1. 24 1月, 2016 37 次提交
  2. 22 1月, 2016 3 次提交
    • H
      MIPS: hpet: Choose a safe value for the ETIME check · 5610b125
      Huacai Chen 提交于
      This patch is borrowed from x86 hpet driver and explaind below:
      
      Due to the overly intelligent design of HPETs, we need to workaround
      the problem that the compare value which we write is already behind
      the actual counter value at the point where the value hits the real
      compare register. This happens for two reasons:
      
      1) We read out the counter, add the delta and write the result to the
         compare register. When a NMI hits between the read out and the write
         then the counter can be ahead of the event already.
      
      2) The write to the compare register is delayed by up to two HPET
         cycles in AMD chipsets.
      
      We can work around this by reading back the compare register to make
      sure that the written value has hit the hardware. But that is bad
      performance wise for the normal case where the event is far enough in
      the future.
      
      As we already know that the write can be delayed by up to two cycles
      we can avoid the read back of the compare register completely if we
      make the decision whether the delta has elapsed already or not based
      on the following calculation:
      
        cmp = event - actual_count;
      
      If cmp is less than 64 HPET clock cycles, then we decide that the event
      has happened already and return -ETIME. That covers the above #1 and #2
      problems which would cause a wait for HPET wraparound (~306 seconds).
      Signed-off-by: NHuacai Chen <chenhc@lemote.com>
      Cc: Aurelien Jarno <aurelien@aurel32.net>
      Cc: Steven J. Hill <Steven.Hill@imgtec.com>
      Cc: Fuxin Zhang <zhangfx@lemote.com>
      Cc: Zhangjin Wu <wuzhangjin@gmail.com>
      Cc: Huacai Chen <chenhc@lemote.com>
      Cc: linux-mips@linux-mips.org
      Cc: stable@vger.kernel.org
      Patchwork: https://patchwork.linux-mips.org/patch/12162/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>
      5610b125
    • H
      MIPS: Loongson-3: Fix SMP_ASK_C0COUNT IPI handler · 57548432
      Huacai Chen 提交于
      When Core-0 handle SMP_ASK_C0COUNT IPI, we should make other cores to
      see the result as soon as possible (especially when Store-Fill-Buffer
      is enabled). Otherwise, C0_Count syncronization makes no sense.
      
      BTW, array is more suitable than per-cpu variable for syncronization,
      and there is a corner case should be avoid: C0_Count of Core-0 can be
      really 0.
      Signed-off-by: NHuacai Chen <chenhc@lemote.com>
      Cc: Aurelien Jarno <aurelien@aurel32.net>
      Cc: Steven J. Hill <Steven.Hill@imgtec.com>
      Cc: Fuxin Zhang <zhangfx@lemote.com>
      Cc: Zhangjin Wu <wuzhangjin@gmail.com>
      Cc: Huacai Chen <chenhc@lemote.com>
      Cc: linux-mips@linux-mips.org
      Cc: <stable@vger.kernel.org>
      Patchwork: https://patchwork.linux-mips.org/patch/12160/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>
      57548432
    • H
      MIPS: Loongson-3: Improve -march option and move it to Platform · 5188129b
      Huacai Chen 提交于
      If GCC >= 4.9 and Binutils >=2.25, we use -march=loongson3a, otherwise
      we use -march=mips64r2, this can slightly improve performance. Besides,
      arch/mips/loongson64/Platform is a better location rather than arch/
      mips/Makefile.
      Signed-off-by: NHuacai Chen <chenhc@lemote.com>
      Cc: Aurelien Jarno <aurelien@aurel32.net>
      Cc: Steven J. Hill <Steven.Hill@imgtec.com>
      Cc: Fuxin Zhang <zhangfx@lemote.com>
      Cc: Zhangjin Wu <wuzhangjin@gmail.com>
      Cc: linux-mips@linux-mips.org
      Patchwork: https://patchwork.linux-mips.org/patch/12161/Signed-off-by: NRalf Baechle <ralf@linux-mips.org>
      5188129b