Re: [Qemu-devel] [PATCH v2] target/xtensa: rework zero overhead loops implementation

[Richard Henderson]

On 1/12/19 9:01 AM, Max Filippov wrote:
> Don't invalidate TB with the end of zero overhead loop when LBEG or LEND
> change. Instead encode the distance from the start of the page where the
> TB starts to the LEND in the TB cs_base and generate loopback code when
> the next PC matches encoded LEND. Distance to a destination within the
> same page and up to a maximum instruction length into the next page is
> encoded literally, otherwise it's zero. The distance from LEND to LBEG
> is also encoded in the cs_base: it's encoded literally when less than
> 256 or as 0 otherwise. This allows for TB chaining for the loopback
> branch at the end of a loop for the most common loop sizes.
> 
> With this change the resulting emulation speed is about 10% higher in
> softmmu mode on uClibc-ng and LTP tests. Emulation speed in linux
> user mode is a few percent lower because there's no direct TB chaining
> between different memory pages. Testing with lower limit on direct TB
> chainig range shows gradual slowdown to ~15% for the block size of 64
> bytes and ~50% for the block size of 32 bytes.
> 
> Signed-off-by: Max Filippov <address@hidden>

Reviewed-by: Richard Henderson <address@hidden>


> +        /*
> +         * 0 in the csbase_lend field means that there may not be a loopback
> +         * for any instruction that starts inside this page. Any other value
> +         * means that an instruction that ends at this offset from the page
> +         * start may loop back.

Nit: s/may/will/g

Using "may" makes it seem like we may have missed a case that should have
looped back.


r~