* [RFC 2/7] Unite all PCI-e on 85xx and 86xx under one codebase
@ 2007-02-16 2:44 Andy Fleming
2007-02-16 3:43 ` Kumar Gala
0 siblings, 1 reply; 6+ messages in thread
From: Andy Fleming @ 2007-02-16 2:44 UTC (permalink / raw)
To: linuxppc-dev, Kumar Gala, jdl
Add 85xx PCI express memory map register.
Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
Acked-by: Andy Fleming <afleming@freescale.com>
---
include/asm-ppc/immap_85xx.h | 75 ++++++++++++++++++++++++++++++++++++++++++
1 files changed, 75 insertions(+), 0 deletions(-)
diff --git a/include/asm-ppc/immap_85xx.h b/include/asm-ppc/immap_85xx.h
index b35cd6a..adef8c7 100644
--- a/include/asm-ppc/immap_85xx.h
+++ b/include/asm-ppc/immap_85xx.h
@@ -87,6 +87,81 @@ typedef struct ccsr_pci {
char res19[472];
} ccsr_pci_t;
+/* PCI Express Registers */
+typedef struct ccsr_pex {
+ uint pex_config_addr; /* 0x.000 - PCI Express Configuration Address Register */
+ uint pex_config_data; /* 0x.004 - PCI Express Configuration Data Register */
+ char res1[4];
+ uint pex_otb_cpl_tor; /* 0x.00c - PCI Express Outbound completion timeout register */
+ uint pex_conf_tor; /* 0x.010 - PCI Express configuration timeout register */
+ char res2[12];
+ uint pex_pme_mes_dr; /* 0x.020 - PCI Express PME and message detect register */
+ uint pex_pme_mes_disr; /* 0x.024 - PCI Express PME and message disable register */
+ uint pex_pme_mes_ier; /* 0x.028 - PCI Express PME and message interrupt enable register */
+ uint pex_pmcr; /* 0x.02c - PCI Express power management command register */
+ char res3[3024];
+ uint pexotar0; /* 0x.c00 - PCI Express outbound translation address register 0 */
+ uint pexotear0; /* 0x.c04 - PCI Express outbound translation extended address register 0*/
+ char res4[8];
+ uint pexowar0; /* 0x.c10 - PCI Express outbound window attributes register 0*/
+ char res5[12];
+ uint pexotar1; /* 0x.c20 - PCI Express outbound translation address register 1 */
+ uint pexotear1; /* 0x.c24 - PCI Express outbound translation extended address register 1*/
+ uint pexowbar1; /* 0x.c28 - PCI Express outbound window base address register 1*/
+ char res6[4];
+ uint pexowar1; /* 0x.c30 - PCI Express outbound window attributes register 1*/
+ char res7[12];
+ uint pexotar2; /* 0x.c40 - PCI Express outbound translation address register 2 */
+ uint pexotear2; /* 0x.c44 - PCI Express outbound translation extended address register 2*/
+ uint pexowbar2; /* 0x.c48 - PCI Express outbound window base address register 2*/
+ char res8[4];
+ uint pexowar2; /* 0x.c50 - PCI Express outbound window attributes register 2*/
+ char res9[12];
+ uint pexotar3; /* 0x.c60 - PCI Express outbound translation address register 3 */
+ uint pexotear3; /* 0x.c64 - PCI Express outbound translation extended address register 3*/
+ uint pexowbar3; /* 0x.c68 - PCI Express outbound window base address register 3*/
+ char res10[4];
+ uint pexowar3; /* 0x.c70 - PCI Express outbound window attributes register 3*/
+ char res11[12];
+ uint pexotar4; /* 0x.c80 - PCI Express outbound translation address register 4 */
+ uint pexotear4; /* 0x.c84 - PCI Express outbound translation extended address register 4*/
+ uint pexowbar4; /* 0x.c88 - PCI Express outbound window base address register 4*/
+ char res12[4];
+ uint pexowar4; /* 0x.c90 - PCI Express outbound window attributes register 4*/
+ char res13[12];
+ char res14[256];
+ uint pexitar3; /* 0x.da0 - PCI Express inbound translation address register 3 */
+ char res15[4];
+ uint pexiwbar3; /* 0x.da8 - PCI Express inbound window base address register 3 */
+ uint pexiwbear3; /* 0x.dac - PCI Express inbound window base extended address register 3 */
+ uint pexiwar3; /* 0x.db0 - PCI Express inbound window attributes register 3 */
+ char res16[12];
+ uint pexitar2; /* 0x.dc0 - PCI Express inbound translation address register 2 */
+ char res17[4];
+ uint pexiwbar2; /* 0x.dc8 - PCI Express inbound window base address register 2 */
+ uint pexiwbear2; /* 0x.dcc - PCI Express inbound window base extended address register 2 */
+ uint pexiwar2; /* 0x.dd0 - PCI Express inbound window attributes register 2 */
+ char res18[12];
+ uint pexitar1; /* 0x.de0 - PCI Express inbound translation address register 2 */
+ char res19[4];
+ uint pexiwbar1; /* 0x.de8 - PCI Express inbound window base address register 2 */
+ uint pexiwbear1; /* 0x.dec - PCI Express inbound window base extended address register 2 */
+ uint pexiwar1; /* 0x.df0 - PCI Express inbound window attributes register 2 */
+ char res20[12];
+ uint pex_err_dr; /* 0x.e00 - PCI Express error detect register */
+ char res21[4];
+ uint pex_err_en; /* 0x.e08 - PCI Express error interrupt enable register */
+ char res22[4];
+ uint pex_err_disr; /* 0x.e10 - PCI Express error disable register */
+ char res23[12];
+ uint pex_err_cap_stat; /* 0x.e20 - PCI Express error capture status register */
+ char res24[4];
+ uint pex_err_cap_r0; /* 0x.e28 - PCI Express error capture register 0 */
+ uint pex_err_cap_r1; /* 0x.e2c - PCI Express error capture register 0 */
+ uint pex_err_cap_r2; /* 0x.e30 - PCI Express error capture register 0 */
+ uint pex_err_cap_r3; /* 0x.e34 - PCI Express error capture register 0 */
+} ccsr_pex_t;
+
/* Global Utility Registers */
typedef struct ccsr_guts {
uint porpllsr; /* 0x.0000 - POR PLL Ratio Status Register */
--
1.4.4
^ permalink raw reply related [flat|nested] 6+ messages in thread
* Re: [RFC 2/7] Unite all PCI-e on 85xx and 86xx under one codebase
2007-02-16 2:44 [RFC 2/7] Unite all PCI-e on 85xx and 86xx under one codebase Andy Fleming
@ 2007-02-16 3:43 ` Kumar Gala
2007-02-26 6:41 ` Zang Roy-r61911
0 siblings, 1 reply; 6+ messages in thread
From: Kumar Gala @ 2007-02-16 3:43 UTC (permalink / raw)
To: Andy Fleming; +Cc: linuxppc-dev, jdl
On Feb 15, 2007, at 8:44 PM, Andy Fleming wrote:
> Add 85xx PCI express memory map register.
>
> Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
> Acked-by: Andy Fleming <afleming@freescale.com>
> ---
> include/asm-ppc/immap_85xx.h | 75 +++++++++++++++++++++++++++++++
> +++++++++++
> 1 files changed, 75 insertions(+), 0 deletions(-)
Are the 86xx and 85xx structs the same? If so, can we pull them out
of immap_*.h into a sysdev/fsl_pcie.h
(and lose the typedef while you're at it).
- k
^ permalink raw reply [flat|nested] 6+ messages in thread
* RE: [RFC 2/7] Unite all PCI-e on 85xx and 86xx under one codebase
2007-02-16 3:43 ` Kumar Gala
@ 2007-02-26 6:41 ` Zang Roy-r61911
2007-02-26 14:52 ` Kumar Gala
0 siblings, 1 reply; 6+ messages in thread
From: Zang Roy-r61911 @ 2007-02-26 6:41 UTC (permalink / raw)
To: Kumar Gala, Fleming Andy-afleming; +Cc: linuxppc-dev, jdl
=20
> On Feb 15, 2007, at 8:44 PM, Andy Fleming wrote:
>=20
> > Add 85xx PCI express memory map register.
> >
> > Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
> > Acked-by: Andy Fleming <afleming@freescale.com>
> > ---
> > include/asm-ppc/immap_85xx.h | 75=20
> +++++++++++++++++++++++++++++++=20
> > +++++++++++
> > 1 files changed, 75 insertions(+), 0 deletions(-)
>=20
> Are the 86xx and 85xx structs the same? If so, can we pull=20
> them out of immap_*.h into a sysdev/fsl_pcie.h
>=20
> (and lose the typedef while you're at it).
They have same structs. But I do not think it is proper to move them
into =20
sysdev/fsl_pcie.h now. sysdev/fsl_pcie.c does not use this header file.
plaforms/85xx/pci.c use it to set up memory map windows.=20
In fact, there is little difference between immap_85xx.h and
immap_86xx.h.=20
After the migration from ppc to powerpc, we can unify them together.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [RFC 2/7] Unite all PCI-e on 85xx and 86xx under one codebase
2007-02-26 6:41 ` Zang Roy-r61911
@ 2007-02-26 14:52 ` Kumar Gala
2007-02-26 15:52 ` Zang Roy-r61911
0 siblings, 1 reply; 6+ messages in thread
From: Kumar Gala @ 2007-02-26 14:52 UTC (permalink / raw)
To: Zang Roy-r61911; +Cc: linuxppc-dev, jdl
On Feb 26, 2007, at 12:41 AM, Zang Roy-r61911 wrote:
>
>> On Feb 15, 2007, at 8:44 PM, Andy Fleming wrote:
>>
>>> Add 85xx PCI express memory map register.
>>>
>>> Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
>>> Acked-by: Andy Fleming <afleming@freescale.com>
>>> ---
>>> include/asm-ppc/immap_85xx.h | 75
>> +++++++++++++++++++++++++++++++
>>> +++++++++++
>>> 1 files changed, 75 insertions(+), 0 deletions(-)
>>
>> Are the 86xx and 85xx structs the same? If so, can we pull
>> them out of immap_*.h into a sysdev/fsl_pcie.h
>>
>> (and lose the typedef while you're at it).
> They have same structs. But I do not think it is proper to move them
> into
> sysdev/fsl_pcie.h now. sysdev/fsl_pcie.c does not use this header
> file.
> plaforms/85xx/pci.c use it to set up memory map windows.
> In fact, there is little difference between immap_85xx.h and
> immap_86xx.h.
> After the migration from ppc to powerpc, we can unify them together.
Why wait, I don't believe we support pcie in arch/ppc for anything
fsl platform and the PCI support is different between arch/powerpc
and arch/ppc.
- k
^ permalink raw reply [flat|nested] 6+ messages in thread
* RE: [RFC 2/7] Unite all PCI-e on 85xx and 86xx under one codebase
2007-02-26 14:52 ` Kumar Gala
@ 2007-02-26 15:52 ` Zang Roy-r61911
2007-02-26 18:18 ` Kumar Gala
0 siblings, 1 reply; 6+ messages in thread
From: Zang Roy-r61911 @ 2007-02-26 15:52 UTC (permalink / raw)
To: Kumar Gala; +Cc: linuxppc-dev, jdl
=20
> On Feb 26, 2007, at 12:41 AM, Zang Roy-r61911 wrote:
>=20
> >
> >> On Feb 15, 2007, at 8:44 PM, Andy Fleming wrote:
> >>
> >>> Add 85xx PCI express memory map register.
> >>>
> >>> Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
> >>> Acked-by: Andy Fleming <afleming@freescale.com>
> >>> ---
> >>> include/asm-ppc/immap_85xx.h | 75
> >> +++++++++++++++++++++++++++++++
> >>> +++++++++++
> >>> 1 files changed, 75 insertions(+), 0 deletions(-)
> >>
> >> Are the 86xx and 85xx structs the same? If so, can we=20
> pull them out=20
> >> of immap_*.h into a sysdev/fsl_pcie.h
> >>
> >> (and lose the typedef while you're at it).
> > They have same structs. But I do not think it is proper to=20
> move them=20
> > into sysdev/fsl_pcie.h now. sysdev/fsl_pcie.c does not use this=20
> > header file.
> > plaforms/85xx/pci.c use it to set up memory map windows.
> > In fact, there is little difference between immap_85xx.h and=20
> > immap_86xx.h.
> > After the migration from ppc to powerpc, we can unify them together.
>=20
> Why wait, I don't believe we support pcie in arch/ppc for=20
> anything fsl platform and the PCI support is different=20
> between arch/powerpc and arch/ppc.
>=20
Wait to unify immap together :-).
I agree to separate pcie struct for arch/powerpc, then the code in
arch/powerpc will not depend header file include/asm-ppc.
I will provide patch to Andy and Jon tomorrow.
Roy
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [RFC 2/7] Unite all PCI-e on 85xx and 86xx under one codebase
2007-02-26 15:52 ` Zang Roy-r61911
@ 2007-02-26 18:18 ` Kumar Gala
0 siblings, 0 replies; 6+ messages in thread
From: Kumar Gala @ 2007-02-26 18:18 UTC (permalink / raw)
To: Zang Roy-r61911; +Cc: linuxppc-dev, jdl
On Feb 26, 2007, at 9:52 AM, Zang Roy-r61911 wrote:
>
>> On Feb 26, 2007, at 12:41 AM, Zang Roy-r61911 wrote:
>>
>>>
>>>> On Feb 15, 2007, at 8:44 PM, Andy Fleming wrote:
>>>>
>>>>> Add 85xx PCI express memory map register.
>>>>>
>>>>> Signed-off-by: Roy Zang <tie-fei.zang@freescale.com>
>>>>> Acked-by: Andy Fleming <afleming@freescale.com>
>>>>> ---
>>>>> include/asm-ppc/immap_85xx.h | 75
>>>> +++++++++++++++++++++++++++++++
>>>>> +++++++++++
>>>>> 1 files changed, 75 insertions(+), 0 deletions(-)
>>>>
>>>> Are the 86xx and 85xx structs the same? If so, can we
>> pull them out
>>>> of immap_*.h into a sysdev/fsl_pcie.h
>>>>
>>>> (and lose the typedef while you're at it).
>>> They have same structs. But I do not think it is proper to
>> move them
>>> into sysdev/fsl_pcie.h now. sysdev/fsl_pcie.c does not use this
>>> header file.
>>> plaforms/85xx/pci.c use it to set up memory map windows.
>>> In fact, there is little difference between immap_85xx.h and
>>> immap_86xx.h.
>>> After the migration from ppc to powerpc, we can unify them together.
>>
>> Why wait, I don't believe we support pcie in arch/ppc for
>> anything fsl platform and the PCI support is different
>> between arch/powerpc and arch/ppc.
>>
> Wait to unify immap together :-).
Don't exactly follow.
> I agree to separate pcie struct for arch/powerpc, then the code in
> arch/powerpc will not depend header file include/asm-ppc.
> I will provide patch to Andy and Jon tomorrow.
ok.
- k
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2007-02-26 18:19 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2007-02-16 2:44 [RFC 2/7] Unite all PCI-e on 85xx and 86xx under one codebase Andy Fleming
2007-02-16 3:43 ` Kumar Gala
2007-02-26 6:41 ` Zang Roy-r61911
2007-02-26 14:52 ` Kumar Gala
2007-02-26 15:52 ` Zang Roy-r61911
2007-02-26 18:18 ` Kumar Gala
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.