1. 20 Jun, 2018 5 commits
    • Andi Kleen's avatar
      x86/speculation/l1tf: Make sure the first page is always reserved · 10a70416
      Andi Kleen authored
      The L1TF workaround doesn't make any attempt to mitigate speculate accesses
      to the first physical page for zeroed PTEs. Normally it only contains some
      data from the early real mode BIOS.
      
      It's not entirely clear that the first page is reserved in all
      configurations, so add an extra reservation call to make sure it is really
      reserved. In most configurations (e.g.  with the standard reservations)
      it's likely a nop.
      Signed-off-by: default avatarAndi Kleen <ak@linux.intel.com>
      Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
      Reviewed-by: default avatarJosh Poimboeuf <jpoimboe@redhat.com>
      Acked-by: default avatarDave Hansen <dave.hansen@intel.com>
      
      10a70416
    • Andi Kleen's avatar
      x86/speculation/l1tf: Protect PROT_NONE PTEs against speculation · 6b28baca
      Andi Kleen authored
      When PTEs are set to PROT_NONE the kernel just clears the Present bit and
      preserves the PFN, which creates attack surface for L1TF speculation
      speculation attacks.
      
      This is important inside guests, because L1TF speculation bypasses physical
      page remapping. While the host has its own migitations preventing leaking
      data from other VMs into the guest, this would still risk leaking the wrong
      page inside the current guest.
      
      This uses the same technique as Linus' swap entry patch: while an entry is
      is in PROTNONE state invert the complete PFN part part of it. This ensures
      that the the highest bit will point to non existing memory.
      
      The invert is done by pte/pmd_modify and pfn/pmd/pud_pte for PROTNONE and
      pte/pmd/pud_pfn undo it.
      
      This assume that no code path touches the PFN part of a PTE directly
      without using these primitives.
      
      This doesn't handle the case that MMIO is on the top of the CPU physical
      memory. If such an MMIO region was exposed by an unpriviledged driver for
      mmap it would be possible to attack some real memory.  However this
      situation is all rather unlikely.
      
      For 32bit non PAE the inversion is not done because there are really not
      enough bits to protect anything.
      
      Q: Why does the guest need to be protected when the HyperVisor already has
         L1TF mitigations?
      
      A: Here's an example:
      
         Physical pages 1 2 get mapped into a guest as
         GPA 1 -> PA 2
         GPA 2 -> PA 1
         through EPT.
      
         The L1TF speculation ignores the EPT remapping.
      
         Now the guest kernel maps GPA 1 to process A and GPA 2 to process B, and
         they belong to different users and should be isolated.
      
         A sets the GPA 1 PA 2 PTE to PROT_NONE to bypass the EPT remapping and
         gets read access to the underlying physical page. Which in this case
         points to PA 2, so it can read process B's data, if it happened to be in
         L1, so isolation inside the guest is broken.
      
         There's nothing the hypervisor can do about this. This mitigation has to
         be done in the guest itself.
      
      [ tglx: Massaged changelog ]
      Signed-off-by: default avatarAndi Kleen <ak@linux.intel.com>
      Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
      Reviewed-by: default avatarJosh Poimboeuf <jpoimboe@redhat.com>
      Acked-by: default avatarMichal Hocko <mhocko@suse.com>
      Acked-by: default avatarVlastimil Babka <vbabka@suse.cz>
      Acked-by: default avatarDave Hansen <dave.hansen@intel.com>
      
      
      6b28baca
    • Linus Torvalds's avatar
      x86/speculation/l1tf: Protect swap entries against L1TF · 2f22b4cd
      Linus Torvalds authored
      With L1 terminal fault the CPU speculates into unmapped PTEs, and resulting
      side effects allow to read the memory the PTE is pointing too, if its
      values are still in the L1 cache.
      
      For swapped out pages Linux uses unmapped PTEs and stores a swap entry into
      them.
      
      To protect against L1TF it must be ensured that the swap entry is not
      pointing to valid memory, which requires setting higher bits (between bit
      36 and bit 45) that are inside the CPUs physical address space, but outside
      any real memory.
      
      To do this invert the offset to make sure the higher bits are always set,
      as long as the swap file is not too big.
      
      Note there is no workaround for 32bit !PAE, or on systems which have more
      than MAX_PA/2 worth of memory. The later case is very unlikely to happen on
      real systems.
      
      [AK: updated description and minor tweaks by. Split out from the original
           patch ]
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      Signed-off-by: default avatarAndi Kleen <ak@linux.intel.com>
      Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
      Tested-by: default avatarAndi Kleen <ak@linux.intel.com>
      Reviewed-by: default avatarJosh Poimboeuf <jpoimboe@redhat.com>
      Acked-by: default avatarMichal Hocko <mhocko@suse.com>
      Acked-by: default avatarVlastimil Babka <vbabka@suse.cz>
      Acked-by: default avatarDave Hansen <dave.hansen@intel.com>
      
      2f22b4cd
    • Linus Torvalds's avatar
      x86/speculation/l1tf: Change order of offset/type in swap entry · bcd11afa
      Linus Torvalds authored
      If pages are swapped out, the swap entry is stored in the corresponding
      PTE, which has the Present bit cleared. CPUs vulnerable to L1TF speculate
      on PTE entries which have the present bit set and would treat the swap
      entry as phsyical address (PFN). To mitigate that the upper bits of the PTE
      must be set so the PTE points to non existent memory.
      
      The swap entry stores the type and the offset of a swapped out page in the
      PTE. type is stored in bit 9-13 and offset in bit 14-63. The hardware
      ignores the bits beyond the phsyical address space limit, so to make the
      mitigation effective its required to start 'offset' at the lowest possible
      bit so that even large swap offsets do not reach into the physical address
      space limit bits.
      
      Move offset to bit 9-58 and type to bit 59-63 which are the bits that
      hardware generally doesn't care about.
      
      That, in turn, means that if you on desktop chip with only 40 bits of
      physical addressing, now that the offset starts at bit 9, there needs to be
      30 bits of offset actually *in use* until bit 39 ends up being set, which
      means when inverted it will again point into existing memory.
      
      So that's 4 terabyte of swap space (because the offset is counted in pages,
      so 30 bits of offset is 42 bits of actual coverage). With bigger physical
      addressing, that obviously grows further, until the limit of the offset is
      hit (at 50 bits of offset - 62 bits of actual swap file coverage).
      
      This is a preparatory change for the actual swap entry inversion to protect
      against L1TF.
      
      [ AK: Updated description and minor tweaks. Split into two parts ]
      [ tglx: Massaged changelog ]
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      Signed-off-by: default avatarAndi Kleen <ak@linux.intel.com>
      Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
      Tested-by: default avatarAndi Kleen <ak@linux.intel.com>
      Reviewed-by: default avatarJosh Poimboeuf <jpoimboe@redhat.com>
      Acked-by: default avatarMichal Hocko <mhocko@suse.com>
      Acked-by: default avatarVlastimil Babka <vbabka@suse.cz>
      Acked-by: default avatarDave Hansen <dave.hansen@intel.com>
      
      bcd11afa
    • Andi Kleen's avatar
      x86/speculation/l1tf: Increase 32bit PAE __PHYSICAL_PAGE_SHIFT · 50896e18
      Andi Kleen authored
      L1 Terminal Fault (L1TF) is a speculation related vulnerability. The CPU
      speculates on PTE entries which do not have the PRESENT bit set, if the
      content of the resulting physical address is available in the L1D cache.
      
      The OS side mitigation makes sure that a !PRESENT PTE entry points to a
      physical address outside the actually existing and cachable memory
      space. This is achieved by inverting the upper bits of the PTE. Due to the
      address space limitations this only works for 64bit and 32bit PAE kernels,
      but not for 32bit non PAE.
      
      This mitigation applies to both host and guest kernels, but in case of a
      64bit host (hypervisor) and a 32bit PAE guest, inverting the upper bits of
      the PAE address space (44bit) is not enough if the host has more than 43
      bits of populated memory address space, because the speculation treats the
      PTE content as a physical host address bypassing EPT.
      
      The host (hypervisor) protects itself against the guest by flushing L1D as
      needed, but pages inside the guest are not protected against attacks from
      other processes inside the same guest.
      
      For the guest the inverted PTE mask has to match the host to provide the
      full protection for all pages the host could possibly map into the
      guest. The hosts populated address space is not known to the guest, so the
      mask must cover the possible maximal host address space, i.e. 52 bit.
      
      On 32bit PAE the maximum PTE mask is currently set to 44 bit because that
      is the limit imposed by 32bit unsigned long PFNs in the VMs. This limits
      the mask to be below what the host could possible use for physical pages.
      
      The L1TF PROT_NONE protection code uses the PTE masks to determine which
      bits to invert to make sure the higher bits are set for unmapped entries to
      prevent L1TF speculation attacks against EPT inside guests.
      
      In order to invert all bits that could be used by the host, increase
      __PHYSICAL_PAGE_SHIFT to 52 to match 64bit.
      
      The real limit for a 32bit PAE kernel is still 44 bits because all Linux
      PTEs are created from unsigned long PFNs, so they cannot be higher than 44
      bits on a 32bit kernel. So these extra PFN bits should be never set. The
      only users of this macro are using it to look at PTEs, so it's safe.
      
      [ tglx: Massaged changelog ]
      Signed-off-by: default avatarAndi Kleen <ak@linux.intel.com>
      Signed-off-by: default avatarThomas Gleixner <tglx@linutronix.de>
      Reviewed-by: default avatarJosh Poimboeuf <jpoimboe@redhat.com>
      Acked-by: default avatarMichal Hocko <mhocko@suse.com>
      Acked-by: default avatarDave Hansen <dave.hansen@intel.com>
      50896e18
  2. 16 Jun, 2018 8 commits
    • Linus Torvalds's avatar
      Linux 4.18-rc1 · ce397d21
      Linus Torvalds authored
      ce397d21
    • Linus Torvalds's avatar
      Merge tag 'for-linus-20180616' of git://git.kernel.dk/linux-block · 265c5596
      Linus Torvalds authored
      Pull block fixes from Jens Axboe:
       "A collection of fixes that should go into -rc1. This contains:
      
         - bsg_open vs bsg_unregister race fix (Anatoliy)
      
         - NVMe pull request from Christoph, with fixes for regressions in
           this window, FC connect/reconnect path code unification, and a
           trace point addition.
      
         - timeout fix (Christoph)
      
         - remove a few unused functions (Christoph)
      
         - blk-mq tag_set reinit fix (Roman)"
      
      * tag 'for-linus-20180616' of git://git.kernel.dk/linux-block:
        bsg: fix race of bsg_open and bsg_unregister
        block: remov blk_queue_invalidate_tags
        nvme-fabrics: fix and refine state checks in __nvmf_check_ready
        nvme-fabrics: handle the admin-only case properly in nvmf_check_ready
        nvme-fabrics: refactor queue ready check
        blk-mq: remove blk_mq_tagset_iter
        nvme: remove nvme_reinit_tagset
        nvme-fc: fix nulling of queue data on reconnect
        nvme-fc: remove reinit_request routine
        blk-mq: don't time out requests again that are in the timeout handler
        nvme-fc: change controllers first connect to use reconnect path
        nvme: don't rely on the changed namespace list log
        nvmet: free smart-log buffer after use
        nvme-rdma: fix error flow during mapping request data
        nvme: add bio remapping tracepoint
        nvme: fix NULL pointer dereference in nvme_init_subsystem
        blk-mq: reinit q->tag_set_list entry only after grace period
      265c5596
    • Linus Torvalds's avatar
      Merge tag 'docs-broken-links' of git://linuxtv.org/mchehab/experimental · 5e7b9212
      Linus Torvalds authored
      Pull documentation fixes from Mauro Carvalho Chehab:
       "This solves a series of broken links for files under Documentation,
        and improves a script meant to detect such broken links (see
        scripts/documentation-file-ref-check).
      
        The changes on this series are:
      
         - can.rst: fix a footnote reference;
      
         - crypto_engine.rst: Fix two parsing warnings;
      
         - Fix a lot of broken references to Documentation/*;
      
         - improve the scripts/documentation-file-ref-check script, in order
           to help detecting/fixing broken references, preventing
           false-positives.
      
        After this patch series, only 33 broken references to doc files are
        detected by scripts/documentation-file-ref-check"
      
      * tag 'docs-broken-links' of git://linuxtv.org/mchehab/experimental: (26 commits)
        fix a series of Documentation/ broken file name references
        Documentation: rstFlatTable.py: fix a broken reference
        ABI: sysfs-devices-system-cpu: remove a broken reference
        devicetree: fix a series of wrong file references
        devicetree: fix name of pinctrl-bindings.txt
        devicetree: fix some bindings file names
        MAINTAINERS: fix location of DT npcm files
        MAINTAINERS: fix location of some display DT bindings
        kernel-parameters.txt: fix pointers to sound parameters
        bindings: nvmem/zii: Fix location of nvmem.txt
        docs: Fix more broken references
        scripts/documentation-file-ref-check: check tools/*/Documentation
        scripts/documentation-file-ref-check: get rid of false-positives
        scripts/documentation-file-ref-check: hint: dash or underline
        scripts/documentation-file-ref-check: add a fix logic for DT
        scripts/documentation-file-ref-check: accept more wildcards at filenames
        scripts/documentation-file-ref-check: fix help message
        media: max2175: fix location of driver's companion documentation
        media: v4l: fix broken video4linux docs locations
        media: dvb: point to the location of the old README.dvb-usb file
        ...
      5e7b9212
    • Linus Torvalds's avatar
      Merge tag 'fsnotify_for_v4.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs · dbb2816f
      Linus Torvalds authored
      Pull fsnotify updates from Jan Kara:
       "fsnotify cleanups unifying handling of different watch types.
      
        This is the shortened fsnotify series from Amir with the last five
        patches pulled out. Amir has modified those patches to not change
        struct inode but obviously it's too late for those to go into this
        merge window"
      
      * tag 'fsnotify_for_v4.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
        fsnotify: add fsnotify_add_inode_mark() wrappers
        fanotify: generalize fanotify_should_send_event()
        fsnotify: generalize send_to_group()
        fsnotify: generalize iteration of marks by object type
        fsnotify: introduce marks iteration helpers
        fsnotify: remove redundant arguments to handle_event()
        fsnotify: use type id to identify connector object type
      dbb2816f
    • Linus Torvalds's avatar
      Merge tag 'fbdev-v4.18' of git://github.com/bzolnier/linux · 644f2639
      Linus Torvalds authored
      Pull fbdev updates from Bartlomiej Zolnierkiewicz:
       "There is nothing really major here, few small fixes, some cleanups and
        dead drivers removal:
      
         - mark omapfb drivers as orphans in MAINTAINERS file (Tomi Valkeinen)
      
         - add missing module license tags to omap/omapfb driver (Arnd
           Bergmann)
      
         - add missing GPIOLIB dependendy to omap2/omapfb driver (Arnd
           Bergmann)
      
         - convert savagefb, aty128fb & radeonfb drivers to use msleep & co.
           (Jia-Ju Bai)
      
         - allow COMPILE_TEST build for viafb driver (media part was reviewed
           by media subsystem Maintainer)
      
         - remove unused MERAM support from sh_mobile_lcdcfb and shmob-drm
           drivers (drm parts were acked by shmob-drm driver Maintainer)
      
         - remove unused auo_k190xfb drivers
      
         - misc cleanups (Souptick Joarder, Wolfram Sang, Markus Elfring, Andy
           Shevchenko, Colin Ian King)"
      
      * tag 'fbdev-v4.18' of git://github.com/bzolnier/linux: (26 commits)
        fb_omap2: add gpiolib dependency
        video/omap: add module license tags
        MAINTAINERS: make omapfb orphan
        video: fbdev: pxafb: match_string() conversion fixup
        video: fbdev: nvidia: fix spelling mistake: "scaleing" -> "scaling"
        video: fbdev: fix spelling mistake: "frambuffer" -> "framebuffer"
        video: fbdev: pxafb: Convert to use match_string() helper
        video: fbdev: via: allow COMPILE_TEST build
        video: fbdev: remove unused sh_mobile_meram driver
        drm: shmobile: remove unused MERAM support
        video: fbdev: sh_mobile_lcdcfb: remove unused MERAM support
        video: fbdev: remove unused auo_k190xfb drivers
        video: omap: Improve a size determination in omapfb_do_probe()
        video: sm501fb: Improve a size determination in sm501fb_probe()
        video: fbdev-MMP: Improve a size determination in path_init()
        video: fbdev-MMP: Delete an error message for a failed memory allocation in two functions
        video: auo_k190x: Delete an error message for a failed memory allocation in auok190x_common_probe()
        video: sh_mobile_lcdcfb: Delete an error message for a failed memory allocation in two functions
        video: sh_mobile_meram: Delete an error message for a failed memory allocation in sh_mobile_meram_probe()
        video: fbdev: sh_mobile_meram: Drop SUPERH platform dependency
        ...
      644f2639
    • Linus Torvalds's avatar
      Merge branch 'afs-proc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs · 35773c93
      Linus Torvalds authored
      Pull AFS updates from Al Viro:
       "Assorted AFS stuff - ended up in vfs.git since most of that consists
        of David's AFS-related followups to Christoph's procfs series"
      
      * 'afs-proc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
        afs: Optimise callback breaking by not repeating volume lookup
        afs: Display manually added cells in dynamic root mount
        afs: Enable IPv6 DNS lookups
        afs: Show all of a server's addresses in /proc/fs/afs/servers
        afs: Handle CONFIG_PROC_FS=n
        proc: Make inline name size calculation automatic
        afs: Implement network namespacing
        afs: Mark afs_net::ws_cell as __rcu and set using rcu functions
        afs: Fix a Sparse warning in xdr_decode_AFSFetchStatus()
        proc: Add a way to make network proc files writable
        afs: Rearrange fs/afs/proc.c to remove remaining predeclarations.
        afs: Rearrange fs/afs/proc.c to move the show routines up
        afs: Rearrange fs/afs/proc.c by moving fops and open functions down
        afs: Move /proc management functions to the end of the file
      35773c93
    • Linus Torvalds's avatar
      Merge branch 'work.compat' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs · 29d6849d
      Linus Torvalds authored
      Pull compat updates from Al Viro:
       "Some biarch patches - getting rid of assorted (mis)uses of
        compat_alloc_user_space().
      
        Not much in that area this cycle..."
      
      * 'work.compat' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
        orangefs: simplify compat ioctl handling
        signalfd: lift sigmask copyin and size checks to callers of do_signalfd4()
        vmsplice(): lift importing iovec into vmsplice(2) and compat counterpart
      29d6849d
    • Linus Torvalds's avatar
      Merge branch 'work.aio' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs · a5b729ea
      Linus Torvalds authored
      Pull aio fixes from Al Viro:
       "Assorted AIO followups and fixes"
      
      * 'work.aio' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
        eventpoll: switch to ->poll_mask
        aio: only return events requested in poll_mask() for IOCB_CMD_POLL
        eventfd: only return events requested in poll_mask()
        aio: mark __aio_sigset::sigmask const
      a5b729ea
  3. 15 Jun, 2018 27 commits