Displaying 4 results from an estimated 4 matches for "gf100_ram_train".
2017 Apr 10
0
[PATCH 02/11] nvkm/ramgf100: Calculate timings
...;
+ timing[4] = (cur4 & ~0x001f8000) |
+ (T(RRD) << 15);
+
+ nvkm_debug(subdev, "Entry: 290: %08x %08x %08x %08x\n",
+ timing[0], timing[1], timing[2], timing[3]);
+ nvkm_debug(subdev, " 2a0: %08x\n",
+ timing[4]);
+ return 0;
+}
+#undef T
+
static void
gf100_ram_train(struct gf100_ramfuc *fuc, u32 magic)
{
@@ -136,10 +176,11 @@ gf100_ram_calc(struct nvkm_ram *base, u32 freq)
struct nvkm_ram_data *next;
u8 ver, hdr, cnt, len, strap;
u32 data;
+ u32 timing[5];
int ref, div, out;
int from, mode;
int N1, M1, P;
- int ret;
+ int i, ret;
next = &...
2017 Apr 10
11
Preparations for Fermi DRAM clock changes
No, no, these will not implement Fermi reclocking. This set of patches
contains some of the preparatory work that I deem stable enough to
move upstream. Notable changes
- Training pattern upload routines from GK104+ now shared with GT215+
- Timing calculation for Fermi
- GDDR5 MR calculation from VBIOS timing table v1.0. Also useful for that
pesky GT 240.
- A routine to translate a VBIOS init
2017 Apr 10
0
[PATCH 03/11] nvkm/gddr5: MR calculation for timing table v1.0
...ram_wr32(fuc, 0x10f610, 0x40044f77);
- ram_wr32(fuc, 0x10f344, 0x00600009);
+ ram_wr32(fuc, mr[6], 0x00600009);
ram_nsec(fuc, 1000);
- ram_wr32(fuc, 0x10f348, 0x00700008);
+ ram_wr32(fuc, mr[7], 0x00700008);
ram_wr32(fuc, 0x61c140, 0x19240000);
ram_wr32(fuc, 0x10f830, 0x00300017);
gf100_ram_train(fuc, 0x80021001);
gf100_ram_train(fuc, 0x80081001);
- ram_wr32(fuc, 0x10f340, 0x00500004);
+ ram_wr32(fuc, mr[5], 0x00500004);
ram_nsec(fuc, 1000);
ram_wr32(fuc, 0x10f830, 0x01300017);
ram_wr32(fuc, 0x10f830, 0x00300017);
@@ -412,11 +419,11 @@ gf100_ram_calc(struct nvkm_ram *base, u32...
2017 Apr 10
14
RESEND Preparations for Fermi DRAM clock changes
Two patches went missing as a result of PEBCAK. No v2 marks as nothing
changed really. Just resending for easier enforcement of patch order
in other people's trees. Sorry for the noise.
Original message:
No, no, these will not implement Fermi reclocking. This set of patches
contains some of the preparatory work that I deem stable enough to
move upstream. Notable changes
- Training pattern