Re: [PATCH v9 16/18] mtd: spinand: negotiate optimal controller operating point before dirmap creation

Hello Miquel,

On 04/09/26 21:53, Miquel Raynal wrote:

Hello Santhosh,

quoted

+static bool spinand_try_ranked_variant(struct spinand_device *spinand,
+				       struct spi_mem *mem,
+				       enum spinand_bus_interface iface,
+				       u32 *tried_mask)
+{
+	const struct spinand_op_variants *variants = spinand->all_read_variants;
+	struct spinand_mem_ops *templates = iface == ODTR ?
+		&spinand->odtr_op_templates : &spinand->ssdr_op_templates;
+	const struct spi_mem_op *best;
+	int ret;
+
+	if (!variants)
+		return false;
+
+	while ((best = spinand_op_find_best_variant(spinand, variants, iface,
+						    *tried_mask, NULL))) {
+		*tried_mask |= BIT(best - variants->ops);
+		spinand->max_read_op = *best;
+		spinand->max_read_op.max_freq = 0;
+		spinand->max_write_op.max_freq = 0;
+		ret = spi_mem_execute_tuning(mem, &spinand->max_read_op,
+					     &spinand->max_write_op);
+		if (ret && ret != -EOPNOTSUPP)
+			dev_dbg(&mem->spi->dev, "%s optimization failed: %d\n",
+				iface == ODTR ? "ODTR" : "SSDR", ret);
+		if (!ret && spinand->max_read_op.max_freq) {
+			templates->read_cache = best;
+			if (spinand->all_cont_read_variants) {
+				templates->cont_read_cache =
+					spinand_find_cont_read_variant(spinand, best, iface);
+				spinand->cont_read_possible =
+					!!templates->cont_read_cache;

I believe you almost addressed the concerns raised in v7 about
cont_read_possible, except what you do here is still a bit dangerous
since you could re-enable continuous reads on no_cs_assertion capable
controllers. I believe you should clear the flag only, never risk to set
it, something like:

if (!templates->cont_read_cache)
     spinand->cont_read_possible = false;

With this fixed,

Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>

Thank you for the review.

Fixed as you suggested - spinand_try_ranked_variant() now only clears
cont_read_possible when the newly selected variant has no matching
cont_read counterpart; Will carry this into v10 along with the other
fixes.

Thanks,
Santhosh.

I believe I will take this patch only after Mark takes the batch before,
which looks mostly good for my non-spi-expert eyes. Either Mark shares
an immutable tag, or I wait one cycle.

Thanks,
Miquèl