On 10/30/2012 08:24 AM, Anuj Phogat wrote:
On Tue, Oct 30, 2012 at 7:22 AM, Brian Paul <bri...@vmware.com> wrote:
On 10/29/2012 07:37 PM, Ian Romanick wrote:
On 10/25/2012 03:17 PM, Brian Paul wrote:
On 10/24/2012 02:22 PM, Anuj Phogat wrote:
Signed-off-by: Anuj Phogat<anuj.pho...@gmail.com>
---
src/mesa/main/format_unpack.c | 9 +++++++++
src/mesa/main/formats.c | 13 +++++++++++++
src/mesa/main/formats.h | 1 +
src/mesa/main/glformats.c | 2 ++
src/mesa/main/texcompress.c | 9 +++++++++
src/mesa/main/texformat.c | 3 +++
src/mesa/main/teximage.c | 1 +
src/mesa/main/texstore.c | 2 ++
8 files changed, 40 insertions(+), 0 deletions(-)
diff --git a/src/mesa/main/format_unpack.c
b/src/mesa/main/format_unpack.c
index 1a8f113..d31d2fa 100644
--- a/src/mesa/main/format_unpack.c
+++ b/src/mesa/main/format_unpack.c
@@ -1392,6 +1392,13 @@ unpack_ETC2_RGB8_PUNCHTHROUGH_ALPHA1(const void
*src, GLfloat dst[][4],
}
static void
+unpack_ETC2_SRGB8_PUNCHTHROUGH_ALPHA1(const void *src, GLfloat
dst[][4],
+ GLuint n)
+{
+ /* XXX to do */
+}
+
+static void
unpack_SIGNED_A8(const void *src, GLfloat dst[][4], GLuint n)
{
const GLbyte *s = ((const GLbyte *) src);
@@ -1650,6 +1657,8 @@ get_unpack_rgba_function(gl_format format)
table[MESA_FORMAT_ETC2_SIGNED_RG11_EAC] =
unpack_ETC2_SIGNED_RG11_EAC;
table[MESA_FORMAT_ETC2_RGB8_PUNCHTHROUGH_ALPHA1] =
unpack_ETC2_RGB8_PUNCHTHROUGH_ALPHA1;
+ table[MESA_FORMAT_ETC2_RGB8_PUNCHTHROUGH_ALPHA1] =
+ unpack_ETC2_SRGB8_PUNCHTHROUGH_ALPHA1;
table[MESA_FORMAT_SIGNED_A8] = unpack_SIGNED_A8;
table[MESA_FORMAT_SIGNED_L8] = unpack_SIGNED_L8;
table[MESA_FORMAT_SIGNED_AL88] = unpack_SIGNED_AL88;
diff --git a/src/mesa/main/formats.c b/src/mesa/main/formats.c
index cba51ee..8aa4f28 100644
--- a/src/mesa/main/formats.c
+++ b/src/mesa/main/formats.c
@@ -1488,6 +1488,16 @@ static struct gl_format_info
format_info[MESA_FORMAT_COUNT] =
4, 4, 8 /* 8 bytes per 4x4 block */
},
+ {
+ MESA_FORMAT_ETC2_SRGB8_PUNCHTHROUGH_ALPHA1,
+ "MESA_FORMAT_ETC2_SRGB8_PUNCHTHROUGH_ALPHA1",
+ GL_RGBA,
+ GL_UNSIGNED_NORMALIZED,
+ 8, 8, 8, 1,
+ 0, 0, 0, 0, 0,
+ 4, 4, 8 /* 8 bytes per 4x4 block */
+ },
+
Anytime we add a new format to the format_info[] array we have to
add/update the corresponding entry in texfetch_funcs[] array in
swrast/s_texfetch.c
There's a runtime assertion that'll detect missing entries, but only if
you run swrast.
I misspoke. There's a compile-time static assertion. Perhaps you're not
compiling swrast, Anuj.
-Brian
I was compiling swrast and i965 drivers. Didn't get any compile time assertion.
I'm not sure if i have the required code for compile time assertion in my etc2
branch. I took my branch at commit 299acac. Issues with etc2 decoding on
swrast are now fixed on my etc2-v12.1 branch. Found few bugs in my
*fetch_texel functions during this process.
Ian, if we have a run time assertion, do we still need unit tests to be run by
'make check'?
This sounds like it's actually a (broken) compile-time assertion, so it
should be fine without 'make check'. If there's a run-time assertion
that only occurs in some drivers, we should make it into a 'make check'
test.
_______________________________________________
mesa-dev mailing list
mesa-dev@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/mesa-dev
