radeon: Make cache flushing for R300 series less invasive.

Instead of calling the DRM CP idle ioctl, just emit the cache flush commands
into the CP stream.

3 files changed, 18 insertions, 43 deletions

diff --git a/src/radeon.h b/src/radeon.h
index d13beeb..4c3cbcc 100644
--- a/src/radeon.h
+++ b/src/radeon.h
@@ -621,6 +621,7 @@ typedef struct {
     int               CPMode;           /* CP mode that server/clients use */
     int               CPFifoSize;       /* Size of the CP command FIFO */
     int               CPusecTimeout;    /* CP timeout in usecs */
+    Bool              needCacheFlush;
 
 				/* CP ring buffer data */
     unsigned long     ringStart;        /* Offset into GART space */
@@ -986,6 +987,11 @@ do {									\
 #define RADEONCP_REFRESH(pScrn, info)					\
 do {									\
     if (!info->CPInUse) {						\
+	if (info->needCacheFlush) {					\
+	    RADEON_PURGE_CACHE();					\
+	    RADEON_PURGE_ZCACHE();					\
+	    info->needCacheFlush = FALSE;				\
+	}								\
 	RADEON_WAIT_UNTIL_IDLE();					\
 	BEGIN_RING(6);							\
 	OUT_RING_REG(RADEON_RE_TOP_LEFT,     info->re_top_left);	\
@@ -1132,6 +1138,12 @@ do {									\
     ADVANCE_RING();							\
 } while (0)
 
+#define RADEON_PURGE_ZCACHE()						\
+do {									\
+    OUT_RING(CP_PACKET0(RADEON_RB3D_ZCACHE_CTLSTAT, 0));		\
+    OUT_RING(RADEON_RB3D_ZC_FLUSH_ALL);					\
+} while (0)
+
 #endif /* XF86DRI */
 
 static __inline__ void RADEON_MARK_SYNC(RADEONInfoPtr info, ScrnInfoPtr pScrn)
diff --git a/src/radeon_dri.c b/src/radeon_dri.c
index 70ff288..e223825 100644
--- a/src/radeon_dri.c
+++ b/src/radeon_dri.c
@@ -357,50 +357,10 @@ static void RADEONEnterServer(ScreenPtr pScreen)
     RADEON_MARK_SYNC(info, pScrn);
 
     pSAREAPriv = DRIGetSAREAPrivate(pScrn->pScreen);
-    if (pSAREAPriv->ctxOwner != DRIGetContext(pScrn->pScreen))
+    if (pSAREAPriv->ctxOwner != DRIGetContext(pScrn->pScreen)) {
 	info->XInited3D = FALSE;
-
-
-    /* TODO: Fix this more elegantly.
-     * Sometimes (especially with multiple DRI clients), this code
-     * runs immediately after a DRI client issues a rendering command.
-     *
-     * The accel code regularly inserts WAIT_UNTIL_IDLE into the
-     * command buffer that is sent with the indirect buffer below.
-     * The accel code fails to set the 3D cache flush registers for
-     * the R300 before sending WAIT_UNTIL_IDLE. Sending a cache flush
-     * on these new registers is not necessary for pure 2D functionality,
-     * but it *is* necessary after 3D operations.
-     * Without the cache flushes before WAIT_UNTIL_IDLE, the R300 locks up.
-     *
-     * The CP_IDLE call into the DRM indirectly flushes all caches and
-     * thus avoids the lockup problem, but the solution is far from ideal.
-     * Better solutions could be:
-     *  - always flush caches when entering the X server
-     *  - track the type of rendering commands somewhere and issue
-     *    cache flushes when they change
-     * However, I don't feel confident enough with the control flow
-     * inside the X server to implement either fix. -- nh
-     */
-    
-    /* On my computer (Radeon Mobility M10)
-       The fix below results in x11perf -shmput500 rate of 245.0/sec
-       which is lower than 264.0/sec I get without it.
-       
-       Doing the same each time before indirect buffer is submitted
-       results in x11perf -shmput500 rate of 225.0/sec.
-       
-       On the other hand, not using CP acceleration at all benchmarks
-       at 144.0/sec.
-      
-       For now let us accept this as a lesser evil, especially as the
-       DRM driver for R300 is still in flux.
-       
-       Once the code is more stable this should probably be moved into DRM driver.
-    */ 
-     
-    if (info->ChipFamily>=CHIP_FAMILY_R300)
-        drmCommandNone(info->drmFD, DRM_RADEON_CP_IDLE);
+	info->needCacheFlush = (info->ChipFamily >= CHIP_FAMILY_R300);
+    }
 
 #ifdef DAMAGE
     if (!info->pDamage && info->allowPageFlip) {
diff --git a/src/radeon_reg.h b/src/radeon_reg.h
index 9503dde..a6bbe69 100644
--- a/src/radeon_reg.h
+++ b/src/radeon_reg.h
@@ -1280,6 +1280,9 @@
 #       define RADEON_RB2D_DC_BUSY          (1 << 31)
 #define RADEON_RB2D_DSTCACHE_MODE           0x3428
 
+#define RADEON_RB3D_ZCACHE_MODE             0x3250
+#define RADEON_RB3D_ZCACHE_CTLSTAT          0x3254
+#       define RADEON_RB3D_ZC_FLUSH_ALL     0x5
 #define RADEON_RB3D_DSTCACHE_MODE           0x3258
 # define RADEON_RB3D_DC_CACHE_ENABLE            (0)
 # define RADEON_RB3D_DC_2D_CACHE_DISABLE        (1)