Skip to contentSkip to conversation

hw/ssi: imx_spi: Use a macro for number of chip selects supported

PATCH Superseded Bin Meng 3 participants 13 messages Dec 2, 2020 – Jan 8, 2021 Arm i.MX and Layerscape

v2 of 9 Superseded Latest is v9 →

lore
Share & tools
lore

Bin Meng

author
#0 lore raw

PATCH v2 1/2 hw/ssi: imx_spi: Use a macro for number of chip selects supported

From: Bin Meng <[email protected]>

Avoid using a magic number (4) everywhere for the number of chip selects supported.

---

(no changes since v1)

 hw/ssi/imx_spi.c         | 4 ++--
 include/hw/ssi/imx_spi.h | 5 ++++-
 2 files changed, 6 insertions(+), 3 deletions(-)
hw/ssi/imx_spi.c +2 -2
index d8885ae454..e605049a21 100644
@@ -361,7 +361,7 @@ static void imx_spi_write(void *opaque, hwaddr offset, uint64_t value,
 
             /* We are in master mode */
 
-            for (i = 0; i < 4; i++) {
+            for (i = 0; i < ECSPI_NUM_CS; i++) {
                 qemu_set_irq(s->cs_lines[i],
                              i == imx_spi_selected_channel(s) ? 0 : 1);
             }
@@ -424,7 +424,7 @@ static void imx_spi_realize(DeviceState *dev, Error **errp)
     sysbus_init_mmio(SYS_BUS_DEVICE(dev), &s->iomem);
     sysbus_init_irq(SYS_BUS_DEVICE(dev), &s->irq);
 
-    for (i = 0; i < 4; ++i) {
+    for (i = 0; i < ECSPI_NUM_CS; ++i) {
         sysbus_init_irq(SYS_BUS_DEVICE(dev), &s->cs_lines[i]);
     }
 
include/hw/ssi/imx_spi.h +4 -1
index b82b17f364..eeaf49bbac 100644
@@ -77,6 +77,9 @@
 
 #define EXTRACT(value, name) extract32(value, name##_SHIFT, name##_LENGTH)
 
+/* number of chip selects supported */
+#define ECSPI_NUM_CS 4
+
 #define TYPE_IMX_SPI "imx.spi"
 OBJECT_DECLARE_SIMPLE_TYPE(IMXSPIState, IMX_SPI)
 
@@ -89,7 +92,7 @@ struct IMXSPIState {
 
     qemu_irq irq;
 
-    qemu_irq cs_lines[4];
+    qemu_irq cs_lines[ECSPI_NUM_CS];
 
     SSIBus *bus;
 
Signature · 1 line
2.25.1
author
#1 lore raw

PATCH v2 2/2 hw/ssi: imx_spi: Disable chip selects in imx_spi_reset()

From: Xuzhou Cheng <[email protected]>

When a write to ECSPI_CONREG register to disable the SPI controller, imx_spi_reset() is called to reset the controller, during which CS lines should have been disabled, otherwise the state machine of any devices (e.g.: SPI flashes) connected to the SPI master is stuck to its last state and responds incorrectly to any follow-up commands.

Fixes c906a3a01582 ("i.MX: Add the Freescale SPI Controller")

---

Changes in v2:
- Fix the "Fixes" tag in the commit message

 hw/ssi/imx_spi.c | 5 +++++
 1 file changed, 5 insertions(+)
hw/ssi/imx_spi.c +5 -0
index e605049a21..85c172e815 100644
@@ -231,6 +231,7 @@ static void imx_spi_flush_txfifo(IMXSPIState *s)
 static void imx_spi_reset(DeviceState *dev)
 {
     IMXSPIState *s = IMX_SPI(dev);
+    int i;
 
     DPRINTF("\n");
 
@@ -243,6 +244,10 @@ static void imx_spi_reset(DeviceState *dev)
 
     imx_spi_update_irq(s);
 
+    for (i = 0; i < ECSPI_NUM_CS; i++) {
+        qemu_set_irq(s->cs_lines[i], 1);
+    }
+
     s->burst_length = 0;
 }
 
Signature · 1 line
2.25.1
#2 lore raw

Re: [PATCH v2 2/2] hw/ssi: imx_spi: Disable chip selects in imx_spi_reset()

Quoted 35 lines · Bin Meng

From: Xuzhou Cheng <[email protected]>

When a write to ECSPI_CONREG register to disable the SPI controller, imx_spi_reset() is called to reset the controller, during which CS lines should have been disabled, otherwise the state machine of any devices (e.g.: SPI flashes) connected to the SPI master is stuck to its last state and responds incorrectly to any follow-up commands.

Fixes c906a3a01582 ("i.MX: Add the Freescale SPI Controller")

---

Changes in v2:
- Fix the "Fixes" tag in the commit message

 hw/ssi/imx_spi.c | 5 +++++
 1 file changed, 5 insertions(+)
hw/ssi/imx_spi.c +1 -0
index e605049a21..85c172e815 100644
@@ -231,6 +231,7 @@ static void imx_spi_flush_txfifo(IMXSPIState *s)
 static void imx_spi_reset(DeviceState *dev)
 {
     IMXSPIState *s = IMX_SPI(dev);
+    int i;
 
     DPRINTF("\n");
 

hw/ssi/imx_spi.c static void imx_spi_reset(DeviceState *dev) · line 249

Quoted 6 lines
hw/ssi/imx_spi.c +3 -0
@@ -243,6 +244,10 @@ static void imx_spi_reset(DeviceState *dev)
 
     imx_spi_update_irq(s);
 
+    for (i = 0; i < ECSPI_NUM_CS; i++) {
+        qemu_set_irq(s->cs_lines[i], 1);
+    }

Calling qemu_set_irq() in a device reset function is a bad idea, because you don't know whether the thing on the other end of the IRQ line (a) has already reset before you or
(b) is going to reset after you. If you need to do this then I think you need to convert this device (and perhaps whatever it's connected to) to the 3-phase-reset API. (But you probably don't, see below.)

Usually the approach is that the device on the other end of the line is going to reset its state anyway, so there's no need to actively signal an irq line change.

If this is required only for the case of "guest requested a controller reset via the ECSPI_CONREG register" and not for full system reset, then you can handle that by having an imx_spi_soft_reset() which calls imx_spi_reset() and then does the qemu_set_irq() calls, so full system (power-cycle) reset still goes to imx_spi_reset() but guest-commanded reset via the register interface calls imx_spi_soft_reset().

thanks
-- PMM

author
#3 lore raw

Re: [PATCH v2 2/2] hw/ssi: imx_spi: Disable chip selects in imx_spi_reset()

Quoted 40 lines · Peter Maydell
Quoted 38 lines · Bin Meng

From: Xuzhou Cheng <[email protected]>

When a write to ECSPI_CONREG register to disable the SPI controller, imx_spi_reset() is called to reset the controller, during which CS lines should have been disabled, otherwise the state machine of any devices (e.g.: SPI flashes) connected to the SPI master is stuck to its last state and responds incorrectly to any follow-up commands.

Fixes c906a3a01582 ("i.MX: Add the Freescale SPI Controller")

---

Changes in v2:
- Fix the "Fixes" tag in the commit message

 hw/ssi/imx_spi.c | 5 +++++
 1 file changed, 5 insertions(+)
hw/ssi/imx_spi.c +1 -0
index e605049a21..85c172e815 100644
@@ -231,6 +231,7 @@ static void imx_spi_flush_txfifo(IMXSPIState *s)
 static void imx_spi_reset(DeviceState *dev)
 {
     IMXSPIState *s = IMX_SPI(dev);
+    int i;
 
     DPRINTF("\n");
 
@@ -243,6 +244,10 @@ static void imx_spi_reset(DeviceState *dev)
 
     imx_spi_update_irq(s);
 
Quoted 12 lines

hw/ssi/imx_spi.c static void imx_spi_reset(DeviceState *dev) · line 249

Quoted 3 lines
hw/ssi/imx_spi.c +3 -0
+    for (i = 0; i < ECSPI_NUM_CS; i++) {
+        qemu_set_irq(s->cs_lines[i], 1);
+    }

Calling qemu_set_irq() in a device reset function is a bad idea, because you don't know whether the thing on the other end of the IRQ line (a) has already reset before you or
(b) is going to reset after you. If you need to do this then I think you need to convert this device (and perhaps whatever it's connected to) to the 3-phase-reset API. (But you probably don't, see below.)

Thanks for the review. What about the imx_spi_update_irq() in the imx_spi_reset()? Should we remove that from the imx_spi_reset() as well?

Quoted 4 lines

Usually the approach is that the device on the other end of the line is going to reset its state anyway, so there's no need to actively signal an irq line change.

Quoted 8 lines

If this is required only for the case of "guest requested a controller reset via the ECSPI_CONREG register" and not for full system reset, then you can handle that by having an imx_spi_soft_reset() which calls imx_spi_reset() and then does the qemu_set_irq() calls, so full system (power-cycle) reset still goes to imx_spi_reset() but guest-commanded reset via the register interface calls imx_spi_soft_reset().

Regards,
Bin

#4 lore raw

Re: [PATCH v2 2/2] hw/ssi: imx_spi: Disable chip selects in imx_spi_reset()

Quoted 11 lines · Bin Meng
Quoted 8 lines · Peter Maydell

Calling qemu_set_irq() in a device reset function is a bad idea, because you don't know whether the thing on the other end of the IRQ line (a) has already reset before you or
(b) is going to reset after you. If you need to do this then I think you need to convert this device (and perhaps whatever it's connected to) to the 3-phase-reset API. (But you probably don't, see below.)

Quoted 3 lines

Thanks for the review. What about the imx_spi_update_irq() in the imx_spi_reset()? Should we remove that from the imx_spi_reset() as well?

Yes, I think so.

thanks
-- PMM

author
#5 lore raw
In reply to Bin Meng

Hi Alistair, Peter,

Quoted 8 lines · Bin Meng

From: Bin Meng <[email protected]>

Avoid using a magic number (4) everywhere for the number of chip selects supported.

Quoted 8 lines

---

(no changes since v1)

 hw/ssi/imx_spi.c         | 4 ++--
 include/hw/ssi/imx_spi.h | 5 ++++-
 2 files changed, 6 insertions(+), 3 deletions(-)

Ping, not sure who is going to pick up this series?

Regards,
Bin

author
#6 lore raw
Quoted 14 lines · Bin Meng

Hi Alistair, Peter,

Quoted 10 lines · Bin Meng

From: Bin Meng <[email protected]>

Avoid using a magic number (4) everywhere for the number of chip selects supported.

---

Quoted 8 lines
Quoted 6 lines

(no changes since v1)

 hw/ssi/imx_spi.c         | 4 ++--
 include/hw/ssi/imx_spi.h | 5 ++++-
 2 files changed, 6 insertions(+), 3 deletions(-)

Ping, not sure who is going to pick up this series?

Ping?

author
#7 lore raw
Quoted 18 lines · Bin Meng
Quoted 16 lines · Bin Meng

Hi Alistair, Peter,

Quoted 12 lines · Bin Meng

From: Bin Meng <[email protected]>

Avoid using a magic number (4) everywhere for the number of chip selects supported.

---

(no changes since v1)

Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
 hw/ssi/imx_spi.c         | 4 ++--
 include/hw/ssi/imx_spi.h | 5 ++++-
 2 files changed, 6 insertions(+), 3 deletions(-)

Ping, not sure who is going to pick up this series?

Ping?

Ping?

#8 lore raw
In reply to Bin Meng
Quoted 14 lines · Bin Meng

Hi Alistair, Peter,

Quoted 10 lines · Bin Meng

From: Bin Meng <[email protected]>

Avoid using a magic number (4) everywhere for the number of chip selects supported.

---

Quoted 8 lines
Quoted 6 lines

(no changes since v1)

 hw/ssi/imx_spi.c         | 4 ++--
 include/hw/ssi/imx_spi.h | 5 ++++-
 2 files changed, 6 insertions(+), 3 deletions(-)

Ping, not sure who is going to pick up this series?

It should be reviewed by Jean-Christophe and then probably go via the ARM tree.

Alistair

Quoted 4 lines

Regards,
Bin

#9 lore raw
Quoted 18 lines · Alistair Francis
Quoted 16 lines · Bin Meng

Hi Alistair, Peter,

Quoted 12 lines · Bin Meng

From: Bin Meng <[email protected]>

Avoid using a magic number (4) everywhere for the number of chip selects supported.

---

(no changes since v1)

Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
 hw/ssi/imx_spi.c         | 4 ++--
 include/hw/ssi/imx_spi.h | 5 ++++-
 2 files changed, 6 insertions(+), 3 deletions(-)

Ping, not sure who is going to pick up this series?

It should be reviewed by Jean-Christophe and then probably go via the ARM tree.

It doesn't seem to have been sent with a cover letter. Multi-patch patchsets without a cover letter tend to get missed because when I scan through my email looking for patches I should review or pick up I'm looking for either (a) single patches or (b) the 00/nn cover letter email...

thanks
-- PMM

author
#10 lore raw
Quoted 27 lines · Peter Maydell
Quoted 25 lines · Alistair Francis
Quoted 22 lines · Bin Meng

Hi Alistair, Peter,

Quoted 8 lines · Bin Meng

From: Bin Meng <[email protected]>

Avoid using a magic number (4) everywhere for the number of chip selects supported.

Quoted 8 lines

---

(no changes since v1)

 hw/ssi/imx_spi.c         | 4 ++--
 include/hw/ssi/imx_spi.h | 5 ++++-
 2 files changed, 6 insertions(+), 3 deletions(-)

Ping, not sure who is going to pick up this series?

Quoted 7 lines
Quoted 1 line

It should be reviewed by Jean-Christophe and then probably go via the ARM tree.

It doesn't seem to have been sent with a cover letter. Multi-patch patchsets without a cover letter tend to get missed because when I scan through my email looking for patches I should review or pick up I'm looking for either (a) single patches or (b) the 00/nn cover letter email...

Thanks. I will resend it with a cover-letter.

Regards,
Bin

author
#11 lore raw
Quoted 31 lines · Bin Meng
Quoted 29 lines · Peter Maydell
Quoted 26 lines · Alistair Francis
Quoted 14 lines · Bin Meng

Hi Alistair, Peter,

Quoted 10 lines · Bin Meng

From: Bin Meng <[email protected]>

Avoid using a magic number (4) everywhere for the number of chip selects supported.

---

Quoted 8 lines
Quoted 6 lines

(no changes since v1)

 hw/ssi/imx_spi.c         | 4 ++--
 include/hw/ssi/imx_spi.h | 5 ++++-
 2 files changed, 6 insertions(+), 3 deletions(-)

Ping, not sure who is going to pick up this series?

It should be reviewed by Jean-Christophe and then probably go via the ARM tree.

Quoted 7 lines
Quoted 5 lines

It doesn't seem to have been sent with a cover letter. Multi-patch patchsets without a cover letter tend to get missed because when I scan through my email looking for patches I should review or pick up I'm looking for either (a) single patches or (b) the 00/nn cover letter email...

Thanks. I will resend it with a cover-letter.

This series is resent and included in the following series with a cover letter:
http://patchwork.ozlabs.org/project/qemu-devel/list/?series=222931

Regards,
Bin

#12 lore raw
In reply to Alistair Francis
Quoted 18 lines · Alistair Francis
Quoted 16 lines · Bin Meng

Hi Alistair, Peter,

Quoted 12 lines · Bin Meng

From: Bin Meng <[email protected]>

Avoid using a magic number (4) everywhere for the number of chip selects supported.

---

(no changes since v1)

Quoted 8 lines
Quoted 6 lines
Quoted 4 lines
 hw/ssi/imx_spi.c         | 4 ++--
 include/hw/ssi/imx_spi.h | 5 ++++-
 2 files changed, 6 insertions(+), 3 deletions(-)

Ping, not sure who is going to pick up this series?

It should be reviewed by Jean-Christophe and then probably go via the ARM tree.

Ah, I just realised I apparently have hw/ssi/* listed in MAINTAINERS. As this file is specifically mentioned somewhere else in MAINTAINERS and I know very little about IMX I'm going to leave this alone still.

Alistair

Quoted 7 lines

Alistair

Quoted 4 lines

Regards,
Bin