[PATCH 1/2] intel: Add accessor to get HW context ID from a drm_intel_context

[PATCH 1/2] intel: Add accessor to get HW context ID from a drm_intel_context

[Ian Romanick]

From: Ian Romanick <ian.d.romanick@intel.com>

The drm_intel_context structure is, wisely, opaque.  However, libdrm
users may want to know the hardware context ID associated with the
structure.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: Ben Widawsky <ben@bwidawsk.net>
---
 intel/intel_bufmgr.h     | 1 +
 intel/intel_bufmgr_gem.c | 6 ++++++
 2 files changed, 7 insertions(+)

diff --git a/intel/intel_bufmgr.h b/intel/intel_bufmgr.h
index 15f818e..7b28a70 100644
--- a/intel/intel_bufmgr.h
+++ b/intel/intel_bufmgr.h
@@ -194,6 +194,7 @@ drm_intel_context *drm_intel_gem_context_create(drm_intel_bufmgr *bufmgr);
 void drm_intel_gem_context_destroy(drm_intel_context *ctx);
 int drm_intel_gem_bo_context_exec(drm_intel_bo *bo, drm_intel_context *ctx,
 				  int used, unsigned int flags);
+unsigned int drm_intel_gem_context_get_hw_context_id(const drm_intel_context *);
 
 int drm_intel_bo_gem_export_to_prime(drm_intel_bo *bo, int *prime_fd);
 drm_intel_bo *drm_intel_bo_gem_create_from_prime(drm_intel_bufmgr *bufmgr,
diff --git a/intel/intel_bufmgr_gem.c b/intel/intel_bufmgr_gem.c
index 029ca5d..5b64a7f 100644
--- a/intel/intel_bufmgr_gem.c
+++ b/intel/intel_bufmgr_gem.c
@@ -3020,6 +3020,12 @@ drm_intel_gem_context_destroy(drm_intel_context *ctx)
 	free(ctx);
 }
 
+unsigned int
+drm_intel_gem_context_get_hw_context_id(const drm_intel_context *ctx)
+{
+	return ctx->ctx_id;
+}
+
 int
 drm_intel_reg_read(drm_intel_bufmgr *bufmgr,
 		   uint32_t offset,
-- 
1.8.1.4

[PATCH 2/2] intel: Silence warning in non-Valgrind build

[Ian Romanick]

From: Ian Romanick <ian.d.romanick@intel.com>

Previously GCC was squaking about:

intel_bufmgr_gem.c: In function 'drm_intel_gem_bo_map_unsynchronized':
intel_bufmgr_gem.c:1325:20: warning: unused variable 'bo_gem' [-Wunused-variable]

Wrapping with VG(); replaced that warning with:

intel_bufmgr_gem.c: In function 'drm_intel_gem_bo_map_unsynchronized':
intel_bufmgr_gem.c:1326:2: warning: ISO C90 forbids mixed declarations and code [-Wdeclaration-after-statement]

Which, looking at the definition of the VG macro, seems a bit weird.
It's caused by the dangling ";" left from the empty macro.

Signed-off-by: Ian Romanick <ian.d.romanick@intel.com>
Cc: Chia-I Wu <olvaffe@gmail.com>
---
 intel/intel_bufmgr_gem.c | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/intel/intel_bufmgr_gem.c b/intel/intel_bufmgr_gem.c
index 5b64a7f..dbadc52 100644
--- a/intel/intel_bufmgr_gem.c
+++ b/intel/intel_bufmgr_gem.c
@@ -1322,7 +1322,9 @@ int drm_intel_gem_bo_map_gtt(drm_intel_bo *bo)
 int drm_intel_gem_bo_map_unsynchronized(drm_intel_bo *bo)
 {
 	drm_intel_bufmgr_gem *bufmgr_gem = (drm_intel_bufmgr_gem *) bo->bufmgr;
+#ifdef HAVE_VALGRIND
 	drm_intel_bo_gem *bo_gem = (drm_intel_bo_gem *) bo;
+#endif
 	int ret;
 
 	/* If the CPU cache isn't coherent with the GTT, then use a
-- 
1.8.1.4

Re: [PATCH 1/2] intel: Add accessor to get HW context ID from a drm_intel_context

[Eric Anholt]

[-- Attachment #1.1: Type: text/plain, Size: 563 bytes --]

Ian Romanick <idr@freedesktop.org> writes:

> From: Ian Romanick <ian.d.romanick@intel.com>
>
> The drm_intel_context structure is, wisely, opaque.  However, libdrm
> users may want to know the hardware context ID associated with the
> structure.

We've had a bunch of our other structures be partially transparent.  The
context id to be passed to the kernel could easily be public just like
the gem handle in a BO is public.  I would lean slightly toward that.

But I don't feel strongly either way, so these two are:

Reviewed-by: Eric Anholt <eric@anholt.net>

[-- Attachment #1.2: Type: application/pgp-signature, Size: 835 bytes --]

[-- Attachment #2: Type: text/plain, Size: 159 bytes --]

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

Re: [PATCH 1/2] intel: Add accessor to get HW context ID from a drm_intel_context

[Ben Widawsky]

On Mon, Nov 11, 2013 at 11:58:59AM -0800, Eric Anholt wrote:
> Ian Romanick <idr@freedesktop.org> writes:
> 
> > From: Ian Romanick <ian.d.romanick@intel.com>
> >
> > The drm_intel_context structure is, wisely, opaque.  However, libdrm
> > users may want to know the hardware context ID associated with the
> > structure.
> 
> We've had a bunch of our other structures be partially transparent.  The
> context id to be passed to the kernel could easily be public just like
> the gem handle in a BO is public.  I would lean slightly toward that.
> 
> But I don't feel strongly either way, so these two are:
> 
> Reviewed-by: Eric Anholt <eric@anholt.net>

I think my preference would be to add a "context" argument to a libdrm
get_hangstats function, but if you feel this way is better, it is fine
with me.

My only [slight, unjustified] concern is that once you make the id
transparent, we can't play any games. Since the DDX doesn't use libdrm
however, I think it is fairly moot.

Acked-by: Ben Widawsky <ben@bwidawsk.net>

-- 
Ben Widawsky, Intel Open Source Technology Center