Re: [Tarantool-patches] [PATCH luajit] test: fix misclib-getmetrics-lapi.test.lua

[Sergey Bronnikov via Tarantool-patches <tarantool-patches@dev.tarantool.org>]

[-- Attachment #1: Type: text/plain, Size: 2642 bytes --]

Hi, Sergey,

thanks for the patch! LGTM

On 28.09.2024 13:33, Sergey Kaplun wrote:
> When run under Tarantool, the test may fail if Tarantool already has
> performed some finalize steps. This patch fixes it by making the test
> more general.
> ---
>
> Branch:https://github.com/tarantool/luajit/tree/skaplun/fix-getmetrics-lapi-test
>
>   test/tarantool-tests/misclib-getmetrics-lapi.test.lua | 11 ++++++-----
>   1 file changed, 6 insertions(+), 5 deletions(-)
>
> diff --git a/test/tarantool-tests/misclib-getmetrics-lapi.test.lua b/test/tarantool-tests/misclib-getmetrics-lapi.test.lua
> index 741900b6..89aecf7b 100644
> --- a/test/tarantool-tests/misclib-getmetrics-lapi.test.lua
> +++ b/test/tarantool-tests/misclib-getmetrics-lapi.test.lua
> @@ -132,8 +132,9 @@ test:test("gc-steps", function(subtest)
>       subtest:ok(oldm.gc_steps_atomic > 0)
>       subtest:ok(oldm.gc_steps_sweepstring > 0)
>       subtest:ok(oldm.gc_steps_sweep > 0)
> -    -- Nothing to finalize, skipped.
> -    subtest:is(oldm.gc_steps_finalize, 0)
> +    -- Nothing to finalize, skipped. Tarantool may perform some
> +    -- finalize steps already, so just use the base test.
> +    subtest:ok(oldm.gc_steps_finalize >= 0)
>   
>       -- As long as we stopped the GC, consequent call
>       -- should return the same values:
> @@ -144,7 +145,7 @@ test:test("gc-steps", function(subtest)
>       subtest:is(newm.gc_steps_sweepstring - oldm.gc_steps_sweepstring, 0)
>       subtest:is(newm.gc_steps_sweep - oldm.gc_steps_sweep, 0)
>       -- Nothing to finalize, skipped.
> -    subtest:is(newm.gc_steps_finalize, 0)
> +    subtest:is(newm.gc_steps_finalize - oldm.gc_steps_finalize, 0)
>       oldm = newm
>   
>       -- Now the last phase: run full GC once and make sure that
> @@ -158,7 +159,7 @@ test:test("gc-steps", function(subtest)
>       subtest:ok(newm.gc_steps_sweepstring - oldm.gc_steps_sweepstring >= 1)
>       subtest:ok(newm.gc_steps_sweep  - oldm.gc_steps_sweep >= 1)
>       -- Nothing to finalize, skipped.
> -    subtest:is(newm.gc_steps_finalize, 0)
> +    subtest:is(newm.gc_steps_finalize - oldm.gc_steps_finalize, 0)
>       oldm = newm
>   
>       -- Now let's run three GC cycles to ensure that increment
> @@ -174,7 +175,7 @@ test:test("gc-steps", function(subtest)
>       subtest:ok(newm.gc_steps_sweepstring - oldm.gc_steps_sweepstring >= 3)
>       subtest:ok(newm.gc_steps_sweep  - oldm.gc_steps_sweep >= 3)
>       -- Nothing to finalize, skipped.
> -    subtest:is(newm.gc_steps_finalize, 0)
> +    subtest:is(newm.gc_steps_finalize - oldm.gc_steps_finalize, 0)
>   end)
>   
>   test:test("objcount", function(subtest)

[-- Attachment #2: Type: text/html, Size: 3124 bytes --]