1. 08 Aug, 2016 1 commit
    • Ville Syrjälä's avatar
      x86/hweight: Don't clobber %rdi · 65ea11ec
      Ville Syrjälä authored
      The caller expects %rdi to remain intact, push+pop it make that happen.
      
      Fixes the following kind of explosions on my core2duo machine when
      trying to reboot or shut down:
      
        general protection fault: 0000 [#1] PREEMPT SMP
        Modules linked in: i915 i2c_algo_bit drm_kms_helper cfbfillrect syscopyarea cfbimgblt sysfillrect sysimgblt fb_sys_fops cfbcopyarea drm netconsole configfs binfmt_misc iTCO_wdt psmouse pcspkr snd_hda_codec_idt e100 coretemp hwmon snd_hda_codec_generic i2c_i801 mii i2c_smbus lpc_ich mfd_core snd_hda_intel uhci_hcd snd_hda_codec snd_hwdep snd_hda_core ehci_pci 8250 ehci_hcd snd_pcm 8250_base usbcore evdev serial_core usb_common parport_pc parport snd_timer snd soundcore
        CPU: 0 PID: 3070 Comm: reboot Not tainted 4.8.0-rc1-perf-dirty #69
        Hardware name:                  /D946GZIS, BIOS TS94610J.86A.0087.2007.1107.1049 11/07/2007
        task: ffff88012a0b4080 task.stack: ffff880123850000
        RIP: 0010:[<ffffffff81003c92>]  [<ffffffff81003c92>] x86_perf_event_update+0x52/0xc0
        RSP: 0018:ffff880123853b60  EFLAGS: 00010087
        RAX: 0000000000000001 RBX: ffff88012fc0a3c0 RCX: 000000000000001e
        RDX: 0000000000000000 RSI: 0000000040000000 RDI: ffff88012b014800
        RBP: ffff880123853b88 R08: ffffffffffffffff R09: 0000000000000000
        R10: ffffea0004a012c0 R11: ffffea0004acedc0 R12: ffffffff80000001
        R13: ffff88012b0149c0 R14: ffff88012b014800 R15: 0000000000000018
        FS:  00007f8b155cd700(0000) GS:ffff88012fc00000(0000) knlGS:0000000000000000
        CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
        CR2: 00007f8b155f5000 CR3: 000000012a2d7000 CR4: 00000000000006f0
        Stack:
         ffff88012fc0a3c0 ffff88012b014800 0000000000000004 0000000000000001
         ffff88012fc1b750 ffff880123853bb0 ffffffff81003d59 ffff88012b014800
         ffff88012fc0a3c0 ffff88012b014800 ffff880123853bd8 ffffffff81003e13
        Call Trace:
         [<ffffffff81003d59>] x86_pmu_stop+0x59/0xd0
         [<ffffffff81003e13>] x86_pmu_del+0x43/0x140
         [<ffffffff8111705d>] event_sched_out.isra.105+0xbd/0x260
         [<ffffffff8111738d>] __perf_remove_from_context+0x2d/0xb0
         [<ffffffff8111745d>] __perf_event_exit_context+0x4d/0x70
         [<ffffffff810c8826>] generic_exec_single+0xb6/0x140
         [<ffffffff81117410>] ? __perf_remove_from_context+0xb0/0xb0
         [<ffffffff81117410>] ? __perf_remove_from_context+0xb0/0xb0
         [<ffffffff810c898f>] smp_call_function_single+0xdf/0x140
         [<ffffffff81113d27>] perf_event_exit_cpu_context+0x87/0xc0
         [<ffffffff81113d73>] perf_reboot+0x13/0x40
         [<ffffffff8107578a>] notifier_call_chain+0x4a/0x70
         [<ffffffff81075ad7>] __blocking_notifier_call_chain+0x47/0x60
         [<ffffffff81075b06>] blocking_notifier_call_chain+0x16/0x20
         [<ffffffff81076a1d>] kernel_restart_prepare+0x1d/0x40
         [<ffffffff81076ae2>] kernel_restart+0x12/0x60
         [<ffffffff81076d56>] SYSC_reboot+0xf6/0x1b0
         [<ffffffff811a823c>] ? mntput_no_expire+0x2c/0x1b0
         [<ffffffff811a83e4>] ? mntput+0x24/0x40
         [<ffffffff811894fc>] ? __fput+0x16c/0x1e0
         [<ffffffff811895ae>] ? ____fput+0xe/0x10
         [<ffffffff81072fc3>] ? task_work_run+0x83/0xa0
         [<ffffffff81001623>] ? exit_to_usermode_loop+0x53/0xc0
         [<ffffffff8100105a>] ? trace_hardirqs_on_thunk+0x1a/0x1c
         [<ffffffff81076e6e>] SyS_reboot+0xe/0x10
         [<ffffffff814c4ba5>] entry_SYSCALL_64_fastpath+0x18/0xa3
        Code: 7c 4c 8d af c0 01 00 00 49 89 fe eb 10 48 09 c2 4c 89 e0 49 0f b1 55 00 4c 39 e0 74 35 4d 8b a6 c0 01 00 00 41 8b 8e 60 01 00 00 <0f> 33 8b 35 6e 02 8c 00 48 c1 e2 20 85 f6 7e d2 48 89 d3 89 cf
        RIP  [<ffffffff81003c92>] x86_perf_event_update+0x52/0xc0
         RSP <ffff880123853b60>
        ---[ end trace 7ec95181faf211be ]---
        note: reboot[3070] exited with preempt_count 2
      
      Cc: Borislav Petkov <bp@suse.de>
      Cc: H. Peter Anvin <hpa@zytor.com>
      Cc: Andy Lutomirski <luto@amacapital.net>
      Cc: Brian Gerst <brgerst@gmail.com>
      Cc: Denys Vlasenko <dvlasenk@redhat.com>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Thomas Gleixner <tglx@linutronix.de>
      Cc: Ingo Molnar <mingo@kernel.org>
      Fixes: f5967101
      
       ("x86/hweight: Get rid of the special calling convention")
      Signed-off-by: default avatarVille Syrjälä <ville.syrjala@linux.intel.com>
      Signed-off-by: default avatarLinus Torvalds <torvalds@linux-foundation.org>
      65ea11ec
  2. 08 Jun, 2016 1 commit
    • Borislav Petkov's avatar
      x86/hweight: Get rid of the special calling convention · f5967101
      Borislav Petkov authored
      
      
      People complained about ARCH_HWEIGHT_CFLAGS and how it throws a wrench
      into kcov, lto, etc, experimentations.
      
      Add asm versions for __sw_hweight{32,64}() and do explicit saving and
      restoring of clobbered registers. This gets rid of the special calling
      convention. We get to call those functions on !X86_FEATURE_POPCNT CPUs.
      
      We still need to hardcode POPCNT and register operands as some old gas
      versions which we support, do not know about POPCNT.
      
      Btw, remove redundant REX prefix from 32-bit POPCNT because alternatives
      can do padding now.
      Suggested-by: default avatarH. Peter Anvin <hpa@zytor.com>
      Signed-off-by: default avatarBorislav Petkov <bp@suse.de>
      Acked-by: default avatarPeter Zijlstra (Intel) <peterz@infradead.org>
      Cc: Andy Lutomirski <luto@amacapital.net>
      Cc: Borislav Petkov <bp@alien8.de>
      Cc: Brian Gerst <brgerst@gmail.com>
      Cc: Denys Vlasenko <dvlasenk@redhat.com>
      Cc: Linus Torvalds <torvalds@linux-foundation.org>
      Cc: Peter Zijlstra <peterz@infradead.org>
      Cc: Thomas Gleixner <tglx@lin...
      f5967101