From: Sami Uddin
Prevent usage of shared memory regions where the length is zero,
as such configurations are not valid and may lead to unexpected behavior.
Signed-off-by: Sami Uddin
---
v2:
- Fixed coding style issue: added space after 'if' statement
include/linux/virtio_config.h
From: Sami Uddin
Prevent usage of shared memory regions where the length is zero,
as such configurations are not valid and may lead to unexpected behavior.
Signed-off-by: Sami Uddin
---
include/linux/virtio_config.h | 2 ++
1 file changed, 2 insertions(+)
diff --git a/include/linux
From: Sami Uddin
Prevent usage of shared memory regions where the length is zero,
as such configurations are not valid and may lead to unexpected behavior.
Signed-off-by: Sami Uddin
---
v3:
- Use idiomatic 'if (!region->len)' as suggested by reviewer
v2:
- Fixed coding style issue
Fix tegra_smmu_probe() to initialize client_lock spinlocks in
per-address-space structures.
Signed-off-by: Sami Liedes
---
drivers/iommu/tegra-smmu.c |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/drivers/iommu/tegra-smmu.c b/drivers/iommu/tegra-smmu.c
index f08dbcd
From: Sami Rahman
Kernel Version: 3.5.0-34-generic
Added support for MMB Networks and Planet Innovation Ingeni ZigBee USB devices
using customized Silicon Labs' CP210x.c USB to UART bridge drivers with PIDs:
88A4, 88A5.
Signed-off-by: Sami Rahman
Tested-by: Sami Rahman
---
---
g the concept
makes sense and could be considered part of upstream kernel (assuming
my coding meets the usual quality criteria).
Any thoughts, comments, recommendations?
--
Sami Kerola
http://www.iki.fi/kerolasa/
--
To unsubscribe from this list: send the line "unsubscribe linux-kern
he boot time as precisily as it can.
CC: Karel Zak
CC: Lennart Poettering
CC: Kay Sievers
CC: Eric W. Biederman
References: http://www.spinics.net/lists/util-linux-ng/msg08092.html
Measurement: https://plus.google.com/u/0/117341944650953092607/posts/SHy5qu1HpSU
Signed-off-by: Sami Kerola
--
On 29 June 2013 15:23, Eric W. Biederman wrote:
>
> Nacked-by: "Eric W. Biederman"
>
> The pach does not solve the problem described.
Which might be true, and I am open to better proposals.
> Sami Kerola writes:
>
>> The kernel does not expose precise
On 30 June 2013 16:10, Kay Sievers wrote:
> On Sat, Jun 29, 2013 at 10:53 PM, Sami Kerola wrote:
>
>> BTW having a way to measure effect of suspend/resume could lead to a
>> way to fix time time distortion.
>
>> Perhaps there is better alternative to fix user space p
Make resulted to untraced files, which where the binaries make
produced, such need to be ignored.
Signed-off-by: Sami Kerola
---
tools/power/x86/turbostat/.gitignore | 1 +
tools/power/x86/x86_energy_perf_policy/.gitignore | 1 +
2 files changed, 2 insertions(+)
create mode 100644
On Tue, Nov 20, 2012 at 12:21 PM, Karel Zak wrote:
> On Mon, Nov 19, 2012 at 04:12:50PM -0800, Andrew Morton wrote:
>> Where's the benefit in switching ipcs over to using /proc?
>>
>> procfs reads are probably slower than the syscalls?
>
> 32bit userspace and 64bit kernel (RHEL ppc64), it seems m
cture is fine. Also overlay looks good
if I keep the window size small enough.
Sami
--
Whoa...I did a 'zcat /vmlinuz > /dev/audio' and I think I heard God...
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTEC
On Thu, Oct 11, 2007 at 16:36:10 +0200, Philippe Elie wrote:
> On Sat, 29 Sep 2007 at 20:05 +0000, Sami Farin wrote:
>
> > > x86_64 SMP kernel v2.6.22.6 (not using callgraph).
> > > sometimes oprofile works for a longer time... but not this time.
> > >
>
In write_pool(), isn't cond_resched() needed after call to
add_entropy_words() because otherwise there can be large latencies
(think of command "dd if=/dev/zero of=/dev/random bs=1" ) ?
--
Do what you love because life is too short for anything else.
-
To unsubscribe from this list: send
On Tue, Nov 20, 2007 at 13:52:52 +0100, Alessandro Suardi wrote:
> On Nov 20, 2007 7:52 AM, Herbert Xu <[EMAIL PROTECTED]> wrote:
> > On Mon, Nov 19, 2007 at 10:47:59PM -0800, H. Peter Anvin wrote:
> > >
> > > This one is definitely messy. There is absolutely no way to know what
> > > gcc has misc
On Thu, Feb 17, 2005 at 01:59:47AM +0100, Guennadi Liakhovetski wrote:
> Hi
>
> The machine was running updatedb, while I was trying to burn on an ATAPI
> CDR (hdc) and read a SCSI DVD-ROM not very intensively, ran a couple of
> random applications, when my /home partition (hda7) became unaccess
On Tue, Jan 18, 2005 at 05:22:44PM -0800, Daniel McNeil wrote:
> Andrew,
>
> This is a patch to generic_file_buffered_write() to correctly
> handle partial O_DIRECT writes (because of unallocated blocks)
> when there is more than 1 iovec. Without this patch, the code is
> writing the wrong iovec
On Wed, Jan 19, 2005 at 08:55:40AM -0800, Daniel McNeil wrote:
> On Tue, 2005-01-18 at 18:31, Sami Farin wrote:
...
> > I have Linux 2.6.10-ac9 + bio clone memory corruption -patch,
> > and dio_bug does not give errors (without your patch).
>
> I should have mentioned that m
Using SMP kernel 2.6.22.6pre-CFS-v20.5 on Pentium D (IA-32).
I think this bug (or whatever you want to call it) got triggered
when you first allocate several megabytes of memory in a kernel module
and then free them, and then run e.g. X and when memory gets tight,
you end up with this situation...
On Tue, Sep 04, 2007 at 21:37:35 -0400, Rik van Riel wrote:
> Sami Farin wrote:
>> Using SMP kernel 2.6.22.6pre-CFS-v20.5 on Pentium D (IA-32).
>> I think this bug (or whatever you want to call it) got triggered
>> when you first allocate several megabytes of memory in a kern
large margin to continue putting equal
> pressure on all zones in normal situations.
>
> Sami, could you try out this patch to see if it helps your
> situation?
Thanks, Rik. bzImage is ready, I probably reboot inside one
month for a reason or other 8-)
--
Do what you love because
On Mon, Oct 22, 2007 at 19:38:10 -0700, Linus Torvalds wrote:
>
> This set of two patches look ok by me, but I'd like sign-offs.. Also, were
> they tested and found to fix the problem by Sami?
>
> Linus
The previous patch I tested by Philippe, oprof-fix-
On Tue, Oct 23, 2007 at 18:13:21 +0200, Philippe Elie wrote:
> On Tue, 23 Oct 2007 at 13:10 +0000, Sami Farin wrote:
>
> > On Mon, Oct 22, 2007 at 19:38:10 -0700, Linus Torvalds wrote:
> > >
> > > This set of two patches look ok by me, but I'd like sign-o
Would be neat if randomized brk and setrlimit(RLIMIT_DATA, ...)
worked in a predictable way:
$ gcc brk.c -fPIC -pie -m64;./a.out;./a.out;./a.out
sbrk=0x7f721b815000 main=0x7f721af04860
sbrk succeeded (brk=0x7f721b909240)
sbrk=0x7fc3d77e2000 main=0x7fc3d66fa860
sbrk failed: Cannot allocate memory (
gt; regardless of what you do.
>
> I tend to agree here with Arjan. However it probably would make no harm to
> make at least a little bit consisten behavior of setrlimit(), though it
> has a little use in such cases.
>
> Sami, does the patch below work for you?
Thanks, Jiri, now
On Fri, Sep 14, 2007 at 20:17:46 +0300, Sami Farin wrote:
> On Wed, Sep 05, 2007 at 18:48:51 -0400, Rik van Riel wrote:
> > Sami Farin wrote:
> >> On Wed, Sep 05, 2007 at 12:24:26 -0400, Rik van Riel wrote:
> >> ...
> >>>> *shrug*
> >>> The a
I know this may be entirely my fault but I have tried reversing
all of my _own_ patches I applied to 2.6.19.2 but can't find what broke this.
I did three times "netcat 127.0.0.69 42", notice the different
port numbers.
First, if someone could attempt this on 2.6.19.2 or 2.6.20-rc* ,
and tell it wo
On Mon, Jan 15, 2007 at 23:55:15 +0200, Sami Farin wrote:
> I know this may be entirely my fault but I have tried reversing
> all of my _own_ patches I applied to 2.6.19.2 but can't find what broke this.
> I did three times "netcat 127.0.0.69 42", notice the different
>
On Tue, Jan 16, 2007 at 14:57:56 +0100, Jan Engelhardt wrote:
>
> >Subject: Re: I broke my port numbers :(
> >
> >On Mon, Jan 15, 2007 at 23:55:15 +0200, Sami Farin wrote:
> >> I know this may be entirely my fault but I have tried reversing
> >> all of my
On Wed, Jan 17, 2007 at 14:43:29 +1100, David Chinner wrote:
...
> > > Subject: BUG: at mm/truncate.c:60 cancel_dirty_page() (XFS)
> > > References : http://lkml.org/lkml/2007/1/5/308
> > > Submitter : Sami Farin <[EMAIL PROTECTED]>
> > > Ha
On Tue, Dec 12, 2006 at 15:10:56 -0500, Chuck Ebbert wrote:
> In-Reply-To: <[EMAIL PROTECTED]>
>
> On Tue, 5 Dec 2006 19:25:13 +0200, Sami Farin wrote:
>
> > BUG: unable to handle kernel paging request at virtual address fffb9dc0
>
> > eax: fffb8000 eb
just a simple test I did...
xfs_freeze -f /mnt/newtest
cp /etc/fstab /mnt/newtest
xfs_freeze -u /mnt/newtest
2007-01-04 01:44:30.341979500 <4>BUG: warning at
kernel/mutex-debug.c:80/debug_mutex_unlock()
2007-01-04 01:44:30.385771500 <4> [] dump_trace+0x215/0x21a
2007-01-04 01:44:30.385774500 <4>
On Fri, Feb 23, 2007 at 17:05:27 -0800, Stephen Hemminger wrote:
> Since there already two users of full 64 bit division in the kernel,
> and other places maybe hiding out as well. Add a full 64/64 bit divide.
>
> Yes this expensive, but there are places where it is necessary.
> It is not clear if
I setup namespace for /tmp and /var/tmp
( pam_namespace.so into /etc/pam.d/{su,login} )
and something did not like something I did:
[322593.844838] 0x0: 00 00 00 00 2b 00 00 11 20 21 00 00 00 68 ff ff
[322593.844854] Filesystem "sda8": XFS internal error xfs_da_do_buf(2) at line
2087 of file fs
On Wed, Dec 06, 2006 at 11:41:38 +0200, Sami Farin wrote:
> On Tue, Dec 05, 2006 at 22:39:45 -0800, Andrew Morton wrote:
> > On Tue, 5 Dec 2006 19:25:13 +0200
> > Sami Farin <[EMAIL PROTECTED]> wrote:
> >
> > > [1.] PROBLEM: 2.6.19 + highmem = BUG at do_wp_pag
On Tue, Dec 12, 2006 at 12:46:01 +0200, Sami Farin wrote:
> On Wed, Dec 06, 2006 at 11:41:38 +0200, Sami Farin wrote:
> > On Tue, Dec 05, 2006 at 22:39:45 -0800, Andrew Morton wrote:
> > > On Tue, 5 Dec 2006 19:25:13 +0200
> > > Sami Farin <[EMAIL PROTECTED]> w
On Tue, Dec 12, 2006 at 15:10:56 -0500, Chuck Ebbert wrote:
> In-Reply-To: <[EMAIL PROTECTED]>
>
> On Tue, 5 Dec 2006 19:25:13 +0200, Sami Farin wrote:
>
> > BUG: unable to handle kernel paging request at virtual address fffb9dc0
>
> > eax: fffb8000 eb
Kernel 2.6.19.1 SMP on Pentium D. I ran command restorecon -R /wrk.
After a while or two programs stopped responding and I had to reboot.
I'm not sure is this bug or feature...
I upgraded selinux policy before running restorecon.
2007-01-04 22:41:55.360538500 <4>softlimit D 61707865 0 18
On Thu, Jan 04, 2007 at 16:03:34 -0800, Andrew Morton wrote:
> On Fri, 5 Jan 2007 00:26:42 +0200
> Sami Farin <[EMAIL PROTECTED]> wrote:
>
> > Kernel 2.6.19.1 SMP on Pentium D. I ran command restorecon -R /wrk.
> > After a while or two programs stopped res
Linux 2.6.19.1 SMP [2] on Pentium D...
I was running dt-15.14 [2] and I ran
"cinfo datafile" (it does mincore()).
Well it went OK but when I ran "strace cinfo datafile"...:
04:18:48.062466 mincore(0x37f1f000, 2147266560,
...
2007-01-06 04:19:03.788181500 <4>BUG: warning at
mm/truncate.c:60/cancel
On Sat, Jan 06, 2007 at 04:39:07 +0200, Sami Farin wrote:
> Linux 2.6.19.1 SMP [2] on Pentium D...
> I was running dt-15.14 [2] and I ran
> "cinfo datafile" (it does mincore()).
> Well it went OK but when I ran "strace cinfo datafile"...:
> 04:18:48.062466 minco
On Sat, Jan 06, 2007 at 21:11:07 +, Hugh Dickins wrote:
> On Sat, 6 Jan 2007, Sami Farin wrote:
>
> > Linux 2.6.19.1 SMP [2] on Pentium D...
> > I was running dt-15.14 [2] and I ran
> > "cinfo datafile" (it does mincore()).
> > Well it went OK but when
> mutexes complain loudly when a the process unlocking the mutex is
> not the process that locked it.
>
> Basically, the generic code is broken - the bd_mount_mutex needs to
> be reverted back to a semaphore because it is locked and unlocked
> by different processes. The following
On Mon, Jan 08, 2007 at 10:04:36 +1100, David Chinner wrote:
> On Sun, Jan 07, 2007 at 02:48:12PM -0800, Andrew Morton wrote:
> > On Mon, 8 Jan 2007 09:23:41 +1100
> > David Chinner <[EMAIL PROTECTED]> wrote:
> >
> > > How are you supposed to invalidate a range of pages in a mapping for
> > > this
Linux 2.6.19.1 SMP on Pentium D, Intel DQ965GF mobo.
Got this while bittorrenting knoppix:
2007-01-09 22:53:40.020693500 <4>NETFILTER drop IN=eth0 OUT=
MAC=00:19:d1:00:5f:01:00:05:00:1c:58:1c:08:00 SRC=83.46.5.76 DST=80.223.106.128
LEN=121 TOS=0x00 PREC=0x00 TTL=112 ID=53273 PROTO=ICMP TYPE=3 CO
On Tue, Jan 09, 2007 at 15:59:30 -0800, Auke Kok wrote:
> Sami Farin wrote:
...
> >I do "ethtool -K eth0 tso off" now and check if I get the hang again. =)
>
> I'm unsure whether v7.2.x already automatically disables TSO for 100mbit
> speed link, probably not.
On Tue, Jan 09, 2007 at 17:48:04 -0800, Auke Kok wrote:
> Sami Farin wrote:
> >On Tue, Jan 09, 2007 at 15:59:30 -0800, Auke Kok wrote:
> >>Sami Farin wrote:
> >...
> >>>I do "ethtool -K eth0 tso off" now and check if I get the hang again. =)
>
After I got this error [1], system got real slow, like 386 having 32 MB of RAM
and swapping constantly.
My system is P4 SMP with 1GB of RAM.
I got this same behavior with 2.6.19, too, but then I used GNU cp v6.9
which had micro-optimization which did not bother doing read() for regular
files, like
On Wed, Mar 07, 2007 at 00:24:35 +0200, Sami Farin wrote:
> On Tue, Mar 06, 2007 at 23:53:49 +0200, Sami Farin wrote:
> ...
> > And I found bug in gcc-4.1.2, it gave 0 for ncubic results
> > when doing 1000 loops test... gcc-4.0.3 works.
>
> Found it.
>
> --- cbrt
Not sure if this is oprofile's fault, but I got this
when I started opreport.
10 min after BUG system needed to be rebooted (cycling power).
With earlier kernel (without oprofile) everything worked for a month
without hiccups.
[295318.822839] BUG: unable to handle kernel NULL pointer dereference a
On Tue, Mar 06, 2007 at 16:00:55 -0800, Stephen Hemminger wrote:
...
> > Now Linux 2.6 does not have "memory" in fls, maybe it causes
> > some gcc funnies some people are seeing.
> >
>
> That code was copy-paste from:
> include/asm-x86_64/bitops.h
>
> So shouldn't both fls() and ffs() be f
On Tue, Mar 06, 2007 at 10:29:41 -0800, Stephen Hemminger wrote:
> Don't count the existing Newton-Raphson out. It turns out that to get enough
> precision for 32 bits, only 4 iterations are needed. By unrolling those, it
> gets much better timing.
>
> Slightly gross test program (with original cu
On Tue, Mar 06, 2007 at 23:53:49 +0200, Sami Farin wrote:
...
> And I found bug in gcc-4.1.2, it gave 0 for ncubic results
> when doing 1000 loops test... gcc-4.0.3 works.
Found it.
--- cbrt-test.c~2007-03-07 00:20:54.735248105 +0200
+++ cbrt-test.c 2007-03-07 00:21:03.964864343
On Tue, Mar 06, 2007 at 21:20:33 +0100, Eric Dumazet wrote:
...
> I rewrote the reciprocal_div() for i386 so that one multiply is used.
>
> static inline u32 reciprocal_divide(u32 A, u32 R)
> {
> #if __i386
> unsigned int edx, eax;
> asm("mul %2":"=a" (eax), "=d" (edx):"rm" (R), "0
On Wed, Mar 07, 2007 at 11:11:49 -0500, Chuck Ebbert wrote:
> Sami Farin wrote:
> > On Tue, Mar 06, 2007 at 23:53:49 +0200, Sami Farin wrote:
> > ...
> >> And I found bug in gcc-4.1.2, it gave 0 for ncubic results
> >> when doing 1000 loops test... g
is there room for improvement in security_port_sid() ?
little test with dns queries (dnsfilter (the client) on local host
using poll() and dnscache (the server) using epoll (at max 4000 concurrent
queries):
(stats for only vmlinux)
CPU: P4 / Xeon, speed 2797.32 MHz (estimated)
Counted GLOBAL_POWE
On Tue, Mar 27, 2007 at 09:40:23 -0400, Stephen Smalley wrote:
> On Tue, 2007-03-27 at 13:06 +0300, Sami Farin wrote:
> > is there room for improvement in security_port_sid() ?
>
> Yes, lots of room. Also, it won't get called per-packet if you enable
> secmark (echo 0 &
With CONFIG_CLANG_LTO, clang generates LLVM IR instead of ELF object
files. As empty.o is used for probing target properties, disable LTO
for it to produce an object file instead.
Signed-off-by: Sami Tolvanen
---
scripts/mod/Makefile | 1 +
1 file changed, 1 insertion(+)
diff --git a/scripts
GNU gold may require different flags than GNU ld. Add a macro for
detecting the linker and conditionally add gold specific flags from
LDFLAGS_GOLD.
Signed-off-by: Sami Tolvanen
---
Makefile | 5 +
scripts/Kbuild.include | 4
2 files changed, 9 insertions(+)
diff --git a
when
RANDOMIZE_MODULE_REGION_FULL is selected,
2) and disables RANDOMIZE_MODULE_REGION_FULL with clang LTO.
Signed-off-by: Sami Tolvanen
---
arch/arm64/Kconfig | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 43ccddb7c8ec..a301fdc90319 1
CONFIG_CLANG_LTO requires the use of clang's integrated assembler, which
doesn't understand the inline assembly in aes-ce-cipher.c. Disable LTO for
the file to work around the issue.
Signed-off-by: Sami Tolvanen
---
arch/arm64/crypto/Makefile | 2 +-
1 file changed, 1 insertion(+),
when
RANDOMIZE_MODULE_REGION_FULL is selected,
2) and disables RANDOMIZE_MODULE_REGION_FULL with clang LTO.
Signed-off-by: Sami Tolvanen
---
arch/arm64/Kconfig | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index 43ccddb7c8ec..a301fdc90319 1
9
As binutils doesn't allow macros to be redefined, this change adds C
preprocessor macros that define the assembly macros globally for binutils
and locally for clang's integrated assembler.
Suggested-by: Greg Hackmann
Suggested-by: Nick Desaulniers
Signed-off-by: Sami Tolvanen
---
arc
Allow CONFIG_CLANG_LTO to be enabled for the architecture.
Signed-off-by: Sami Tolvanen
---
arch/arm64/Kconfig | 1 +
1 file changed, 1 insertion(+)
diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index a301fdc90319..6ae31f0c58ee 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
CONFIG_CLANG_LTO depends on GNU gold, which can generate ADR_PREL_PG_HI21
relocations even with --fix-cortex-a53-843419.
Since ARM64_ERRATUM_843419 disables kernel support for these relocations,
disable the erratum when LTO is used.
Signed-off-by: Sami Tolvanen
---
arch/arm64/Kconfig | 1 +
1
9
As binutils doesn't allow macros to be redefined, this change adds C
preprocessor macros that define the assembly macros globally for binutils
and locally for clang's integrated assembler.
Suggested-by: Greg Hackmann
Suggested-by: Nick Desaulniers
Signed-off-by: Sami Tolvanen
---
arc
e the erratum fix is not
used even if the linker is configured to enable it by default.
Signed-off-by: Sami Tolvanen
---
arch/arm64/Makefile | 5 +
1 file changed, 5 insertions(+)
diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile
index c16bd1ab37f8..0c1fb297c98e 100644
--- a/arch/arm64/Mak
e the erratum fix is not
used even if the linker is configured to enable it by default.
Signed-off-by: Sami Tolvanen
---
arch/arm64/Makefile | 5 +
1 file changed, 5 insertions(+)
diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile
index c16bd1ab37f8..0c1fb297c98e 100644
--- a/arch/arm64/Mak
Make sure the linker doesn't remove .altinstructions or
.altinstr_replacement when CONFIG_LD_DEAD_CODE_DATA_ELIMINATION is
enabled.
Signed-off-by: Sami Tolvanen
---
arch/arm64/kernel/vmlinux.lds.S | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/arch/arm64/k
This change fixes the following error message when linking with GNU
gold:
ld.gold: error: unrecognized emulation aarch64linux
Signed-off-by: Sami Tolvanen
---
arch/arm64/Makefile | 4
1 file changed, 4 insertions(+)
diff --git a/arch/arm64/Makefile b/arch/arm64/Makefile
index
C=clang LD=ld.gold LD_FINAL_VMLINUX=ld
Recommended versions are >= 5.0 for clang, and >= 2.27 for binutils.
Signed-off-by: Sami Tolvanen
---
.gitignore | 2 ++
Makefile | 16 -
arch/Kconfig | 32 +
scripts/Makef
From: Greg Hackmann
LLVM bug 30792 causes clang's AArch64 backend to crash compiling
arch/arm64/crypto/aes-ce-cipher.c. Replacing -mgeneral-regs-only with
-mno-implicit-float is the suggested workaround.
Signed-off-by: Greg Hackmann
Cc: Matthias Kaehlcke
Signed-off-by: Sami Tol
With CONFIG_CLANG_LTO, we produce LLVM IR instead of object files. Since LTO
is not really needed here and the Makefile assumes we produce an object file,
disable LTO for libstub.
Signed-off-by: Sami Tolvanen
---
drivers/firmware/efi/libstub/Makefile | 3 ++-
1 file changed, 2 insertions(+), 1
gnore, and
a __norecordmcount attribute for moving functions to this section.
Signed-off-by: Sami Tolvanen
---
arch/Kconfig | 2 +-
include/asm-generic/vmlinux.lds.h | 2 +-
include/linux/compiler-clang.h| 7 +++
include/linux/compiler.h | 4
kernel/
With CONFIG_CLANG_LTO, we produce LLVM IR instead of object files. Since LTO
is not really needed here and the Makefile assumes we produce an object file,
disable LTO for libstub.
Signed-off-by: Sami Tolvanen
---
drivers/firmware/efi/libstub/Makefile | 3 ++-
1 file changed, 2 insertions(+), 1
Don't remove .head.text or .exitcall.exit when linking with --gc-sections,
and include .init.text.* in .init.text and .init.rodata.* in .init.rodata.
Signed-off-by: Sami Tolvanen
---
include/asm-generic/vmlinux.lds.h | 8
1 file changed, 4 insertions(+), 4 deletions(-)
diff --
ld issues
with KASLR, for example.
Greg Hackmann (1):
arm64: use -mno-implicit-float instead of -mgeneral-regs-only
Sami Tolvanen (14):
kbuild: add ld-name macro and support for GNU gold
kbuild: fix LD_DEAD_CODE_DATA_ELIMINATION with GNU gold
kbuild: add support for clang LTO
kbuild: fix
-DLLVM_BINUTILS_INCDIR=/path/to/binutils-2.27/include \
-DLLVM_TARGETS_TO_BUILD="X86;AArch64" && \
ninja
Sami
On Fri, Nov 03, 2017 at 07:26:35PM +, Mark Rutland wrote:
> I guess that in Google you haven't tested on a platform with EL2
> available?
Correct. I'll look into this and include a fix in v2. Does this work on a
clang build without LTO?
Sami
eam clang
doesn't generate any worse code with -mno-implicit-float, so there should
be no harm in using this workaround with all versions.
This might change in future, of course, so I don't have objections to
adding a check for clang version < 6.0 here.
Sami
. I'll see if I can
find a better workaround for this for v2.
Sami
te, but at the same time,
it supports a number of emulations, just with different names than ld.bfd.
I'll take a closer look at this and see if there's a different flag we
should be passing here instead. Looks like I also have some 0-day issues
to address for v2.
Sami
On Mon, Apr 9, 2018 at 10:02 AM Greg Hackmann wrote:
> Sami, what are the plans for upstreaming this work?
CFI is a clang-specific feature that depends on LTO. Based on the earlier
LTO discussion, we decided to collect some more evidence that clang's LTO
doesn't actually break an
:
VCPU:000804fc20001221
This change adds -fno-jump-tables to arm64/hyp to work around the
bug.
Suggested-by: AKASHI Takahiro
Signed-off-by: Sami Tolvanen
---
arch/arm64/kvm/hyp/Makefile | 4
1 file changed, 4 insertions(+)
diff --git a/arch/arm64/kvm/hyp/Makefile b/arch
set_reset_devices() to the variable .init.text:$x.0
This change adds a call to is_valid_name to find_elf_symbol, similarly
to how it's already used in find_elf_symbol2.
Signed-off-by: Sami Tolvanen
---
scripts/mod/modpost.c | 50 ++-
1 file changed, 26 inser
const char *name)
> +{
> + /*
> +* Since all syscall functions have __riscv_ prefix, we must skip it.
> +* However, as we described above, we decided to ignore compat
> +* syscalls, so we don't care about __riscv_compat_ prefix here.
> +*/
> + return !strcmp(sym + 8, name);
> +}
Good catch, thanks for fixing this!
Reviewed-by: Sami Tolvanen
Sami
compatibility.
Signed-off-by: Sami Tolvanen
---
tools/Makefile | 11 +--
tools/gendwarfksyms/Build | 2 +
tools/gendwarfksyms/Makefile| 49 +++
tools/gendwarfksyms/gendwarfksyms.c | 128
tools/gendwarfksyms
l.org/lkml/2023061155.1349375-1-g...@garyguo.net/
[2]
https://lore.kernel.org/rust-for-linux/20231118025748.2778044-1-mmau...@google.com/
[3]
https://lore.kernel.org/rust-for-linux/cagsqo005hriuzdeppcifdqg9zfdjrwahpble4x7-myfjscn...@mail.gmail.com/
Sami
Sami Tolvanen (15):
tools: Add gen
ols, where only one name appears in the
debugging information.
Signed-off-by: Sami Tolvanen
---
tools/gendwarfksyms/Build | 1 +
tools/gendwarfksyms/gendwarfksyms.c | 2 +
tools/gendwarfksyms/gendwarfksyms.h | 17
tools/gendwarfksyms/symbols.c
emove13assert_failed
0x985f94dd
...
Note that type strings are not yet expanded, so the CRCs can only
distinguish between functions and variables for now.
Signed-off-by: Sami Tolvanen
---
tools/gendwarfksyms/Build | 1 +
tools/gendwarfksyms/crc32.c | 69
ug vmlinux.o
Output:
gendwarfksyms: symbol_read_list: adding { 4b0, "loops_per_jiffy" }
gendwarfksyms: process_modules: vmlinux.o
gendwarfksyms: process_exported_symbols: loops_per_jiffy (@ 4b0)
variable base_type unsigned long byte_size(8);
#SYMVER loops_per_jiffy 0xc694aefc
Signed-off
Basic types in DWARF repeat frequently and traversing the DIEs using
libdw is relatively slow. Add a simple hashtable based cache for the
processed DIEs.
Signed-off-by: Sami Tolvanen
---
tools/gendwarfksyms/Build | 1 +
tools/gendwarfksyms/cache.c | 147
Add support for expanding DWARF type modifiers, such as pointers,
const values etc., and typedefs. These types all have DW_AT_type
attribute pointing the underlying type, and thus produce similar
output.
Signed-off-by: Sami Tolvanen
---
tools/gendwarfksyms/gendwarfksyms.h | 3 ++
tools
byte_size(1)};
Signed-off-by: Sami Tolvanen
---
tools/gendwarfksyms/cache.c | 13 +
tools/gendwarfksyms/gendwarfksyms.c | 3 +++
tools/gendwarfksyms/gendwarfksyms.h | 5 -
tools/gendwarfksyms/types.c | 22 ++
4 files changed, 42 insertions(+), 1
oid;
Signed-off-by: Sami Tolvanen
---
tools/gendwarfksyms/gendwarfksyms.h | 1 +
tools/gendwarfksyms/types.c | 58 -
2 files changed, 58 insertions(+), 1 deletion(-)
diff --git a/tools/gendwarfksyms/gendwarfksyms.h
b/tools/gendwarfksyms/gendwarfksyms.h
in
Add support for expanding DW_TAG_array_type, and the subrange type
indicating array size.
Example output with --debug:
variable array_type [34] {
pointer_type {
const_type {
base_type char byte_size(1)
}
}
};
Signed-off-by: Sami Tolvanen
---
tools
levels
inside the pointer. This should be plenty for catching ABI differences
and stops us from pulling in half the kernel for structs that contain
pointers to large structs like task_struct.
Signed-off-by: Sami Tolvanen
---
tools/gendwarfksyms/cache.c | 49
tools
Understanding the operation and caching of the tool can be somewhat
challenging without a debugger. Add inline debugging information with
the --inline-debug command, which adds highlighted tags to the --debug
output with information about cache states etc.
Signed-off-by: Sami Tolvanen
---
tools
if needed.
Suggested-by: Matthew Maurer
Signed-off-by: Sami Tolvanen
---
scripts/mod/Makefile | 4 +-
scripts/mod/modpost.c | 20 ++-
scripts/mod/modpost.h | 20 +++
scripts/mod/symhash.c | 327 ++
4 files changed, 364 insertions(+), 7 deletions(-)
When checking modversions for symbol names longer than MODULE_NAME_LEN,
look for the hashed symbol name instead. This is needed for Rust modules,
which frequently use symbol names longer than the maximum length
supported by struct modversion_info.
Suggested-by: Matthew Maurer
Signed-off-by: Sami
1 - 100 of 1151 matches
Mail list logo
