Hi Paul,
After merging the audit tree, today's linux-next build (i386 defconfig
and others) produced this warning:
security/lsm_audit.c: In function 'dump_common_audit_data':
security/lsm_audit.c:340:9: warning: passing argument 1 of 'unix_sk' discards
'const' qualifier from pointer target type
Hi Andrew,
Today's linux-next merge of the akpm-current tree got conflicts in:
arch/mips/include/uapi/asm/unistd.h
arch/mips/kernel/scall32-o32.S
arch/mips/kernel/scall64-n32.S
arch/mips/kernel/scall64-64.S
arch/mips/kernel/scall64-o32.S
between commit:
96fc7a9cee67 ("MIPS: Wire up
On Mon, Oct 05, 2015 at 12:38:55PM +0200, Geert Uytterhoeven wrote:
> Below is the list of build error/warning regressions/improvements in
> v4.3-rc4[1] compared to v4.2[2].
>
> 8 improvements:
> - /home/kisskb/slave/src/drivers/net/ethernet/via/via-rhine.c: error:
> implicit declaration of
On Mon, Oct 05, 2015 at 05:32:05PM -0700, Dmitry Torokhov wrote:
> On Sat, Oct 03, 2015 at 02:54:56PM +0530, Sudip Mukherjee wrote:
> > The variable i is used to check the port to attach to and we are
> > supposed to save the reference of struct db9 in the location given by
> > db9_base[i]. But
MINSIGSTKSZ and SIGSTKSZ for ARM64 are not correctly set in latest kernel.
This patch fixes this issue.
This issue is reported in LTP (testcase: sigaltstack02.c).
Testcase failed when sigaltstack() called with stack size "MINSIGSTKSZ - 1"
Since in Glibc-2.22, MINSIGSTKSZ is set to 5120 but in
On Tuesday 06 October 2015 03:40 AM, Andrew Morton wrote:
> On Sat, 3 Oct 2015 18:25:13 +0530 Vineet Gupta
> wrote:
>
>> Hi,
>>
>> I noticed increased boot time when enabling highmem for ARC. Turns out that
>> freeing highmem pages into buddy allocator is done page at a time, while it
>> is
>>
Hi all,
After merging the net-next tree, today's linux-next build (i386 defconfig)
failed like this:
net/built-in.o: In function `fib_rebalance':
fib_semantics.c:(.text+0x9d6d4): undefined reference to `__divdi3'
Caused by commit
0e884c78ee19 ("ipv4: L3 hash-based multipath")
I have
h8300 builds fail with
lib/string.c:31:32: fatal error: asm/word-at-a-time.h: No such file or directory
Fixes: 30035e45753b ("string: provide strscpy()")
Cc: Chris Metcalf
Signed-off-by: Guenter Roeck
---
arch/h8300/include/asm/Kbuild | 1 +
1 file changed, 1 insertion(+)
diff --git
On Tue, 2015-10-06 at 06:20 +0100, gre...@linuxfoundation.org wrote:
> On Tue, Oct 06, 2015 at 06:12:40AM +0100, gre...@linuxfoundation.org wrote:
> > On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> > > On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > > > On Tue,
On Tue, Oct 06, 2015 at 06:12:40AM +0100, gre...@linuxfoundation.org wrote:
> On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> > On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > > On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > > > From: Sudeep Dutt
> > >
On Mon, Oct 05, 2015 at 07:12:17PM -0700, Andrew Morton wrote:
> On Mon, 5 Oct 2015 12:29:36 -0700 Andrew Morton
> wrote:
>
> > Maybe it would be better to add the gfp_t argument to the
> > address_space_operations. At a minimum, writepage(), readpage(),
> > writepages(), readpages(). What a
On Mon, Oct 05, 2015 at 10:38:43AM -0700, Sudeep Dutt wrote:
> On Mon, 2015-10-05 at 03:50 -0700, Woodhouse, David wrote:
> > On Tue, 2015-09-29 at 18:09 -0700, Ashutosh Dixit wrote:
> > > From: Sudeep Dutt
> > >
> > > iova is a library which can be built without IOMMU_SUPPORT
> > >
> > >
Hi Suzuki,
[auto build test ERROR on v4.3-rc4 -- if it's inappropriate base, please ignore]
config: arm64-alldefconfig (attached as .config)
reproduce:
wget
https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross
-O ~/bin/make.cross
chmod +x
From: Jacob Kiefer
This patch fixes the following sparse errors:
CHECK drivers/staging/rtl8723au/hal/rtl8723a_cmd.c
...
drivers/staging/rtl8723au/hal/rtl8723a_cmd.c:118:25: \
warning: incorrect type in assignment (different base types)
drivers/staging/rtl8723au/hal/rtl8723a_cmd.c:118:25: \
On Mon, Oct 05, 2015 at 09:53:26PM +0200, Radim Krčmář wrote:
> 2015-09-28 13:38+0800, Haozhong Zhang:
> > Both VMX and SVM propagate virtual_tsc_khz in the same way, so this
> > patch removes the call-back set_tsc_khz() and replaces it with a common
> > function.
> >
> > Signed-off-by: Haozhong
On Sat, 2015-10-03 at 17:19 +0200, Christoph Hellwig wrote:
> This ensures the dma mask that is supported by the driver is recorded
> in the device structure.
>
> Signed-off-by: Christoph Hellwig
> ---
> drivers/net/ethernet/amd/pcnet32.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
On 2015/09/22 15:23, Ingo Molnar wrote:
> So when memory hotplug removes a piece of physical memory from pagetable
> mappings, it also frees the underlying PGD entry.
>
> This complicates PGD management, so don't do this. We can keep the
> PGD mapped and the PUD table all clear - it's only a
On Fri, 2015-10-02 at 13:11 -0700, Nishanth Aravamudan wrote:
> Drivers like NVMe need to be able to determine the page size used for
> DMA transfers. Add a new API that defaults to return PAGE_SHIFT on all
> architectures.
>
> Signed-off-by: Nishanth Aravamudan
>
> diff --git
On Sat, 2015-10-03 at 04:33 +0800, kbuild test robot wrote:
> Hi Nishanth,
>
> [auto build test results on v4.3-rc3 -- if it's inappropriate base, please
> ignore]
>
> config: powerpc-defconfig (attached as .config)
> reproduce:
> wget
>
On Fri, 2015-10-02 at 08:43 -0700, Laura Abbott wrote:
> Hi,
>
> We received a report (https://bugzilla.redhat.com/show_bug.cgi?id=1267395) of
> bad assembly
> when compiling on powerpc with little endian
...
> After some discussion with the binutils folks, it turns out that the tlbie
>
On Tue, Oct 06, 2015 at 03:22:40AM +0200, Marcin Wojtas wrote:
> The newest revisions of A388-GP (v1.5 and higher) support only
> DAT3-based card detection, which is enabled by this commit. Hitherto
> revisions, without such modification, will be impacted with a broken
> card detection - in order
On Fri, Oct 02, 2015 at 10:18:00AM -0700, Nishanth Aravamudan wrote:
> We will leverage this macro in the NVMe driver, which needs to know the
> configured IOMMU page shift to properly configure its device's page
> size.
>
> Signed-off-by: Nishanth Aravamudan
>
> ---
> Given this is available,
Replace (1 << x) by BIT(x)
Signed-off-by: Ronit halder
---
drivers/staging/fbtft/fb_ssd1351.c | 4 +--
drivers/staging/fbtft/fbtft.h | 56 +++---
2 files changed, 30 insertions(+), 30 deletions(-)
diff --git a/drivers/staging/fbtft/fb_ssd1351.c
Roger,
On 2015년 10월 05일 19:49, Roger Quadros wrote:
> Chanwoo,
>
> On 05/10/15 12:04, Chanwoo Choi wrote:
>> Hi Roger,
>>
>> On 2015년 10월 05일 18:01, Chanwoo Choi wrote:
>>> Hi Roger,
>>>
>>> On 2015년 10월 05일 17:26, Roger Quadros wrote:
Chanwoo,
On 04/10/15 06:45, Chanwoo Choi
Hello Will,
Thank you so much for the review. I have thought it over and it makes
sense not to have that comment in cmpxchg.h, I will also change !defined to
#ifndef and quickly post a v3.
Regards,
Sarbojit
--- Original Message ---
Sender : Will Deacon
Date : Oct 05, 2015 21:30
Hi all,
Today's linux-next merge of the tip tree got a conflict in:
arch/arm64/kernel/acpi.c
between commit:
f26527b1428f ("irqchip / GIC: Convert the GIC driver to ACPI probing")
from the pm tree and commit:
89e44b51cc0d ("arm64, acpi/apei: Implement arch_apei_get_mem_attributes()")
> -Original Message-
> From: Andy Lutomirski [mailto:l...@amacapital.net]
> Sent: Tuesday, October 06, 2015 3:06 AM
>
> >> And what if cap_hdr isn't written yet?
> >
> > This design mainly targeting a simplest interface that user could
> > upload efi capsule in a single command action: cat
work-simple.c uses EXPORT_SYMBOL_GPL(), defined in .
Signed-off-by: Ben Hutchings
Cc: stable...@vger.kernel.org
---
kernel/work-simple.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/kernel/work-simple.c b/kernel/work-simple.c
index c996f75..e57a052 100644
--- a/kernel/work-simple.c
+++
The type of the slowfn parameter is declared incorrectly in some of
the backports, resulting in:
kernel/rtmutex.c: In function 'rt_mutex_timed_futex_lock':
kernel/rtmutex.c:1730:12: warning: passing argument 5 of
'rt_mutex_timed_fastlock' from incompatible pointer type [enabled by default]
gcc warns about casting directly from a 32-bit pointer to u64. In the
current rt patch series we cast to unsigned long to avoid the warning,
but this fix hasn't been backported to all versions.
Signed-off-by: Ben Hutchings
Cc: # 3.2 3.4
---
kernel/irq/manage.c | 3 ++-
1 file changed, 2
The timer handler was named kvm_timer_fn, not apic_timer_fn, before
Linux 3.7.
Signed-off-by: Ben Hutchings
Cc: # 3.2 3.4
---
arch/x86/kvm/lapic.c | 4 +---
1 file changed, 1 insertion(+), 3 deletions(-)
diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
index 7198be0..c661909 100644
Hi Paul,
Today's linux-next merge of the audit tree got a conflict in:
include/linux/lsm_audit.h
between commit:
671a2781ff01 ("security: add ioctl specific auditing to lsm_audit")
from Linus' tree and commit:
43cfd5e38587 ("audit: constify parts of common_audit_data and
On Mon, Oct 5, 2015 at 6:52 PM, Frank Rowand wrote:
> On 9/21/2015 7:03 AM, Tomeu Vizoso wrote:
>> Some initcalls in the late level assume that some devices will have
>> already probed without explicitly checking for that.
>>
>> After the recent move to defer most device probes when they are
>>
On Tue, 2015-10-06 at 08:48 +1100, paul.sz...@sydney.edu.au wrote:
> The Linux CFS scheduler prefers pinned tasks and unfairly
> gives more CPU time to tasks that have set CPU affinity.
> This effect is observed with or without CGROUP controls.
>
> To demonstrate: on an otherwise idle machine, as
These patches fix various warnings I got when building 3.2.71-rt104;
some of them are applicable to other RT branches as well.
Ben.
Ben Hutchings (4):
KVM: lapic: Fix name of timer handler before 3.7
rtmutex: Fix rt_mutex_timed_fastlock() slowfn type mismatch up to 3.10
genirq: Fix pointer
On Fri, Sep 25 2015 at 07:04 -0600, Marc Titinger wrote:
From: Marc Titinger
fake path to start testing, eventually move this out of /arch/.
incidently enable PM_GENERIC_DOMAINS for VExpress.
In fact, this could be moved out of ARM. My last series moved it to
drivers/base/power/.
-- Lina
- On Oct 5, 2015, at 7:21 PM, Rusty Russell ru...@ozlabs.org wrote:
> Mathieu Desnoyers writes:
>> Hi Andrew,
>>
>> Here is a repost of sys_membarrier, rebased on top of Linus commit
>> c4b5fd3fb2058b650447372472ad24e2a989f9f6 without any change since the
>> last v19 post other that
On Mon, 5 Oct 2015 12:29:36 -0700 Andrew Morton
wrote:
> Maybe it would be better to add the gfp_t argument to the
> address_space_operations. At a minimum, writepage(), readpage(),
> writepages(), readpages(). What a pickle.
I'm being dumb. All we need to do is to add a new
On 10/05/2015 03:11 AM, Benjamin Gaignard wrote:
version 4 changes:
- rebased on kernel 4.3-rc3
- fix missing EXPORT_SYMBOL for smaf_create_handle()
version 3 changes:
- Remove ioctl for allocator selection instead provide the name of
the targeted allocator with allocation request.
Nicolas Boichat writes:
> The section mismatch warning can be easy to miss during the kernel build
> process. Allow it to be marked as fatal to be easily caught and prevent
> bugs from slipping in.
>
> Setting CONFIG_SECTION_MISMATCH_WARN_ONLY=y causes these warnings to be
> non-fatal, since
Mathieu Desnoyers writes:
> Hi Andrew,
>
> Here is a repost of sys_membarrier, rebased on top of Linus commit
> c4b5fd3fb2058b650447372472ad24e2a989f9f6 without any change since the
> last v19 post other that proceeding to further testing. When merging
> with other system calls, system call
On 10/05/2015 03:11 AM, Benjamin Gaignard wrote:
diff --git a/drivers/smaf/smaf-core.c b/drivers/smaf/smaf-core.c
new file mode 100644
index 000..37914e7
--- /dev/null
+++ b/drivers/smaf/smaf-core.c
@@ -0,0 +1,736 @@
+/*
+ * smaf.c
+ *
+ * Copyright (C) Linaro SA 2015
+ * Author: Benjamin
On Fri, Sep 25, 2015 at 2:31 PM, James Liao wrote:
> In kernel late init, it turns off all unused clocks, which
> needs to access subsystem registers such as VENC and VENC_LT.
>
> Accessing MT8173 VENC registers needs two top clocks, mm_sel and
> venc_sel. Accessing VENC_LT registers needs mm_sel
On Wed, Sep 30, 2015 at 4:35 PM, Dave Chinner wrote:
> On Tue, Sep 29, 2015 at 08:41:36PM -0400, Dan Williams wrote:
>> The DAX implementation needs to protect new calls to ->direct_access()
>> and usage of its return value against unbind of the underlying block
>> device. Use
Dear Kukjin,
This is an updated pull request with DTS related changes for 4.4.
Description along with a tag.
You can find them also on the lists with my reviewed-by.
Best regards,
Krzysztof
The following changes since commit 6ff33f3902c3b1c5d0db6b1e2c70b6d76fba357f:
Linux 4.3-rc1
On Mon, Oct 05, 2015 at 09:26:30PM +0200, Radim Krčmář wrote:
> 2015-09-28 13:38+0800, Haozhong Zhang:
> > This patch moves the field of TSC scaling ratio from the architecture
> > struct vcpu_svm to the common struct kvm_vcpu_arch.
> >
> > Signed-off-by: Haozhong Zhang
> > ---
> > diff --git
Hi Stephen,
[auto build test ERROR on next-20151002 -- if it's inappropriate base, please
ignore]
config: i386-randconfig-x006-201540 (attached as .config)
reproduce:
# save the attached .config to linux build tree
make ARCH=i386
All errors (new ones prefixed by >>):
On 06.10.2015 10:47, Kukjin Kim wrote:
> Krzysztof Kozlowski wrote:
>>
>> Dear Kukjin,
>>
>> One ARMv8 DTS change for 4.4.
>>
>> Best regards,
>> Krzysztof
>>
>> The following changes since commit 6ff33f3902c3b1c5d0db6b1e2c70b6d76fba357f:
>>
>> Linux 4.3-rc1 (2015-09-12 16:35:56 -0700)
>>
>> are
Krzysztof Kozlowski wrote:
>
> Dear Kukjin,
>
> One ARMv8 DTS change for 4.4.
>
> Best regards,
> Krzysztof
>
> The following changes since commit 6ff33f3902c3b1c5d0db6b1e2c70b6d76fba357f:
>
> Linux 4.3-rc1 (2015-09-12 16:35:56 -0700)
>
> are available in the git repository at:
>
>
On 06.10.2015 10:42, Kukjin Kim wrote:
> Krzysztof Kozlowski wrote:
>>
>> Dear Kukjin,
>>
> Hi,
>
>> DTS related changes for 4.4.
>>
>> Description along with a tag.
>> You can find them also on the lists with my reviewed-by.
>>
> I have small comments. Then I could just pull your branch...
>
>>
Hi Jens,
After merging the block tree, today's linux-next build (arm
multi_v7_defconfig) failed like this:
drivers/block/loop.c: In function 'lo_rw_aio_complete':
drivers/block/loop.c:474:2: error: too few arguments to function
'blk_mq_complete_request'
blk_mq_complete_request(rq);
^
In
Krzysztof Kozlowski wrote:
>
> Dear Kukjin,
>
Hi,
> DTS related changes for 4.4.
>
> Description along with a tag.
> You can find them also on the lists with my reviewed-by.
>
I have small comments. Then I could just pull your branch...
> Best regards,
> Krzysztof
>
>
> The following
On Tue, Oct 6, 2015 at 4:59 AM, Felipe Balbi wrote:
> Vaishali Thakkar writes:
>
>> On Mon, Aug 24, 2015 at 2:29 PM, David Laight
>> wrote:
>>> From: Vaishali Thakkar [mailto:vthakkar1...@gmail.com]
Sent: 22 August 2015 02:57
>>> ...
>> - .bcdADC =
drivers/input/touchscreen/ft6236.c:316:3-8: No need to set .owner here. The
core will do it.
Remove .owner field if calls are used which set it automatically
Generated by: scripts/coccinelle/api/platform_no_drv_owner.cocci
CC: Noralf Trønnes
Signed-off-by: Fengguang Wu
---
ft6236.c |1
Hi Jens,
Today's linux-next merge of the block tree got a conflict in:
drivers/block/loop.c
between commit:
f4829a9b7a61 ("blk-mq: fix racy updates of rq->errors")
from Linus' tree and commit:
bc07c10a3603 ("block: loop: support DIO & AIO")
from the block tree.
I fixed it up (see
Hi Rasmus,
On Sun, Oct 4, 2015 at 9:09 AM, Rasmus Villemoes
wrote:
> Subject: [PATCH 2/2] scsi: reduce CONFIG_SCSI_CONSTANTS=y impact by 8k
>
> On 64 bit, struct error_info has 6 bytes of padding, which amounts to
> over 4k of wasted space in the additional[] array. We could easily get
> rid of
On Tue, Oct 06, 2015 at 12:01:19AM +0200, Andreas Gruenbacher wrote:
> On Mon, Oct 5, 2015 at 11:17 PM, Dave Chinner wrote:
> > On Mon, Oct 05, 2015 at 08:45:40PM +0200, Andreas Gruenbacher wrote:
> >> On Sun, Oct 4, 2015 at 8:23 AM, Christoph Hellwig
> >> wrote:
> >> > After that the wire up
From: Nadav Haklai
According to 'FE-2946959' erratum the clock inversion option is
needed to support slow frequencies when the card input hold time
requirement is high. This setting is not required for high speed
MMC and might cause timing violation.
Signed-off-by: Nadav Haklai
Cc: # v4.2
---
When resuming from suspend on Armada 38x SoC MBus windows have to be
re-configured and for that purpose mv_conf_mbus_windows function needed
rework. MBus windows register base address obtaining was moved to
armada_38x_quirks function in order to be kept in pxa global structure,
because it is used
In case of armada_38x_quirks error, all clocks should be cleaned-up, same
as after mv_conf_mbus_windows failure.
Signed-off-by: Marcin Wojtas
Cc: # v4.2
---
drivers/mmc/host/sdhci-pxav3.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/mmc/host/sdhci-pxav3.c
Hi Maxime,
On Tue, Oct 6, 2015 at 1:23 AM, Maxime Ripard
wrote:
> The C.H.I.P. is a small SBC with an Allwinner R8, 8GB of NAND, 512MB of
> RAM, USB host and OTG, a wifi / bluetooth combo chip, an audio/video jack
> and two connectors to plug additional boards on top of it.
Sorry for the late
The newest revisions of A388-GP (v1.5 and higher) support only
DAT3-based card detection, which is enabled by this commit. Hitherto
revisions, without such modification, will be impacted with a broken
card detection - in order to operate the cards have to be present
during kernel boot.
On 10/05/2015 04:05 PM, Sean O. Stalley wrote:
On Fri, Oct 02, 2015 at 08:16:48PM -0700, Yinghai Lu wrote:
On Fri, Oct 2, 2015 at 3:37 PM, David Daney wrote:
From: David Daney
PCI Enhanced Allocation is a new method of allocating MMIO & IO
resources for PCI devices & bridges. It can be used
Some sdhci hosts may require handling quirks during card initialization at
the time when its type is already known. Hence a new callback (init_card)
is added in sdhci_ops.
Signed-off-by: Marcin Wojtas
---
drivers/mmc/host/sdhci.c | 9 +
drivers/mmc/host/sdhci.h | 1 +
2 files changed,
On Marvell Armada 38x SoC's the MMC_CARD bit in SD_CE_ATA_1 register must
be set to 0x1 when a MMC card is supposed to work in DDR mode, or when
commands CMD11, CMD14 and CMD20 are used.
This commit enables the above for all MMC cards by modifying the host
registers during card initialization. It
Recently an issue was reported that was difficult to detect except
by tweaking the adjtimex tick value, and noticing how quickly the
adjustment took to be made:
https://lkml.org/lkml/2015/9/1/488
Thus this patch introduces a new test which manipulates the adjtimex
tick value and validates
Marvell Armada 38x SDHCI controller enable using DAT3 pin as a hardware
card detection. According to the SD sdandard this signal can be used for
this purpose combined with a pull-up resistor, implying inverted (active
low) polarization of a card detect. MMC standard does not support this
feature
Hi,
This series brings a couple of fixes and improvements to Armada 38x SDHCI
controller driver. First four patches are fixes, of which three are stable
CC'ed.
Another two add DAT3-pin based hardware card detection in the driver, what
should be used by the newest revisions of A388-GP boards.
From: Nadav Haklai
shci-pxav3 driver is enabling by default the
SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN quirk. However this quirk is not
required for Armada 38x and leads to wrong clock setting in the divider.
Signed-off-by: Nadav Haklai
Signed-off-by: Marcin Wojtas
Cc: # v4.2
---
Hi Dave,
After merging the drm tree, today's linux-next build (arm
multi_v7_defconfig) failed like this:
drivers/gpu/drm/drm_fb_helper.c: In function 'restore_fbdev_mode':
drivers/gpu/drm/drm_fb_helper.c:448:5: error: 'error' undeclared (first use in
this function)
error = true;
^
The first two patches are optimizations that I'm surprised we didn't
already have. I noticed them when I was looking at the generated
asm.
The next two patches are tests and some old stuff. There's a test
that validates the vDSO AT_SYSINFO annotations. There's also a test
that exercises some
This should improve code quality a bit. It also shrinks the kernel
text.
Before:
textdata bss dec hex filename
218283795194760 1277952 283010911afd723 vmlinux
textdata bss dec hex filename
218279975194760 1277952 28300709
Rather than worrying about exactly where LOCKDEP_SYS_EXIT should go
in the asm code, add it to prepare_exit_from_usermode and remove all
of the asm calls that are followed by prepare_exit_to_usermode.
LOCKDEP_SYS_EXIT now appears only in the syscall fast paths.
Signed-off-by: Andy Lutomirski
While the kernel itself doesn't use DWARF unwinding, user code
expects to be able to unwind the vDSO. The vsyscall (AT_SYSINFO)
entry is manually CFI-annotated, and this tests that it unwinds
correctly.
I tested the test by incorrectly annotating __kernel_vsyscall, and
the test indeed fails if I
Clearing NT is part of the prologue, whereas loading up arg6 makes
more sense to think about as part of syscall processing. Reorder
them.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/entry_64_compat.S | 18 +-
1 file changed, 9 insertions(+), 9 deletions(-)
diff --git
Before we start calling execve in contexts that honor the full
pt_regs, we need to teach it to initialize all registers.
Signed-off-by: Andy Lutomirski
---
arch/x86/include/asm/elf.h | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/arch/x86/include/asm/elf.h
The goal is to integrate the SYSENTER and SYSCALL32 entry paths with
the INT80 path. SYSENTER clobbers ESP and EIP. SYSCALL32 clobbers
ECX (and, invisibly, R11). SYSRETL (long mode to compat mode)
clobbers ECX and, invisibly, R11. SYSEXIT (which we only need for
native 32-bit) clobbers ECX and
32-bit userspace will now always see the same vDSO, which is exactly
what used to be the int80 vDSO. Subsequent patches will clean it up
and make it support SYSENTER and SYSCALL using alternatives.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/vdso/Makefile | 35 +++---
Commit 131484c8da97 ("x86/debug: Remove perpetually broken,
unmaintainable dwarf annotations") removed all the manual DWARF
annotations outside the vDSO. It also removed the macros we used
for the manual annotations.
Re-add the macros so that we can clean up the vDSO annotations.
Signed-off-by:
These audit optimizations are messy and hard to maintain. We'll
get a similar effect from opportunistic sysret when fast compat
system calls are re-implemented.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/entry_64_compat.S | 98 +---
1 file changed, 2
For the vDSO, user code wants runtime unwind info. Make sure that,
if we use .cfi directives, we generate it.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/vdso/Makefile | 4 ++--
arch/x86/include/asm/dwarf2.h | 13 ++---
2 files changed, 12 insertions(+), 5 deletions(-)
diff
We've disabled the vDSO helpers to call them, so turn off the
entries entirely (temporarily) in preparation for cleaning them up.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/entry_64_compat.S | 13 +
1 file changed, 13 insertions(+)
diff --git
The header was missing some compat declarations.
Also make sys_call_ptr_t have a consistent type.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/syscall_32.c | 5 +
arch/x86/include/asm/syscall.h | 12 +++-
arch/x86/um/sys_call_table_32.c | 3 +--
Maintaining the current CFI annotations written in R'lyehian is
difficult for most of us. Translate them to something a little
closer to English.
This will remove the CFI data for kernels built with extremely old
versions of binutils. I think this is a fair tradeoff for the
ability for mortals
We now have only one code path that calls through the compat syscall
table. This will make it much more pleasant to change the pt_regs
vs register calling convention, which we need to do to move the call
into C.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/entry_64_compat.S | 246
On Mon, Oct 05, 2015 at 03:44:07PM +0100, Will Deacon wrote:
> On Thu, Oct 01, 2015 at 07:03:01PM +0100, Paul E. McKenney wrote:
> > On Thu, Oct 01, 2015 at 07:13:04PM +0200, Peter Zijlstra wrote:
> > > On Thu, Oct 01, 2015 at 08:09:09AM -0700, Paul E. McKenney wrote:
> > > > On Thu, Oct 01, 2015
System calls are really quite simple. Add a helper to call
a 32-bit system call.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/common.c | 43 +++
1 file changed, 43 insertions(+)
diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
index
This handles both SYSENTER and SYSCALL. The asm glue will take care
of the differences.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/common.c | 43 +++
1 file changed, 43 insertions(+)
diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
This is conceptually simpler. More importantly, it eliminates the
PTREGSCALL and execve stubs, which were not compatible with the C
ABI. This means that C code can call through the compat syscall
table.
The execve stubs are a bit subtle. They did two things: they
cleared some registers and
Syscalls are asmlinkage functions (on 32-bit kernels), take six args
of type unsigned long, and return long. Note that uml could
probably be slightly cleaned up on top of this patch.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/syscall_32.c | 4 ++--
arch/x86/entry/syscall_64.c |
What, you didn't realize that SYSENTER and SYSCALL were actually the
same thing? :)
Unlike the old code, this actually passes the ptrace_syscall_32 test
on AMD systems.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/entry_64_compat.S | 90 +++-
syscall_exit is going away, and return tracing is just a function
call now, so open-code the two non-syscall 32-bit users.
While we're at it, update the big register layout comment.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/entry_32.S | 18 +++---
1 file changed, 15
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/entry_32.S | 34 +-
1 file changed, 13 insertions(+), 21 deletions(-)
diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
index 36b6beb181f6..02881e528945 100644
--- a/arch/x86/entry/entry_32.S
+++
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/entry_64.S| 2 +-
arch/x86/entry/entry_64_compat.S | 56 +---
2 files changed, 19 insertions(+), 39 deletions(-)
diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
index
This is slightly messy, but it eliminates an unnecessary cli;sti pair.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/common.c | 18 +++---
1 file changed, 11 insertions(+), 7 deletions(-)
diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
index
These checks are quite slow. Disable them in non-lockdep kernels to
reduce the performance hit.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/common.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
index
On 10/5/15 3:14 PM, Daniel Borkmann wrote:
One scenario that comes to mind ... what happens when there are kernel
pointers stored in skb->cb[] (either from the current layer or an old
one from a different layer that the skb went through previously, but
which did not get overwritten)?
Socket
we're following a 32-bit pointer, and the uaccess code isn't smart
enough to figure out that the access_ok check isn't needed.
This saves about three cycles on a cache-hot fast syscall.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/common.c | 16 ++--
1 file changed, 14
This shaves a few cycles off the slow paths.
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/common.c | 6 --
1 file changed, 4 insertions(+), 2 deletions(-)
diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
index 3ae6191ad1a2..88966fbf7125 100644
---
It generates simpler and faster code than current_thread_info().
Signed-off-by: Andy Lutomirski
---
arch/x86/entry/common.c | 22 +++---
1 file changed, 11 insertions(+), 11 deletions(-)
diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
index
1 - 100 of 1788 matches
Mail list logo
