Results 1 to 6 of 6

Thread: my 3D status & questions

  1. #1
    Join Date
    Nov 2008
    Posts
    57

    Default my 3D status & questions

    system info: x86_64, gcc 4.4.2 git 20090805, kernel 2.6.30-zen3.5, git libdrm-20090820 (r600: emit timestamp AFTER the ib), git mesa-20090820 (st/egl: Open the first drm node), git ati-20090820 (rs600: add support for DisplayPriority HIGH).

    When compiling linux-core *.ko modules these warnings display but the modules still build successfully:

    Code:
    /tmp/gitdrm/drm/linux-core/drm_bufs.c: In function 'drm_rmmap_locked':                         
    /tmp/gitdrm/drm/linux-core/drm_bufs.c:402: warning: enumeration value '_DRM_GEM' not handled in switch
    /tmp/gitdrm/drm/linux-core/drm_proc.c: In function 'drm_gem_one_name_info':                           
    /tmp/gitdrm/drm/linux-core/drm_proc.c:602: warning: format '%d' expects type 'int', but argument 3 has type 'size_t'                                                                                            
    /tmp/gitdrm/drm/linux-core/drm_proc.c:610: warning: format '%9d' expects type 'int', but argument 4 has type 'size_t'                                                                                           
    /tmp/gitdrm/drm/linux-core/drm_proc.c: In function 'drm__vma_info':                                     
    /tmp/gitdrm/drm/linux-core/drm_proc.c:691: warning: format '%08lx' expects type 'long unsigned int', but argument 5 has type 'phys_addr_t'                                                                      
    /tmp/gitdrm/drm/linux-core/drm_vm.c: In function 'drm_vm_shm_close':                                    
    /tmp/gitdrm/drm/linux-core/drm_vm.c:261: warning: enumeration value '_DRM_GEM' not handled in switch    
    /tmp/gitdrm/drm/linux-core/ati_pcigart.c: In function 'drm_ati_pcigart_init':
    /tmp/gitdrm/drm/linux-core/ati_pcigart.c:154: warning: format '%08X' expects type 'unsigned int', but argument 3 has type 'dma_addr_t'
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c: In function 'r600_check_config_reg':
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c:175: warning: format '%016lx' expects type 'long unsigned int', but argument 4 has type 'u64'
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c: In function 'r600_check_context_reg':
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c:220: warning: format '%016lx' expects type 'long unsigned int', but argument 4 has type 'u64'
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c: In function 'r600_cs_packet3':
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c:442: warning: format '%016lx' expects type 'long unsigned int', but argument 3 has type 'u64'
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c:460: warning: format '%016lx' expects type 'long unsigned int', but argument 3 has type 'u64'
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c:479: warning: format '%016lx' expects type 'long unsigned int', but argument 3 has type 'u64'
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c: In function 'r600_do_cp_cmdbuf':
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c:660: warning: unused variable 'buf_size'
    /tmp/gitdrm/drm/linux-core/r600_cmdbuf.c:659: warning: unused variable 'offset'
    /tmp/gitdrm/drm/linux-core/radeon_cs.c: In function 'r600_nomm_relocate':
    /tmp/gitdrm/drm/linux-core/radeon_cs.c:308: warning: format '%lx' expects type 'long unsigned int', but argument 3 has type 'uint64_t'
    __________________
    When configuring git mesa/mesa a line displays, "checking for LIBDRM_RADEON... no". Am I missing something?

    Also Mesa will not successfully compile without the configure switch, "--disable-gallium". Is gallium needed?
    __________________
    When configuring git ati there is also a line that says, "checking for LIBDRM_RADEON... no".
    __________________
    In this machine's Xorg log there is a line that says, "AIGLX: Screen 0 is not DRI2 capable". Is this to be expected for now? There are no other error or warning messages related to video.
    __________________
    Opengl accerlation does appear to be partially working in KDE4.
    -----
    Thanks for the replies.
    Last edited by charlie; 08-20-2009 at 10:21 PM.

  2. #2
    Join Date
    Dec 2007
    Posts
    2,329

    Default

    Quote Originally Posted by charlie View Post
    When configuring git mesa/mesa a line displays, "checking for LIBDRM_RADEON... no". Am I missing something?
    No. libdrm_radeon is only available in libdrm git master and is only needed for kms.

    Quote Originally Posted by charlie View Post
    Also Mesa will not successfully compile without the configure switch, "--disable-gallium". Is gallium needed?
    No.

    Quote Originally Posted by charlie View Post
    When configuring git ati there is also a line that says, "checking for LIBDRM_RADEON... no".
    Only needed for kms. see above.

    Quote Originally Posted by charlie View Post
    In this machine's Xorg log there is a line that says, "AIGLX: Screen 0 is not DRI2 capable". Is this to be expected for now? There are no other error or warning messages related to video.
    DRI2 only works with radeon kms.

  3. #3
    Join Date
    Oct 2007
    Location
    Toronto-ish
    Posts
    7,386

    Default

    One last point -- 6xx/7xx 3D is not integrated with KMS yet, so you need to either run 6xx/7xx 3D *or* KMS, not both.

  4. #4
    Join Date
    Oct 2007
    Posts
    1,259

    Default

    Quick question (and I'm sorry if it was covered before in the larger threads): Is EGL needed when building mesa for r600 3D, or can we just configure with --disable-egl ?

  5. #5
    Join Date
    Dec 2007
    Posts
    2,329

    Default

    Quote Originally Posted by DanL View Post
    Quick question (and I'm sorry if it was covered before in the larger threads): Is EGL needed when building mesa for r600 3D, or can we just configure with --disable-egl ?
    egl isn't needed. you can disable it.

  6. #6
    Join Date
    Aug 2008
    Location
    Finland
    Posts
    1,577

    Default

    Those format warnings look like something that might be a bit tiresome to fix in a platform-independent way. (I'd assume quite a few of them happen due to casting to a smaller type of an integer than the original value was)
    Maybe better just to keep thumbs up for those being fixed in current Linus' tree and leave them alone. :3
    Same for the unused variables. Probably stuff that makes sense to be fixed only when migrating to Linus' tree.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •