From: "Dixit, Ashutosh" <ashutosh.dixit@intel.com>
To: "Zbigniew Kempczyński" <zbigniew.kempczynski@intel.com>
Cc: igt-dev@lists.freedesktop.org, Petri Latvala <petri.latvala@intel.com>
Subject: Re: [igt-dev] [PATCH i-g-t] tests/gem_exec_await: Adopt to no-reloc mode
Date: Fri, 22 Oct 2021 16:13:02 -0700 [thread overview]
Message-ID: <87fssswt4x.wl-ashutosh.dixit@intel.com> (raw)
In-Reply-To: <20211021114400.28238-1-zbigniew.kempczynski@intel.com>
On Thu, 21 Oct 2021 04:44:00 -0700, Zbigniew Kempczyński wrote:
>
> In previous series most of tests were already adopted to no-reloc
> mode but some still requires adoption. This change adds softpinning
> with offset acquired from an allocator.
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
> Signed-off-by: Zbigniew Kempczyński <zbigniew.kempczynski@intel.com>
> Cc: Ashutosh Dixit <ashutosh.dixit@intel.com>
> Cc: Petri Latvala <petri.latvala@intel.com>
> ---
> tests/i915/gem_exec_await.c | 31 +++++++++++++++++++++++++++----
> 1 file changed, 27 insertions(+), 4 deletions(-)
>
> diff --git a/tests/i915/gem_exec_await.c b/tests/i915/gem_exec_await.c
> index bea57c61b..bf8eb6822 100644
> --- a/tests/i915/gem_exec_await.c
> +++ b/tests/i915/gem_exec_await.c
> @@ -72,6 +72,7 @@ static void wide(int fd, const intel_ctx_t *ctx, int ring_size,
> unsigned engines[I915_EXEC_RING_MASK + 1], nengine;
> unsigned long count;
> double time;
> + uint64_t ahnd = get_reloc_ahnd(fd, 0); /* just offset provider */
>
> nengine = 0;
> for_each_ctx_engine(fd, ctx, engine) {
> @@ -97,8 +98,13 @@ static void wide(int fd, const intel_ctx_t *ctx, int ring_size,
> for (unsigned n = 0; n < ring_size; n++) {
> exec[e].obj[n].handle = gem_create(fd, 4096);
> exec[e].obj[n].flags = EXEC_OBJECT_WRITE;
> + exec[e].obj[n].offset = get_offset(ahnd, exec[e].obj[n].handle,
> + 4096, 0);
> + if (ahnd)
> + exec[e].obj[n].flags |= EXEC_OBJECT_PINNED;
>
> obj[e*ring_size + n].handle = exec[e].obj[n].handle;
> + obj[e*ring_size + n].offset = exec[e].obj[n].offset;
> }
>
> exec[e].execbuf.buffers_ptr = to_user_pointer(exec[e].exec);
> @@ -115,6 +121,11 @@ static void wide(int fd, const intel_ctx_t *ctx, int ring_size,
> }
>
> exec[e].exec[0].handle = gem_create(fd, 4096);
> + exec[e].exec[0].offset = get_offset(ahnd, exec[e].exec[0].handle,
> + 4096, 0);
> + if (ahnd)
> + exec[e].exec[0].flags = EXEC_OBJECT_PINNED;
> +
> exec[e].cmd = gem_mmap__wc(fd, exec[e].exec[0].handle,
> 0, 4096, PROT_WRITE);
>
> @@ -133,11 +144,17 @@ static void wide(int fd, const intel_ctx_t *ctx, int ring_size,
> exec[e].reloc.delta = 1;
>
> exec[e].exec[1].relocs_ptr = to_user_pointer(&exec[e].reloc);
> - exec[e].exec[1].relocation_count = 1;
> + exec[e].exec[1].relocation_count = !ahnd ? 1 : 0;
> }
> +
> obj[nengine*ring_size].handle = gem_create(fd, 4096);
> gem_write(fd, obj[nengine*ring_size].handle, 0, &bbe, sizeof(bbe));
>
> + obj[nengine*ring_size].offset = get_offset(ahnd, obj[nengine*ring_size].handle,
> + 4096, 0);
> + if (ahnd)
> + obj[nengine*ring_size].flags |= EXEC_OBJECT_PINNED;
> +
> memset(&execbuf, 0, sizeof(execbuf));
> execbuf.buffers_ptr = to_user_pointer(&obj[nengine*ring_size]);
> execbuf.buffer_count = 1;
> @@ -222,16 +239,22 @@ static void wide(int fd, const intel_ctx_t *ctx, int ring_size,
> if (flags & CONTEXTS)
> intel_ctx_destroy(fd, exec[e].ctx);
>
> - for (unsigned n = 0; n < ring_size; n++)
> + for (unsigned n = 0; n < ring_size; n++) {
> gem_close(fd, exec[e].obj[n].handle);
> + put_offset(ahnd, exec[e].obj[n].handle);
> + }
> free(exec[e].obj);
>
> munmap(exec[e].cmd, 4096);
> gem_close(fd, exec[e].exec[1].handle);
> + put_offset(ahnd, exec[e].exec[1].handle);
> }
> free(exec);
> + put_ahnd(ahnd);
> }
>
> +#define TIMEOUT 20
> +
> igt_main
> {
> int ring_size = 0;
> @@ -254,11 +277,11 @@ igt_main
> }
>
> igt_subtest("wide-all")
> - wide(device, ctx, ring_size, 20, 0);
> + wide(device, ctx, ring_size, TIMEOUT, 0);
>
> igt_subtest("wide-contexts") {
> gem_require_contexts(device);
> - wide(device, ctx, ring_size, 20, CONTEXTS);
> + wide(device, ctx, ring_size, TIMEOUT, CONTEXTS);
> }
>
> igt_fixture {
> --
> 2.26.0
>
prev parent reply other threads:[~2021-10-22 23:13 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-10-21 11:44 [igt-dev] [PATCH i-g-t] tests/gem_exec_await: Adopt to no-reloc mode Zbigniew Kempczyński
2021-10-21 13:01 ` [igt-dev] ✓ Fi.CI.BAT: success for " Patchwork
2021-10-21 15:00 ` [igt-dev] ✗ Fi.CI.IGT: failure " Patchwork
2021-10-22 4:08 ` Zbigniew Kempczyński
2021-10-22 23:13 ` Dixit, Ashutosh [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87fssswt4x.wl-ashutosh.dixit@intel.com \
--to=ashutosh.dixit@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=petri.latvala@intel.com \
--cc=zbigniew.kempczynski@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.