From 62341f7bc2ccb7c024eb023a05714d798ba6437d Mon Sep 17 00:00:00 2001 From: Sunil Khatri Date: Wed, 31 Jul 2024 18:37:42 +0530 Subject: [PATCH] drm/amdgpu: optimize the padding for gfx_v9_4_3 MIME-Version: 1.0 Content-Type: text/plain; charset=utf8 Content-Transfer-Encoding: 8bit Adding NOP packets one by one in the ring does not use the CP efficiently. Solution: Use CP optimization while adding NOP packet's so PFP can discard NOP packets based on information of count from the Header instead of fetching all NOP packets one by one. Reviewed-by: Christian König Cc: Christian König Cc: Pierre-Eric Pelloux-Prayer Cc: Tvrtko Ursulin Cc: Marek Olšák Signed-off-by: Sunil Khatri Signed-off-by: Alex Deucher --- drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c | 20 +++++++++++++++++++- 1 file changed, 19 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c index 98fe6c40da64..c4832a5725c3 100644 --- a/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c +++ b/drivers/gpu/drm/amd/amdgpu/gfx_v9_4_3.c @@ -4178,6 +4178,24 @@ static void gfx_v9_4_3_enable_watchdog_timer(struct amdgpu_device *adev) amdgpu_gfx_ras_error_func(adev, NULL, gfx_v9_4_3_inst_enable_watchdog_timer); } +static void gfx_v9_4_3_ring_insert_nop(struct amdgpu_ring *ring, uint32_t num_nop) +{ + int i; + + /* Header itself is a NOP packet */ + if (num_nop == 1) { + amdgpu_ring_write(ring, ring->funcs->nop); + return; + } + + /* Max HW optimization till 0x3ffe, followed by remaining one NOP at a time*/ + amdgpu_ring_write(ring, PACKET3(PACKET3_NOP, min(num_nop - 2, 0x3ffe))); + + /* Header is at index 0, followed by num_nops - 1 NOP packet's */ + for (i = 1; i < num_nop; i++) + amdgpu_ring_write(ring, ring->funcs->nop); +} + static const struct amd_ip_funcs gfx_v9_4_3_ip_funcs = { .name = "gfx_v9_4_3", .early_init = gfx_v9_4_3_early_init, @@ -4227,7 +4245,7 @@ static const struct amdgpu_ring_funcs gfx_v9_4_3_ring_funcs_compute = { .emit_hdp_flush = gfx_v9_4_3_ring_emit_hdp_flush, .test_ring = gfx_v9_4_3_ring_test_ring, .test_ib = gfx_v9_4_3_ring_test_ib, - .insert_nop = amdgpu_ring_insert_nop, + .insert_nop = gfx_v9_4_3_ring_insert_nop, .pad_ib = amdgpu_ring_generic_pad_ib, .emit_wreg = gfx_v9_4_3_ring_emit_wreg, .emit_reg_wait = gfx_v9_4_3_ring_emit_reg_wait, -- 2.49.0