Author: cognet
Date: Mon Apr 13 23:16:32 2020
New Revision: 359909
URL: https://svnweb.freebsd.org/changeset/base/359909
Log:
Remove FreeBSD/armv4 specific bits from CK.
Now that armv4/v5 is gone, remove the bits that implemented atomic operations
by disabling interrupts.
Those were
Author: cognet
Date: Mon Feb 24 16:25:11 2020
New Revision: 358290
URL: https://svnweb.freebsd.org/changeset/base/358290
Log:
debug_monitor: Avoid setting the PSR_D flag for 32bits binaries.
In dbg_monitor_exit(), avoid setting the PSR_D bit if the process is
a 32bits binary. PSR_D is an
00:00:00 1970 (empty, because file is newly added)
> > +++ head/sys/arm64/arm64/machdep_boot.c Thu Nov 14 21:58:40
> > 2019(r354714)
> > @@ -0,0 +1,330 @@
> > +/*-
> > + * Copyright (c) 2019 Juniper Networks, Inc
> > + * Copyright (c) 2004 Oliv
Author: cognet
Date: Mon Nov 11 00:21:05 2019
New Revision: 354602
URL: https://svnweb.freebsd.org/changeset/base/354602
Log:
linprocfs: Make sure to report -1 as tty when we have no controlling tty.
When reporting a process' stats, we can't just provide the tty as an
unsigned long, as
Author: cognet
Date: Fri Oct 11 13:34:09 2019
New Revision: 353435
URL: https://svnweb.freebsd.org/changeset/base/353435
Log:
Document that aarch64 can now run armv6/armv7 binaries, but won't however
run armv5 binaries.
Modified:
head/share/man/man7/arch.7
Modified:
Author: cognet
Date: Wed Jun 26 22:06:40 2019
New Revision: 349444
URL: https://svnweb.freebsd.org/changeset/base/349444
Log:
In get_fpcontext32() and set_fpcontext32(), we can't just use memcpy() to
copy the VFP registers.
arvm7 VFP uses 32 64bits fp registers (but those could be used in
Author: cognet
Date: Wed Jun 26 16:56:56 2019
New Revision: 349426
URL: https://svnweb.freebsd.org/changeset/base/349426
Log:
Fix debugging of 32bits arm binaries on arm64.
In set_regs32()/fill_regs32(), we have to get/set SP and LR from/to
tf_x[13] and tf_x[14].
set_regs() and
Author: cognet
Date: Sun Jan 13 23:41:56 2019
New Revision: 343007
URL: https://svnweb.freebsd.org/changeset/base/343007
Log:
Don't forget to add the needed #includes.
Pointy hat to:cognet
Modified:
head/sys/arm64/arm64/cpufunc_asm.S
Modified: head/sys/arm64/arm64/cpufunc_asm.S
Author: cognet
Date: Sun Jan 13 23:29:46 2019
New Revision: 343006
URL: https://svnweb.freebsd.org/changeset/base/343006
Log:
Introduce cpu_icache_sync_range_checked(), that does the same thing as
cpu_icache_sync_range(), except that it sets pcb_onfault to catch any page
fault, as doing
Olivier Houchard
+ * All rights reserved.
*
+ * This software was developed by Andrew Turner under
+ * sponsorship from the FreeBSD Foundation.
+ *
+ * Portions of this software were developed by Konstantin Belousov
+ * under sponsorship from the FreeBSD Foundation.
+ *
* Redistribution and use
Author: cognet
Date: Sun Jan 13 00:37:31 2019
New Revision: 342979
URL: https://svnweb.freebsd.org/changeset/base/342979
Log:
amd64 is the only arch that doesn't require padding for 32bits syscalls, so
instead of listing every arch thar requires it, just exclude amd64.
Modified:
Author: cognet
Date: Sun Jan 13 00:38:55 2019
New Revision: 342980
URL: https://svnweb.freebsd.org/changeset/base/342980
Log:
Regenerate sysent files after having modified syscalls.master.
Modified:
head/sys/compat/freebsd32/freebsd32_proto.h
head/sys/compat/freebsd32/freebsd32_syscalls.c
Author: cognet
Date: Sun Jan 13 00:19:15 2019
New Revision: 342978
URL: https://svnweb.freebsd.org/changeset/base/342978
Log:
Instead of using an incomplete list of platforms that uses 64bits time_t
in 32bits mode, special case amd64, as i386 is the only arch that still
uses 32bits time_t.
Author: cognet
Date: Thu Aug 9 12:11:49 2018
New Revision: 337533
URL: https://svnweb.freebsd.org/changeset/base/337533
Log:
Import CK as of commit 08813496570879fbcc2adcdd9ddc0a054361bfde, mostly
to avoid using lwsync on ppc32.
Modified:
head/sys/contrib/ck/include/gcc/ppc/ck_pr.h
Hi,
On Fri, Jun 15, 2018 at 12:23:36AM +0200, Emmanuel Vadot wrote:
>
> Hi Jonathan,
>
[...]
> This brake module loading on armv7 and arm64
> kevans log on armv7 :
> https://gist.github.com/kevans91/d0ffcd6c94a0e4cdfdc5433612fce2e8
> mine on arm64 : http://dpaste.com/3VVBZDV
>
> Do you
Author: cognet
Date: Thu Jun 14 23:10:10 2018
New Revision: 335182
URL: https://svnweb.freebsd.org/changeset/base/335182
Log:
Use M_EXEC when calling malloc() to allocate the memory to store the module,
as it'll contain executable code.
Modified:
head/sys/kern/link_elf.c
Modified:
Author: cognet
Date: Thu May 24 21:38:18 2018
New Revision: 334189
URL: https://svnweb.freebsd.org/changeset/base/334189
Log:
Import CK as of commit 0f017230ccc86929f56bf44ef2dca93d7df8076b.
This brings us the renaming of fields in ck_queue, so that our own
LIST/SLIST/TAILQ/etc won't
Author: cognet
Date: Fri May 18 13:28:02 2018
New Revision: 333791
URL: https://svnweb.freebsd.org/changeset/base/333791
Log:
Instead of ignoring the VFP registers, set the dumppcb's pcb_fpusaved
field, so that they are saved, as they may be used in the kernel, in the
EFI and the crypto
Author: cognet
Date: Thu May 17 22:40:22 2018
New Revision: 333772
URL: https://svnweb.freebsd.org/changeset/base/333772
Log:
In pmap_get_tables(), check that the L2 is indeed a table before attempting
to get the l3.
Modified:
head/sys/arm64/arm64/pmap.c
Modified:
Author: cognet
Date: Thu May 17 22:38:16 2018
New Revision: 333771
URL: https://svnweb.freebsd.org/changeset/base/333771
Log:
In vfp_save_state(), don't bother trying to save the VFP registers if the
provided PCB doesn't have a pcb_fpusaved. All PCBs associated to a thread
should have one,
Author: cognet
Date: Thu May 17 21:03:36 2018
New Revision: 333764
URL: https://svnweb.freebsd.org/changeset/base/333764
Log:
Import CK as of commit deca119d14bfffd440770eb67cbdbeaf7b57eb7b.
This is mostly a noop, for mergeinfo purpose, because the relevant changes
were committed directly.
Author: cognet
Date: Mon Apr 2 23:40:50 2018
New Revision: 331898
URL: https://svnweb.freebsd.org/changeset/base/331898
Log:
Import CK as of commit b19ed4c6a56ec93215ab567ba18ba61bf1cfbac8
It should fix ck_pr_[load|store]_ptr on mips and riscv, make sure no
*fence instructions are used on
Author: cognet
Date: Fri Mar 23 17:25:19 2018
New Revision: 331441
URL: https://svnweb.freebsd.org/changeset/base/331441
Log:
In __sync_bool_compare_and_swap(), return true if the returned value is the
same as the expected one, not the desired one.
Pointy hat to:cognet
MFC
Author: cognet
Date: Mon Feb 26 13:12:51 2018
New Revision: 330018
URL: https://svnweb.freebsd.org/changeset/base/330018
Log:
In do_ast, make sure the interrupts are enabled before calling ast().
We can reach that point with IRQs disabled, and calling ast() with IRQs
disabled can lead to
Author: cognet
Date: Sat Feb 24 14:34:23 2018
New Revision: 329908
URL: https://svnweb.freebsd.org/changeset/base/329908
Log:
Use NULL as a mtx type instead of "", as it otherwise confuses WITNESS.
Modified:
head/sys/arm/arm/gic.c
Modified: head/sys/arm/arm/gic.c
Author: cognet
Date: Fri Feb 16 17:50:06 2018
New Revision: 329388
URL: https://svnweb.freebsd.org/changeset/base/329388
Log:
Define CK_MD_TSO for the relevant arches (i386, amd64 and sparc64).
Defaulting to CK_MD_RMO has the unfortunate side effect of generating
memory barriers that are
Author: cognet
Date: Thu Feb 15 15:46:14 2018
New Revision: 329317
URL: https://svnweb.freebsd.org/changeset/base/329317
Log:
Rename the ACPI variant of the gicv2m driver from "gicv2m" to "gicv2m_acpi".
The FDT variant is called "gicv2m" too, and as both would try to register
on gic, only
Author: cognet
Date: Tue Sep 19 23:41:55 2017
New Revision: 323784
URL: https://svnweb.freebsd.org/changeset/base/323784
Log:
Define CPU_XSCALE_CORE3 when relevant.
It was lost when cpuconf.h was deobirted.
Modified:
head/sys/arm/include/pmap-v4.h
Modified: head/sys/arm/include/pmap-v4.h
Author: cognet
Date: Tue Sep 19 20:33:22 2017
New Revision: 323777
URL: https://svnweb.freebsd.org/changeset/base/323777
Log:
i81342 is little endian, not big endian.
Modified:
head/sys/arm/xscale/i8134x/std.i81342
Modified: head/sys/arm/xscale/i8134x/std.i81342
Author: cognet
Date: Tue Sep 12 10:43:02 2017
New Revision: 323490
URL: https://svnweb.freebsd.org/changeset/base/323490
Log:
Some devices come with the same name as TI devices, so we can't rely on the
"probe" method of those drivers to mean we're on e TI SoC. Introduce a new
function,
Author: cognet
Date: Tue Jun 13 16:19:32 2017
New Revision: 319904
URL: https://svnweb.freebsd.org/changeset/base/319904
Log:
style(9) fixes.
Reported by: cem
Modified:
head/usr.bin/yes/yes.c
Modified: head/usr.bin/yes/yes.c
Author: cognet
Date: Fri Jun 2 14:17:14 2017
New Revision: 319496
URL: https://svnweb.freebsd.org/changeset/base/319496
Log:
- Don't bother flushing the data cache for pages we're about to unmap, there's
no need to.
- Remove pmap_is_current(), pmap_[pte|l3]_valid_cacheable as there were
Author: cognet
Date: Wed Apr 26 16:13:22 2017
New Revision: 317441
URL: https://svnweb.freebsd.org/changeset/base/317441
Log:
Check if the device is marked as dma-coherent in the FDT, and if so, let
busdma know, so that on architectures where dma isn't always coherent, we
know we don't have
Author: cognet
Date: Tue Apr 25 23:46:53 2017
New Revision: 317428
URL: https://svnweb.freebsd.org/changeset/base/317428
Log:
In arm_gicv2m_alloc_msi(), if we found a suitable irq range, leave the loop
before we increase irq again, or we'd end up choosing an irq, and then
really using the
Author: cognet
Date: Thu Apr 20 15:06:16 2017
New Revision: 317195
URL: https://svnweb.freebsd.org/changeset/base/317195
Log:
Update comment ot reflect realilty, we know also take care of CPUs that
provide a enable-method.
Suggested by: jhibbits
Modified:
head/sys/dev/ofw/ofw_cpu.c
Author: cognet
Date: Thu Apr 20 13:56:06 2017
New Revision: 317191
URL: https://svnweb.freebsd.org/changeset/base/317191
Log:
ePAPR states that any non-boot CPU will come in "disabled" state. So we should
not consider a "disabled" cpu as a CPU we have to ignore, and we should use
them if
Author: cognet
Date: Tue Apr 11 13:31:27 2017
New Revision: 316702
URL: https://svnweb.freebsd.org/changeset/base/316702
Log:
This file is also used in libcompiler_rt, so bring in stdbool.h if we're not
in the kernel.
Modified:
head/sys/mips/mips/stdatomic.c
Modified:
On Mon, Apr 10, 2017 at 04:30:18PM -0400, Ed Maste wrote:
> On 9 April 2017 at 17:02, Olivier Houchard <cog...@freebsd.org> wrote:
> >
> > Author: cognet
> > Date: Sun Apr 9 21:02:05 2017
> > New Revision: 316665
> > URL: https://svnweb.freebsd.org/changese
Author: cognet
Date: Mon Apr 10 21:44:13 2017
New Revision: 316691
URL: https://svnweb.freebsd.org/changeset/base/316691
Log:
Provide some kind of __sync_bool_compare_and_swap_4(), as it is used by CK,
and our gcc is too ancient to provide it.
This should fix the build on mips.
Modified:
On Sun, Apr 09, 2017 at 03:49:57PM -0700, Conrad Meyer wrote:
> On Sun, Apr 9, 2017 at 2:02 PM, Olivier Houchard <cog...@freebsd.org> wrote:
> > Author: cognet
> > Date: Sun Apr 9 21:02:05 2017
> > New Revision: 316665
> > URL: https://svnweb.freebsd.org/
Author: cognet
Date: Sun Apr 9 21:02:05 2017
New Revision: 316665
URL: https://svnweb.freebsd.org/changeset/base/316665
Log:
Import CK as of commit 6b141c0bdd21ce8b3e14147af8f87f22b20ecf32
This brings us changes we needed in ck_epoch.
Modified:
Author: cognet
Date: Tue Feb 28 23:30:14 2017
New Revision: 314435
URL: https://svnweb.freebsd.org/changeset/base/314435
Log:
Merge CK as of commit 24d26965d1a28039062ba3bcf9433b623f3d2c5e, to get
a fix in ck_epoch.
Modified:
head/sys/contrib/ck/include/ck_md.h
On Thu, Feb 23, 2017 at 07:23:36PM -0500, Pedro Giffuni wrote:
> Hi;
>
> > Il giorno 23 feb 2017, alle ore 19:05, Ian Lepore ha
> > scritto:
> >
> > On Thu, 2017-02-23 at 23:48 +, Pedro F. Giffuni wrote:
> >> Author: pfg
> >> Date: Thu Feb 23 23:48:44 2017
> >> New
Author: cognet
Date: Mon Feb 13 20:51:08 2017
New Revision: 313708
URL: https://svnweb.freebsd.org/changeset/base/313708
Log:
Remove debugging code that was probably unused since before the arm code
was initially committed.
Reported by: Alexandre Martins
Author: cognet
Date: Sun Jan 29 13:31:56 2017
New Revision: 312949
URL: https://svnweb.freebsd.org/changeset/base/312949
Log:
Correct the IT instruction in atomic_fcmpset_64().
Reported by: andrew
Modified:
head/sys/arm/include/atomic-v6.h
Modified: head/sys/arm/include/atomic-v6.h
Author: cognet
Date: Sat Jan 28 17:48:33 2017
New Revision: 312933
URL: https://svnweb.freebsd.org/changeset/base/312933
Log:
Remove useless labels.
Modified:
head/sys/arm/include/atomic-v6.h
Modified: head/sys/arm/include/atomic-v6.h
Author: cognet
Date: Sat Jan 28 17:46:04 2017
New Revision: 312932
URL: https://svnweb.freebsd.org/changeset/base/312932
Log:
Use strexeq instead of needlessly branch.
Suggested by: ian
Modified:
head/sys/arm/include/atomic-v6.h
Modified: head/sys/arm/include/atomic-v6.h
Author: cognet
Date: Sat Jan 28 16:24:06 2017
New Revision: 312925
URL: https://svnweb.freebsd.org/changeset/base/312925
Log:
Implement atomic_fcmpset_* for arm and arm64.
Modified:
head/sys/arm/include/atomic-v4.h
head/sys/arm/include/atomic-v6.h
head/sys/arm/include/atomic.h
Author: cognet
Date: Tue Jan 3 17:24:56 2017
New Revision: 311159
URL: https://svnweb.freebsd.org/changeset/base/311159
Log:
In the netfront_rxq struct, we should use NET_RX_RING_SIZE, not
NET_TX_RING_SIZE.
Reviewed by: royger
Modified:
head/sys/dev/xen/netfront/netfront.c
" \
-"stxr" W " %w2, %w1, [%3];"\
- "cbnz %w2, 1b;"\
- : "=" (previous), \
-
Author: cognet
Date: Mon Nov 28 21:16:03 2016
New Revision: 309270
URL: https://svnweb.freebsd.org/changeset/base/309270
Log:
Add a FREEBSD-Xlist file for CK.
Added:
head/sys/contrib/ck/FREEBSD-Xlist
Added: head/sys/contrib/ck/FREEBSD-Xlist
Author: cognet
Date: Mon Nov 28 20:44:12 2016
New Revision: 309268
URL: https://svnweb.freebsd.org/changeset/base/309268
Log:
Hook CK to the kernel build.
Modified:
head/sys/conf/files
Modified: head/sys/conf/files
newly added)
+++ head/sys/contrib/ck/include/gcc/arm/ck_pr_armv4.h Mon Nov 28 20:33:30
2016 (r309267)
@@ -0,0 +1,350 @@
+/*
+ * Copyright 2009-2016 Samy Al Bahra.
+ * Copyright 2016 Olivier Houchard.
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or wi
Author: cognet
Date: Mon Nov 28 20:27:58 2016
New Revision: 309266
URL: https://svnweb.freebsd.org/changeset/base/309266
Log:
Import Concurrency Kit in the kernel.
CK is a toolkit providing different lockfree algorithms/data structures.
More information can be found here :
Author: cognet
Date: Sat Nov 19 15:43:22 2016
New Revision: 308840
URL: https://svnweb.freebsd.org/changeset/base/308840
Log:
The only remaining offender that used ti_chip() without checking for
compatibility first was the gpio code, so change that, and re-assert
that the TI chip is a known
Author: cognet
Date: Sat Nov 19 15:35:40 2016
New Revision: 308837
URL: https://svnweb.freebsd.org/changeset/base/308837
Log:
Don't panic if it's not a TI chip, this code can be called when it is not.
Modified:
head/sys/arm/ti/ti_cpuid.h
Modified: head/sys/arm/ti/ti_cpuid.h
Author: cognet
Date: Sat Nov 19 15:35:10 2016
New Revision: 308836
URL: https://svnweb.freebsd.org/changeset/base/308836
Log:
Don't attempt to get the chip revision if it's not a supported TI chip
Modified:
head/sys/arm/ti/ti_cpuid.c
Modified: head/sys/arm/ti/ti_cpuid.c
Author: cognet
Date: Sat Nov 19 13:11:03 2016
New Revision: 308828
URL: https://svnweb.freebsd.org/changeset/base/308828
Log:
Move the pandaboard initialization from the probe to the attach method.
Use ofw_bus_node_is_compatible instead of fdt_is_compatible, as the
later is deprecated.
Author: cognet
Date: Sat Nov 19 01:51:56 2016
New Revision: 308822
URL: https://svnweb.freebsd.org/changeset/base/308822
Log:
Resolv the remaining conflicting symbols between omap4 and am335x, and
add omap4/pandaboard into the GENERIC kernel.
Modified:
head/sys/arm/conf/GENERIC
Sat Nov 19 00:55:46
2016 (r308819)
@@ -0,0 +1,32 @@
+/*-
+ * Copyright (c) 2016 Olivier Houchard <cog...@freebsd.org>
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions
+ * ar
Author: cognet
Date: Fri Nov 18 23:48:20 2016
New Revision: 308818
URL: https://svnweb.freebsd.org/changeset/base/308818
Log:
Guess the TI chip based on the PLATFORM infos, instead of relying on the
kernel config file.
Modified:
head/sys/arm/ti/ti_cpuid.h
head/sys/arm/ti/ti_machdep.c
+ head/sys/arm/ti/omap4/omap4_mp.hFri Nov 18 22:58:47 2016
(r308816)
@@ -0,0 +1,33 @@
+/*-
+ * Copyright (c) 2016 Olivier Houchard <cog...@freebsd.org>
+ * All rights reserved.
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are pe
Author: cognet
Date: Tue Aug 30 23:32:38 2016
New Revision: 305097
URL: https://svnweb.freebsd.org/changeset/base/305097
Log:
Nuke obio_bs_tag, it was used before it was initialized, and
arm_base_bs_tag is the same, anyway.
Modified:
head/sys/arm/xscale/i8134x/crb_machdep.c
Author: cognet
Date: Tue Aug 30 23:30:26 2016
New Revision: 305096
URL: https://svnweb.freebsd.org/changeset/base/305096
Log:
Some old arm ports don't load the kernel at the beginning of the memory,
because the bootloader, ie redboot, won't let them do so, and so used the
memory before the
Author: cognet
Date: Tue Aug 30 22:32:33 2016
New Revision: 305094
URL: https://svnweb.freebsd.org/changeset/base/305094
Log:
Garbage collect bits forgotten in r295267.
Modified:
head/sys/arm/arm/cpufunc_asm_xscale_c3.S
Modified: head/sys/arm/arm/cpufunc_asm_xscale_c3.S
Author: cognet
Date: Fri Nov 6 20:12:31 2015
New Revision: 290465
URL: https://svnweb.freebsd.org/changeset/base/290465
Log:
Include opt_platform.h to get FDT defined.
Modified:
head/sys/arm/ti/ti_common.c
Modified: head/sys/arm/ti/ti_common.c
Author: cognet
Date: Thu Nov 5 22:03:42 2015
New Revision: 290422
URL: https://svnweb.freebsd.org/changeset/base/290422
Log:
Make if_macb work with FDT.
Modified:
head/sys/arm/at91/if_macb.c
Modified: head/sys/arm/at91/if_macb.c
Author: cognet
Date: Thu Nov 5 22:03:20 2015
New Revision: 290420
URL: https://svnweb.freebsd.org/changeset/base/290420
Log:
Make at91_pmc probe any at91 pmc device we support, not just at91rm9200.
MFC after:1 week
Modified:
head/sys/arm/at91/at91_pmc.c
Modified:
Author: cognet
Date: Sat Mar 21 15:32:59 2015
New Revision: 280324
URL: https://svnweb.freebsd.org/changeset/base/280324
Log:
When waiting on PTE allocation, another thread could free the l2_dtable while
we're not looking at it.
Fix this by increasing l2-l2_occupancy before we try to alloc
Author: cognet
Date: Sat Mar 21 16:16:17 2015
New Revision: 280325
URL: https://svnweb.freebsd.org/changeset/base/280325
Log:
error is only used if MAC is defined, so make its declaration conditional
as well.
Modified:
head/sys/kern/sysv_shm.c
Modified: head/sys/kern/sysv_shm.c
Author: cognet
Date: Thu May 29 16:54:15 2014
New Revision: 266849
URL: http://svnweb.freebsd.org/changeset/base/266849
Log:
For old CPUs, map the 64 first MB of RAM as it used to be. Some ports
(XScale mainly) expects the memory located before the kernel to be mapped,
and use it to
Author: cognet
Date: Thu May 29 16:56:39 2014
New Revision: 266850
URL: http://svnweb.freebsd.org/changeset/base/266850
Log:
Do not hand the VM the memory used for stacks/page tables/etc.
Modified:
head/sys/arm/xscale/i80321/ep80219_machdep.c
head/sys/arm/xscale/i80321/iq31244_machdep.c
Author: cognet
Date: Thu May 29 19:57:51 2014
New Revision: 266855
URL: http://svnweb.freebsd.org/changeset/base/266855
Log:
In the grab function, keep the bit 6 on in the IER, on XScale, using 0
turns the UART off, which is unfortunate if one want to use it as a console.
Modified:
On Thu, May 29, 2014 at 01:44:54PM -0700, John-Mark Gurney wrote:
Olivier Houchard wrote this message on Thu, May 29, 2014 at 19:57 +:
Author: cognet
Date: Thu May 29 19:57:51 2014
New Revision: 266855
URL: http://svnweb.freebsd.org/changeset/base/266855
Log:
In the grab
Author: cognet
Date: Thu May 29 21:00:06 2014
New Revision: 266858
URL: http://svnweb.freebsd.org/changeset/base/266858
Log:
In uart_bus_grab(), use the ier_mask instead of a custom hack for XScale.
Suggested by: jmg
Modified:
head/sys/dev/uart/uart_dev_ns8250.c
Modified:
Author: cognet
Date: Thu Mar 6 21:07:13 2014
New Revision: 262870
URL: http://svnweb.freebsd.org/changeset/base/262870
Log:
When calculating the MPU freq, make sure not to overflow by using a uint64_t.
PR: arm/187223
Submitted by: Svatopluk Kraus onw...@gmail.com
Modified:
Author: cognet
Date: Sun Feb 2 20:45:41 2014
New Revision: 261414
URL: http://svnweb.freebsd.org/changeset/base/261414
Log:
Don't call device_set_ivars() for the mmchs, it doesn't seem to be used,
and it overrides the ivars set by the simplebus.
Modified:
head/sys/arm/ti/ti_mmchs.c
Author: cognet
Date: Sun Feb 2 22:26:30 2014
New Revision: 261418
URL: http://svnweb.freebsd.org/changeset/base/261418
Log:
Invalidate cachelines for bounce pages on PREREAD too, there may still be
stale entries from a previous transfer.
Modified:
head/sys/arm/arm/busdma_machdep-v6.c
Author: cognet
Date: Sun Feb 2 23:29:51 2014
New Revision: 261419
URL: http://svnweb.freebsd.org/changeset/base/261419
Log:
Only use the CPU ID register if SMP is defined. Some non-MPCore armv6 cpu,
such as the one found in the RPi, don't have it, and just hang when we try
to access it.
Author: cognet
Date: Thu Nov 21 20:39:53 2013
New Revision: 258443
URL: http://svnweb.freebsd.org/changeset/base/258443
Log:
Kill ARM_VFP_SUPPORT, it's been removed some time ago.
Modified:
head/sys/conf/options.arm
Modified: head/sys/conf/options.arm
Author: cognet
Date: Wed Nov 20 23:06:54 2013
New Revision: 258410
URL: http://svnweb.freebsd.org/changeset/base/258410
Log:
In pmap_unmapdev(), remember the size, and use that as an argument to
kva_free(), or we'd end up always passing it a size of 0, and for some
strange reason it doesn't
Author: cognet
Date: Mon Oct 28 20:26:34 2013
New Revision: 257270
URL: http://svnweb.freebsd.org/changeset/base/257270
Log:
Include sys/ktr.h, since we need it if ATH_DEBUG is defined.
Modified:
head/sys/dev/ath/if_ath_tx.c
Modified: head/sys/dev/ath/if_ath_tx.c
Author: cognet
Date: Sun Oct 27 22:15:50 2013
New Revision: 257231
URL: http://svnweb.freebsd.org/changeset/base/257231
Log:
Make sure the PCB is aligned on 8 bytes, we may use ldrd/strd to access it,
which may have strong alignment requirements.
Modified:
head/sys/arm/include/pcb.h
Author: cognet
Date: Sun Oct 27 23:48:59 2013
New Revision: 257233
URL: http://svnweb.freebsd.org/changeset/base/257233
Log:
Use the size of the MACHINE_ARCH string instead of sizeof(uint32_t). It can
happen sizeof(MACHINE_ARCH) is more than 4 bytes, and bad things would
happen. This should
Author: cognet
Date: Tue Oct 22 21:47:34 2013
New Revision: 256941
URL: http://svnweb.freebsd.org/changeset/base/256941
Log:
Try to make sure the frame is indeed in the kernel memory.
Modified:
head/sys/arm/arm/stack_machdep.c
Modified: head/sys/arm/arm/stack_machdep.c
Author: cognet
Date: Tue Oct 22 21:49:58 2013
New Revision: 256942
URL: http://svnweb.freebsd.org/changeset/base/256942
Log:
Typo fix.
Modified:
head/sys/arm/xscale/ixp425/ixp425_mem.c
Modified: head/sys/arm/xscale/ixp425/ixp425_mem.c
Author: cognet
Date: Tue Oct 22 21:51:07 2013
New Revision: 256943
URL: http://svnweb.freebsd.org/changeset/base/256943
Log:
- Use bus_dmamap_unload(), it is not optional.
- The new allocator won't return coherent memory for any size PAGE_SIZE,
so don't assume we have coherent memory, and
Author: cognet
Date: Fri Oct 18 17:21:47 2013
New Revision: 256748
URL: http://svnweb.freebsd.org/changeset/base/256748
Log:
KERNBASE is unsigned, so we'd better use hs instead of ge.
Pointy hat to:cognet
Suggested by: ian
Modified:
head/sys/arm/arm/fusu.S
Modified:
Author: cognet
Date: Fri Oct 18 22:47:10 2013
New Revision: 256756
URL: http://svnweb.freebsd.org/changeset/base/256756
Log:
There's no need to guard pmap_extract(), it won't be called until well after
the VM has been properly initialized.
Spotted out by: alc
Modified:
Author: cognet
Date: Thu Oct 17 09:57:09 2013
New Revision: 256672
URL: http://svnweb.freebsd.org/changeset/base/256672
Log:
If we avoid to use the page at addr 0, we should adjust the size to reflect
it.
Modified:
head/sys/arm/arm/machdep.c
Modified: head/sys/arm/arm/machdep.c
Author: cognet
Date: Thu Oct 17 17:11:15 2013
New Revision: 256691
URL: http://svnweb.freebsd.org/changeset/base/256691
Log:
Make casuword() atomic for armv6
Modified:
head/sys/arm/arm/fusu.S
Modified: head/sys/arm/arm/fusu.S
Author: cognet
Date: Thu Oct 17 21:06:19 2013
New Revision: 256707
URL: http://svnweb.freebsd.org/changeset/base/256707
Log:
- Switch to use WBWA mappings for page tables on armv6, this is needed for
SMP.
- Fix PTE_SYNC() for PIPT L2 caches, using the virtual address wasn't so
useful.
-
Author: cognet
Date: Thu Oct 17 22:12:32 2013
New Revision: 256712
URL: http://svnweb.freebsd.org/changeset/base/256712
Log:
Increase the KVA available for xscale CPUs.
Modified:
head/sys/arm/xscale/i80321/ep80219_machdep.c
head/sys/arm/xscale/i80321/iq31244_machdep.c
Author: cognet
Date: Sun Aug 18 18:08:12 2013
New Revision: 254495
URL: http://svnweb.freebsd.org/changeset/base/254495
Log:
Increase the max KVA available for general consumption on the Exynos 5.
Submitted by: Ruslan Bukin b...@bsdpad.com
Modified:
Author: cognet
Date: Sun Aug 11 21:21:02 2013
New Revision: 254229
URL: http://svnweb.freebsd.org/changeset/base/254229
Log:
Only allocate 2 bounce pages for maps that can only use them for buffers that
are unaligned on cache lines boundary, as we will never need more.
Modified:
Author: cognet
Date: Fri Aug 9 21:53:02 2013
New Revision: 254164
URL: http://svnweb.freebsd.org/changeset/base/254164
Log:
Make sure vm_kmem_size is aligned on a page boundary, since that's what vmem
expects.
Modified:
head/sys/arm/include/vmparam.h
Modified:
Author: cognet
Date: Fri Aug 9 21:56:28 2013
New Revision: 254165
URL: http://svnweb.freebsd.org/changeset/base/254165
Log:
- The address lies in the bus space handle, not in the cookie
- Use the right address when calling kva_free()
Modified:
head/sys/arm/arm/bus_space_generic.c
Author: cognet
Date: Fri Aug 9 22:30:54 2013
New Revision: 254166
URL: http://svnweb.freebsd.org/changeset/base/254166
Log:
Instead of just trying to do it for arm, make sure vm_kmem_size is properly
aligned in kmeminit(), where it'll work for any arch.
Suggested by: alc
Modified:
Author: cognet
Date: Fri Aug 9 23:13:52 2013
New Revision: 254167
URL: http://svnweb.freebsd.org/changeset/base/254167
Log:
Don't call sleepinit() from proc0_init(), make it a SYSINIT instead.
vmem needs the sleepq locks to be initialized when free'ing kva, so we want it
called as early as
On Fri, Aug 09, 2013 at 10:27:48AM +0200, Zbigniew Bodek wrote:
2013/8/8 Zbyszek Bodek z...@semihalf.com
On 07.08.2013 20:55, Jeff Roberson wrote:
On Wed, 7 Aug 2013, Zbyszek Bodek wrote:
On 07.08.2013 08:21, Jeff Roberson wrote:
Author: jeff
Date: Wed Aug 7 06:21:20 2013
1 - 100 of 170 matches
Mail list logo