summaryrefslogtreecommitdiff
path: root/drivers/platform/x86/intel/ifs/load.c
AgeCommit message (Collapse)Author
2023-05-23platform/x86/intel/ifs: Annotate work queue on stack so object debug does ↵David Arcari
not complain Object Debug results in the following warning while attempting to load ifs firmware: [ 220.007422] ODEBUG: object 000000003bf952db is on stack 00000000e843994b, but NOT annotated. [ 220.007459] ------------[ cut here ]------------ [ 220.007461] WARNING: CPU: 0 PID: 11774 at lib/debugobjects.c:548 __debug_object_init.cold+0x22e/0x2d5 [ 220.137476] RIP: 0010:__debug_object_init.cold+0x22e/0x2d5 [ 220.254774] Call Trace: [ 220.257641] <TASK> [ 220.265606] scan_chunks_sanity_check+0x368/0x5f0 [intel_ifs] [ 220.288292] ifs_load_firmware+0x2a3/0x400 [intel_ifs] [ 220.332793] current_batch_store+0xea/0x160 [intel_ifs] [ 220.357947] kernfs_fop_write_iter+0x355/0x530 [ 220.363048] new_sync_write+0x28e/0x4a0 [ 220.381226] vfs_write+0x62a/0x920 [ 220.385160] ksys_write+0xf9/0x1d0 [ 220.399421] do_syscall_64+0x59/0x90 [ 220.440635] entry_SYSCALL_64_after_hwframe+0x63/0xcd [ 220.566845] ---[ end trace 3a01b299db142b41 ]--- Correct this by calling INIT_WORK_ONSTACK instead of INIT_WORK. Fixes: 684ec215706d ("platform/x86/intel/ifs: Authenticate and copy to secured memory") Signed-off-by: David Arcari <darcari@redhat.com> Cc: Jithu Joseph <jithu.joseph@intel.com> Cc: Ashok Raj <ashok.raj@intel.com> Cc: Tony Luck <tony.luck@intel.com> Cc: Hans de Goede <hdegoede@redhat.com> Cc: Mark Gross <markgross@kernel.org> Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Thomas Gleixner <tglx@linutronix.de> Cc: Dan Williams <dan.j.williams@intel.com> Cc: linux-kernel@vger.kernel.org Cc: stable@vger.kernel.org Link: https://lore.kernel.org/r/20230523105400.674152-1-darcari@redhat.com Reviewed-by: Hans de Goede <hdegoede@redhat.com> Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2023-03-27platform/x86/intel/ifs: Reorganize driver dataJithu Joseph
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>
2023-03-27platform/x86/intel/ifs: Separate ifs_pkg_auth from ifs_dataJithu Joseph
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>
2022-11-19platform/x86/intel/ifs: Add current_batch sysfs entryJithu Joseph
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
2022-11-19platform/x86/intel/ifs: Add metadata validationJithu Joseph
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
2022-11-19platform/x86/intel/ifs: Use generic microcode headers and functionsJithu Joseph
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
2022-11-19platform/x86/intel/ifs: Add metadata supportAshok Raj
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
2022-11-18platform/x86/intel/ifs: Remove memory allocation from load pathJithu Joseph
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
2022-11-18platform/x86/intel/ifs: Return a more appropriate error codeJithu Joseph
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
2022-05-12platform/x86/intel/ifs: Authenticate and copy to secured memoryJithu Joseph
The IFS image contains hashes that will be used to authenticate the ifs test chunks. First, use WRMSR to copy the hashes and enumerate the number of test chunks, chunk size and the maximum number of cores that can run scan test simultaneously. Next, use WRMSR to authenticate each and every scan test chunk which is stored in the IFS image. The CPU will check if the test chunks match the hashes, otherwise failure is indicated to system software. If the test chunk is authenticated, it is automatically copied to secured memory. Use schedule_work_on() to perform the hash copy and authentication. Note this needs only be done on the first logical cpu of each socket. Reviewed-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jithu Joseph <jithu.joseph@intel.com> Co-developed-by: Tony Luck <tony.luck@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> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lore.kernel.org/r/20220506225410.1652287-8-tony.luck@intel.com Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2022-05-12platform/x86/intel/ifs: Check IFS Image sanityJithu Joseph
IFS image is designed specifically for a given family, model and stepping of the processor. Like Intel microcode header, the IFS image has the Processor Signature, Checksum and Processor Flags that must be matched with the information returned by the CPUID. Reviewed-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jithu Joseph <jithu.joseph@intel.com> Co-developed-by: Tony Luck <tony.luck@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> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lore.kernel.org/r/20220506225410.1652287-7-tony.luck@intel.com Signed-off-by: Hans de Goede <hdegoede@redhat.com>
2022-05-12platform/x86/intel/ifs: Read IFS firmware imageJithu Joseph
Driver probe routine allocates structure to communicate status and parameters between functions in the driver. Also call load_ifs_binary() to load the scan image file. There is a separate scan image file for each processor family, model, stepping combination. This is read from the static path: /lib/firmware/intel/ifs/{ff-mm-ss}.scan Step 1 in loading is to generate the correct path and use request_firmware_direct() to load into memory. Subsequent patches will use the IFS MSR interfaces to copy the image to BIOS reserved memory and validate the SHA256 checksums. Reviewed-by: Dan Williams <dan.j.williams@intel.com> Signed-off-by: Jithu Joseph <jithu.joseph@intel.com> Co-developed-by: Tony Luck <tony.luck@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> Reviewed-by: Thomas Gleixner <tglx@linutronix.de> Link: https://lore.kernel.org/r/20220506225410.1652287-6-tony.luck@intel.com Signed-off-by: Hans de Goede <hdegoede@redhat.com>