// Tile size: 5x2
	// Accumulators: 0-9
	// Col regs: ymm10-13
	// Row regs: ymm14-15

	vmovaps			ymm10,	[rax]
	vbroadcastss	ymm14,	dword ptr [rcx + 0]
	vbroadcastss	ymm15,	dword ptr [rcx + 4]
	vmovaps			ymm11,	[rax + 32]

	// NB stepping column-wise
	vfmadd231ps		ymm0,	ymm10, ymm14
	vfmadd231ps		ymm5,	ymm10, ymm15

	vmovaps			ymm12,	[rax + 64]

	vfmadd231ps		ymm1,	ymm11, ymm14
	vfmadd231ps		ymm6,	ymm11, ymm15

	vmovaps			ymm13,	[rax + 96]

	vfmadd231ps		ymm2,	ymm12, ymm14
	vfmadd231ps		ymm7,	ymm12, ymm15

	vmovaps			ymm10,	[rax + 128]

	vfmadd231ps		ymm3,	ymm13, ymm14
	vfmadd231ps		ymm8,	ymm13, ymm15

	vmovaps			ymm11,	[rax + 160]

	vfmadd231ps		ymm4,	ymm10, ymm14
	vfmadd231ps		ymm9,	ymm10, ymm15

	vbroadcastss	ymm14,	dword ptr [rcx + 8]
	vbroadcastss	ymm15,	dword ptr [rcx + 12]

	vmovaps			ymm12,	[rax + 192]

	// NB stepping column-wise
	vfmadd231ps		ymm0,	ymm11, ymm14
	vfmadd231ps		ymm5,	ymm11, ymm15

	vmovaps			ymm13,	[rax + 224]

	vfmadd231ps		ymm1,	ymm12, ymm14
	vfmadd231ps		ymm6,	ymm12, ymm15

	vmovaps			ymm10,	[rax + 256]

	vfmadd231ps		ymm2,	ymm13, ymm14
	vfmadd231ps		ymm7,	ymm13, ymm15

	vmovaps			ymm11,	[rax + 288]

	vfmadd231ps		ymm3,	ymm10, ymm14
	vfmadd231ps		ymm8,	ymm10, ymm15

	vfmadd231ps		ymm4,	ymm11, ymm14
	vfmadd231ps		ymm9,	ymm11, ymm15

	add rax, 320
	add rcx, 16