Skip site navigation (1)Skip section navigation (2)
Date:      Fri, 25 Jul 2008 11:17:51 -0700
From:      Andrew Thompson <thompsa@FreeBSD.org>
To:        freebsd-arm@FreeBSD.org
Subject:   ARM mfc list
Message-ID:  <20080725181751.GB13638@citylink.fud.org.nz>

next in thread | raw e-mail | index | archive | help
Hi,

I would like to do a large merge of arm fixes to 7-stable, some bugfixes
and many just to reduce diffs. In particular 174051+177457+177459 fix
usb on the avila board.

The commit logs and diffs can also be found at
 http://people.freebsd.org/~thompsa/arm-mfc/

Please review this list and let me know if anything shouldnt be merged
(or if you would like to handle your own commits).


r172568 (kevlo) Spelling fix for interupt -> interrupt

r172613 (cognet) Define _ARM_ARCH_5E too, so that we know if
        pld/strd/ldrd are available.

r172614 (cognet) Do not use __XSCALE__ to detect if pld/strd/ldrd is
        available, use_ARM_ARCH_5E instead.

r172738 (imp) Merge support from p4 (from NetBSD) for arm9e and arm10,
        arm11 cores.

r172739 (imp) correct guard variable names.

r173155 (imp) kill commented out line of code.

r173215 (kevlo) Don't define get_cachetype() for CPU_ARM9E unless it's
        going to be used.

r173249 (kevlo) __CPU_XSCALE_PXA2XX -> CPU_XSCALE_PXA2X0

r173336 (cognet) Remove a staled comment, NPE-C should work fine.

r173442 (cognet) Add entries for the L2 cache-related functions for armv5.

r174051 (cognet) Correct the logic : we can just invalidate the cache
        lines, and notwrite-back them, only if PREWRITE is not set, and
        if the buffer iscache-line aligned.

r174058 (cognet) Fixes for ARM9/ARM10 :Call uma_sel_align() there at
        well.Set CPU_CONTROL_VECRELOC if we're using the high vectors page.

r174176 (cognet) Move the strongarm-specific files from conf/files.arm
        to sa11x0/files.sa11xO.

r174700 (kevlo) Use M_NOWAIT instead of M_WAITOK to cause malloc() to
        return NULL

r174781 (imp) Actually program the interrupt controller for priorities.
        As wesupport more AT91 platforms, we'll need to move this into
        someplatform init routine.

r175120 (cognet) Add a missing \n.

r176759 (kevlo) Convert to be a 2-clause bsd-only license.

r177103 (raj) Improve ARM bus_dmamap_load_buffer() error handling.

r177105 (raj) Respect RF_SHAREABLE flag in ARM nexus_setup_intr()

r177457 (sam) Correct cache handling for xfer requests marked
        URQ_REQUEST: many (if notall uses) involve a read but
        usbd_start_transfer only does a PREWRITE; changethis to
        BUS_DMASYNC_PREREAD | BUS_DMASYNC_PREWRITE as I'm not sure if
        anyusers do write+read.

r177459 (sam) Workaround design botch in usb: blindly mixing bus_dma
        with PIO does notwork on architectures with a write-back cache
        as the PIO writes end upin the cache which the
        sync(BUS_DMASYNC_POSTREAD) in usb_transfer_completethen
        discards; compensate in the xfer methods that do PIO by pushing
        thewrites out of the cache before usb_transfer_complete is
        called.

r177505 (sam) Improve mac+phy configuration so that hints can be used to
        describelayouts different than the defaults:o
        hint.npe.0.mac="A", "B", etc. specifies the window for MAC
        register accesseso hint.npe.0.mii="A", "B", etc. specifies PHY
        registerso hint.npe.1.phy=%d specifies the PHY to map to a port

r177874 (imp) KERNBASE + 0x00200000 is the same thing as KERNVIRTADDR on
        thisplatform, so use the latter in preference to the former.
        This makesthe fake_preload setup be the same between
        kb920x_machdep.c andavila_machdep.c....

r177883 (imp) Take the first baby step towards unifying and cleaning up
        arminit():
        - Pull all the code to deal with the trampoline stuff into one
          centeralized place and use it from everywhere.
        - Some minor style tidiness

r177886 (raj) Fix AVILA build.

r177887 (raj) Refactor certain ARM bus space methods: instead of having
        multiple copies ofthe same code introduce
        sys/arm/arm/bus_space_generic.c for a shared set ofroutines.

r177888 (raj) Now really add the bus_space_generic.c file...

r177916 (raj) Make kernel.tramp build properly on ARM9E.

r177943 (cognet) Add bus_space_generic.c for the i81342 as well.

r177944 (cognet) Remove bus_space_generic.c from the per-plarform files.
        Having it in theper-cpu files should be enough.

r178001 (kevlo) Remove some long-dead code

r179375 (imp) Release the resources for the registers for the TWI device
        withSYS_RES_MEMORY to match how we allocate them...

r179664 (kevlo) Remove sa1_cache_clean_addr

r179693 (wkoszek) Since we create a DMA tag "mtag" for TX map with
        bus_dmamap_create(),we must synchronize such a map against
        "mtag" with bus_dmamap_sync(),not the tag designated for RX map.

r179745 (kevlo) Fix a typo: i80321_pci_probe -> i81342_pci_probe

r180486 (ticso) fix multicast hash register definition

r180584 (alc) Eliminate unused global variables.  (These global
        variables became fields ofstruct kva_md_info many years ago.)




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20080725181751.GB13638>