1. 23 Jun, 2018 1 commit
  2. 21 Jun, 2018 17 commits
  3. 20 Jun, 2018 1 commit
    • Soby Mathew's avatar
      ARM Platforms: Update CNTFRQ register in CNTCTLBase frame · 342d6220
      Soby Mathew authored
      
      
      Currently TF-A doesn't initialise CNTFRQ register in CNTCTLBase
      frame of the system timer. ARM ARM states that "The instance of
      the register in the CNTCTLBase frame must be programmed with this
      value as part of system initialization."
      
      The psci_arch_setup() updates the CNTFRQ system register but
      according to the ARM ARM, this instance of the register is
      independent of the memory mapped instance. This is only an issue
      for Normal world software which relies on the memory mapped
      instance rather than the system register one.
      
      This patch resolves the issue for ARM platforms.
      
      The patch also solves a related issue on Juno, wherein
      CNTBaseN.CNTFRQ can be written and does not reflect the value of
      the register in CNTCTLBase frame. Hence this patch additionally
      updates CNTFRQ register in the Non Secure frame of the CNTBaseN.
      
      Fixes ARM-Software/tf-issues#593
      
      Change-Id: I09cebb6633688b34d5b1bc349fbde4751025b350
      Signed-off-by: default avatarSoby Mathew <soby.mathew@arm.com>
      342d6220
  4. 19 Jun, 2018 17 commits
  5. 15 Jun, 2018 4 commits
    • John Tsichritzis's avatar
      Panic in BL1 when TB_FW_CONFIG is invalid · 355e0967
      John Tsichritzis authored
      
      
      In Arm platforms, when using dynamic configuration, the necessary
      parameters are made available as a DTB. The DTB is loaded by BL1 and,
      later on, is parsed by BL1, BL2 or even both, depending on when
      information from the DTB is needed.
      
      When the DTB is going to be parsed, it must be validated first, to
      ensure that it is properly structured. If an invalid DTB is detected
      then:
        - BL1 prints a diagnostic but allows execution to continue,
        - BL2 prints a diagnostic and panics.
      
      Now the behaviour of BL1 is changed so for it also to panic. Thus, the
      behaviour of BL1 and BL2 is now similar.
      
      Keep in mind that if BL1 only loads the DTB but it doesn't need to
      read/write it, then it doesn't validate it. The validation is done only
      when the DTB is actually going to be accessed.
      
      Change-Id: Idcae6092e6dbeab7248dd5e041d6cbb7784fe410
      Signed-off-by: default avatarJohn Tsichritzis <john.tsichritzis@arm.com>
      355e0967
    • Andre Przywara's avatar
      allwinner: Add security setup · acb8b3ca
      Andre Przywara authored
      
      
      Some peripherals are TrustZone aware, so they need to be configured to
      be accessible from non-secure world, as we don't need any of them being
      exclusive to the secure world.
      This affects some clocks, DMA channels and the Secure Peripheral
      Controller (SPC). The latter controls access to most devices, but is not
      active unless booting with the secure boot fuse burnt.
      Signed-off-by: default avatarAndre Przywara <andre.przywara@arm.com>
      acb8b3ca
    • Samuel Holland's avatar
      allwinner: Add platform PSCI functions required for SMP · 560581ec
      Samuel Holland authored
      
      
      The reset vector entry point is preserved across CPU resets, so it only
      needs to be set once at boot.
      
      Hotplugged CPUs are not actually powered down, but are put in a wfi with
      the GIC disconnected.
      
      With this commit, Linux is able to enable, hotplug and use all four CPUs.
      Signed-off-by: default avatarSamuel Holland <samuel@sholland.org>
      Signed-off-by: default avatarAndre Przywara <andre.przywara@arm.com>
      560581ec
    • Samuel Holland's avatar
      allwinner: Add functions to control CPU power/reset · 333d66cf
      Samuel Holland authored
      
      
      sun50i_cpu_on will be used by the PSCI implementation to initialize
      secondary cores for SMP. Unfortunately, sun50i_cpu_off is not usable by
      PSCI directly, because it is not possible for a CPU to use this function
      to power itself down. Power cannot be shut off until the outputs are
      clamped, and MMIO does not work once the outputs are clamped.
      
      But at least CPU0 can shutdown the other cores early in the BL31 boot
      process and before shutting down the system.
      Signed-off-by: default avatarSamuel Holland <samuel@sholland.org>
      Signed-off-by: default avatarAndre Przywara <andre.przywara@arm.com>
      333d66cf