* [PATCH v2] mtd: rawnand: cadence: support 64-bit slave dma interface
@ 2022-10-17 15:26 Valentin Korenblit
2022-10-17 18:56 ` Arnd Bergmann
0 siblings, 1 reply; 2+ messages in thread
From: Valentin Korenblit @ 2022-10-17 15:26 UTC (permalink / raw)
To: linux-mtd; +Cc: miquel.raynal, arnd, ye.xingchen, Valentin Korenblit
32-bit accesses on 64-bit sdma trigger sdma_err in intr_status register.
Check dma capabilities before reading/writing from/to sdma interface.
Link to discussion:
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org/thread/3NMACGIM5NDUBPXRT5RTBZON6LQE5A3B/
Signed-off-by: Valentin Korenblit <vkorenblit@sequans.com>
---
Changes v1 -> v2:
- Replaced ioread64_rep by cadence_nand_readsq (suggested by Arnd)
- Replaced iowrite64_rep by cadence_nand_writesq (suggested by Arnd)
- Do not try to access 64-bit sdma if __raw_readq/__raw_writeq are not defined
---
.../mtd/nand/raw/cadence-nand-controller.c | 96 ++++++++++++++++---
1 file changed, 84 insertions(+), 12 deletions(-)
diff --git a/drivers/mtd/nand/raw/cadence-nand-controller.c b/drivers/mtd/nand/raw/cadence-nand-controller.c
index 9dac3ca69d57..0309a6b0d3d3 100644
--- a/drivers/mtd/nand/raw/cadence-nand-controller.c
+++ b/drivers/mtd/nand/raw/cadence-nand-controller.c
@@ -1869,6 +1869,46 @@ static int cadence_nand_slave_dma_transfer(struct cdns_nand_ctrl *cdns_ctrl,
return -EIO;
}
+static inline void cadence_nand_readsq(struct cdns_nand_ctrl *cdns_ctrl,
+ void *buffer, unsigned int count)
+{
+#ifdef __raw_readq
+ const volatile void __iomem *addr = cdns_ctrl->io.virt;
+
+ if (count) {
+ u64 *buf = buffer;
+
+ do {
+ u64 x = __raw_readq(addr);
+ *buf++ = x;
+ } while (--count);
+ }
+#else
+ dev_err(cdns_ctrl->dev,
+ "cannot read 64-bit sdma on !64-bit architectures");
+#endif
+}
+
+static inline void cadence_nand_writesq(struct cdns_nand_ctrl *cdns_ctrl,
+ const void *buffer,
+ unsigned int count)
+{
+#ifdef __raw_writeq
+ volatile void __iomem *addr = cdns_ctrl->io.virt;
+
+ if (count) {
+ const u64 *buf = buffer;
+
+ do {
+ __raw_writeq(*buf++, addr);
+ } while (--count);
+ }
+#else
+ dev_err(cdns_ctrl->dev,
+ "cannot write 64-bit sdma on !64-bit architectures");
+#endif
+}
+
static int cadence_nand_read_buf(struct cdns_nand_ctrl *cdns_ctrl,
u8 *buf, int len)
{
@@ -1882,17 +1922,33 @@ static int cadence_nand_read_buf(struct cdns_nand_ctrl *cdns_ctrl,
return status;
if (!cdns_ctrl->caps1->has_dma) {
- int len_in_words = len >> 2;
+ u8 data_dma_width = cdns_ctrl->caps2.data_dma_width;
+
+ int len_in_words = (data_dma_width == 4) ? len >> 2 : len >> 3;
/* read alingment data */
- ioread32_rep(cdns_ctrl->io.virt, buf, len_in_words);
+ if (data_dma_width == 4)
+ ioread32_rep(cdns_ctrl->io.virt, buf, len_in_words);
+ else
+ cadence_nand_readsq(cdns_ctrl, buf, len_in_words);
+
if (sdma_size > len) {
+ int read_bytes = (data_dma_width == 4) ?
+ len_in_words << 2 : len_in_words << 3;
+
/* read rest data from slave DMA interface if any */
- ioread32_rep(cdns_ctrl->io.virt, cdns_ctrl->buf,
- sdma_size / 4 - len_in_words);
+ if (data_dma_width == 4)
+ ioread32_rep(cdns_ctrl->io.virt,
+ cdns_ctrl->buf,
+ sdma_size / 4 - len_in_words);
+ else
+ cadence_nand_readsq(cdns_ctrl,
+ cdns_ctrl->buf,
+ sdma_size / 8 - len_in_words);
+
/* copy rest of data */
- memcpy(buf + (len_in_words << 2), cdns_ctrl->buf,
- len - (len_in_words << 2));
+ memcpy(buf + read_bytes, cdns_ctrl->buf,
+ len - read_bytes);
}
return 0;
}
@@ -1936,16 +1992,32 @@ static int cadence_nand_write_buf(struct cdns_nand_ctrl *cdns_ctrl,
return status;
if (!cdns_ctrl->caps1->has_dma) {
- int len_in_words = len >> 2;
+ u8 data_dma_width = cdns_ctrl->caps2.data_dma_width;
+
+ int len_in_words = (data_dma_width == 4) ? len >> 2 : len >> 3;
+
+ if (data_dma_width == 4)
+ iowrite32_rep(cdns_ctrl->io.virt, buf, len_in_words);
+ else
+ cadence_nand_writesq(cdns_ctrl, buf, len_in_words);
- iowrite32_rep(cdns_ctrl->io.virt, buf, len_in_words);
if (sdma_size > len) {
+ int written_bytes = (data_dma_width == 4) ?
+ len_in_words << 2 : len_in_words << 3;
+
/* copy rest of data */
- memcpy(cdns_ctrl->buf, buf + (len_in_words << 2),
- len - (len_in_words << 2));
+ memcpy(cdns_ctrl->buf, buf + written_bytes,
+ len - written_bytes);
+
/* write all expected by nand controller data */
- iowrite32_rep(cdns_ctrl->io.virt, cdns_ctrl->buf,
- sdma_size / 4 - len_in_words);
+ if (data_dma_width == 4)
+ iowrite32_rep(cdns_ctrl->io.virt,
+ cdns_ctrl->buf,
+ sdma_size / 4 - len_in_words);
+ else
+ cadence_nand_writesq(cdns_ctrl,
+ cdns_ctrl->buf,
+ sdma_size / 8 - len_in_words);
}
return 0;
--
2.20.1
-- IMPORTANT NOTICE:
The contents of this email and any attachments are confidential and may also be privileged. If you are not the intended recipient, please notify the sender immediately and do not disclose the contents to any other person, use it for any purpose, or store or copy the information in any medium.
Thank you.
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply related [flat|nested] 2+ messages in thread* Re: [PATCH v2] mtd: rawnand: cadence: support 64-bit slave dma interface
2022-10-17 15:26 [PATCH v2] mtd: rawnand: cadence: support 64-bit slave dma interface Valentin Korenblit
@ 2022-10-17 18:56 ` Arnd Bergmann
0 siblings, 0 replies; 2+ messages in thread
From: Arnd Bergmann @ 2022-10-17 18:56 UTC (permalink / raw)
To: Valentin Korenblit, linux-mtd; +Cc: Miquel Raynal, ye.xingchen
On Mon, Oct 17, 2022, at 5:26 PM, Valentin Korenblit wrote:
> 32-bit accesses on 64-bit sdma trigger sdma_err in intr_status register.
>
> Check dma capabilities before reading/writing from/to sdma interface.
>
> Link to discussion:
> https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org/thread/3NMACGIM5NDUBPXRT5RTBZON6LQE5A3B/
>
> Signed-off-by: Valentin Korenblit <vkorenblit@sequans.com>
Seems reasonable overall based on the discussion so far.
Minor comments:
> Changes v1 -> v2:
> - Replaced ioread64_rep by cadence_nand_readsq (suggested by Arnd)
> - Replaced iowrite64_rep by cadence_nand_writesq (suggested by Arnd)
> - Do not try to access 64-bit sdma if __raw_readq/__raw_writeq are not defined
> +static inline void cadence_nand_readsq(struct cdns_nand_ctrl
> *cdns_ctrl,
> + void *buffer, unsigned int count)
> +{
> +#ifdef __raw_readq
> + const volatile void __iomem *addr = cdns_ctrl->io.virt;
> +
> + if (count) {
> + u64 *buf = buffer;
> +
> + do {
> + u64 x = __raw_readq(addr);
> + *buf++ = x;
> + } while (--count);
> + }
While this is correct, I suppose you don't actually need that
loop any more, since architectures that have __raw_readq()
should also also have ioread64_rep() or readsq(). We actually
just had a few issues with that in another context, but I think
starting with 6.1-rc1 you should be able to rely on that ;-)
> +#else
> + dev_err(cdns_ctrl->dev,
> + "cannot read 64-bit sdma on !64-bit architectures");
> +#endif
It would be nice to print an error at probe time and fail attaching
the device rather than waiting for the first transfer to continue
with incorrect data.
Arnd
______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2022-10-17 18:57 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2022-10-17 15:26 [PATCH v2] mtd: rawnand: cadence: support 64-bit slave dma interface Valentin Korenblit
2022-10-17 18:56 ` Arnd Bergmann
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox