur frequently enough to prevent this operation from ever
completing. Disable irq on local cpu around the aforementioned WRMSR to
allow the operation to complete.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ashok Raj <ashok.raj@intel.com>
Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Link: https://lore.kernel.org/r/20240412172349.544064-4-jithu.joseph@intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
"Scan controller error" means that scan hardware encountered an error
prior to doing an actual test on the target CPU. It does not mean that
there is an actual cpu/core failure. "scan signature failure" indicates
that the test result on the target core did not match the expected value
and should be treated as a cpu failure.
Current driver classifies both these scenarios as failures. Modify
the driver to classify this situation with a more appropriate "untested"
status instead of "fail" status.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ashok Raj <ashok.raj@intel.com>
Reviewed-by: Kuppuswamy Sathyanarayanan <sathyanarayanan.kuppuswamy@linux.intel.com>
Link: https://lore.kernel.org/r/20240412172349.544064-2-jithu.joseph@intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Highlights:
- WMI bus driver fixes
- Second attempt (previously reverted) at P2SB PCI rescan deadlock fix
- AMD PMF driver improvements
- MAINTAINERS updates
- Misc. other small fixes and hw-id additions
The following is an automated git shortlog grouped by driver:
MAINTAINERS:
- remove defunct acpi4asus project info from asus notebooks section
- add Luke Jones as maintainer for asus notebooks
- Remove Perry Yuan as DELL WMI HARDWARE PRIVACY SUPPORT maintainer
intel-uncore-freq:
- Fix types in sysfs callbacks
intel-wmi-sbl-fw-update:
- Fix function name in error message
p2sb:
- Use pci_resource_n() in p2sb_read_bar0()
- Allow p2sb_bar() calls during PCI device probe
platform/mellanox:
- mlxbf-pmc: Fix offset calculation for crspace events
- mlxbf-tmfifo: Drop Tx network packet when Tx TmFIFO is full
platform/x86/amd/pmf:
- Fix memory leak in amd_pmf_get_pb_data()
- Get ambient light information from AMD SFH driver
- Get Human presence information from AMD SFH driver
platform/x86/intel/ifs:
- Call release_firmware() when handling errors.
silicom-platform:
- Add missing "Description:" for power_cycle sysfs attr
touchscreen_dmi:
- Add info for the TECLAST X16 Plus tablet
wmi:
- Fix wmi_dev_probe()
- Fix notify callback locking
- Decouple legacy WMI notify handlers from wmi_block_list
- Return immediately if an suitable WMI event is found
- Fix error handling in legacy WMI notify handler functions
-----BEGIN PGP SIGNATURE-----
iQFIBAABCAAyFiEEuvA7XScYQRpenhd+kuxHeUQDJ9wFAmW1JScUHGhkZWdvZWRl
QHJlZGhhdC5jb20ACgkQkuxHeUQDJ9xNVwf/YXSuNEw+ztLH0pEySBUATHrcIbO7
gOpW2ZISf6IzRe7HFw7Ea1IJxrvysPn8VEribT3Sot9Ka+Pzd6H/TVA64sfyE7oG
wEke2Uxpnfie65Yo2IYNADhfLTOyAL7mvchScQz5hTE+gBq5Fdac2ykK+ox1dpTs
BqPg1/yG06L1SRX2Id0UNNYGMBsmjUH6v2b+M8Rcba+qcdznGMRe7l8T1Q2fY+nl
P6+tz3rYdfrGn1j+35Wo2bgKaB8l6rrtOscIvpke+CxZ95+6UxqZfLOBCg8u/njA
QbWqfZGjmbRGrbNo4C3fAHjj6SzQNyNfsm4gd4eJzl8X1CR9gzM8kb/xmg==
=yoe7
-----END PGP SIGNATURE-----
gpgsig -----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQSCSUwRdwTNL2MhaBlZrE9hU+XOMQUCZcTBOwAKCRBZrE9hU+XO
MWqxAQCI45uMAAJyya3FEVUaUNO1fgC1q8gIG9KoRMACys+HnAEAzR9JNjQOJ8g2
ccMpCtfVJxxlBJw3U8L6p+2FWkrt5gU=
=5exp
-----END PGP SIGNATURE-----
Merge tag 'platform-drivers-x86-v6.8-2' into pdx/for-next
Merge tag 'platform-drivers-x86-v6.8-2' fixes into pdf86/for-next
because of WMI fixes. The WMI changes done in for-next already created
a minor conflict with the fixes and WMI is actively being improved
currently so besides resolving the current conflict, this is also to
avoid further conflicts.
The activation for Scan at Field (SAF) includes a parameter to make
microcode wait for both threads to join. It's preferable to perform an
entry rendezvous before the activation to ensure that they start the
`wrmsr` close enough to each other. In some cases it has been observed
that one of the threads might be just a bit late to arrive. An entry
rendezvous reduces the likelihood of these cases occurring.
Add an entry rendezvous to ensure the activation on both threads happen
close enough to each other.
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20240125082254.424859-6-ashok.raj@intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
ARRAY_BIST requires the test to be invoked only from one of the HT
siblings of a core. If the other sibling was in mwait(), that didn't
permit the test to complete and resulted in several retries before the
test could finish.
The exit rendezvous was introduced to keep the HT sibling busy until
the primary CPU completed the test to avoid those retries. What is
actually needed is to ensure that both the threads rendezvous *before*
the wrmsr to trigger the test to give good chance to complete the test.
The `stop_machine()` function returns only after all the CPUs complete
running the function, and provides an exit rendezvous implicitly.
In kernel/stop_machine.c::multi_cpu_stop(), every CPU in the mask
needs to complete reaching MULTI_STOP_RUN. When all CPUs complete, the
state machine moves to next state, i.e MULTI_STOP_EXIT. Thus the
underlying API stop_core_cpuslocked() already provides an exit
rendezvous.
Add the rendezvous earlier in order to ensure the wrmsr is triggered
after all CPUs reach the do_array_test(). Remove the exit rendezvous
since stop_core_cpuslocked() already guarantees that.
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20240125082254.424859-5-ashok.raj@intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Add the current batch number in the trace output. When there are
failures, it's important to know which test content resulted in failure.
# TASK-PID CPU# ||||| TIMESTAMP FUNCTION
# | | | ||||| | |
migration/0-18 [000] d..1. 527287.084668: ifs_status: batch: 02, start: 0000, stop: 007f, status: 0000000000007f80
migration/128-785 [128] d..1. 527287.084669: ifs_status: batch: 02, start: 0000, stop: 007f, status: 0000000000007f80
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20240125082254.424859-4-ashok.raj@intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Enable the trace function on all HT threads. Currently, the trace is
called from some arbitrary CPU where the test was invoked.
This change gives visibility to the exact errors as seen by each
participating HT threads, and not just what was seen from the primary
thread.
Sample output below.
# TASK-PID CPU# ||||| TIMESTAMP FUNCTION
# | | | ||||| | |
migration/0-18 [000] d..1. 527287.084668: start: 0000, stop: 007f, status: 0000000000007f80
migration/128-785 [128] d..1. 527287.084669: start: 0000, stop: 007f, status: 0000000000007f80
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20240125082254.424859-3-ashok.raj@intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Missing release_firmware() due to error handling blocked any future image
loading.
Fix the return code and release_fiwmare() to release the bad image.
Fixes: 25a76dbb36 ("platform/x86/intel/ifs: Validate image size")
Reported-by: Pengfei Xu <pengfei.xu@intel.com>
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Link: https://lore.kernel.org/r/20240125082254.424859-2-ashok.raj@intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Gleixner:
- Restructure the code needed for it and add a temporary initrd mapping
on 32-bit so that the loader can access the microcode blobs. This in
itself is a preparation for the next major improvement:
- Do not load microcode on 32-bit before paging has been enabled.
Handling this has caused an endless stream of headaches, issues, ugly
code and unnecessary hacks in the past. And there really wasn't any
sensible reason to do that in the first place. So switch the 32-bit
loading to happen after paging has been enabled and turn the loader
code "real purrty" again
- Drop mixed microcode steppings loading on Intel - there, a single patch
loaded on the whole system is sufficient
- Rework late loading to track which CPUs have updated microcode
successfully and which haven't, act accordingly
- Move late microcode loading on Intel in NMI context in order to
guarantee concurrent loading on all threads
- Make the late loading CPU-hotplug-safe and have the offlined threads
be woken up for the purpose of the update
- Add support for a minimum revision which determines whether late
microcode loading is safe on a machine and the microcode does not
change software visible features which the machine cannot use anyway
since feature detection has happened already. Roughly, the minimum
revision is the smallest revision number which must be loaded
currently on the system so that late updates can be allowed
- Other nice leanups, fixess, etc all over the place
-----BEGIN PGP SIGNATURE-----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=JOZu
-----END PGP SIGNATURE-----
Merge tag 'x86_microcode_for_v6.7_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip
Pull x86 microcode loading updates from Borislac Petkov:
"Major microcode loader restructuring, cleanup and improvements by
Thomas Gleixner:
- Restructure the code needed for it and add a temporary initrd
mapping on 32-bit so that the loader can access the microcode
blobs. This in itself is a preparation for the next major
improvement:
- Do not load microcode on 32-bit before paging has been enabled.
Handling this has caused an endless stream of headaches, issues,
ugly code and unnecessary hacks in the past. And there really
wasn't any sensible reason to do that in the first place. So switch
the 32-bit loading to happen after paging has been enabled and turn
the loader code "real purrty" again
- Drop mixed microcode steppings loading on Intel - there, a single
patch loaded on the whole system is sufficient
- Rework late loading to track which CPUs have updated microcode
successfully and which haven't, act accordingly
- Move late microcode loading on Intel in NMI context in order to
guarantee concurrent loading on all threads
- Make the late loading CPU-hotplug-safe and have the offlined
threads be woken up for the purpose of the update
- Add support for a minimum revision which determines whether late
microcode loading is safe on a machine and the microcode does not
change software visible features which the machine cannot use
anyway since feature detection has happened already. Roughly, the
minimum revision is the smallest revision number which must be
loaded currently on the system so that late updates can be allowed
- Other nice leanups, fixess, etc all over the place"
* tag 'x86_microcode_for_v6.7_rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (40 commits)
x86/microcode/intel: Add a minimum required revision for late loading
x86/microcode: Prepare for minimal revision check
x86/microcode: Handle "offline" CPUs correctly
x86/apic: Provide apic_force_nmi_on_cpu()
x86/microcode: Protect against instrumentation
x86/microcode: Rendezvous and load in NMI
x86/microcode: Replace the all-in-one rendevous handler
x86/microcode: Provide new control functions
x86/microcode: Add per CPU control field
x86/microcode: Add per CPU result state
x86/microcode: Sanitize __wait_for_cpus()
x86/microcode: Clarify the late load logic
x86/microcode: Handle "nosmt" correctly
x86/microcode: Clean up mc_cpu_down_prep()
x86/microcode: Get rid of the schedule work indirection
x86/microcode: Mop up early loading leftovers
x86/microcode/amd: Use cached microcode for AP load
x86/microcode/amd: Cache builtin/initrd microcode early
x86/microcode/amd: Cache builtin microcode too
x86/microcode/amd: Use correct per CPU ucode_cpu_info
...
Highlights:
- asus-wmi: Support for screenpad and solve brightness key
press duplication
- int3472: Eliminate the last use of deprecated GPIO functions
- mlxbf-pmc: New HW support
- msi-ec: Support new EC configurations
- thinkpad_acpi: Support reading aux MAC address during passthrough
- wmi: Fixes & improvements
- x86-android-tablets: Detection fix and avoid use of GPIO private APIs
- Debug & metrics interface improvements
- Miscellaneous cleanups / fixes / improvements
The following is an automated shortlog grouped by driver:
acer-wmi:
- Remove void function return
amd/hsmp:
- add support for metrics tbl
- create plat specific struct
- Fix iomem handling
- improve the error log
amd/pmc:
- Add dump_custom_stb module parameter
- Add PMFW command id to support S2D force flush
- Handle overflow cases where the num_samples range is higher
- Use flex array when calling amd_pmc_stb_debugfs_open_v2()
asus-wireless:
- Replace open coded acpi_match_acpi_device()
asus-wmi:
- add support for ASUS screenpad
- Do not report brightness up/down keys when also reported by acpi_video
gpiolib: acpi:
- Add a ignore interrupt quirk for Peaq C1010
- Check if a GPIO is listed in ignore_interrupt earlier
hp-bioscfg:
- Annotate struct bios_args with __counted_by
inspur-platform-profile:
- Add platform profile support
int3472:
- Add new skl_int3472_fill_gpiod_lookup() helper
- Add new skl_int3472_gpiod_get_from_temp_lookup() helper
- Stop using gpiod_toggle_active_low()
- Switch to devm_get_gpiod()
intel: bytcrc_pwrsrc:
- Convert to platform remove callback returning void
intel/ifs:
- Add new CPU support
- Add new error code
- ARRAY BIST for Sierra Forest
- Gen2 scan image loading
- Gen2 Scan test support
- Metadata validation for start_chunk
- Refactor image loading code
- Store IFS generation number
- Validate image size
intel_speed_select_if:
- Remove hardcoded map size
- Use devm_ioremap_resource
intel/tpmi:
- Add debugfs support for read/write blocked
- Add defines to get version information
intel-uncore-freq:
- Ignore minor version change
ISST:
- Allow level 0 to be not present
- Ignore minor version change
- Use fuse enabled mask instead of allowed levels
mellanox:
- Fix misspelling error in routine name
- Rename some init()/exit() functions for consistent naming
mlxbf-bootctl:
- Convert to platform remove callback returning void
mlxbf-pmc:
- Add support for BlueField-3
mlxbf-tmfifo:
- Convert to platform remove callback returning void
mlx-Convert to platform remove callback returning void:
- mlx-Convert to platform remove callback returning void
mlxreg-hotplug:
- Convert to platform remove callback returning void
mlxreg-io:
- Convert to platform remove callback returning void
mlxreg-lc:
- Convert to platform remove callback returning void
msi-ec:
- Add more EC configs
- rename fn_super_swap
nvsw-sn2201:
- Convert to platform remove callback returning void
sel3350-Convert to platform remove callback returning void:
- sel3350-Convert to platform remove callback returning void
siemens: simatic-ipc-batt-apollolake:
- Convert to platform remove callback returning void
siemens: simatic-ipc-batt:
- Convert to platform remove callback returning void
siemens: simatic-ipc-batt-elkhartlake:
- Convert to platform remove callback returning void
siemens: simatic-ipc-batt-f7188x:
- Convert to platform remove callback returning void
siemens: simatic-ipc-batt:
- Simplify simatic_ipc_batt_remove()
surface: acpi-notify:
- Convert to platform remove callback returning void
surface: aggregator:
- Annotate struct ssam_event with __counted_by
surface: aggregator-cdev:
- Convert to platform remove callback returning void
surface: aggregator-registry:
- Convert to platform remove callback returning void
surface: dtx:
- Convert to platform remove callback returning void
surface: gpe:
- Convert to platform remove callback returning void
surface: hotplug:
- Convert to platform remove callback returning void
surface: surface3-wmi:
- Convert to platform remove callback returning void
think-lmi:
- Add bulk save feature
- Replace kstrdup() + strreplace() with kstrdup_and_replace()
- Use strreplace() to replace a character by nul
thinkpad_acpi:
- Add battery quirk for Thinkpad X120e
- replace deprecated strncpy with memcpy
- sysfs interface to auxmac
tools/power/x86/intel-speed-select:
- Display error for core-power support
- Increase max CPUs in one request
- No TRL for non compute domains
- Sanitize integer arguments
- turbo-mode enable disable swapped
- Update help for TRL
- Use cgroup isolate for CPU 0
- v1.18 release
wmi:
- Decouple probe deferring from wmi_block_list
- Decouple WMI device removal from wmi_block_list
- Fix opening of char device
- Fix probe failure when failing to register WMI devices
- Fix refcounting of WMI devices in legacy functions
x86-android-tablets:
- Add a comment about x86_android_tablet_get_gpiod()
- Create a platform_device from module_init()
- Drop "linux,power-supply-name" from lenovo_yt3_bq25892_0_props[]
- Fix Lenovo Yoga Tablet 2 830F/L vs 1050F/L detection
- Remove invalid_aei_gpiochip from Peaq C1010
- Remove invalid_aei_gpiochip support
- Stop using gpiolib private APIs
- Use platform-device as gpio-keys parent
xo15-ebook:
- Replace open coded acpi_match_acpi_device()
Merges:
- Merge branch 'pdx86/platform-drivers-x86-int3472' into review-ilpo
- Merge branch 'pdx86/platform-drivers-x86-mellanox-init' into review-ilpo
- Merge remote-tracking branch 'intel-speed-select/intel-sst' into review-ilpo
- Merge remote-tracking branch 'pdx86/platform-drivers-x86-android-tablets' into review-hans
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQSCSUwRdwTNL2MhaBlZrE9hU+XOMQUCZT+lBwAKCRBZrE9hU+XO
Mck0AQCFU7dYLCF4d1CXtHf1eZhSXLpYdhcO+C08JGGoM+MqSgD+Jyb9KJHk4pxE
FvKG51I9neyAne9lvNrLodHRzxCYgAo=
=duM8
-----END PGP SIGNATURE-----
Merge tag 'platform-drivers-x86-v6.7-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86
Pull x86 platform driver updates from Ilpo Järvinen:
- asus-wmi: Support for screenpad and solve brightness key press
duplication
- int3472: Eliminate the last use of deprecated GPIO functions
- mlxbf-pmc: New HW support
- msi-ec: Support new EC configurations
- thinkpad_acpi: Support reading aux MAC address during passthrough
- wmi: Fixes & improvements
- x86-android-tablets: Detection fix and avoid use of GPIO private APIs
- Debug & metrics interface improvements
- Miscellaneous cleanups / fixes / improvements
* tag 'platform-drivers-x86-v6.7-1' of git://git.kernel.org/pub/scm/linux/kernel/git/pdx86/platform-drivers-x86: (80 commits)
platform/x86: inspur-platform-profile: Add platform profile support
platform/x86: thinkpad_acpi: Add battery quirk for Thinkpad X120e
platform/x86: wmi: Decouple WMI device removal from wmi_block_list
platform/x86: wmi: Fix opening of char device
platform/x86: wmi: Fix probe failure when failing to register WMI devices
platform/x86: wmi: Fix refcounting of WMI devices in legacy functions
platform/x86: wmi: Decouple probe deferring from wmi_block_list
platform/x86/amd/hsmp: Fix iomem handling
platform/x86: asus-wmi: Do not report brightness up/down keys when also reported by acpi_video
platform/x86: thinkpad_acpi: replace deprecated strncpy with memcpy
tools/power/x86/intel-speed-select: v1.18 release
tools/power/x86/intel-speed-select: Use cgroup isolate for CPU 0
tools/power/x86/intel-speed-select: Increase max CPUs in one request
tools/power/x86/intel-speed-select: Display error for core-power support
tools/power/x86/intel-speed-select: No TRL for non compute domains
tools/power/x86/intel-speed-select: turbo-mode enable disable swapped
tools/power/x86/intel-speed-select: Update help for TRL
tools/power/x86/intel-speed-select: Sanitize integer arguments
platform/x86: acer-wmi: Remove void function return
platform/x86/amd/pmc: Add dump_custom_stb module parameter
...
Take a cpu_signature argument and work from there. Move the match()
helper next to the callsite as there is no point for having it in
a header.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20231002115902.797820205@linutronix.de
Nothing needs struct ucode_cpu_info. Make it take struct cpu_signature,
let it return a boolean and simplify the implementation. Rename it now
that the silly name clash with collect_cpu_info() is gone.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20231017211722.851573238@linutronix.de
Array BIST MSR addresses, bit definition and semantics are different for
Sierra Forest. Branch into a separate Array BIST flow on Sierra Forest
when user invokes Array Test.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Link: https://lore.kernel.org/r/20231005195137.3117166-10-jithu.joseph@intel.com
[ij: ARRAY_GEN_* -> ARRAY_GEN* for consistency]
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Make driver aware of a newly added error code so that it can provide a
more appropriate error message.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Link: https://lore.kernel.org/r/20231005195137.3117166-9-jithu.joseph@intel.com
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Add Granite Rapids(GNR) and Sierra Forest(SRF) cpuids to x86 match table
so that IFS driver can be loaded for those.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Link: https://lore.kernel.org/r/20231005195137.3117166-8-jithu.joseph@intel.com
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Add an additional check to validate IFS image metadata field prior to
loading the test image.
If start_chunk is not a multiple of chunks_per_stride error out.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Link: https://lore.kernel.org/r/20231005195137.3117166-7-jithu.joseph@intel.com
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Perform additional validation prior to loading IFS image.
Error out if the size of the file being loaded doesn't match the size
specified in the header.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Link: https://lore.kernel.org/r/20231005195137.3117166-6-jithu.joseph@intel.com
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Width of chunk related bitfields is ACTIVATE_SCAN and SCAN_STATUS MSRs
are different in newer IFS generation compared to gen0.
Make changes to scan test flow such that MSRs are populated
appropriately based on the generation supported by hardware.
Account for the 8/16 bit MSR bitfield width differences between gen0 and
newer generations for the scan test trace event too.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Link: https://lore.kernel.org/r/20231005195137.3117166-5-jithu.joseph@intel.com
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Scan image loading flow for newer IFS generations are slightly different
from that of current generation. In newer schemes, loading need not be
done once for each socket as was done in gen0.
Also the width of NUM_CHUNKS bitfield in SCAN_HASHES_STATUS MSR has
increased from 8 -> 16 bits. Similarly there are width differences for
CHUNK_AUTHENTICATION_STATUS too.
Further the parameter to AUTHENTICATE_AND_COPY_CHUNK is passed
differently in newer generations.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Link: https://lore.kernel.org/r/20231005195137.3117166-4-jithu.joseph@intel.com
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
IFS image loading flow is slightly different for newer IFS generations.
In preparation for adding support for newer IFS generations, refactor
portions of existing image loading code for reuse.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Link: https://lore.kernel.org/r/20231005195137.3117166-3-jithu.joseph@intel.com
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
IFS generation number is reported via MSR_INTEGRITY_CAPS. As IFS
support gets added to newer CPUs, some differences are expected during
IFS image loading and test flows.
Define MSR bitmasks to extract and store the generation in driver data,
so that driver can modify its MSR interaction appropriately.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Tested-by: Pengfei Xu <pengfei.xu@intel.com>
Link: https://lore.kernel.org/r/20231005195137.3117166-2-jithu.joseph@intel.com
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com>
Couple of error paths in do_core_test() was returning directly without
doing a necessary cpus_read_unlock().
Following lockdep warning was observed when exercising these scenarios
with PROVE_RAW_LOCK_NESTING enabled:
[ 139.304775] ================================================
[ 139.311185] WARNING: lock held when returning to user space!
[ 139.317593] 6.6.0-rc2ifs01+ #11 Tainted: G S W I
[ 139.324499] ------------------------------------------------
[ 139.330908] bash/11476 is leaving the kernel with locks still held!
[ 139.338000] 1 lock held by bash/11476:
[ 139.342262] #0: ffffffffaa26c930 (cpu_hotplug_lock){++++}-{0:0}, at:
do_core_test+0x35/0x1c0 [intel_ifs]
Fix the flow so that all scenarios release the lock prior to returning
from the function.
Fixes: 5210fb4e18 ("platform/x86/intel/ifs: Sysfs interface for Array BIST")
Cc: stable@vger.kernel.org
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Link: https://lore.kernel.org/r/20230927184824.2566086-1-jithu.joseph@intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Rename get_datasize() to intel_microcode_get_datasize() and make it an inline.
[ tglx: Make the argument typed and fix up the IFS code ]
Suggested-by: Boris Petkov <bp@alien8.de>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230812195727.894165745@linutronix.de
Currently vendor specific headers are included explicitly when used in
common code. Instead, include the vendor specific headers in
microcode.h, and include that in all usages.
No functional change.
Suggested-by: Boris Petkov <bp@alien8.de>
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Link: https://lore.kernel.org/r/20230812195727.776541545@linutronix.de
The summary of the changes for this pull requests is:
* Song Liu's new struct module_memory replacement
* Nick Alcock's MODULE_LICENSE() removal for non-modules
* My cleanups and enhancements to reduce the areas where we vmalloc
module memory for duplicates, and the respective debug code which
proves the remaining vmalloc pressure comes from userspace.
Most of the changes have been in linux-next for quite some time except
the minor fixes I made to check if a module was already loaded
prior to allocating the final module memory with vmalloc and the
respective debug code it introduces to help clarify the issue. Although
the functional change is small it is rather safe as it can only *help*
reduce vmalloc space for duplicates and is confirmed to fix a bootup
issue with over 400 CPUs with KASAN enabled. I don't expect stable
kernels to pick up that fix as the cleanups would have also had to have
been picked up. Folks on larger CPU systems with modules will want to
just upgrade if vmalloc space has been an issue on bootup.
Given the size of this request, here's some more elaborate details
on this pull request.
The functional change change in this pull request is the very first
patch from Song Liu which replaces the struct module_layout with a new
struct module memory. The old data structure tried to put together all
types of supported module memory types in one data structure, the new
one abstracts the differences in memory types in a module to allow each
one to provide their own set of details. This paves the way in the
future so we can deal with them in a cleaner way. If you look at changes
they also provide a nice cleanup of how we handle these different memory
areas in a module. This change has been in linux-next since before the
merge window opened for v6.3 so to provide more than a full kernel cycle
of testing. It's a good thing as quite a bit of fixes have been found
for it.
Jason Baron then made dynamic debug a first class citizen module user by
using module notifier callbacks to allocate / remove module specific
dynamic debug information.
Nick Alcock has done quite a bit of work cross-tree to remove module
license tags from things which cannot possibly be module at my request
so to:
a) help him with his longer term tooling goals which require a
deterministic evaluation if a piece a symbol code could ever be
part of a module or not. But quite recently it is has been made
clear that tooling is not the only one that would benefit.
Disambiguating symbols also helps efforts such as live patching,
kprobes and BPF, but for other reasons and R&D on this area
is active with no clear solution in sight.
b) help us inch closer to the now generally accepted long term goal
of automating all the MODULE_LICENSE() tags from SPDX license tags
In so far as a) is concerned, although module license tags are a no-op
for non-modules, tools which would want create a mapping of possible
modules can only rely on the module license tag after the commit
8b41fc4454 ("kbuild: create modules.builtin without Makefile.modbuiltin
or tristate.conf"). Nick has been working on this *for years* and
AFAICT I was the only one to suggest two alternatives to this approach
for tooling. The complexity in one of my suggested approaches lies in
that we'd need a possible-obj-m and a could-be-module which would check
if the object being built is part of any kconfig build which could ever
lead to it being part of a module, and if so define a new define
-DPOSSIBLE_MODULE [0]. A more obvious yet theoretical approach I've
suggested would be to have a tristate in kconfig imply the same new
-DPOSSIBLE_MODULE as well but that means getting kconfig symbol names
mapping to modules always, and I don't think that's the case today. I am
not aware of Nick or anyone exploring either of these options. Quite
recently Josh Poimboeuf has pointed out that live patching, kprobes and
BPF would benefit from resolving some part of the disambiguation as
well but for other reasons. The function granularity KASLR (fgkaslr)
patches were mentioned but Joe Lawrence has clarified this effort has
been dropped with no clear solution in sight [1].
In the meantime removing module license tags from code which could never
be modules is welcomed for both objectives mentioned above. Some
developers have also welcomed these changes as it has helped clarify
when a module was never possible and they forgot to clean this up,
and so you'll see quite a bit of Nick's patches in other pull
requests for this merge window. I just picked up the stragglers after
rc3. LWN has good coverage on the motivation behind this work [2] and
the typical cross-tree issues he ran into along the way. The only
concrete blocker issue he ran into was that we should not remove the
MODULE_LICENSE() tags from files which have no SPDX tags yet, even if
they can never be modules. Nick ended up giving up on his efforts due
to having to do this vetting and backlash he ran into from folks who
really did *not understand* the core of the issue nor were providing
any alternative / guidance. I've gone through his changes and dropped
the patches which dropped the module license tags where an SPDX
license tag was missing, it only consisted of 11 drivers. To see
if a pull request deals with a file which lacks SPDX tags you
can just use:
./scripts/spdxcheck.py -f \
$(git diff --name-only commid-id | xargs echo)
You'll see a core module file in this pull request for the above,
but that's not related to his changes. WE just need to add the SPDX
license tag for the kernel/module/kmod.c file in the future but
it demonstrates the effectiveness of the script.
Most of Nick's changes were spread out through different trees,
and I just picked up the slack after rc3 for the last kernel was out.
Those changes have been in linux-next for over two weeks.
The cleanups, debug code I added and final fix I added for modules
were motivated by David Hildenbrand's report of boot failing on
a systems with over 400 CPUs when KASAN was enabled due to running
out of virtual memory space. Although the functional change only
consists of 3 lines in the patch "module: avoid allocation if module is
already present and ready", proving that this was the best we can
do on the modules side took quite a bit of effort and new debug code.
The initial cleanups I did on the modules side of things has been
in linux-next since around rc3 of the last kernel, the actual final
fix for and debug code however have only been in linux-next for about a
week or so but I think it is worth getting that code in for this merge
window as it does help fix / prove / evaluate the issues reported
with larger number of CPUs. Userspace is not yet fixed as it is taking
a bit of time for folks to understand the crux of the issue and find a
proper resolution. Worst come to worst, I have a kludge-of-concept [3]
of how to make kernel_read*() calls for modules unique / converge them,
but I'm currently inclined to just see if userspace can fix this
instead.
[0] https://lore.kernel.org/all/Y/kXDqW+7d71C4wz@bombadil.infradead.org/
[1] https://lkml.kernel.org/r/025f2151-ce7c-5630-9b90-98742c97ac65@redhat.com
[2] https://lwn.net/Articles/927569/
[3] https://lkml.kernel.org/r/20230414052840.1994456-3-mcgrof@kernel.org
-----BEGIN PGP SIGNATURE-----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=56WK
-----END PGP SIGNATURE-----
Merge tag 'modules-6.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/mcgrof/linux
Pull module updates from Luis Chamberlain:
"The summary of the changes for this pull requests is:
- Song Liu's new struct module_memory replacement
- Nick Alcock's MODULE_LICENSE() removal for non-modules
- My cleanups and enhancements to reduce the areas where we vmalloc
module memory for duplicates, and the respective debug code which
proves the remaining vmalloc pressure comes from userspace.
Most of the changes have been in linux-next for quite some time except
the minor fixes I made to check if a module was already loaded prior
to allocating the final module memory with vmalloc and the respective
debug code it introduces to help clarify the issue. Although the
functional change is small it is rather safe as it can only *help*
reduce vmalloc space for duplicates and is confirmed to fix a bootup
issue with over 400 CPUs with KASAN enabled. I don't expect stable
kernels to pick up that fix as the cleanups would have also had to
have been picked up. Folks on larger CPU systems with modules will
want to just upgrade if vmalloc space has been an issue on bootup.
Given the size of this request, here's some more elaborate details:
The functional change change in this pull request is the very first
patch from Song Liu which replaces the 'struct module_layout' with a
new 'struct module_memory'. The old data structure tried to put
together all types of supported module memory types in one data
structure, the new one abstracts the differences in memory types in a
module to allow each one to provide their own set of details. This
paves the way in the future so we can deal with them in a cleaner way.
If you look at changes they also provide a nice cleanup of how we
handle these different memory areas in a module. This change has been
in linux-next since before the merge window opened for v6.3 so to
provide more than a full kernel cycle of testing. It's a good thing as
quite a bit of fixes have been found for it.
Jason Baron then made dynamic debug a first class citizen module user
by using module notifier callbacks to allocate / remove module
specific dynamic debug information.
Nick Alcock has done quite a bit of work cross-tree to remove module
license tags from things which cannot possibly be module at my request
so to:
a) help him with his longer term tooling goals which require a
deterministic evaluation if a piece a symbol code could ever be
part of a module or not. But quite recently it is has been made
clear that tooling is not the only one that would benefit.
Disambiguating symbols also helps efforts such as live patching,
kprobes and BPF, but for other reasons and R&D on this area is
active with no clear solution in sight.
b) help us inch closer to the now generally accepted long term goal
of automating all the MODULE_LICENSE() tags from SPDX license tags
In so far as a) is concerned, although module license tags are a no-op
for non-modules, tools which would want create a mapping of possible
modules can only rely on the module license tag after the commit
8b41fc4454 ("kbuild: create modules.builtin without
Makefile.modbuiltin or tristate.conf").
Nick has been working on this *for years* and AFAICT I was the only
one to suggest two alternatives to this approach for tooling. The
complexity in one of my suggested approaches lies in that we'd need a
possible-obj-m and a could-be-module which would check if the object
being built is part of any kconfig build which could ever lead to it
being part of a module, and if so define a new define
-DPOSSIBLE_MODULE [0].
A more obvious yet theoretical approach I've suggested would be to
have a tristate in kconfig imply the same new -DPOSSIBLE_MODULE as
well but that means getting kconfig symbol names mapping to modules
always, and I don't think that's the case today. I am not aware of
Nick or anyone exploring either of these options. Quite recently Josh
Poimboeuf has pointed out that live patching, kprobes and BPF would
benefit from resolving some part of the disambiguation as well but for
other reasons. The function granularity KASLR (fgkaslr) patches were
mentioned but Joe Lawrence has clarified this effort has been dropped
with no clear solution in sight [1].
In the meantime removing module license tags from code which could
never be modules is welcomed for both objectives mentioned above. Some
developers have also welcomed these changes as it has helped clarify
when a module was never possible and they forgot to clean this up, and
so you'll see quite a bit of Nick's patches in other pull requests for
this merge window. I just picked up the stragglers after rc3. LWN has
good coverage on the motivation behind this work [2] and the typical
cross-tree issues he ran into along the way. The only concrete blocker
issue he ran into was that we should not remove the MODULE_LICENSE()
tags from files which have no SPDX tags yet, even if they can never be
modules. Nick ended up giving up on his efforts due to having to do
this vetting and backlash he ran into from folks who really did *not
understand* the core of the issue nor were providing any alternative /
guidance. I've gone through his changes and dropped the patches which
dropped the module license tags where an SPDX license tag was missing,
it only consisted of 11 drivers. To see if a pull request deals with a
file which lacks SPDX tags you can just use:
./scripts/spdxcheck.py -f \
$(git diff --name-only commid-id | xargs echo)
You'll see a core module file in this pull request for the above, but
that's not related to his changes. WE just need to add the SPDX
license tag for the kernel/module/kmod.c file in the future but it
demonstrates the effectiveness of the script.
Most of Nick's changes were spread out through different trees, and I
just picked up the slack after rc3 for the last kernel was out. Those
changes have been in linux-next for over two weeks.
The cleanups, debug code I added and final fix I added for modules
were motivated by David Hildenbrand's report of boot failing on a
systems with over 400 CPUs when KASAN was enabled due to running out
of virtual memory space. Although the functional change only consists
of 3 lines in the patch "module: avoid allocation if module is already
present and ready", proving that this was the best we can do on the
modules side took quite a bit of effort and new debug code.
The initial cleanups I did on the modules side of things has been in
linux-next since around rc3 of the last kernel, the actual final fix
for and debug code however have only been in linux-next for about a
week or so but I think it is worth getting that code in for this merge
window as it does help fix / prove / evaluate the issues reported with
larger number of CPUs. Userspace is not yet fixed as it is taking a
bit of time for folks to understand the crux of the issue and find a
proper resolution. Worst come to worst, I have a kludge-of-concept [3]
of how to make kernel_read*() calls for modules unique / converge
them, but I'm currently inclined to just see if userspace can fix this
instead"
Link: https://lore.kernel.org/all/Y/kXDqW+7d71C4wz@bombadil.infradead.org/ [0]
Link: https://lkml.kernel.org/r/025f2151-ce7c-5630-9b90-98742c97ac65@redhat.com [1]
Link: https://lwn.net/Articles/927569/ [2]
Link: https://lkml.kernel.org/r/20230414052840.1994456-3-mcgrof@kernel.org [3]
* tag 'modules-6.4-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/mcgrof/linux: (121 commits)
module: add debugging auto-load duplicate module support
module: stats: fix invalid_mod_bytes typo
module: remove use of uninitialized variable len
module: fix building stats for 32-bit targets
module: stats: include uapi/linux/module.h
module: avoid allocation if module is already present and ready
module: add debug stats to help identify memory pressure
module: extract patient module check into helper
modules/kmod: replace implementation with a semaphore
Change DEFINE_SEMAPHORE() to take a number argument
module: fix kmemleak annotations for non init ELF sections
module: Ignore L0 and rename is_arm_mapping_symbol()
module: Move is_arm_mapping_symbol() to module_symbol.h
module: Sync code of is_arm_mapping_symbol()
scripts/gdb: use mem instead of core_layout to get the module address
interconnect: remove module-related code
interconnect: remove MODULE_LICENSE in non-modules
zswap: remove MODULE_LICENSE in non-modules
zpool: remove MODULE_LICENSE in non-modules
x86/mm/dump_pagetables: remove MODULE_LICENSE in non-modules
...
Fundamentally semaphores are a counted primitive, but
DEFINE_SEMAPHORE() does not expose this and explicitly creates a
binary semaphore.
Change DEFINE_SEMAPHORE() to take a number argument and use that in the
few places that open-coded it using __SEMAPHORE_INITIALIZER().
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
[mcgrof: add some tribal knowledge about why some folks prefer
binary sempahores over mutexes]
Reviewed-by: Sergey Senozhatsky <senozhatsky@chromium.org>
Reviewed-by: Davidlohr Bueso <dave@stgolabs.net>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
Array BIST is the second test supported by IFS. Modify IFS doc
entry to be more general.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20230322003359.213046-9-jithu.joseph@intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Array BIST test (for a particular core) is triggered by writing
to MSR_ARRAY_BIST from one sibling of the core.
This will initiate a test for all supported arrays on that
CPU. Array BIST test may be aborted before completing all the
arrays in the event of an interrupt or other reasons.
In this case, kernel will restart the test from that point
onwards. Array test will also be aborted when the test fails,
in which case the test is stopped immediately without further
retry.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20230322003359.213046-8-jithu.joseph@intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
The interface to trigger Array BIST test and obtain its result
is similar to the existing scan test. The only notable
difference is that, Array BIST doesn't require any test content
to be loaded. So binary load related options are not needed for
this test.
Add sysfs interface for array BIST test, the testing support will
be added by subsequent patch.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20230322003359.213046-7-jithu.joseph@intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Array BIST is a new type of core test introduced under the Intel Infield
Scan (IFS) suite of tests.
Emerald Rapids (EMR) is the first CPU to support Array BIST.
Array BIST performs tests on some portions of the core logic such as
caches and register files. These are different portions of the silicon
compared to the parts tested by the first test type
i.e Scan at Field (SAF).
Make changes in the device driver init flow to register this new test
type with the device driver framework. Each test will have its own
sysfs directory (intel_ifs_0 , intel_ifs_1) under misc hierarchy to
accommodate for the differences in test type and how they are initiated.
Upcoming patches will add actual support.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20230322003359.213046-6-jithu.joseph@intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
The struct holding device driver data contained both read only(ro)
and read write(rw) fields.
Separating ro fields from rw fields was recommended as
a preferable design pattern during review[1].
Group ro fields into a separate const struct. Associate it to
the miscdevice being registered by keeping its pointer in the
same container struct as the miscdevice.
Link: https://lore.kernel.org/lkml/Y+9H9otxLYPqMkUh@kroah.com/ [1]
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20230322003359.213046-3-jithu.joseph@intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
In preparation to supporting additional tests, remove ifs_pkg_auth
from per-test scope, as it is only applicable for one test type.
This will simplify ifs_init() flow when multiple tests are added.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Link: https://lore.kernel.org/r/20230322003359.213046-2-jithu.joseph@intel.com
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Document the test_num member of struct ifs_data.
Reported-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Acked-by: Randy Dunlap <rdunlap@infradead.org>
Link: https://lore.kernel.org/lkml/774fd22a-aaee-758d-8195-77bac783ecbc@infradead.org/
Initial implementation assumed a single IFS test image file with a
fixed name ff-mm-ss.scan. (where ff, mm, ss refers to family, model and
stepping of the core).
Subsequently, it became evident that supporting more than one test
image file is needed to provide more comprehensive test coverage. (Test
coverage in this scenario refers to testing more transistors in the core
to identify faults).
The other alternative of increasing the size of a single scan test image
file would not work as the upper bound is limited by the size of memory
area reserved by BIOS for loading IFS test image.
Introduce "current_batch" file which accepts a number. Writing a
number to the current_batch file would load the test image file by
name ff-mm-ss-<xy>.scan, where <xy> is the number written to the
"current_batch" file in hex. Range check of the input is done to verify
it not greater than 0xff.
For e.g if the scan test image comprises of 6 files, they would be named:
06-8f-06-01.scan
06-8f-06-02.scan
06-8f-06-03.scan
06-8f-06-04.scan
06-8f-06-05.scan
06-8f-06-06.scan
And writing 3 to current_batch would result in loading 06-8f-06-03.scan
above. The file can also be read to know the currently loaded file.
And testing a system looks like:
for each scan file
do
load the IFS test image file (write to the batch file)
for each core
do
test the core with this set of tests
done
done
Qualify few error messages with the test image file suffix to provide
better context.
[ bp: Massage commit message. Add link to the discussion. ]
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20221107225323.2733518-13-jithu.joseph@intel.com
Reload sysfs entry will be replaced by current_batch, drop it.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20221117035935.4136738-14-jithu.joseph@intel.com
The data portion of a IFS test image file contains a metadata region
containing possibly multiple metadata structures in addition to test
data and hashes.
IFS Metadata layout
+----------------------+ 0
|META_TYPE_IFS (=1) |
+----------------------+
|meta_size |
+----------------------+
|test type |
+----------------------+
|fusa info |
+----------------------+
|total images |
+----------------------+
|current image# |
+----------------------+
|total chunks |
+----------------------+
|starting chunk |
+----------------------+
|size per chunk |
+----------------------+
|chunks per stride |
+----------------------+
|Reserved[54] |
+----------------------+ 256
| |
| Test Data/Chunks |
| |
+----------------------+ meta_size
| META_TYPE_END (=0) |
+----------------------+ meta_size + 4
| size of end (=8) |
+----------------------+ meta_size + 8
Introduce the layout of this meta_data structure and validate
the sanity of certain fields of the new image before loading.
Tweak references to IFS test image chunks to reflect the updated
layout of the test image.
[ bp: Massage commit message. ]
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20221117230408.30331-1-jithu.joseph@intel.com
Existing implementation (broken) of IFS used a header format (for IFS
test images) which was very similar to microcode format, but didn’t
accommodate extended signatures. This meant same IFS test image had to
be duplicated for different steppings and the validation code in the
driver was only looking at the primary header parameters. Going forward,
IFS test image headers have been tweaked to become fully compatible with
the microcode format.
Newer IFS test image headers will use header version 2 in order to
distinguish it from microcode images and older IFS test images.
In light of the above, reuse struct microcode_header_intel directly in
the IFS driver and reuse microcode functions for validation and sanity
checking.
[ bp: Massage commit message. ]
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20221117225039.30166-1-jithu.joseph@intel.com
One of the existing reserved fields in the microcode header has been
allocated to indicate the size of metadata structures.
The location of metadata section within microcode header is as shown
below:
Microcode Blob Format
+----------------------+ Base
|Header Version |
+----------------------+
|Update revision |
+----------------------+
|Date DDMMYYYY |
+----------------------+
|Sig |
+----------------------+
|Checksum |
+----------------------+
|Loader Version |
+----------------------+
|Processor Flags |
+----------------------+
|Data Size |
+----------------------+
|Total Size |
+----------------------+
|Meta Size |
+----------------------+
|Reserved |
+----------------------+
|Reserved |
+----------------------+ Base+48
| |
| Microcode |
| Data |
| |
+----------------------+ Base+48+data_size-
| | meta_size
| Meta Data |
| structure(s) |
| |
+----------------------+ Base+48+data_size
| |
| Extended Signature |
| Table |
| |
+----------------------+ Base+total_size
Add an accessor function which will return a pointer to the start of a
specific meta_type being queried.
[ bp: Massage commit message. ]
Signed-off-by: Ashok Raj <ashok.raj@intel.com>
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20221117035935.4136738-11-jithu.joseph@intel.com
IFS requires tests to be authenticated once for each CPU socket on a
system.
scan_chunks_sanity_check() was dynamically allocating memory to store
the state of whether tests have been authenticated on each socket for
every load operation.
Move the memory allocation to init path and store the pointer in
ifs_data struct.
Also rearrange the adjacent error checking in init for a more simplified
and natural flow.
Suggested-by: Borislav Petkov <bp@alien8.de>
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20221117195957.28225-1-jithu.joseph@intel.com
IFS test image is unnecessarily loaded during driver initialization.
Drop image loading during ifs_init() and improve module load time. With
this change, user has to load one when starting the tests.
As a consequence, make ifs_sem static as it is only used within sysfs.c
Suggested-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20221117035935.4136738-4-jithu.joseph@intel.com
scan_chunks_sanity_check() returns -ENOMEM if it encounters an error
while copying IFS test image from memory to Secure Memory.
Return -EIO in this scenario, as it is more appropriate.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20221117035935.4136738-3-jithu.joseph@intel.com
CONFIG_INTEL_IFS_DEVICE is not used anywhere. The selection in
Kconfig is therefore pointless. Delete it.
Signed-off-by: Jithu Joseph <jithu.joseph@intel.com>
Signed-off-by: Borislav Petkov <bp@suse.de>
Reviewed-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Sohil Mehta <sohil.mehta@intel.com>
Reviewed-by: Hans de Goede <hdegoede@redhat.com>
Link: https://lore.kernel.org/r/20221117035935.4136738-2-jithu.joseph@intel.com
The driver is using functions from a compilation unit which is enabled
by CONFIG_CPU_SUP_INTEL. Add that dependency to Kconfig explicitly
otherwise:
drivers/platform/x86/intel/ifs/load.o: in function `ifs_load_firmware':
load.c:(.text+0x3b8): undefined reference to `intel_cpu_collect_info'
Reported-by: Randy Dunlap <rdunlap@infradead.org>
Signed-off-by: Borislav Petkov <bp@suse.de>
Link: https://lore.kernel.org/r/YoZay8YR0zRGyVu+@zn.tnic
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Add documentation for In-Field Scan (IFS). This documentation
describes the basics of IFS, the loading IFS image, chunk
authentication, running scan and how to check result via sysfs.
The CORE_CAPABILITIES MSR enumerates whether IFS is supported.
The full github location for distributing the IFS images is
still being decided. So just a placeholder included for now
in the documentation.
Future CPUs will support more than one type of test. Plan for
that now by using a "_0" suffix on the ABI directory names.
Additional test types will use "_1", etc.
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Link: https://lore.kernel.org/r/20220506225410.1652287-13-tony.luck@intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Add tracing support which may be useful for debugging systems that fail to complete
In Field Scan tests.
Acked-by: Steven Rostedt (Google) <rostedt@goodmis.org>
Reviewed-by: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
Acked-by: Hans de Goede <hdegoede@redhat.com>
Reviewed-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Link: https://lore.kernel.org/r/20220506225410.1652287-11-tony.luck@intel.com
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
