usb: gadget: udc: atmel: Endpoint allocation scheme fixes
Subsystems (3)
[email protected]
maintainerv3 cover usb: gadget: udc: atmel: Endpoint allocation scheme fixes
From: Cristian Birsan <[email protected]>
This patch series provides fixes, based on the feedback received on the mailing list, for the following:
- fifo table parameters validation against device tree values
- coding style
- message display for EP configuration error
- Kconfig comments for fifo_mode=0
Changes since v1:
- Removed static for usba_config_fifo_table() function from "Check fifo configuration values against device tree" patch
Changes since v2:
- Use shorter warning message if the fifo table size is greather than device tree value
- replace dev_warn() with dev_err() for hardware fifo allocation error
Cristian Birsan (4):
usb: gadget: udc: atmel: Check fifo configuration values against
device tree
usb: gadget: udc: atmel: Minor code cleanup
usb: gadget: udc: atmel: Use dev_err() to display EP configuration
error
usb: gadget: udc: atmel: Update Kconfig help for fifo_mode = 0
drivers/usb/gadget/udc/Kconfig | 5 ++-- drivers/usb/gadget/udc/atmel_usba_udc.c | 49 ++++++++++++++++++++++----------- 2 files changed, 36 insertions(+), 18 deletions(-)
Signature · 1 line
2.7.4
[email protected]
maintainerPATCH v3 1/4 usb: gadget: udc: atmel: Check fifo configuration values against device tree
From: Cristian Birsan <[email protected]>
Check fifo configuration values against device tree values for endpoint fifo in auto configuration mode (fifo_mode=0).
--- drivers/usb/gadget/udc/atmel_usba_udc.c | 24 ++++++++++++++++++++++-- 1 file changed, 22 insertions(+), 2 deletions(-)
drivers/usb/gadget/udc/atmel_usba_udc.c +22 -2
@@ -2118,14 +2118,34 @@ static struct usba_ep * atmel_udc_of_init(struct platform_device *pdev,dev_err (&pdev->dev,"of_probe: fifo-size error(%d)\n" , ret); goto err; } - ep->fifo_size = fifo_mode ? udc->fifo_cfg[i].fifo_size : val; +if (fifo_mode) { +if (val < udc->fifo_cfg[i].fifo_size) { +dev_warn (&pdev->dev, +"Using max fifo-size value from DT\n" ); + ep->fifo_size = val; + }else { + ep->fifo_size = udc->fifo_cfg[i].fifo_size; + } + }else { + ep->fifo_size = val; + } ret =of_property_read_u32 (pp,"atmel,nb-banks" , &val);if (ret) {dev_err (&pdev->dev,"of_probe: nb-banks error(%d)\n" , ret); goto err; } - ep->nr_banks = fifo_mode ? udc->fifo_cfg[i].nr_banks : val; +if (fifo_mode) { +if (val < udc->fifo_cfg[i].nr_banks) { +dev_warn (&pdev->dev, +"Using max nb-banks value from DT\n" ); + ep->nr_banks = val; + }else { + ep->nr_banks = udc->fifo_cfg[i].nr_banks; + } + }else { + ep->nr_banks = val; + } ep->can_dma =of_property_read_bool (pp,"atmel,can-dma" ); ep->can_isoc =of_property_read_bool (pp,"atmel,can-isoc" );
Signature · 1 line
2.7.4
[email protected]
maintainerPATCH v3 2/4 usb: gadget: udc: atmel: Minor code cleanup
In reply to [email protected]From: Cristian Birsan <[email protected]>
Minor code cleanup based on feedback received on mailinglist.
--- drivers/usb/gadget/udc/atmel_usba_udc.c | 17 +++++++---------- 1 file changed, 7 insertions(+), 10 deletions(-)
drivers/usb/gadget/udc/atmel_usba_udc.c +7 -10
@@ -321,7 +321,6 @@ static inline void usba_cleanup_debugfs(struct usba_udc *udc)static ushort fifo_mode; -/* "modprobe ... fifo_mode=1" etc */ module_param (fifo_mode, ushort,0x0 );MODULE_PARM_DESC (fifo_mode,"Endpoint configuration mode" );
@@ -371,7 +370,7 @@ static struct usba_fifo_cfg mode_4_cfg[] = { };/* Add additional configurations here */ -int usba_config_fifo_table (struct usba_udc *udc) +static int usba_config_fifo_table (struct usba_udc *udc) {int n;
@@ -1076,11 +1075,9 @@ static int atmel_usba_start(struct usb_gadget *gadget, struct usb_gadget_driver *driver);static int atmel_usba_stop (struct usb_gadget *gadget); -static struct usb_ep *atmel_usba_match_ep ( -struct usb_gadget *gadget, -struct usb_endpoint_descriptor *desc, -struct usb_ss_ep_comp_descriptor *ep_comp -) +static struct usb_ep *atmel_usba_match_ep (struct usb_gadget *gadget, +struct usb_endpoint_descriptor *desc, +struct usb_ss_ep_comp_descriptor *ep_comp) {struct usb_ep *_ep;struct usba_ep *ep;
@@ -1100,7 +1097,6 @@ static struct usb_ep *atmel_usba_match_ep( ep =to_usba_ep (_ep);switch (usb_endpoint_type (desc)) { -case USB_ENDPOINT_XFER_CONTROL :break ;
@@ -1141,7 +1137,7 @@ static struct usb_ep *atmel_usba_match_ep( ep->udc->configured_ep++; } -return _ep; +return _ep; }static const struct usb_gadget_ops usba_udc_ops = {
@@ -2089,8 +2085,9 @@ static struct usba_ep * atmel_udc_of_init(struct platform_device *pdev,while ((pp =of_get_next_child (np, pp))) udc->num_ep++; udc->configured_ep =1 ; - }else + }else { udc->num_ep =usba_config_fifo_table (udc); + } eps =devm_kzalloc (&pdev->dev,sizeof (struct usba_ep ) * udc->num_ep,GFP_KERNEL );
Signature · 1 line
2.7.4
[email protected]
maintainerPATCH v3 3/4 usb: gadget: udc: atmel: Use dev_err() to display EP configuration error
In reply to [email protected]From: Cristian Birsan <[email protected]>
Use dev_err() to display EP configuration error to avoid silent failure.
--- drivers/usb/gadget/udc/atmel_usba_udc.c | 8 ++++---- 1 file changed, 4 insertions(+), 4 deletions(-)
drivers/usb/gadget/udc/atmel_usba_udc.c +4 -4
@@ -1851,8 +1851,8 @@ static irqreturn_t usba_udc_irq(int irq, void *devid)* but it's clearly harmless... */ if (!(usba_ep_readl (ep0,CFG ) &USBA_EPT_MAPPED )) -dev_dbg (&udc->pdev->dev, -"ODD: EP0 configuration is invalid!\n" ); +dev_err (&udc->pdev->dev, +"ODD: EP0 configuration is invalid!\n" );/* Preallocate other endpoints */ n = fifo_mode ? udc->num_ep : udc->configured_ep;
@@ -1860,8 +1860,8 @@ static irqreturn_t usba_udc_irq(int irq, void *devid) ep = &udc->usba_ep[i];usba_ep_writel (ep,CFG , ep->ept_cfg);if (!(usba_ep_readl (ep,CFG ) &USBA_EPT_MAPPED )) -dev_dbg (&udc->pdev->dev, -"ODD: EP%d configuration is invalid!\n" , i); +dev_err (&udc->pdev->dev, +"ODD: EP%d configuration is invalid!\n" , i); } }
Signature · 1 line
2.7.4
[email protected]
maintainerPATCH v3 4/4 usb: gadget: udc: atmel: Update Kconfig help for fifo_mode = 0
In reply to [email protected]From: Cristian Birsan <[email protected]>
Update Kconfig help for fifo_mode = 0 to explain the behavior better.
--- drivers/usb/gadget/udc/Kconfig | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
drivers/usb/gadget/udc/Kconfig +3 -2
@@ -62,8 +62,9 @@ config USB_ATMEL_USBA The fifo_mode parameter is used to select endpoint allocation mode. fifo_mode = 0 is used to let the driver autoconfigure the endpoints. - In this case 2 banks are allocated for isochronous endpoints and - only one bank is allocated for the rest of the endpoints. + In this case, for ep1 2 banks are allocated if it works in isochronous + mode and only 1 bank otherwise. For the rest of the endpoints + only 1 bank is allocated. fifo_mode = 1 is a generic maximum fifo size (1024 bytes) configuration allowing the usage of ep1 - ep6
Signature · 1 line
2.7.4
Re: [PATCH linux-next v3 2/4] usb: gadget: udc: atmel: Minor code cleanup
In reply to [email protected]Hi,
Quoted 6 lines · cristian.birsan at microchip.com
From: Cristian Birsan <[email protected]>
Minor code cleanup based on feedback received on mailinglist.
Signed-off-by Cristian Birsan <[email protected]>Acked-by Nicolas Ferre <[email protected]>
patch 1 applied fine, patch 2 didn't. Please rebase on testing/next
Signature · 8 lines
balbi -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 832 bytes Desc: not available URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20170405/23ce3984/attachment-0001.sig>
m91496
maintainerRe: [PATCH linux-next v3 2/4] usb: gadget: udc: atmel: Minor code cleanup
Hi,
The patch 2 does not apply because it's already on testing/next.
I'll send a new patch series version that will apply clean on top of testing branch.
Regards,
Cristi
Quoted 13 lines · Felipe Balbi
Hi,
Quoted 6 lines · cristian.birsan at microchip.com
From: Cristian Birsan <[email protected]>
Minor code cleanup based on feedback received on mailinglist.
Signed-off-by Cristian Birsan <[email protected]>Acked-by Nicolas Ferre <[email protected]>patch 1 applied fine, patch 2 didn't. Please rebase on testing/next