There are no return value checking when using kzalloc() and kcalloc() for
memory allocation. so add it.
Signed-off-by: Yunfeng Ye
---
v1 -> v2:
- return error code when memory allocation failure
arch/arm64/kernel/armv8_deprecated.c | 57 +++-
1 file changed, 43
On 06/10/2019 18:18, Greg Kroah-Hartman wrote:
> This is the start of the stable review cycle for the 4.4.196 release.
> There are 36 patches in this series, all will be posted as a response
> to this one. If anyone has any issues with these being applied, please
> let me know.
>
> Responses sh
On 06/10/2019 18:20, Greg Kroah-Hartman wrote:
> This is the start of the stable review cycle for the 4.9.196 release.
> There are 47 patches in this series, all will be posted as a response
> to this one. If anyone has any issues with these being applied, please
> let me know.
>
> Responses sh
On 06/10/2019 18:20, Greg Kroah-Hartman wrote:
> This is the start of the stable review cycle for the 4.19.78 release.
> There are 106 patches in this series, all will be posted as a response
> to this one. If anyone has any issues with these being applied, please
> let me know.
>
> Responses s
On 06/10/2019 18:19, Greg Kroah-Hartman wrote:
> This is the start of the stable review cycle for the 5.2.20 release.
> There are 137 patches in this series, all will be posted as a response
> to this one. If anyone has any issues with these being applied, please
> let me know.
>
> Responses sh
On Fri 04-10-19 07:32:40, Jens Axboe wrote:
> On 10/4/19 5:01 AM, Mika Westerberg wrote:
> > On Fri, Oct 04, 2019 at 11:59:26AM +0200, Rafael J. Wysocki wrote:
> >> On Friday, October 4, 2019 10:03:40 AM CEST Mika Westerberg wrote:
> >>> On Thu, Oct 03, 2019 at 09:50:33AM -0700, Tejun Heo wrote:
>
On 06/10/2019 18:29:19+0800, YueHaibing wrote:
> devm_platform_ioremap_resource() internally have platform_get_resource()
> and devm_ioremap_resource() in it. So instead of calling them separately
> use devm_platform_ioremap_resource() directly.
>
> YueHaibing (34):
> rtc: asm9260: use devm_plat
On 06/10/2019 18:19, Greg Kroah-Hartman wrote:
> This is the start of the stable review cycle for the 5.3.5 release.
> There are 166 patches in this series, all will be posted as a response
> to this one. If anyone has any issues with these being applied, please
> let me know.
>
> Responses sho
On 06/10/2019 18:20, Greg Kroah-Hartman wrote:
> This is the start of the stable review cycle for the 4.14.148 release.
> There are 68 patches in this series, all will be posted as a response
> to this one. If anyone has any issues with these being applied, please
> let me know.
>
> Responses s
Add initial support for Analog Devices ADUX1020 Photometric sensor.
Only proximity mode has been enabled for now.
Signed-off-by: Manivannan Sadhasivam
---
drivers/iio/light/Kconfig| 11 +
drivers/iio/light/Makefile | 1 +
drivers/iio/light/adux1020.c | 783 ++
Add devicetree binding for Analog Devices ADUX1020 Photometric
sensor.
Signed-off-by: Manivannan Sadhasivam
---
.../bindings/iio/light/adux1020.txt | 22 +++
1 file changed, 22 insertions(+)
create mode 100644 Documentation/devicetree/bindings/iio/light/adux1020.txt
d
Hello,
This patchset adds initial IIO driver support for ADUX1020 Photometric
sensor from Analog Devices. This sensor is usable for multiple optical
measurement applications, including gesture control and proximity sensing.
This initial driver includes support for only proximity mode with event
b
When sendmsg() finds a call to continue on with, if the call is in an
inappropriate state, it doesn't release the ref it just got on that call
before returning an error.
This causes the following symptom to show up with kasan:
BUG: KASAN: use-after-free in rxrpc_send_keepalive+0x8a2/0x940
On Mon, Oct 7, 2019 at 3:35 PM Maxime Ripard wrote:
>
> On Sat, Oct 05, 2019 at 04:15:57PM +0530, Amit Kucheria wrote:
> > On Thu, Oct 3, 2019 at 5:22 PM Maxime Ripard wrote:
> > >
> > > Hi,
> > >
> > > On Thu, Oct 03, 2019 at 04:52:24PM +0530, Amit Kucheria wrote:
> > > > This new binding fails
are tagged here:
git://git.kernel.org/pub/scm/linux/kernel/git/dhowells/linux-fs.git
rxrpc-fixes-20191007
and can also be found on the following branch:
http://git.kernel.org/cgit/linux/kernel/git/dhowells/linux-fs.git/log/?h=rxrpc-fixes
David
---
David Howells (6):
rxrpc: Fi
rxrpc_put_peer() calls trace_rxrpc_peer() after it has done the decrement
of the refcount - which looks at the debug_id in the peer record. But
unless the refcount was reduced to zero, we no longer have the right to
look in the record and, indeed, it may be deleted by some other thread.
Fix this
Fix the cleanup of the crypto state on a call after the call has been
disconnected. As the call has been disconnected, its connection ref has
been discarded and so we can't go through that to get to the security ops
table.
Fix this by caching the security ops pointer in the rxrpc_call struct and
The rxrpc_peer record needs to hold a reference on the rxrpc_local record
it points as the peer is used as a base to access information in the
rxrpc_local record.
This can cause problems in __rxrpc_put_peer(), where we need the network
namespace pointer, and in rxrpc_send_keepalive(), where we nee
On Mon 30-09-19 18:42:33, Dave Chinner wrote:
> On Wed, Sep 25, 2019 at 04:46:03PM -0700, Ira Weiny wrote:
> > On Tue, Sep 24, 2019 at 08:26:20AM +1000, Dave Chinner wrote:
> > > Hence, AFIACT, the above definition of a F_RDLCK|F_LAYOUT lease
> > > doesn't appear to be compatible with the semantics
On Thu 26-09-19 16:34:08, Chengguang Xu wrote:
> It is meaningless to increase DQST_LOOKUPS number while iterating
> over dirty/inuse list, so just avoid it.
>
> Signed-off-by: Chengguang Xu
Yeah, makes sense. I've queued up your patch. Thanks!
On Mon 23-09-19 21:52:23, Chengguang Xu wrote:
> Code cleanup for hash bits calculation by
> calling ilog2().
>
> Signed-off-by: Chengguang Xu
Thanks! I've queued your patch.
Honza
> ---
> v1->v2:
> - Calculate hash bits by direct
rxrpc_put_call() calls trace_rxrpc_call() after it has done the decrement
of the refcount - which looks at the debug_id in the call record. But
unless the refcount was reduced to zero, we no longer have the right to
look in the record and, indeed, it may be deleted by some other thread.
Fix this
rxrpc_put_*conn() calls trace_rxrpc_conn() after they have done the
decrement of the refcount - which looks at the debug_id in the connection
record. But unless the refcount was reduced to zero, we no longer have the
right to look in the record and, indeed, it may be deleted by some other
thread.
On 04.10.2019 23:39, Uwe Kleine-König wrote:
> External E-Mail
>
>
> On Fri, Oct 04, 2019 at 09:35:23AM +, claudiu.bez...@microchip.com wrote:
>> Hi Kamel,
>>
>> On 02.10.2019 17:46, Kamel Bouhara wrote:
>>> +static int at91_init_twi_recovery_info(struct platform_device *pdev,
>>> +
On Sun, 2019-10-06 at 10:06 +0100, Jonathan Cameron wrote:
> [External]
>
> On Sun, 6 Oct 2019 09:53:33 +0100
> Jonathan Cameron wrote:
>
> > On Thu, 26 Sep 2019 14:18:03 +0300
> > Alexandru Ardelean wrote:
> >
> > > The lock can be extended a bit to protect other elements that are not
> > > p
From: Colin Ian King
Don't populate the array st6422_bridge_init on the stack but instead
make it static. Makes the object code smaller by 231 bytes.
Before:
textdata bss dec hex filename
3419 752 644235108b gspca/stv06xx/stv06xx_st6422.o
After:
text
On Mon, 2019-10-07 at 15:40 +0530, Manivannan Sadhasivam wrote:
> [External]
>
> Add devicetree binding for Analog Devices ADUX1020 Photometric
> sensor.
>
Hey,
Thanks for the patches.
This dt-binding docs is in text format.
dt-binding docs now need to be in YAML format.
Also, patches for dt-
This commit fixes id path of allwinner,sun4i-a10-csi.yaml location.
Fixes: c5e8f4ccd775 ("media: dt-bindings: media: Add Allwinner A10 CSI binding")
Signed-off-by: Alexandre Torgue
diff --git
a/Documentation/devicetree/bindings/media/allwinner,sun4i-a10-csi.yaml
b/Documentation/devicetree/bind
This commit fixes an issue seen during yaml check ("make dt_binding_check").
Each enum were not declared as uint32.
"Documentation/devicetree/bindings/net/adi,adin.yaml:
properties:adi,rx-internal-delay-ps:
..., 'enum': [1600, 1800, 2000, 2200, 2400], 'default': 2000}
is not valid under any of the
Using "make dt_binding_check" on top v5.4-rc1 some errors are
reported in several schemas. Those 3 patches allow to execute
"make dt_binding_check" without issues. Most probably those patches
have alread been sent.
regards
Alexandre
Alexandre Torgue (3):
dt-bindings: media: Fix id path for su
This commit fixes an issue seen during yaml check ("make dt_binding_check").
Compatible didn't seem to be seen as a string.
Reported issue:
"properties:compatible:enum:0: {'const': 'regulator-fixed'}
is not of type 'string'"
And
"properties:compatible:enum:1: {'const': 'regulator-fixed-clock'}
is
On Sun, Oct 06, 2019 at 04:10:20PM -0300, Gabriela Bittencourt wrote:
> Cleans up CHECKs of "Alignment should match open parenthesis"
>
> Signed-off-by: Gabriela Bittencourt
> ---
> drivers/staging/vt6656/rxtx.c | 63 +++
> 1 file changed, 41 insertions(+), 22 del
On Sun, Oct 06, 2019 at 07:20:15PM -0300, Gabriela Bittencourt wrote:
> Cleans up checks of: "Alignment should match open parenthesis"
>
> Signed-off-by: Gabriela Bittencourt
> ---
> drivers/staging/rtl8712/osdep_service.h | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
You sent 2 diffe
On Sun, Oct 06, 2019 at 05:34:20PM -0300, Gabriela Bittencourt wrote:
> Cleans up warnings of "Block comments should align the * on each line"
>
> Signed-off-by: Gabriela Bittencourt
> ---
> drivers/staging/rtl8712/recv_linux.c | 4 ++--
> 1 file changed, 2 insertions(+), 2 deletions(-)
>
> dif
Hi Christian,
On Mon, Oct 07, 2019 at 01:52:16AM +0200, Christian Brauner wrote:
> When assiging and testing taskstats in taskstats_exit() there's a race
> when writing and reading sig->stats when a thread-group with more than
> one thread exits:
>
> cpu0:
> thread catches fatal signal and whole
On 04 October 2019 16:05, Krzysztof Kozlowski wrote:
> platform_get_irq_byname() might return -errno which later would be
> cast to an unsigned int and used in request_irq().
>
> Signed-off-by: Krzysztof Kozlowski
This doesn't break the existing RTC functionality when running up on my test
boar
Once again RPL checks have been introduced which don't account for a
32-bit kernel living in ring 1 when running in a PV Xen domain. The
case in FIXUP_FRAME has been preventing boot; adjust BUG_IF_WRONG_CR3
as well just in case.
Fixes: 3c88c692c287 ("x86/stackframe/32: Provide consistent pt_regs")
> On Oct 7, 2019, at 5:18 AM, Vlastimil Babka wrote:
>
> The commit 8974558f49a6 ("mm, page_owner, debug_pagealloc: save and dump
> freeing stack trace") enhanced page_owner to also store freeing stack trace,
> when debug_pagealloc is also enabled. KASAN would also like to do this [1] to
> imp
On 05 October 2019 00:10, Dmitry Torokhov wrote:
> devm_gpiod_get_from_of_node() is being retired in favor of
> devm_fwnode_gpiod_get_index(), that behaves similar to
> devm_gpiod_get_index(), but can work with arbitrary firmware node. It
> will also be able to support secondary software nodes.
>
Hi Jernej,
On 9/29/19 10:00 PM, Jernej Skrabec wrote:
> This series adds support for decoding multi-slice H264 frames along with
> support for V4L2_DEC_CMD_FLUSH and V4L2_BUF_FLAG_M2M_HOLD_CAPTURE_BUF.
>
> Code was tested by modified ffmpeg, which can be found here:
> https://github.com/jernejsk/
On 10/6/19 10:23 AM, Shawn Guo wrote:
On Thu, Sep 12, 2019 at 10:29:28PM +0200, Parthiban Nallathambi wrote:
Extend Phycore i.MX6UL SoM for i.MX6ULL with on board eMMC. Phycore
i.MX6ULL is deployed with same carrier board Segin as the pins are
compatible with UL version.
Signed-off-by: Parth
Both multi_cpu_stop() and set_state() access multi_stop_data::state
racily using plain accesses. These are subject to compiler
transformations which could break the intended behaviour of the code,
and this situation is detected by KCSAN on both arm64 and x86 (splats
below).
Let's improve matters b
On Mon, Oct 07, 2019 at 10:02:21AM +0300, Adrian Hunter wrote:
> slow_copyfile() opens the file by name, so "write" permissions must not
> be removed in copyfile_mode_ns() before calling slow_copyfile().
>
> Example:
>
> Before:
> $ sudo chmod +r /proc/kcore
> $ sudo setcap "cap_sys_admin,ca
On Mon, Oct 07, 2019 at 03:45:29PM +0530, Amit Kucheria wrote:
> On Mon, Oct 7, 2019 at 3:35 PM Maxime Ripard wrote:
> >
> > On Sat, Oct 05, 2019 at 04:15:57PM +0530, Amit Kucheria wrote:
> > > On Thu, Oct 3, 2019 at 5:22 PM Maxime Ripard wrote:
> > > >
> > > > Hi,
> > > >
> > > > On Thu, Oct 03,
On Mon, Oct 07, 2019 at 12:40:39PM +0200, Andrea Parri wrote:
> Hi Christian,
>
> On Mon, Oct 07, 2019 at 01:52:16AM +0200, Christian Brauner wrote:
> > When assiging and testing taskstats in taskstats_exit() there's a race
> > when writing and reading sig->stats when a thread-group with more than
On Mon, Oct 7, 2019 at 11:50 AM Walter Wu wrote:
>
> On Mon, 2019-10-07 at 17:28 +0800, Walter Wu wrote:
> > On Mon, 2019-10-07 at 11:10 +0200, Dmitry Vyukov wrote:
> > > On Mon, Oct 7, 2019 at 11:03 AM Walter Wu
> > > wrote:
> > > >
> > > > On Mon, 2019-10-07 at 10:54 +0200, Dmitry Vyukov wrote
On Sat, Oct 05, 2019 at 11:04:46PM +0200, Hans de Goede wrote:
> Hi All,
>
> Here is a fix for the false-positive dev_err in platform_get_irq_byname()
> discussed recently and reported here:
> https://bugzilla.kernel.org/show_bug.cgi?id=205037
>
> Since patch 2 depends on patch 1, I think it migh
On 2019-10-03 18:13, Jason Baron wrote:
On 9/30/19 7:55 AM, Roman Penyaev wrote:
On 2019-09-28 04:29, Andrew Morton wrote:
On Wed, 25 Sep 2019 09:56:03 +0800 hev wrote:
From: Heiher
Take the case where we have:
t0
| (ew)
e0
| (et)
e1
| (l
From: Colin Ian King
Don't populate the arrays port_map and sl_map on the stack but
instead make them static. Makes the object code smaller by 64 bytes.
Before:
textdata bss dec hex filename
495756872 64 56511dcbf hisilicon/hns/hns_dsaf_main.o
After:
text
On 2019-10-07 10:30:37 [+0200], Daniel Wagner wrote:
> Hi,
Hi Daniel,
> On Fri, Oct 04, 2019 at 06:30:42PM +0200, Sebastian Andrzej Siewior wrote:
> > On 2019-10-04 18:20:41 [+0200], Uladzislau Rezki wrote:
> > > If we have migrate_disable/enable, then, i think preempt_enable/disable
> > > should
Hi Alexandre,
On Mon, Oct 07, 2019 at 12:25:50PM +0200, Alexandre Torgue wrote:
> This commit fixes id path of allwinner,sun4i-a10-csi.yaml location.
>
> Fixes: c5e8f4ccd775 ("media: dt-bindings: media: Add Allwinner A10 CSI
> binding")
> Signed-off-by: Alexandre Torgue
I just merged a patch ad
When assiging and testing taskstats in taskstats_exit() there's a race
when writing and reading sig->stats when a thread-group with more than
one thread exits:
cpu0:
thread catches fatal signal and whole thread-group gets taken down
do_exit()
do_group_exit()
taskstats_exit()
taskstats_tgid_all
On Mon, 7 Oct 2019 at 12:45, Mark Rutland wrote:
>
> Both multi_cpu_stop() and set_state() access multi_stop_data::state
> racily using plain accesses. These are subject to compiler
> transformations which could break the intended behaviour of the code,
> and this situation is detected by KCSAN on
Hello Eric,
On 9/30/19 2:42 PM, Eric W. Biederman wrote:
"Michael Kerrisk (man-pages)" writes:
Hello Eric,
A ping on my question below. Could you take a look please?
Thanks,
Michael
The concern from our conversation at the container mini-summit was that
there is a pathology if in your in
On Mon 2019-08-19 02:37:57, Andrew Lunn wrote:
> > Yes, I believe the integration is neccessary. Using same binding is
> > neccessary for that, but not sufficient. For example, we need
> > compatible trigger names, too.
>
> Hi Pavel
>
> Please could you explain what you mean by compatible trigger
On Sat, Oct 05, 2019 at 12:28:59AM +0800, JC Kuo wrote:
> Hi,
>
> This series introduces support for Tegra194 XUSB host and pad
> controller. Tegra194 XUSB host and pad controller are highly
> similar to the controllers found on Tegra186. Therefore, it's
> possible to resue xhci-tegra.c and xusb-t
From: Colin Ian King
Don't populate the array tick_array on the stack but instead make it
static. Makes the object code smaller by 29 bytes.
Before:
textdata bss dec hex filename
19191 432 0 196234ca7 hisilicon/hns3/hns3pf/hclge_tm.o
After:
textdata
Hello,
I am Austin S. Miller. I need your assistance
Thank you in advance
Sincerely,
Austin S. Miller
On Mon, Oct 07, 2019 at 11:09:04AM +0100, Jon Hunter wrote:
>
> On 06/10/2019 18:19, Greg Kroah-Hartman wrote:
> > This is the start of the stable review cycle for the 5.3.5 release.
> > There are 166 patches in this series, all will be posted as a response
> > to this one. If anyone has any issu
On Fri, Oct 04, 2019 at 11:31:21AM +0300, Adrian Hunter wrote:
SNIP
> +}
> +
> static int record__mmap_evlist(struct record *rec,
> struct evlist *evlist)
> {
> @@ -1383,6 +1417,12 @@ static int __cmd_record(struct record *rec, int argc,
> const char **argv)
>
On Fri, Oct 04, 2019 at 11:31:21AM +0300, Adrian Hunter wrote:
SNIP
> bool strict_freq;
> bool sample_id;
> bool no_bpf_event;
> + bool kcore;
> unsigned int freq;
> unsigned int mmap_pages;
> unsigned int auxtrace_mma
On Fri, Oct 04, 2019 at 11:31:20AM +0300, Adrian Hunter wrote:
SNIP
> u8 pad[8] = {0};
>
> - if (!perf_data__is_pipe(data) && !perf_data__is_dir(data)) {
> + if (!perf_data__is_pipe(data) && perf_data__is_single_file(data)) {
> off_t file_offset;
> int
Ftrace was one of the last W^X violators; these patches move it over to the
generic text_poke() interface and thereby get rid of this oddity.
Nothing ensures the module exists while we're iterating
mod->jump_entries in __jump_label_mod_text_reserved(), take a module
reference to ensure the module sticks around.
Signed-off-by: Peter Zijlstra (Intel)
---
kernel/jump_label.c | 10 --
1 file changed, 8 insertions(+), 2 deletions
Similar to how we disallow kprobes on any other dynamic text
(ftrace/jump_label) also disallow kprobes on inline static_call()s.
Signed-off-by: Peter Zijlstra (Intel)
---
arch/x86/kernel/kprobes/opt.c |4 +-
include/linux/static_call.h | 11 +++
kernel/kprobes.c |2 +
Signed-off-by: Peter Zijlstra (Intel)
---
arch/Kconfig |6 ++
kernel/static_call.c | 28
2 files changed, 34 insertions(+)
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -103,6 +103,12 @@ config STATIC_KEYS_SELFTEST
help
Boot time self
Adding another text_poke_bp_batch() user made me realize the interface
is all sorts of wrong. The text poke vector should be internal to the
implementation.
This then results in a trivial interface:
text_poke_queue() - which has the 'normal' text_poke_bp() interface
text_poke_finish() - whic
These here patches are something I've been poking at for a while, enabling
jump_label to use 2 byte jumps/nops.
It _almost_ works :-/
That is, you can build some kernels with it (x86_64-defconfig for example works
just fine).
The problem comes when GCC generates a branch into another section, mo
The current notifiers have the following error handling pattern all
over the place:
int err, nr;
err = __foo_notifier_call_chain(&chain, val_up, v, -1, &nr);
if (err & NOTIFIER_STOP_MASK)
__foo_notifier_call_chain(&chain, val_down, v, nr-1, NULL)
And aside
While auditing all module notifiers I noticed a whole bunch of fail
wrt the return value. Notifiers have a 'special' return semantics.
As is; NOTIFY_DONE vs NOTIFY_OK is a bit vague; but
notifier_from_errno(0) results in NOTIFY_OK and NOTIFY_DONE has a
comment that says "Don't care".
>From this I
From: Josh Poimboeuf
Add infrastructure for an arch-specific CONFIG_HAVE_STATIC_CALL_INLINE
option, which is a faster version of CONFIG_HAVE_STATIC_CALL. At
runtime, the static call sites are patched directly, rather than using
the out-of-line trampolines.
Compared to out-of-line static calls,
Now that notifiers got unbroken; use the proper interface to handle
notifier errors and propagate them.
There were already MODULE_STATE_COMING notifiers that failed; notably:
- jump_label_module_notifier()
- tracepoint_module_notify()
- bpf_event_notify()
By propagating this error, we fix tho
Enable and emit short JMP/NOP jump_label entries.
A lot of the jumps are in fact short, like around tracepoints:
0920 : |
0920 :
920: 53 push %rbx|
920: 53
Both arch_static_branch() and arch_static_branch_jump() have the same
blurb to generate the __jump_table entry, share it.
Signed-off-by: Peter Zijlstra (Intel)
---
arch/x86/include/asm/jump_label.h | 24
1 file changed, 12 insertions(+), 12 deletions(-)
--- a/arch/x86
Instead of checking if the emitted (default) NOP is the ideal NOP, and
conditionally rewrite the NOP, just rewrite the NOP.
This shouldn't be a problem because init / module_load uses
text_poke_early() which is cheap and this saves us from having to go
figure out which NOP to compare against when
In preparation for static_call and variable size jump_label support,
teach text_poke_bp() to emulate instructions, namely:
JMP32, JMP8, CALL, NOP2, NOP_ATOMIC5, INT3
The current text_poke_bp() takes a @handler argument which is used as
a jump target when the temporary INT3 is hit by a different
Extend the static_call infrastructure to optimize the following common
pattern:
if (func_ptr)
func_ptr(args...)
For the trampoline (which is in effect a tail-call), we patch the
JMP.d32 into a RET, which then directly consumes the trampoline call.
For the in-line sites we
With the last and only user of these functions gone (ftrace) remove
them as well to avoid ever growing new users.
Signed-off-by: Peter Zijlstra (Intel)
---
arch/x86/include/asm/set_memory.h |2 --
arch/x86/mm/init_32.c | 28
arch/x86/mm/init_64.c
Since variable sized jump_label support is somewhat 'inspired', ensure
objtool validates the tricky bits. Specifically it is important that
the displacement for the 2 byte jump is limited to a single byte --
since this is the hardest part of the whole scheme and relies on
somewhat dodgy GAS tricks.
From: Josh Poimboeuf
The __ADDRESSABLE() macro uses the __LINE__ macro to create a temporary
symbol which has a unique name. However, if the macro is used multiple
times from within another macro, the line number will always be the
same, resulting in duplicate symbols.
Make the temporary symbol
There is only a single usage site left, remove the function and extend
the print to include more information, like the expected text and the
patch type.
Signed-off-by: Peter Zijlstra (Intel)
---
arch/x86/kernel/jump_label.c | 23 ++-
1 file changed, 10 insertions(+), 13 del
From: Josh Poimboeuf
Add the x86 out-of-line static call implementation. For each key, a
permanent trampoline is created which is the destination for all static
calls for the given key. The trampoline has a direct jump which gets
patched by static_call_update() when the destination function cha
Replace many of the indirect calls with static_call().
XXX run performance numbers
Signed-off-by: Peter Zijlstra (Intel)
---
arch/x86/events/core.c | 136 +++--
1 file changed, 98 insertions(+), 38 deletions(-)
--- a/arch/x86/events/core.c
+++ b/arc
These patches came from the desire to propagate MODULE_STATE_COMING errors.
While looking at that I spotted fail with a number of module notifiers
themselves and with the whole notification business.
I will probably add a notifier for MODULE_STATE_UNFORMED, but that is later.
Fixing this error pr
Move ftrace over to using the generic x86 text_poke functions; this
avoids having a second/different copy of that code around.
This also avoids ftrace violating the (new) W^X rule and avoids
fragmenting the kernel text page-tables, due to no longer having to
toggle them RW.
Cc: Steven Rostedt
Cc
In order to use static_call() to wire up x86_pmu, we need to
initialize earlier; copy some of the tricks from jump_label to enable
this.
Primarily we overload key->next to store a sites pointer when there
are no modules, this avoids having to use kmalloc() to initialize the
sites and allows us to
From: Josh Poimboeuf
Static calls are a replacement for global function pointers. They use
code patching to allow direct calls to be used instead of indirect
calls. They give the flexibility of function pointers, but with
improved performance. This is especially important for cases where
retpo
From: Josh Poimboeuf
Add the inline static call implementation for x86-64. For each key, a
temporary trampoline is created, named __static_call_tramp_. The
trampoline has an indirect jump to the destination function.
Objtool uses the trampoline naming convention to detect all the call
sites.
GCC can turn our static_call(name)(args...) into a tail call, in which
case we get a JMP.d32 into the trampoline (which then does a further
tail-call).
Teach objtool to recognise and mark these in .static_call_sites and
adjust the code patching to deal with this.
Signed-off-by: Peter Zijlstra (In
Because typing on a Monday is hard, here's a resend with the LKML address
corrected.
---
Hi All,
Here are 4 series that have accumulated in my queue for a while and I
figured I'd 'finish' and send out again.
The first series is rewriting x86/ftrace to use the normal text_poke()
interfaces we
On Mon, Sep 16, 2019 at 12:43:53PM +0200, Gilles DOFFE wrote:
> The pca9535 gpio expander is present on the Rex baseboard, but missing
> from the dtsi.
> The pca9535 is on i2c2 bus which is common to the three SOM
> variants (Basic/Pro/Ultra), thus it is activated by default.
>
> Add also the new
In prepration for variable size jump_label support; remove all ASM
bits that are not used.
Signed-off-by: Peter Zijlstra (Intel)
---
arch/x86/entry/calling.h |2 +-
arch/x86/include/asm/jump_label.h | 28
2 files changed, 5 insertions(+), 25 deletions(
Update the comment now that we've merged x86_32 support.
Signed-off-by: Peter Zijlstra (Intel)
---
arch/x86/include/asm/text-patching.h |3 +++
1 file changed, 3 insertions(+)
--- a/arch/x86/include/asm/text-patching.h
+++ b/arch/x86/include/asm/text-patching.h
@@ -85,6 +85,9 @@ static inli
Provide a simple helper function to create common instruction
encodings.
Signed-off-by: Peter Zijlstra (Intel)
Cc: Steven Rostedt
Cc: Daniel Bristot de Oliveira
Cc: Masami Hiramatsu
---
arch/x86/include/asm/text-patching.h |2 +
arch/x86/kernel/alternative.c| 36
This allows the patching to to emit 2 byte JMP/NOP instruction in
addition to the 5 byte JMP/NOP we already did. This allows for more
compact code.
This code is not yet used, as we don't emit shorter code at compile
time yet.
Signed-off-by: Peter Zijlstra (Intel)
---
arch/x86/include/asm/jump_l
+++ Matthias Maennich [07/10/19 07:46 +0100]:
Hi!
On Sat, Oct 05, 2019 at 04:53:56PM +0900, Masahiro Yamada wrote:
On Sat, Oct 5, 2019 at 3:25 AM Lucas De Marchi
wrote:
On Fri, Oct 4, 2019 at 2:57 AM Matthias Maennich wrote:
depmod in its current version is not aware of symbol namespace i
From: Steven Rostedt
[peterz: updated to new interface]
Signed-off-by: Steven Rostedt (VMware)
Signed-off-by: Peter Zijlstra (Intel)
---
include/linux/tracepoint-defs.h |5 ++
include/linux/tracepoint.h | 75 ++--
include/trace/define_trace.h
[ Sorry, fixed the Cc:lkml line. ]
* Peter Zijlstra wrote:
> These here patches are something I've been poking at for a while,
> enabling jump_label to use 2 byte jumps/nops.
>
> It _almost_ works :-/
>
> That is, you can build some kernels with it (x86_64-defconfig for
> example works jus
This allows architectures to have variable sized jumps.
Signed-off-by: Peter Zijlstra (Intel)
---
arch/x86/include/asm/jump_label.h |4 ++--
arch/x86/kernel/jump_label.c |5 +
include/linux/jump_label.h|9 +
kernel/jump_label.c |2 +-
4 file
x86_64-defconfigx86_64-allmodconfig-UBSAN-KASAN
NOP2164121796
JMP848 114
NOP5101031042
JMP32 29 91
Which results in a possible 3*(1641+48) ~ 5k saving for defconfig
and 3*(21796+114) ~
501 - 600 of 1238 matches
Mail list logo
