N900 mailclient is stupid ... Bear with it please. ----- Původní zpráva -----
Dear Marek Vasut,
In message 1282296691-15910-7-git-send-email-marek.vasut@gmail.com you wrote:
The following hardware is currently supported:
- UART
- USB Host
- FPGA
...
+int usb_board_init(void) +{ + UHCHR = (UHCHR | UHCHR_PCPL | UHCHR_PSPL) & + ~(UHCHR_SSEP0 | UHCHR_SSEP1 | UHCHR_SSEP2 | UHCHR_SSE);
+ UHCHR |= UHCHR_FSBIR;
+ while (UHCHR & UHCHR_FSBIR);
+ UHCHR &= ~UHCHR_SSE; + UHCHIE = (UHCHIE_UPRIE | UHCHIE_RWIE);
+ /* Clear any OTG Pin Hold */ + if (PSSR & PSSR_OTGPH) + PSSR |= PSSR_OTGPH;
+ UHCRHDA &= ~(0x200); + UHCRHDA |= 0x100;
+ /* Set port power control mask bits, only 3 ports. */ + UHCRHDB |= (0x7<<17);
+ /* enable port 2 */ + UP2OCR |= UP2OCR_HXOE | UP2OCR_HXS | UP2OCR_DMPDE | UP2OCR_DPPDE;
Why aren't you using I/O accessors here?
Hey Wolfgang,
we already discussed this somewhere. Check how arch/arm/include/mach-pxa/pxa-regs.h is done. It's ripped off the linux kernel. We already came to a conclusion that it needs to be reworked (remove the _REG() macro crud and use io accessors).
Though I'd like to propose if we could get this stuff in prior to reworking pxa-regs.h as the change will be very intrusive. To have a stable revision where the boards work in case something went wrong.
+void usb_board_stop(void) +{ + UHCHR |= UHCHR_FHR; + udelay(11); + UHCHR &= ~UHCHR_FHR;
+ UHCCOMS |= 1; + udelay(10);
+ CKEN &= ~CKEN10_USBHOST;
And here, and so on?
Best regards,
Wolfgang Denk
-- DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: (+49)-8142-66989-10 Fax: (+49)-8142-66989-80 Email: wd@denx.de Don't tell me how hard you work. Tell me how much you get done. -- James J. Ling