MIPS: pm-cps: Use GPR number macros

Use GPR number macros in uasm code generation parts to
reduce code duplication.

There are functional change due to difference in register
symbolic names between OABI and NABI, while existing code
is only using definitions from OABI.

Code pieces are carefully inspected to ensure register
usages are safe on NABI as well.

We changed register allocation of r_pcohctl from T7 to T8
as T7 is not available on NABI and we just want a caller
saved scratch register here.

Signed-off-by: Jiaxun Yang <[e-mail]>
Signed-off-by: Thomas Bogendoerfer <[e-mail]>

arch/mips/kernel/pm-cps.c

···
 #include <asm/mipsmtregs.h>
 #include <asm/pm.h>
 #include <asm/pm-cps.h>
+#include <asm/regdef.h>
 #include <asm/smp-cps.h>
 #include <asm/uasm.h>
 
···
 /* A somewhat arbitrary number of labels & relocs for uasm */
 static struct uasm_label labels[32];
 static struct uasm_reloc relocs[32];
-
-enum mips_reg {
-	zero, at, v0, v1, a0, a1, a2, a3,
-	t0, t1, t2, t3, t4, t5, t6, t7,
-	s0, s1, s2, s3, s4, s5, s6, s7,
-	t8, t9, k0, k1, gp, sp, fp, ra,
-};
 
 bool cps_pm_support_state(enum cps_pm_state state)
 {
···
 		return;
 
 	/* Load base address */
-	UASM_i_LA(pp, t0, (long)CKSEG0);
+	UASM_i_LA(pp, GPR_T0, (long)CKSEG0);
 
 	/* Calculate end address */
 	if (cache_size < 0x8000)
-		uasm_i_addiu(pp, t1, t0, cache_size);
+		uasm_i_addiu(pp, GPR_T1, GPR_T0, cache_size);
 	else
-		UASM_i_LA(pp, t1, (long)(CKSEG0 + cache_size));
+		UASM_i_LA(pp, GPR_T1, (long)(CKSEG0 + cache_size));
 
 	/* Start of cache op loop */
 	uasm_build_label(pl, *pp, lbl);
···
 	/* Generate the cache ops */
 	for (i = 0; i < unroll_lines; i++) {
 		if (cpu_has_mips_r6) {
-			uasm_i_cache(pp, op, 0, t0);
-			uasm_i_addiu(pp, t0, t0, cache->linesz);
+			uasm_i_cache(pp, op, 0, GPR_T0);
+			uasm_i_addiu(pp, GPR_T0, GPR_T0, cache->linesz);
 		} else {
-			uasm_i_cache(pp, op, i * cache->linesz, t0);
+			uasm_i_cache(pp, op, i * cache->linesz, GPR_T0);
 		}
 	}
 
 	if (!cpu_has_mips_r6)
 		/* Update the base address */
-		uasm_i_addiu(pp, t0, t0, unroll_lines * cache->linesz);
+		uasm_i_addiu(pp, GPR_T0, GPR_T0, unroll_lines * cache->linesz);
 
 	/* Loop if we haven't reached the end address yet */
-	uasm_il_bne(pp, pr, t0, t1, lbl);
+	uasm_il_bne(pp, pr, GPR_T0, GPR_T1, lbl);
 	uasm_i_nop(pp);
 }
 
···
 	 */
 
 	/* Preserve perf counter setup */
-	uasm_i_mfc0(pp, t2, 25, (perf_counter * 2) + 0); /* PerfCtlN */
-	uasm_i_mfc0(pp, t3, 25, (perf_counter * 2) + 1); /* PerfCntN */
+	uasm_i_mfc0(pp, GPR_T2, 25, (perf_counter * 2) + 0); /* PerfCtlN */
+	uasm_i_mfc0(pp, GPR_T3, 25, (perf_counter * 2) + 1); /* PerfCntN */
 
 	/* Setup perf counter to count FSB full pipeline stalls */
-	uasm_i_addiu(pp, t0, zero, (perf_event << 5) | 0xf);
-	uasm_i_mtc0(pp, t0, 25, (perf_counter * 2) + 0); /* PerfCtlN */
+	uasm_i_addiu(pp, GPR_T0, GPR_ZERO, (perf_event << 5) | 0xf);
+	uasm_i_mtc0(pp, GPR_T0, 25, (perf_counter * 2) + 0); /* PerfCtlN */
 	uasm_i_ehb(pp);
-	uasm_i_mtc0(pp, zero, 25, (perf_counter * 2) + 1); /* PerfCntN */
+	uasm_i_mtc0(pp, GPR_ZERO, 25, (perf_counter * 2) + 1); /* PerfCntN */
 	uasm_i_ehb(pp);
 
 	/* Base address for loads */
-	UASM_i_LA(pp, t0, (long)CKSEG0);
+	UASM_i_LA(pp, GPR_T0, (long)CKSEG0);
 
 	/* Start of clear loop */
 	uasm_build_label(pl, *pp, lbl);
 
 	/* Perform some loads to fill the FSB */
 	for (i = 0; i < num_loads; i++)
-		uasm_i_lw(pp, zero, i * line_size * line_stride, t0);
+		uasm_i_lw(pp, GPR_ZERO, i * line_size * line_stride, GPR_T0);
 
 	/*
 	 * Invalidate the new D-cache entries so that the cache will need
···
 	 */
 	for (i = 0; i < num_loads; i++) {
 		uasm_i_cache(pp, Hit_Invalidate_D,
-			     i * line_size * line_stride, t0);
+			     i * line_size * line_stride, GPR_T0);
 		uasm_i_cache(pp, Hit_Writeback_Inv_SD,
-			     i * line_size * line_stride, t0);
+			     i * line_size * line_stride, GPR_T0);
 	}
 
 	/* Barrier ensuring previous cache invalidates are complete */
···
 	uasm_i_ehb(pp);
 
 	/* Check whether the pipeline stalled due to the FSB being full */
-	uasm_i_mfc0(pp, t1, 25, (perf_counter * 2) + 1); /* PerfCntN */
+	uasm_i_mfc0(pp, GPR_T1, 25, (perf_counter * 2) + 1); /* PerfCntN */
 
 	/* Loop if it didn't */
-	uasm_il_beqz(pp, pr, t1, lbl);
+	uasm_il_beqz(pp, pr, GPR_T1, lbl);
 	uasm_i_nop(pp);
 
 	/* Restore perf counter 1. The count may well now be wrong... */
-	uasm_i_mtc0(pp, t2, 25, (perf_counter * 2) + 0); /* PerfCtlN */
+	uasm_i_mtc0(pp, GPR_T2, 25, (perf_counter * 2) + 0); /* PerfCtlN */
 	uasm_i_ehb(pp);
-	uasm_i_mtc0(pp, t3, 25, (perf_counter * 2) + 1); /* PerfCntN */
+	uasm_i_mtc0(pp, GPR_T3, 25, (perf_counter * 2) + 1); /* PerfCntN */
 	uasm_i_ehb(pp);
 
 	return 0;
···
 				struct uasm_reloc **pr,
 				unsigned r_addr, int lbl)
 {
-	uasm_i_lui(pp, t0, uasm_rel_hi(0x80000000));
+	uasm_i_lui(pp, GPR_T0, uasm_rel_hi(0x80000000));
 	uasm_build_label(pl, *pp, lbl);
-	uasm_i_ll(pp, t1, 0, r_addr);
-	uasm_i_or(pp, t1, t1, t0);
-	uasm_i_sc(pp, t1, 0, r_addr);
-	uasm_il_beqz(pp, pr, t1, lbl);
+	uasm_i_ll(pp, GPR_T1, 0, r_addr);
+	uasm_i_or(pp, GPR_T1, GPR_T1, GPR_T0);
+	uasm_i_sc(pp, GPR_T1, 0, r_addr);
+	uasm_il_beqz(pp, pr, GPR_T1, lbl);
 	uasm_i_nop(pp);
 }
 
···
 	struct uasm_label *l = labels;
 	struct uasm_reloc *r = relocs;
 	u32 *buf, *p;
-	const unsigned r_online = a0;
-	const unsigned r_nc_count = a1;
-	const unsigned r_pcohctl = t7;
+	const unsigned r_online = GPR_A0;
+	const unsigned r_nc_count = GPR_A1;
+	const unsigned r_pcohctl = GPR_T8;
 	const unsigned max_instrs = 256;
 	unsigned cpc_cmd;
 	int err;
···
 		 * with the return address placed in v0 to avoid clobbering
 		 * the ra register before it is saved.
 		 */
-		UASM_i_LA(&p, t0, (long)mips_cps_pm_save);
-		uasm_i_jalr(&p, v0, t0);
+		UASM_i_LA(&p, GPR_T0, (long)mips_cps_pm_save);
+		uasm_i_jalr(&p, GPR_V0, GPR_T0);
 		uasm_i_nop(&p);
 	}
 
···
 		/* Increment ready_count */
 		uasm_i_sync(&p, __SYNC_mb);
 		uasm_build_label(&l, p, lbl_incready);
-		uasm_i_ll(&p, t1, 0, r_nc_count);
-		uasm_i_addiu(&p, t2, t1, 1);
-		uasm_i_sc(&p, t2, 0, r_nc_count);
-		uasm_il_beqz(&p, &r, t2, lbl_incready);
-		uasm_i_addiu(&p, t1, t1, 1);
+		uasm_i_ll(&p, GPR_T1, 0, r_nc_count);
+		uasm_i_addiu(&p, GPR_T2, GPR_T1, 1);
+		uasm_i_sc(&p, GPR_T2, 0, r_nc_count);
+		uasm_il_beqz(&p, &r, GPR_T2, lbl_incready);
+		uasm_i_addiu(&p, GPR_T1, GPR_T1, 1);
 
 		/* Barrier ensuring all CPUs see the updated r_nc_count value */
 		uasm_i_sync(&p, __SYNC_mb);
···
 		 * If this is the last VPE to become ready for non-coherence
 		 * then it should branch below.
 		 */
-		uasm_il_beq(&p, &r, t1, r_online, lbl_disable_coherence);
+		uasm_il_beq(&p, &r, GPR_T1, r_online, lbl_disable_coherence);
 		uasm_i_nop(&p);
 
 		if (state < CPS_PM_POWER_GATED) {
···
 			 * has been disabled before proceeding, which it will do
 			 * by polling for the top bit of ready_count being set.
 			 */
-			uasm_i_addiu(&p, t1, zero, -1);
+			uasm_i_addiu(&p, GPR_T1, GPR_ZERO, -1);
 			uasm_build_label(&l, p, lbl_poll_cont);
-			uasm_i_lw(&p, t0, 0, r_nc_count);
-			uasm_il_bltz(&p, &r, t0, lbl_secondary_cont);
+			uasm_i_lw(&p, GPR_T0, 0, r_nc_count);
+			uasm_il_bltz(&p, &r, GPR_T0, lbl_secondary_cont);
 			uasm_i_ehb(&p);
 			if (cpu_has_mipsmt)
-				uasm_i_yield(&p, zero, t1);
+				uasm_i_yield(&p, GPR_ZERO, GPR_T1);
 			uasm_il_b(&p, &r, lbl_poll_cont);
 			uasm_i_nop(&p);
 		} else {
···
 			 */
 			if (cpu_has_mipsmt) {
 				/* Halt the VPE via C0 tchalt register */
-				uasm_i_addiu(&p, t0, zero, TCHALT_H);
-				uasm_i_mtc0(&p, t0, 2, 4);
+				uasm_i_addiu(&p, GPR_T0, GPR_ZERO, TCHALT_H);
+				uasm_i_mtc0(&p, GPR_T0, 2, 4);
 			} else if (cpu_has_vp) {
 				/* Halt the VP via the CPC VP_STOP register */
 				unsigned int vpe_id;
 
 				vpe_id = cpu_vpe_id(&cpu_data[cpu]);
-				uasm_i_addiu(&p, t0, zero, 1 << vpe_id);
-				UASM_i_LA(&p, t1, (long)addr_cpc_cl_vp_stop());
-				uasm_i_sw(&p, t0, 0, t1);
+				uasm_i_addiu(&p, GPR_T0, GPR_ZERO, 1 << vpe_id);
+				UASM_i_LA(&p, GPR_T1, (long)addr_cpc_cl_vp_stop());
+				uasm_i_sw(&p, GPR_T0, 0, GPR_T1);
 			} else {
 				BUG();
 			}
···
 		* defined by the interAptiv & proAptiv SUMs as ensuring that the
 		*  operation resulting from the preceding store is complete.
 		*/
-		uasm_i_addiu(&p, t0, zero, 1 << cpu_core(&cpu_data[cpu]));
-		uasm_i_sw(&p, t0, 0, r_pcohctl);
-		uasm_i_lw(&p, t0, 0, r_pcohctl);
+		uasm_i_addiu(&p, GPR_T0, GPR_ZERO, 1 << cpu_core(&cpu_data[cpu]));
+		uasm_i_sw(&p, GPR_T0, 0, r_pcohctl);
+		uasm_i_lw(&p, GPR_T0, 0, r_pcohctl);
 
 		/* Barrier to ensure write to coherence control is complete */
 		uasm_i_sync(&p, __SYNC_full);
···
 	}
 
 	/* Disable coherence */
-	uasm_i_sw(&p, zero, 0, r_pcohctl);
-	uasm_i_lw(&p, t0, 0, r_pcohctl);
+	uasm_i_sw(&p, GPR_ZERO, 0, r_pcohctl);
+	uasm_i_lw(&p, GPR_T0, 0, r_pcohctl);
 
 	if (state >= CPS_PM_CLOCK_GATED) {
 		err = cps_gen_flush_fsb(&p, &l, &r, &cpu_data[cpu],
···
 		}
 
 		/* Issue the CPC command */
-		UASM_i_LA(&p, t0, (long)addr_cpc_cl_cmd());
-		uasm_i_addiu(&p, t1, zero, cpc_cmd);
-		uasm_i_sw(&p, t1, 0, t0);
+		UASM_i_LA(&p, GPR_T0, (long)addr_cpc_cl_cmd());
+		uasm_i_addiu(&p, GPR_T1, GPR_ZERO, cpc_cmd);
+		uasm_i_sw(&p, GPR_T1, 0, GPR_T0);
 
 		if (state == CPS_PM_POWER_GATED) {
 			/* If anything goes wrong just hang */
···
 	 * will run this. The first will actually re-enable coherence & the
 	 * rest will just be performing a rather unusual nop.
 	 */
-	uasm_i_addiu(&p, t0, zero, mips_cm_revision() < CM_REV_CM3
+	uasm_i_addiu(&p, GPR_T0, GPR_ZERO, mips_cm_revision() < CM_REV_CM3
 				? CM_GCR_Cx_COHERENCE_COHDOMAINEN
 				: CM3_GCR_Cx_COHERENCE_COHEN);
 
-	uasm_i_sw(&p, t0, 0, r_pcohctl);
-	uasm_i_lw(&p, t0, 0, r_pcohctl);
+	uasm_i_sw(&p, GPR_T0, 0, r_pcohctl);
+	uasm_i_lw(&p, GPR_T0, 0, r_pcohctl);
 
 	/* Barrier to ensure write to coherence control is complete */
 	uasm_i_sync(&p, __SYNC_full);
···
 		/* Decrement ready_count */
 		uasm_build_label(&l, p, lbl_decready);
 		uasm_i_sync(&p, __SYNC_mb);
-		uasm_i_ll(&p, t1, 0, r_nc_count);
-		uasm_i_addiu(&p, t2, t1, -1);
-		uasm_i_sc(&p, t2, 0, r_nc_count);
-		uasm_il_beqz(&p, &r, t2, lbl_decready);
-		uasm_i_andi(&p, v0, t1, (1 << fls(smp_num_siblings)) - 1);
+		uasm_i_ll(&p, GPR_T1, 0, r_nc_count);
+		uasm_i_addiu(&p, GPR_T2, GPR_T1, -1);
+		uasm_i_sc(&p, GPR_T2, 0, r_nc_count);
+		uasm_il_beqz(&p, &r, GPR_T2, lbl_decready);
+		uasm_i_andi(&p, GPR_V0, GPR_T1, (1 << fls(smp_num_siblings)) - 1);
 
 		/* Barrier ensuring all CPUs see the updated r_nc_count value */
 		uasm_i_sync(&p, __SYNC_mb);
···
 	}
 
 	/* The core is coherent, time to return to C code */
-	uasm_i_jr(&p, ra);
+	uasm_i_jr(&p, GPR_RA);
 	uasm_i_nop(&p);
 
 gen_done: