[PATCH 3/3] Force resources to be RES_SHARED_VGA

[Tiago Vignatti]

Yes, it is a hack. Some drivers aren't reporting correctly the resources 
being used in post-pciaccess Xorg. Until we don't build a nice scheme to 
arbitrate the resources (and given that I'm lazy to fix all drivers), 
lets assume that _all_ resources are RES_SHARED_VGA.

But relax, it has no side effects in one-card machines.

Signed-off-by: Tiago Vignatti <vignatti at freedesktop.org>
---
  hw/xfree86/common/xf86Helper.c |   11 +++++++++++
  1 files changed, 11 insertions(+), 0 deletions(-)

diff --git a/hw/xfree86/common/xf86Helper.c b/hw/xfree86/common/xf86Helper.c
index 0db7717..6f17eb8 100644
--- a/hw/xfree86/common/xf86Helper.c
+++ b/hw/xfree86/common/xf86Helper.c
@@ -61,6 +61,7 @@
  #include "xf86RAC.h"
  #include "xf86Bus.h"
  #include "xf86Crtc.h"
+#include "xf86Resources.h"

  /* For xf86GetClocks */
  #if defined(CSRG_BASED) || defined(__GNU__)
@@ -2482,12 +2483,22 @@ xf86ConfigPciEntity(ScrnInfoPtr pScrn, int 
scrnFlag, int entityIndex,
      if (xf86IsEntityShared(entityIndex)) {
          return pScrn;
      }
+#if 0
      if (p_chip) {
  	for (p_id = p_chip; p_id->numChipset != -1; p_id++) {
  	    if (pEnt->chipset == p_id->numChipset) break;
  	}
  	xf86ClaimFixedResources(p_id->resList,entityIndex);
      }
+#endif
+    /* XXX: Some drivers aren't reporting correctly the resources being 
used
+     * in post-pciaccess Xorg. Until we don't build a nice scheme to 
arbitrate
+     * the resources (and given that I'm lazy to fix all drivers), lets 
assume
+     * that _all_ resources are RES_SHARED_VGA -- relax, it has no side 
effect
+     * with one-card machines.
+     */
+    xf86ClaimFixedResources((resRange*)RES_SHARED_VGA, entityIndex);
+
      xfree(pEnt);

      xf86ClaimFixedResources(res,entityIndex);
-- 
1.5.4.3