qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [PATCH v3 0/3] target/arm: Reduce overhead of cpu_get_t


From: Emilio G. Cota
Subject: Re: [Qemu-devel] [PATCH v3 0/3] target/arm: Reduce overhead of cpu_get_tb_cpu_state
Date: Fri, 1 Mar 2019 20:49:21 -0500
User-agent: Mutt/1.9.4 (2018-02-28)

On Thu, Feb 21, 2019 at 18:41:03 -0800, Richard Henderson wrote:
> Changes since v2:
>   * Do not cache VECLEN, VECSTRIDE, VFPEN.
>     These variables come from VFP_FPSCR and VFP_FPEXC, not from
>     system control registers.
>   * Move HANDLER and STACKCHECK to rebuild_hflags_a32,
>     instead of building them in rebuild_hflags_common.

Tested-by: Emilio G. Cota <address@hidden>

You might want to add these numbers (I re-ran the benchmarks for v3)
to patch 3's commit log:

                        aarch64-linux-user SPEC06int (train set)
                     Host: Intel(R) Core(TM) i7-4790K CPU @ 4.00GHz
         2.2 +--------------------------------------------------------------+
             |                                                              |
             |                                                after         |
           2 |-+.........................................+-+..............+-|
             |                                            |                 |
             |                                           *|*                |
             |                                           *|*      +-+       |
         1.8 |-+..........+-+............................+-+.......|......+-|
             |             |                             * *       |        |
             |           **|*               +-+     *+-+ * *      *|*       |
         1.6 |-+.........*+-+..............**|*.....*..*.*.*.*+-+.*|*.....+-|
             |           *  *              *+-+     *  * * * *+-+ *|*       |
             |           *  *              *  *     *  * * * *  * +-+       |
         1.4 |-+.........*..*......+-+.....*..*.....*..*.*.*.*..*.*.*.*+-++-|
             |           *  *     **|*     *  *     *  * * * *  * * * *+-+  |
             |  *+-+     *  *     *+-+     *  *     *  * * * *  * * * *  *  |
         1.2 |-+*+-+.....*..*.....*..*.....*..*.....*..*.*.*.*..*.*.*.*..*+-|
             |  *  * +-+ *  *     *  * +-+ *  *     *  * * * *  * * * *  *  |
             |  *  *  |  *  * +-+ *  * +-+ *  * +-+ *  * * * *  * * * *  *  |
             |  *  * *|* *  * *|* *  * * * *  * +-+ *  * * * *  * * * *  *  |
           1 |++*++*++-++*++*++-++*++*+*+*+*++*+*+*+*++*+*+*+*++*+*+*+*++*++|
             |  *  * * * *  * * * *  * * * *  * * * *  * * * *  * * * *  *  |
             |  *  * * * *  * * * *  * * * *  * * * *  * * * *  * * * *  *  |
         0.8 +--------------------------------------------------------------+
     400.perl401.bzi403.429445.456.462.libq464.471.omn483.xalancbgeomean
  png: https://imgur.com/wr4ODMw

Thanks,

                Emilio





reply via email to

[Prev in Thread] Current Thread [Next in Thread]