[Mesa-dev] [PATCH 0/5] Enable up to 24 MRF registers in gen6

[Iago Toral Quiroga]

There are some bug reports about shaders failing to compile in gen6
because MRF 14 is used when we need to spill. For example:
https://bugs.freedesktop.org/show_bug.cgi?id=86469
https://bugs.freedesktop.org/show_bug.cgi?id=90631

Discussion in bugzilla pointed to the fact that gen6 might actually have
24 MRF registers available instead of 16, so we could use other MRF
registers and avoid these conflicts (we still need to investigate why
some shaders need up to MRF 14 anyway, since this is not expected).

Notice that the hardware docs are not clear about this fact:

SNB PRM Vol4 Part2's "Table 5-4. MRF Registers Available in Device
Hardware" says "Number per Thread" - "24 registers"

However, SNB PRM Vol4 Part1, 1.6.1 Message Register File (MRF) says:

"Normal threads should construct their messages in m1..m15. (...)
Regardless of actual hardware implementation, the thread should
not assume th at MRF addresses above m15 wrap to legal MRF registers."

Therefore experimentation was necessary to evaluate if we had these extra
MRF registers available or not. This was tested in gen6 using MRF
registers 21..23 for spilling and doing a full piglit run (all.py) forcing
spilling of everything on the FS backend. It was also tested by doing
spilling of everything on both the FS and the VS backends with a piglit run
of shader.py. In both cases no regressions were observed. In fact, many of
these tests where helped in the cases where we forced spilling, since that
triggered the same underlying problem described in the bug reports. Here are
some results using INTEL_DEBUG=spill_fs,spill_vec4 for a shader.py run on
gen6 hardware:

Using MRFs 13..15 for spilling:
crash: 2, fail: 113, pass: 6621, skip: 5461
    
Using MRFs 21..23 for spilling:
crash: 2, fail: 12, pass: 6722, skip: 5461

We might want to test this further with other instances of gen6 hardware
though... I am not sure that we can safely conclude that all implementations
of gen6 hardware have 24 MRF registers from my tests on just one particular
SandyBridge laptop.

Iago Toral Quiroga (5):
  i965: Move MRF register asserts out of brw_reg.h
  i965: Turn BRW_MAX_MRF into a macro that accepts a hardware generation
  i965/fs: Use MRF registers 21-23 for spilling in gen6
  i965/vec4: Use MRF registers 21-23 for spilling in gen6
  i965: Maximum allowed size of SEND messages is 15 (4 bits)

 src/mesa/drivers/dri/i965/brw_eu_emit.c            | 11 +++++----
 src/mesa/drivers/dri/i965/brw_fs.cpp               |  4 ++--
 src/mesa/drivers/dri/i965/brw_fs_generator.cpp     | 15 ++++++++----
 src/mesa/drivers/dri/i965/brw_fs_reg_allocate.cpp  | 27 ++++++++++++----------
 src/mesa/drivers/dri/i965/brw_inst.h               |  3 +++
 src/mesa/drivers/dri/i965/brw_ir_vec4.h            |  2 +-
 src/mesa/drivers/dri/i965/brw_reg.h                |  9 ++++----
 .../drivers/dri/i965/brw_schedule_instructions.cpp |  4 ++--
 src/mesa/drivers/dri/i965/brw_vec4_generator.cpp   | 10 +++++---
 src/mesa/drivers/dri/i965/brw_vec4_visitor.cpp     | 15 +++++++-----
 10 files changed, 61 insertions(+), 39 deletions(-)

-- 
1.9.1