From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>,
Dale B Stimson <dale.b.stimson@intel.com>,
igt-dev@lists.freedesktop.org, intel-gfx@lists.freedesktop.org,
Bommu Krishnaiah <krishnaiah.bommu@intel.com>,
Sreedhar Telukuntla <sreedhar.telukuntla@intel.com>
Cc: "Melkaveri, Arjun" <arjun.melkaveri@intel.com>,
Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Subject: Re: [igt-dev] [PATCH i-g-t 2/2] DBS: tests/i915/gem_ctx_isolation: use the gem_engine_topology library, part 2
Date: Thu, 23 Jan 2020 15:59:33 +0000 [thread overview]
Message-ID: <c192e00d-1997-c714-50be-91fc64f9cc29@linux.intel.com> (raw)
In-Reply-To: <157977054018.18920.9879323586181021015@skylake-alporthouse-com>
On 23/01/2020 09:09, Chris Wilson wrote:
> Quoting Dale B Stimson (2020-01-22 23:26:57)
>> Switch from simple iteration over all potential engines to using
>> macro __for_each_physical_engine which only returns engines that are
>> actually present.
>>
>> For each context (as it is created) call gem_context_set_all_engines
>> so that execbuf will interpret the engine specification in the new way.
>>
>> Signed-off-by: Dale B Stimson <dale.b.stimson@intel.com>
>> ---
>> tests/i915/gem_ctx_isolation.c | 41 ++++++++++++++++++++++++++--------
>> 1 file changed, 32 insertions(+), 9 deletions(-)
>>
>> diff --git a/tests/i915/gem_ctx_isolation.c b/tests/i915/gem_ctx_isolation.c
>> index 25113b054..31a20ed3a 100644
>> --- a/tests/i915/gem_ctx_isolation.c
>> +++ b/tests/i915/gem_ctx_isolation.c
>> @@ -240,6 +240,25 @@ static bool ignore_register(uint32_t offset)
>> return false;
>> }
>>
>> +/*
>> + * context_create_plus_all_engines - Same as gem_context_create plus setup.
>> + *
>> + * This is a convenience function that may be called instead of the sequence
>> + * of gem_context_create followed by gem_context_set_all_engines.
>> + * If gem_has_engine_topology(), then function gem_context_set_all_engines
>> + * indicates that future execbuf calls for this context should interpret the
>> + * engine specification in a gem_engine_topology-compatible way.
>> + */
>> +static uint32_t context_create_plus_all_engines(int fd)
>> +{
>> + uint32_t ctx;
>> +
>> + ctx = gem_context_create(fd);
>> + gem_context_set_all_engines(fd, ctx);
>> +
>> + return ctx;
>> +}
>
> gem_context_clone_with_engines() so we can stop assuming that
> all-engines is the right answer, because that depends on the conditions
> set up by the iterator on the first context.
gem_context_clone_with_engines was agreed upon in principle some time
ago but never implemented. I have now posted this as
https://patchwork.freedesktop.org/series/72464/ and plan to merge it
once it passes CI.
Dale, Arjun, Krishnaiah and Sreedhar - you have in progress patches
which use gem_context_set_all_engines which will be gone and you will
need to adjust your work accordingly.
Sreedhar specifically for your change in gem_exec_parallel we will need
to add a new helper which transfers the engine map from one fd/context
to another. I will copy you on a patch which will add it.
Regards,
Tvrtko
_______________________________________________
igt-dev mailing list
igt-dev@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/igt-dev
WARNING: multiple messages have this Message-ID (diff)
From: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>,
Dale B Stimson <dale.b.stimson@intel.com>,
igt-dev@lists.freedesktop.org, intel-gfx@lists.freedesktop.org,
Bommu Krishnaiah <krishnaiah.bommu@intel.com>,
Sreedhar Telukuntla <sreedhar.telukuntla@intel.com>
Cc: "Melkaveri, Arjun" <arjun.melkaveri@intel.com>
Subject: Re: [Intel-gfx] [igt-dev] [PATCH i-g-t 2/2] DBS: tests/i915/gem_ctx_isolation: use the gem_engine_topology library, part 2
Date: Thu, 23 Jan 2020 15:59:33 +0000 [thread overview]
Message-ID: <c192e00d-1997-c714-50be-91fc64f9cc29@linux.intel.com> (raw)
In-Reply-To: <157977054018.18920.9879323586181021015@skylake-alporthouse-com>
On 23/01/2020 09:09, Chris Wilson wrote:
> Quoting Dale B Stimson (2020-01-22 23:26:57)
>> Switch from simple iteration over all potential engines to using
>> macro __for_each_physical_engine which only returns engines that are
>> actually present.
>>
>> For each context (as it is created) call gem_context_set_all_engines
>> so that execbuf will interpret the engine specification in the new way.
>>
>> Signed-off-by: Dale B Stimson <dale.b.stimson@intel.com>
>> ---
>> tests/i915/gem_ctx_isolation.c | 41 ++++++++++++++++++++++++++--------
>> 1 file changed, 32 insertions(+), 9 deletions(-)
>>
>> diff --git a/tests/i915/gem_ctx_isolation.c b/tests/i915/gem_ctx_isolation.c
>> index 25113b054..31a20ed3a 100644
>> --- a/tests/i915/gem_ctx_isolation.c
>> +++ b/tests/i915/gem_ctx_isolation.c
>> @@ -240,6 +240,25 @@ static bool ignore_register(uint32_t offset)
>> return false;
>> }
>>
>> +/*
>> + * context_create_plus_all_engines - Same as gem_context_create plus setup.
>> + *
>> + * This is a convenience function that may be called instead of the sequence
>> + * of gem_context_create followed by gem_context_set_all_engines.
>> + * If gem_has_engine_topology(), then function gem_context_set_all_engines
>> + * indicates that future execbuf calls for this context should interpret the
>> + * engine specification in a gem_engine_topology-compatible way.
>> + */
>> +static uint32_t context_create_plus_all_engines(int fd)
>> +{
>> + uint32_t ctx;
>> +
>> + ctx = gem_context_create(fd);
>> + gem_context_set_all_engines(fd, ctx);
>> +
>> + return ctx;
>> +}
>
> gem_context_clone_with_engines() so we can stop assuming that
> all-engines is the right answer, because that depends on the conditions
> set up by the iterator on the first context.
gem_context_clone_with_engines was agreed upon in principle some time
ago but never implemented. I have now posted this as
https://patchwork.freedesktop.org/series/72464/ and plan to merge it
once it passes CI.
Dale, Arjun, Krishnaiah and Sreedhar - you have in progress patches
which use gem_context_set_all_engines which will be gone and you will
need to adjust your work accordingly.
Sreedhar specifically for your change in gem_exec_parallel we will need
to add a new helper which transfers the engine map from one fd/context
to another. I will copy you on a patch which will add it.
Regards,
Tvrtko
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2020-01-23 15:59 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-22 23:26 [igt-dev] [PATCH i-g-t 0/2] i915/gem_ctx_isolation: __for_each_physical_engine + engine mapping Dale B Stimson
2020-01-22 23:26 ` [Intel-gfx] " Dale B Stimson
2020-01-22 23:26 ` [igt-dev] [PATCH i-g-t 1/2] tests/i915/gem_ctx_isolation: use the gem_engine_topology library, part 1 Dale B Stimson
2020-01-22 23:26 ` [Intel-gfx] " Dale B Stimson
2020-01-22 23:26 ` [igt-dev] [PATCH i-g-t 2/2] DBS: tests/i915/gem_ctx_isolation: use the gem_engine_topology library, part 2 Dale B Stimson
2020-01-22 23:26 ` [Intel-gfx] " Dale B Stimson
2020-01-23 9:09 ` [igt-dev] " Chris Wilson
2020-01-23 9:09 ` [Intel-gfx] " Chris Wilson
2020-01-23 15:59 ` Tvrtko Ursulin [this message]
2020-01-23 15:59 ` [Intel-gfx] [igt-dev] " Tvrtko Ursulin
2020-01-23 19:21 ` Dale B Stimson
2020-01-23 19:21 ` [Intel-gfx] " Dale B Stimson
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=c192e00d-1997-c714-50be-91fc64f9cc29@linux.intel.com \
--to=tvrtko.ursulin@linux.intel.com \
--cc=arjun.melkaveri@intel.com \
--cc=chris@chris-wilson.co.uk \
--cc=dale.b.stimson@intel.com \
--cc=igt-dev@lists.freedesktop.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=krishnaiah.bommu@intel.com \
--cc=sreedhar.telukuntla@intel.com \
--cc=tvrtko.ursulin@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.