Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> writes: > Quoting Chris Wilson (2018-02-15 07:37:13) >> if (master_ctl & (GEN8_GT_RCS_IRQ | GEN8_GT_BCS_IRQ)) { >> - gt_iir[0] = I915_READ_FW(GEN8_GT_IIR(0)); >> - if (gt_iir[0]) >> - I915_WRITE_FW(GEN8_GT_IIR(0), gt_iir[0]); >> + gt_iir[0] = readl(regs + i915_mmio_reg_offset(GEN8_GT_IIR(0))); >> + if (likely(gt_iir[0])) >> + writel(gt_iir[0], regs + i915_mmio_reg_offset(GEN8_GT_IIR(0))); > > What do we feel about > > #define raw_reg_read32(mmio, reg) readl(mmio + i915_mmio_reg_offset(reg)) > #define raw_reg_write32(mmio, reg, value) writel(value, mmio + i915_mmio_reg_offset(reg)) I am in favour. Shorter versions would be even more pleasing. reg_read and reg_write. -Mika > > for transitioning away from the implicit argument in I915_READ? > -Chris > _______________________________________________ > Intel-gfx mailing list > Intel-gfx@xxxxxxxxxxxxxxxxxxxxx > https://lists.freedesktop.org/mailman/listinfo/intel-gfx _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx