drm/radeon/kms/blit: fix blit copy for very large buffers

Evergreen and NI blit copy was broken if the buffer maps to a rectangle
whose one dimension is 16384 (max dimension allowed by these chips).
In the mainline kernel, the problem is exposed only when buffers are
very large (1G), but it's still a problem. The problem could be exposed
for smaller buffers if anyone modifies the algorithm for rectangle
construction in r600_blit_create_rect() (the reason why someone would
modify that algorithm is to tune the performance of buffer moves).

The root cause was in i2f() function which only operated on range between
0 and 16383. Fix this by extending the range of i2f() function to 0 to
32767.

While at it improve the function so that the range can be easily
extended in the future (if it becomes necessary), cleanup lines
over 80 characters, and replace in-line comments with one strategic
comment that explains the crux of the function.

Credits to michel@daenzer.net for pointing out the root cause of
the bug.

v2: Fix I2F_MAX_INPUT constant definition goof and warn only once
    if input argument is out of range. Edit the comment a little
    bit to avoid some linguistic confusion and make it look better
    in general.

Signed-off-by: Ilija Hadzic <ihadzic@research.bell-labs.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Reviewed-by: Michel Dänzer <michel@daenzer.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>

diff --git a/drivers/gpu/drm/radeon/r600_blit_kms.c b/drivers/gpu/drm/radeon/r600_blit_kms.c
index d996f43811302fedd6f6a48169d1858b61d11df2..accc032c103fd3f846f2b57e6da4de8c693c5d1e 100644 (file)
--- a/drivers/gpu/drm/radeon/r600_blit_kms.c
+++ b/drivers/gpu/drm/radeon/r600_blit_kms.c
@@ -468,27 +468,42 @@ set_default_state(struct radeon_device *rdev)
        radeon_ring_write(ring, sq_stack_resource_mgmt_2);
 }
 
+#define I2F_MAX_BITS 15
+#define I2F_MAX_INPUT  ((1 << I2F_MAX_BITS) - 1)
+#define I2F_SHIFT (24 - I2F_MAX_BITS)
+
+/*
+ * Converts unsigned integer into 32-bit IEEE floating point representation.
+ * Conversion is not universal and only works for the range from 0
+ * to 2^I2F_MAX_BITS-1. Currently we only use it with inputs between
+ * 0 and 16384 (inclusive), so I2F_MAX_BITS=15 is enough. If necessary,
+ * I2F_MAX_BITS can be increased, but that will add to the loop iterations
+ * and slow us down. Conversion is done by shifting the input and counting
+ * down until the first 1 reaches bit position 23. The resulting counter
+ * and the shifted input are, respectively, the exponent and the fraction.
+ * The sign is always zero.
+ */
 static uint32_t i2f(uint32_t input)
 {
        u32 result, i, exponent, fraction;
 
-       if ((input & 0x3fff) == 0)
-               result = 0; /* 0 is a special case */
+       WARN_ON_ONCE(input > I2F_MAX_INPUT);
+
+       if ((input & I2F_MAX_INPUT) == 0)
+               result = 0;
        else {
-               exponent = 140; /* exponent biased by 127; */
-               fraction = (input & 0x3fff) << 10; /* cheat and only
-                                                     handle numbers below 2^^15 */
-               for (i = 0; i < 14; i++) {
+               exponent = 126 + I2F_MAX_BITS;
+               fraction = (input & I2F_MAX_INPUT) << I2F_SHIFT;
+
+               for (i = 0; i < I2F_MAX_BITS; i++) {
                        if (fraction & 0x800000)
                                break;
                        else {
-                               fraction = fraction << 1; /* keep
-                                                            shifting left until top bit = 1 */
+                               fraction = fraction << 1;
                                exponent = exponent - 1;
                        }
                }
-               result = exponent << 23 | (fraction & 0x7fffff); /* mask
-                                                                   off top bit; assumed 1 */
+               result = exponent << 23 | (fraction & 0x7fffff);
        }
        return result;
 }