tpm: move infineon driver off pci_dev
PATCH 1/1 tpm: move infineon driver off pci_dev
This patch changes the infineon driver from a pci driver to a pnp driver. This is possible because the infineon chip, unlike the other 1.1 chips, is supporting the 1.2 concepts of registering the device in ACPI.
---
drivers/char/tpm/tpm_infineon.c +38 -89
@@ -14,6 +14,7 @@* License. */ +#include <acpi/acpi_bus.h> #include <linux/pnp.h> #include "tpm.h"
@@ -28,10 +29,9 @@#define TPM_MAX_TRIES 5000#define TPM_INFINEON_DEV_VEN_VALUE 0x15D1 -/* These values will be filled after PnP-call */ +/* These values will be filled after ACPI-call */ static int TPM_INF_DATA =0 ;static int TPM_INF_ADDR =0 ; -static int pnp_registered =0 ;/* TPM header definitions */ enum infineon_tpm_header {
@@ -362,31 +362,10 @@ static const struct pnp_device_id tpm_pn {"IFX0102", 0}, {"", 0} }; -MODULE_DEVICE_TABLE (pnp, tpm_pnp_tbl); -static int __devinit tpm_inf_pnp_probe (struct pnp_dev *dev, +static int __devinit tpm_inf_acpi_probe (struct pnp_dev *dev,const struct pnp_device_id *dev_id) { -if (pnp_port_valid (dev,0 )) { -TPM_INF_ADDR = (pnp_port_start (dev,0 ) &0xff ); -TPM_INF_DATA = ((TPM_INF_ADDR +1 ) &0xff ); - tpm_inf.base =pnp_port_start (dev,1 ); -dev_info (&dev->dev,"Found %s with ID %s\n" , - dev->name, dev_id->id); -return 0 ; - } -return -ENODEV ; -} - -static struct pnp_driver tpm_inf_pnp = { - .name ="tpm_inf_pnp" , - .id_table = tpm_pnp_tbl, - .probe = tpm_inf_pnp_probe, -}; - -static int __devinit tpm_inf_probe (struct pci_dev *pci_dev, -const struct pci_device_id *pci_id) -{int rc =0 ;u8 iol, ioh;int vendorid[2 ];
@@ -394,30 +373,19 @@ static int __devinit tpm_inf_probe(strucint productid[2 ];char chipname[20 ]; - rc =pci_enable_device (pci_dev); -if (rc) -return rc; - -dev_info (&pci_dev->dev,"LPC-bus found at 0x%x\n" , pci_id->device); - -/* read IO-ports from PnP */ - rc =pnp_register_driver (&tpm_inf_pnp); -if (rc <0 ) { -dev_err (&pci_dev->dev, -"Error %x from pnp_register_driver!\n" ,rc); - goto error2; - } -if (!rc) { -dev_info (&pci_dev->dev,"No Infineon TPM found!\n" ); - goto error; - }else { - pnp_registered =1 ; - } + +/* read IO-ports from ACPI */ +TPM_INF_ADDR = (pnp_port_start (dev,0 ) &0xff ); +TPM_INF_DATA = ((TPM_INF_ADDR +1 ) &0xff ); + tpm_inf.base =pnp_port_start (dev,1 ); +dev_info (&dev->dev,"Found %s with ID %s\n" , + dev->name, dev_id->id); +if (!((tpm_inf.base >>8 ) &0xff )) + tpm_inf.base =0 ;/* Make sure, we have received valid config ports */ if (!TPM_INF_ADDR ) { -dev_err (&pci_dev->dev,"No valid IO-ports received!\n" ); - goto error; +return -EIO ; }/* query chip for its vendor, its version number a.s.o. */
@@ -437,21 +405,22 @@ static int __devinit tpm_inf_probe(strucswitch ((productid[0 ] <<8 ) | productid[1 ]) {case 6 : -snprintf (chipname,sizeof (chipname)," (SLD 9630 TT 1.1)" ); +sprintf (chipname," (SLD 9630 TT 1.1)" );break ;case 11 : -snprintf (chipname,sizeof (chipname)," (SLB 9635 TT 1.2)" ); +sprintf (chipname," (SLB 9635 TT 1.2)" );break ;default : -snprintf (chipname,sizeof (chipname)," (unknown chip)" ); +sprintf (chipname," (unknown chip)" );break ; } + chipname[19 ] =0 ;if ((vendorid[0 ] <<8 | vendorid[1 ]) == (TPM_INFINEON_DEV_VEN_VALUE )) {if (tpm_inf.base ==0 ) { -dev_err (&pci_dev->dev,"No IO-ports found!\n" ); - goto error; +dev_err (&dev->dev,"No IO-ports found!\n" ); +return -EIO ; }/* configure TPM with IO-ports */ outb (IOLIMH ,TPM_INF_ADDR );
@@ -466,10 +435,10 @@ static int __devinit tpm_inf_probe(struc iol =inb (TPM_INF_DATA );if ((ioh <<8 | iol) != tpm_inf.base) { -dev_err (&pci_dev->dev, +dev_err (&dev->dev,"Could not set IO-ports to %04x\n" , tpm_inf.base); - goto error; +return -EIO ; }/* activate register */
@@ -481,7 +450,7 @@ static int __devinit tpm_inf_probe(strucoutb (RESET_LP_IRQC_DISABLE , tpm_inf.base +CMD );/* Finally, we're done, print some infos */ -dev_info (&pci_dev->dev,"TPM found: " +dev_info (&dev->dev,"TPM found: " "config base 0x%x, " "io base 0x%x, " "chip version %02x%02x, "
@@ -494,62 +463,42 @@ static int __devinit tpm_inf_probe(struc vendorid[0 ], vendorid[1 ], productid[0], productid[1], chipname); - rc =tpm_register_hardware (&pci_dev->dev, &tpm_inf); -if (rc <0 ) - goto error; + rc =tpm_register_hardware (&dev->dev, &tpm_inf); +if (rc <0 ) { +return -ENODEV ; + }return 0 ; }else { -dev_info (&pci_dev->dev,"No Infineon TPM found!\n" ); -error: -pnp_unregister_driver (&tpm_inf_pnp); -error2: -pci_disable_device (pci_dev); - pnp_registered =0 ; +dev_info (&dev->dev,"No Infineon TPM found!\n" );return -ENODEV ; } } -static __devexit void tpm_inf_remove(struct pci_dev* pci_dev) +static __devexit void tpm_inf_remove(struct pnp_dev* dev) { -struct tpm_chip * chip =pci_get_drvdata (pci_dev); +struct tpm_chip * chip =pnp_get_drvdata (dev);if ( chip )tpm_remove_hardware (chip->dev); } -static struct pci_device_id tpm_pci_tbl[]__devinitdata = { - {PCI_DEVICE (PCI_VENDOR_ID_INTEL ,PCI_DEVICE_ID_INTEL_82801BA_0 )}, - {PCI_DEVICE (PCI_VENDOR_ID_INTEL ,PCI_DEVICE_ID_INTEL_82801CA_12 )}, - {PCI_DEVICE (PCI_VENDOR_ID_INTEL ,PCI_DEVICE_ID_INTEL_82801DB_0 )}, - {PCI_DEVICE (PCI_VENDOR_ID_INTEL ,PCI_DEVICE_ID_INTEL_82801DB_12 )}, - {PCI_DEVICE (PCI_VENDOR_ID_INTEL ,PCI_DEVICE_ID_INTEL_82801EB_0 )}, - {PCI_DEVICE (PCI_VENDOR_ID_INTEL ,PCI_DEVICE_ID_INTEL_ICH6_0 )}, - {PCI_DEVICE (PCI_VENDOR_ID_INTEL ,PCI_DEVICE_ID_INTEL_ICH6_1 )}, - {PCI_DEVICE (PCI_VENDOR_ID_INTEL ,PCI_DEVICE_ID_INTEL_ICH6_2 )}, - {0 ,} -}; - -MODULE_DEVICE_TABLE (pci, tpm_pci_tbl); - -static struct pci_driver inf_pci_driver = { - .name ="tpm_inf" , - .id_table = tpm_pci_tbl, - .probe = tpm_inf_probe, - .remove =__devexit_p (tpm_inf_remove), - .suspend = tpm_pm_suspend, - .resume = tpm_pm_resume, +static struct pnp_driver tpm_inf_pnp = { + .name ="tpm_inf_pnp" , + .id_table = tpm_pnp_tbl, + .probe = tpm_inf_acpi_probe, + .remove = tpm_inf_remove, + .driver.suspend = tpm_pm_suspend, + .driver.resume = tpm_pm_resume, };static int __init init_inf (void ) { -return pci_register_driver (&inf_pci_driver); +return pnp_register_driver (&tpm_inf_pnp); } static void __exit cleanup_inf(void) { -if (pnp_registered) -pnp_unregister_driver (&tpm_inf_pnp); -pci_unregister_driver (&inf_pci_driver); +pnp_unregister_driver (&tpm_inf_pnp); }module_init (init_inf);
Hi,
nice to see that the tpm framework was cleanned.
Le Tue, 25 Oct 2005 10:22:18 -0500, Kylene Jo Hall a écrit :
drivers/char/tpm/tpm_infineon.c line 17
Quoted 9 lines
---
drivers/char/tpm/tpm_infineon.c +1
-0@@ -14,6 +14,7 @@* License. */ +#include <acpi/acpi_bus.h>
Why is it needed ?
drivers/char/tpm/tpm_infineon.c line 32
Quoted 2 lines
drivers/char/tpm/tpm_infineon.c +1
-1-/* These values will be filled after PnP-call */ +/* These values will be filled after ACPI-call */
Why not keep PnP ?
drivers/char/tpm/tpm_infineon.c static const struct pnp_device_id tpm_pn · line 365
Quoted 2 lines
drivers/char/tpm/tpm_infineon.c +0
-1-MODULE_DEVICE_TABLE (pnp, tpm_pnp_tbl);
Why removing that ?
This will allow hotplug and udev to auto load the module.
drivers/char/tpm/tpm_infineon.c static int __devinit tpm_inf_probe(struc · line 379
Quoted 4 lines
drivers/char/tpm/tpm_infineon.c +4
-0+ +/* read IO-ports from ACPI */ +TPM_INF_ADDR = (pnp_port_start (dev,0 ) &0xff ); +TPM_INF_DATA = ((TPM_INF_ADDR +1 ) &0xff );
No need to set mask, this is already done by pnp_port_start. And I'll keep PNP instead of ACPI.
drivers/char/tpm/tpm_infineon.c static int __devinit tpm_inf_probe(struc · line 380
Quoted 1 line
drivers/char/tpm/tpm_infineon.c +1
-0+ tpm_inf.base =pnp_port_start (dev,1 );
Can't you be coherent ?
why not using tpm_inf.addr
or TPM_INF_BASE ?
drivers/char/tpm/tpm_infineon.c static int __devinit tpm_inf_probe(struc · line 386
Quoted 6 lines
drivers/char/tpm/tpm_infineon.c +4
-0+dev_info (&dev->dev,"Found %s with ID %s\n" , + dev->name, dev_id->id); +if (!((tpm_inf.base >>8 ) &0xff )) + tpm_inf.base =0 ;/* Make sure, we have received valid config ports */
You should also do :pnp_port_flags(device, 0) & IORESOURCE_DISABLED) in order to check the resources.
drivers/char/tpm/tpm_infineon.c static int __devinit tpm_inf_probe(struc · line 489
Quoted 2 lines
drivers/char/tpm/tpm_infineon.c +2
-0+ .probe = tpm_inf_acpi_probe, + .remove = tpm_inf_remove,
Not coherent : acpi vs nothing.
Again prefer pnp instead of acpi.
Matthieu
Matthieu,
Thanks for the comments. I mistakenly generated this patch which overwrote some of the changes that Marcel had submitted. I am working on going back to the pnp names and I'll additionally take these comments into consideration. Please hold off on applying this patch.
Thanks,
Kylie
Quoted 71 lines · Matthieu CASTET
Hi,
nice to see that the tpm framework was cleanned.
drivers/char/tpm/tpm_infineon.c line 17
Quoted 9 lines · Kylene Jo Hall
---
drivers/char/tpm/tpm_infineon.c +1
-0@@ -14,6 +14,7 @@* License. */ +#include <acpi/acpi_bus.h> Why is it needed ?
drivers/char/tpm/tpm_infineon.c line 32
Quoted 2 lines
drivers/char/tpm/tpm_infineon.c +1
-1-/* These values will be filled after PnP-call */ +/* These values will be filled after ACPI-call */ Why not keep PnP ?
drivers/char/tpm/tpm_infineon.c static const struct pnp_device_id tpm_pn · line 365
Quoted 2 lines
drivers/char/tpm/tpm_infineon.c +0
-1-MODULE_DEVICE_TABLE (pnp, tpm_pnp_tbl);Why removing that ?
This will allow hotplug and udev to auto load the module.drivers/char/tpm/tpm_infineon.c static int __devinit tpm_inf_probe(struc · line 379
Quoted 4 lines
drivers/char/tpm/tpm_infineon.c +4
-0+ +/* read IO-ports from ACPI */ +TPM_INF_ADDR = (pnp_port_start (dev,0 ) &0xff ); +TPM_INF_DATA = ((TPM_INF_ADDR +1 ) &0xff );No need to set mask, this is already done by
pnp_port_start. And I'll keep PNP instead of ACPI.drivers/char/tpm/tpm_infineon.c static int __devinit tpm_inf_probe(struc · line 380
Quoted 1 line
drivers/char/tpm/tpm_infineon.c +1
-0+ tpm_inf.base =pnp_port_start (dev,1 );Can't you be coherent ?
why not using tpm_inf.addr
orTPM_INF_BASE?drivers/char/tpm/tpm_infineon.c static int __devinit tpm_inf_probe(struc · line 386
Quoted 6 lines
drivers/char/tpm/tpm_infineon.c +4
-0+dev_info (&dev->dev,"Found %s with ID %s\n" , + dev->name, dev_id->id); +if (!((tpm_inf.base >>8 ) &0xff )) + tpm_inf.base =0 ;/* Make sure, we have received valid config ports */ You should also do :
pnp_port_flags(device, 0)&IORESOURCE_DISABLED) in order to check the resources.drivers/char/tpm/tpm_infineon.c static int __devinit tpm_inf_probe(struc · line 489
Quoted 2 lines
drivers/char/tpm/tpm_infineon.c +2
-0+ .probe = tpm_inf_acpi_probe, + .remove = tpm_inf_remove,Not coherent : acpi vs nothing.
Again prefer pnp instead of acpi.Matthieu
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/