Re: [PATCH v8 00/22] target/arm: Reduce overhead of cpu_get_tb_cpu_state

[Peter Maydell]

On Tue, 22 Oct 2019 at 13:47, Peter Maydell <address@hidden> wrote:
>
> On Fri, 18 Oct 2019 at 18:44, Richard Henderson
> <address@hidden> wrote:
> >
> > Changes since v7:
> >   * Rebuild hflags for all successful nvic writes (Peter).
> >   * Rebuild hflags for Xscale sctlr writes (Peter).
> >
> > Changes since v6:
> >   * Regen hflags in two more places for m-profile (patch 19).
> >
> > Changes since v5:
> >   * Fix the debug assertion ifdef in the final patch.
> >   * Add more calls to arm_rebuild_hflags: CPSR and M-profile
> >     These become two new patches, 18 & 19.
> >   * Update some comments per review. (Alex)
> >
> > Changes since v4:
> >   * Split patch 1 into 15 smaller patches.
> >   * Cache the new DEBUG_TARGET_EL field.
> >   * Split out m-profile hflags separately from a-profile 32-bit.
> >   * Move around non-cached tb flags as well, avoiding repetitive
> >     checks for m-profile or other mutually exclusive conditions.
> >
> >   I haven't officially re-run the performance test quoted in the
> >   last patch, but I have eyeballed "perf top", and have dug into
> >   the compiled code a bit, which resulted in a few of the new
> >   cleanup patches (e.g. cs_base, arm_mmu_idx_el, and
> >   arm_cpu_data_is_big_endian).
> > ...
> >
> >
> > r~
>
>
>
> Applied to target-arm.next, thanks.

Turns out this asserts in qemu-armeb :-(

/home/petmay01/linaro/qemu-for-merges/build/all-linux-static/armeb-linux-user/qemu-armeb
-L ./gnemul/qemu-armeb armeb/ls -l dummyfile
qemu-armeb: /home/petmay01/linaro/qemu-for-merges/target/arm/helper.c:11267:
cpu_get_tb_cpu_state: Assertion `flags ==
rebuild_hflags_internal(env)' failed.
qemu: uncaught target signal 11 (Segmentation fault) - core dumped
Segmentation fault (core dumped)

Dropping this series again for the moment.

thanks
-- PMM