[Mesa-dev] [PATCH] configure: nuke --with-max-{width,height}
Brian Paul
brianp at vmware.com
Wed Apr 1 12:28:34 PDT 2015
On 04/01/2015 12:21 PM, Emil Velikov wrote:
> On 01/04/15 15:07, Brian Paul wrote:
>> On 04/01/2015 08:51 AM, Emil Velikov wrote:
>>> Unused as of commit 630ab0d27ba(mesa: remove last of MAX_WIDTH,
>>> MAX_HEIGHT). Update all the remaining references to the defines.
>>>
>>> XXX: Does the updated comments in xlib-libGL make sense, or should one
>>> consider SWRAST_MAX_WIDTH ?
>>>
>>> Cc: Brian Paul <brianp at vmware.com>
>>> Signed-off-by: Emil Velikov <emil.l.velikov at gmail.com>
>>> ---
>>> configure.ac | 15 ---------------
>>> include/GL/osmesa.h | 6 ++----
>>> src/gallium/state_trackers/glx/xlib/glx_api.c | 2 +-
>>> src/mesa/drivers/x11/fakeglx.c | 2 +-
>>> 4 files changed, 4 insertions(+), 21 deletions(-)
>>>
>>> diff --git a/configure.ac b/configure.ac
>>> index 3e5b6f5..beb9c31 100644
>>> --- a/configure.ac
>>> +++ b/configure.ac
>>> @@ -1823,21 +1823,6 @@ if ! echo "$egl_platforms" | grep -q 'x11'; then
>>> GL_PC_CFLAGS="$GL_PC_CFLAGS -DMESA_EGL_NO_X11_HEADERS"
>>> fi
>>>
>>> -AC_ARG_WITH([max-width],
>>> - [AS_HELP_STRING([--with-max-width=N],
>>> - [Maximum framebuffer width (4096)])],
>>> - [DEFINES="${DEFINES} -DMAX_WIDTH=${withval}";
>>> - AS_IF([test "${withval}" -gt "4096"],
>>> - [AC_MSG_WARN([Large framebuffer: see s_tritemp.h
>>> comments.])])]
>>> -)
>>> -AC_ARG_WITH([max-height],
>>> - [AS_HELP_STRING([--with-max-height=N],
>>> - [Maximum framebuffer height (4096)])],
>>> - [DEFINES="${DEFINES} -DMAX_HEIGHT=${withval}";
>>> - AS_IF([test "${withval}" -gt "4096"],
>>> - [AC_MSG_WARN([Large framebuffer: see s_tritemp.h
>>> comments.])])]
>>> -)
>>> -
>>> dnl
>>> dnl Gallium LLVM
>>> dnl
>>> diff --git a/include/GL/osmesa.h b/include/GL/osmesa.h
>>> index 16ee89a..ca0d167 100644
>>> --- a/include/GL/osmesa.h
>>> +++ b/include/GL/osmesa.h
>>> @@ -41,10 +41,8 @@
>>> * OSMesaGetIntegerv - return OSMesa state parameters
>>> *
>>> *
>>> - * The limits on the width and height of an image buffer are
>>> MAX_WIDTH and
>>> - * MAX_HEIGHT as defined in Mesa/src/config.h. Defaults are 1280 and
>>> 1024.
>>> - * You can increase them as needed but beware that many temporary
>>> arrays in
>>> - * Mesa are dimensioned by MAX_WIDTH or MAX_HEIGHT.
>>> + * The limits on the width and height of an image buffer can be
>>> retrieved
>>> + * via OSMesaGetIntegerv(OSMESA_MAX_WIDTH/OSMESA_MAX_HEIGHT).
>>> */
>>>
>>>
>>> diff --git a/src/gallium/state_trackers/glx/xlib/glx_api.c
>>> b/src/gallium/state_trackers/glx/xlib/glx_api.c
>>> index f9572b7..62c7f0d 100644
>>> --- a/src/gallium/state_trackers/glx/xlib/glx_api.c
>>> +++ b/src/gallium/state_trackers/glx/xlib/glx_api.c
>>> @@ -1636,7 +1636,7 @@ get_config( XMesaVisual xmvis, int attrib, int
>>> *value, GLboolean fbconfig )
>>> case GLX_MAX_PBUFFER_WIDTH:
>>> if (!fbconfig)
>>> return GLX_BAD_ATTRIBUTE;
>>> - /* XXX or MAX_WIDTH? */
>>> + /* XXX or MAX_VIEWPORT_WIDTH? */
>>> *value = DisplayWidth(xmvis->display, xmvis->visinfo->screen);
>>> break;
>>> case GLX_MAX_PBUFFER_HEIGHT:
>>> diff --git a/src/mesa/drivers/x11/fakeglx.c
>>> b/src/mesa/drivers/x11/fakeglx.c
>>> index 4fd6d75..10b148e 100644
>>> --- a/src/mesa/drivers/x11/fakeglx.c
>>> +++ b/src/mesa/drivers/x11/fakeglx.c
>>> @@ -1720,7 +1720,7 @@ get_config( XMesaVisual xmvis, int attrib, int
>>> *value, GLboolean fbconfig )
>>> case GLX_MAX_PBUFFER_WIDTH:
>>> if (!fbconfig)
>>> return GLX_BAD_ATTRIBUTE;
>>> - /* XXX or MAX_WIDTH? */
>>> + /* XXX or MAX_VIEWPORT_WIDTH? */
>>> *value = DisplayWidth(xmvis->display, xmvis->visinfo->screen);
>>> break;
>>> case GLX_MAX_PBUFFER_HEIGHT:
>>>
>>
>> It should probably be ctx->Const.MaxRenderbufferSize but we don't have
>> access to context data in those functions.
>>
>> Maybe /* XXX should be same as ctx->Const.MaxRenderbufferSize */
>>
>> In any case, the patch looks good.
>>
>> Reviewed-by: Brian Paul <brianp at vmware.com>
>>
> Seems like another reason for moving the constants out of gl_context,
> into the dri device. But for the moment I'll just update the comments
> and commit it.
Yeah, that's all I had in mind for now. The code has been as-is for a
very long time and nobody has complained.
-Brian
> Thanks for the quick review Brian.
>
> Emil
>
> P.S. With this and other libclc-path patch, mesa's configure switches go
> sub 50. One step at a time and we'll get to a reasonable number :-)
>
More information about the mesa-dev
mailing list