Skip to contentSkip to conversation

usb: Prevent using reserved registers on DM36x usb

RFC PATCH Superseded Andrew Murray 2 participants 4 messages Sep 13, 2013 – Sep 29, 2013

Subsystems (2)

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

RFC PATCH 1/1 usb: Prevent using reserved registers on DM36x usb

The musb driver defines and uses MUSB_CSR0_H_DIS_PING, however this bit is reserved on the DM36x. Thus this patch ensures that the reserved bit is not accesssed.

It has been observed that some USB devices will fail to enumerate with errors such as 'error in inquiry' without this patch.

See http://www.ti.com/litv/pdf/sprufh9a for details.

---
 drivers/usb/musb/musb_core.h |    4 ++++
 1 file changed, 4 insertions(+)
drivers/usb/musb/musb_core.h +4 -0
index ec8a038..c9a9d66 100644
@@ -216,7 +216,11 @@ struct musb_regs {
 #define MUSB_CSR0_P_SENTSTALL	0x0004
 
 /* CSR0 in Host mode */
+#if defined(CONFIG_SOC_DM365)
+#define MUSB_CSR0_H_DIS_PING		0x0000
+#else
 #define MUSB_CSR0_H_DIS_PING		0x0800
+#endif
 #define MUSB_CSR0_H_WR_DATATOGGLE	0x0400	/* Set to allow setting: */
 #define MUSB_CSR0_H_DATATOGGLE		0x0200	/* Data toggle control */
 #define MUSB_CSR0_H_NAKTIMEOUT		0x0080
Signature · 1 line
1.7.9.5
#1 lore raw

RFC PATCH 1/1 usb: Prevent using reserved registers on DM36x usb

Dear Andrew Murray,

Quoted 21 lines

The musb driver defines and uses MUSB_CSR0_H_DIS_PING, however this bit is reserved on the DM36x. Thus this patch ensures that the reserved bit is not accesssed.

It has been observed that some USB devices will fail to enumerate with errors such as 'error in inquiry' without this patch.

See http://www.ti.com/litv/pdf/sprufh9a for details.

---
 drivers/usb/musb/musb_core.h |    4 ++++
 1 file changed, 4 insertions(+)
drivers/usb/musb/musb_core.h +0 -0
index ec8a038..c9a9d66 100644

drivers/usb/musb/musb_core.h struct musb_regs { · line 225

Quoted 10 lines
drivers/usb/musb/musb_core.h +4 -0
@@ -216,7 +216,11 @@ struct musb_regs {
 #define MUSB_CSR0_P_SENTSTALL	0x0004
 
 /* CSR0 in Host mode */
+#if defined(CONFIG_SOC_DM365)
+#define MUSB_CSR0_H_DIS_PING		0x0000
+#else
 #define MUSB_CSR0_H_DIS_PING		0x0800
+#endif
 #define MUSB_CSR0_H_WR_DATATOGGLE	0x0400	/* Set to allow setting: */
 #define MUSB_CSR0_H_DATATOGGLE		0x0200	/* Data toggle control 

*/

drivers/usb/musb/musb_core.h struct musb_regs { · line 226

Quoted 1 line
drivers/usb/musb/musb_core.h +0 -0
 #define MUSB_CSR0_H_NAKTIMEOUT		0x0080

It would be much nicer if you avoided using this bit in musb_hcd.c instead of hacking it like this.

Best regards,
Marek Vasut

#2 lore raw

RFC PATCH 1/1 usb: Prevent using reserved registers on DM36x usb

Dear Andrew Murray,

Quoted 41 lines
Quoted 28 lines

The musb driver defines and uses MUSB_CSR0_H_DIS_PING, however this bit is reserved on the DM36x. Thus this patch ensures that the reserved bit is not accesssed.

It has been observed that some USB devices will fail to enumerate with errors such as 'error in inquiry' without this patch.

See http://www.ti.com/litv/pdf/sprufh9a for details.

---

 drivers/usb/musb/musb_core.h |    4 ++++
 1 file changed, 4 insertions(+)
drivers/usb/musb/musb_core.h +0 -0
index ec8a038..c9a9d66 100644
@@ -216,7 +216,11 @@ struct musb_regs {
 
 #define MUSB_CSR0_P_SENTSTALL	0x0004
 
 /* CSR0 in Host mode */
 

drivers/usb/musb/musb_core.h struct musb_regs {

Quoted 10 lines
drivers/usb/musb/musb_core.h +4 -0
+#if defined(CONFIG_SOC_DM365)
+#define MUSB_CSR0_H_DIS_PING		0x0000
+#else
 
 #define MUSB_CSR0_H_DIS_PING		0x0800
 
+#endif
 
 #define MUSB_CSR0_H_WR_DATATOGGLE	0x0400	/* Set to allow setting: */
 #define MUSB_CSR0_H_DATATOGGLE		0x0200	/* Data toggle control

*/

Quoted 4 lines
Quoted 1 line

#define MUSB_CSR0_H_NAKTIMEOUT 0x0080

It would be much nicer if you avoided using this bit in musb_hcd.c instead of hacking it like this.

Bump? Will we get a V2 here ?

Best regards,
Marek Vasut

#3 lore raw

RFC PATCH 1/1 usb: Prevent using reserved registers on DM36x usb

Quoted 4 lines · Marek Vasut
Quoted 2 lines

It would be much nicer if you avoided using this bit in musb_hcd.c instead of hacking it like this.

Bump? Will we get a V2 here ?

I've posted a v2 on the list, apologies for the slow response.

Andrew Murray