7:04 p.m.
Afternoon,
So, NetSurf runs mostly ok on ARMv7 platforms with their backwards
incompatible unaligned LDR behaviour turned on. Tinct, however, appears
to perform some LDRs from unaligned addresses, resulting in crashes.
Here's a backtrace:
Fatal signal received: Segmentation fault
Stack backtrace:
Running thread 0x59e53c
( 5a7ee0) pc: 462860 lr: 74f4c sp: 5a7ee4 __write_backtrace()
( 5a7f10) pc: 74d64 lr: 463298 sp: 5a7f14 ^ro_gui_signal()
( 5a7f38) pc: 463288 lr: 462f58 sp: 5a7f3c __unixlib_exec_sig()
( 5a7fa0) pc: 462a18 lr: 463870 sp: 5a7fa4 __unixlib_raise_signal()
( 5a7fb0) pc: 463774 lr: 5a6c48 sp: 5a6c04 __h_cback()
Register dump at 005a7fb4:
a1: 12f a2: 2d a3: 1f4 a4: 0
v1: 0 v2: 0 v3: 202513b4 v4: 1
v5: 3d5 v6: 0 sl: 41b78d39 fp: 67197124
ip: 20248394 sp: 5a6c04 lr: 5a6c48 pc: 2024609c
cpsr: 80000113
20246088 : .p.â : e21a7003 : ANDS R7,R10,#3
2024608c : .... : 0a000008 : BEQ &202460B4
20246090 : .àgâ : e267e003 : RSB R14,R7,#3
20246094 : ..^á : e15e0000 : CMP R14,R0
20246098 : .à Á : c1a0e000 : MOVGT R14,R0
2024609c : .??å : e59a9000 : LDR R9,[R10,#0]
202460a0 : ..@Ð : d040000e : SUBLE R0,R0,R14
202460a4 : ..?Ò : d2800003 : ADDLE R0,R0,#3
202460a8 : ..?À : c0800007 : ADDGT R0,R0,R7
Invalid pc address bebeec
For reference, Tinct is loaded at 202416B4.
J.
10:26 a.m.
Hi All,
Indeed it does... There is a build option that turns this off IIRC. I
should be back at the house the relevant machine is at in a few days, so
I'll dig it out and /finally/ put it all in SVN.
Thanks!
R.
-----Original Message-----
From: netsurf-dev-bounces(a)netsurf-browser.org
[mailto:netsurf-dev-bounces@netsurf-browser.org] On Behalf Of
John-Mark Bell
Sent: 05 July 2009 18:04
To: netsurf-dev(a)netsurf-browser.org
Subject: NetSurf on ARMv7 platforms
Afternoon,
So, NetSurf runs mostly ok on ARMv7 platforms with their
backwards incompatible unaligned LDR behaviour turned on.
Tinct, however, appears to perform some LDRs from unaligned
addresses, resulting in crashes.
Here's a backtrace:
> Fatal signal received: Segmentation fault
>
> Stack backtrace:
>
> Running thread 0x59e53c
> ( 5a7ee0) pc: 462860 lr: 74f4c sp: 5a7ee4
__write_backtrace()
> ( 5a7f10) pc: 74d64 lr: 463298 sp: 5a7f14
^ro_gui_signal()
> ( 5a7f38) pc: 463288 lr: 462f58 sp: 5a7f3c
__unixlib_exec_sig()
> ( 5a7fa0) pc: 462a18 lr: 463870 sp: 5a7fa4
__unixlib_raise_signal()
> ( 5a7fb0) pc: 463774 lr: 5a6c48 sp: 5a6c04 __h_cback()
>
> Register dump at 005a7fb4:
>
> a1: 12f a2: 2d a3: 1f4 a4: 0
> v1: 0 v2: 0 v3: 202513b4 v4: 1
> v5: 3d5 v6: 0 sl: 41b78d39 fp: 67197124
> ip: 20248394 sp: 5a6c04 lr: 5a6c48 pc: 2024609c
> cpsr: 80000113
>
> 20246088 : .p.â : e21a7003 : ANDS R7,R10,#3
> 2024608c : .... : 0a000008 : BEQ &202460B4
> 20246090 : .àgâ : e267e003 : RSB R14,R7,#3
> 20246094 : ..^á : e15e0000 : CMP R14,R0
> 20246098 : .à Á : c1a0e000 : MOVGT R14,R0
> 2024609c : .??å : e59a9000 : LDR R9,[R10,#0]
> 202460a0 : ..@Ð : d040000e : SUBLE R0,R0,R14
> 202460a4 : ..?Ò : d2800003 : ADDLE R0,R0,#3
> 202460a8 : ..?À : c0800007 : ADDGT R0,R0,R7
>
> Invalid pc address bebeec
For reference, Tinct is loaded at 202416B4.
J.
7:14 p.m.
On Mon, 2009-07-06 at 09:26 +0100, Richard Wilson wrote:
Hi All,
Indeed it does... There is a build option that turns this off IIRC. I
should be back at the house the relevant machine is at in a few days, so
I'll dig it out and /finally/ put it all in SVN.
That's great. Thanks.
J.
12:36 a.m.
On Mon, 2009-07-06 at 18:14 +0100, John-Mark Bell wrote:
On Mon, 2009-07-06 at 09:26 +0100, Richard Wilson wrote:
> Hi All,
> Indeed it does... There is a build option that turns this off IIRC. I
> should be back at the house the relevant machine is at in a few days, so
> I'll dig it out and /finally/ put it all in SVN.
That's great. Thanks.
Any chance of some progress on this? RISC OS on ARMv7 platforms will
abort any attempt to LDR from an unaligned address which, obviously,
reduces the usefulness of NetSurf on those platforms.
J.
10:24 a.m.
-----Original Message-----
From: netsurf-dev-bounces(a)netsurf-browser.org
[mailto:netsurf-dev-bounces@netsurf-browser.org] On Behalf Of
John-Mark Bell
Sent: 15 November 2009 23:37
To: netsurf-dev(a)netsurf-browser.org
Subject: RE: NetSurf on ARMv7 platforms
On Mon, 2009-07-06 at 18:14 +0100, John-Mark Bell wrote:
> On Mon, 2009-07-06 at 09:26 +0100, Richard Wilson wrote:
> > Hi All,
> > Indeed it does... There is a build option that turns
this off IIRC.
> > I should be back at the house the relevant machine is at in a few
> > days, so I'll dig it out and /finally/ put it all in SVN.
>
> That's great. Thanks.
Any chance of some progress on this? RISC OS on ARMv7
platforms will abort any attempt to LDR from an unaligned
address which, obviously, reduces the usefulness of NetSurf
on those platforms.
Sorry, this totally slipped my mind. I'll try to get the source Zip-ped up
over the weekend and across to you for putting into SVN.
Thanks!
R.
--
Richard Wilson
5:56 p.m.
Any updates on this? After hearing that NetSurf was crashing I downloaded
the 2.5 release and tried it myself on my beagleboard. The first crash I
got appeared to be in exactly the same part of Tinct as shown in JMB's
crash dump.
Cheers,
- Jeffrey
On Mon, 6 Jul 2009, Richard Wilson wrote:
Hi All,
Indeed it does... There is a build option that turns this off IIRC. I
should be back at the house the relevant machine is at in a few days, so
I'll dig it out and /finally/ put it all in SVN.
Thanks!
R.
> -----Original Message-----
> From: netsurf-dev-bounces(a)netsurf-browser.org
> [mailto:netsurf-dev-bounces@netsurf-browser.org] On Behalf Of
> John-Mark Bell
> Sent: 05 July 2009 18:04
> To: netsurf-dev(a)netsurf-browser.org
> Subject: NetSurf on ARMv7 platforms
>
> Afternoon,
>
> So, NetSurf runs mostly ok on ARMv7 platforms with their
> backwards incompatible unaligned LDR behaviour turned on.
> Tinct, however, appears to perform some LDRs from unaligned
> addresses, resulting in crashes.
>
> Here's a backtrace:
>
>> Fatal signal received: Segmentation fault
>>
>> Stack backtrace:
>>
>> Running thread 0x59e53c
>> ( 5a7ee0) pc: 462860 lr: 74f4c sp: 5a7ee4
> __write_backtrace()
>> ( 5a7f10) pc: 74d64 lr: 463298 sp: 5a7f14
> ^ro_gui_signal()
>> ( 5a7f38) pc: 463288 lr: 462f58 sp: 5a7f3c
> __unixlib_exec_sig()
>> ( 5a7fa0) pc: 462a18 lr: 463870 sp: 5a7fa4
> __unixlib_raise_signal()
>> ( 5a7fb0) pc: 463774 lr: 5a6c48 sp: 5a6c04 __h_cback()
>>
>> Register dump at 005a7fb4:
>>
>> a1: 12f a2: 2d a3: 1f4 a4: 0
>> v1: 0 v2: 0 v3: 202513b4 v4: 1
>> v5: 3d5 v6: 0 sl: 41b78d39 fp: 67197124
>> ip: 20248394 sp: 5a6c04 lr: 5a6c48 pc: 2024609c
>> cpsr: 80000113
>>
>> 20246088 : .p.â : e21a7003 : ANDS R7,R10,#3
>> 2024608c : .... : 0a000008 : BEQ &202460B4
>> 20246090 : .àgâ : e267e003 : RSB R14,R7,#3
>> 20246094 : ..^á : e15e0000 : CMP R14,R0
>> 20246098 : .à Á : c1a0e000 : MOVGT R14,R0
>> 2024609c : .??å : e59a9000 : LDR R9,[R10,#0]
>> 202460a0 : ..@Ð : d040000e : SUBLE R0,R0,R14
>> 202460a4 : ..?Ò : d2800003 : ADDLE R0,R0,#3
>> 202460a8 : ..?À : c0800007 : ADDGT R0,R0,R7
>>
>> Invalid pc address bebeec
>
> For reference, Tinct is loaded at 202416B4.
>
>
> J.
>
>
6:13 p.m.
On Sun, 30 May 2010 16:56:27 +0100 (GMT Daylight Time)
Jeffrey Lee <me(a)phlamethrower.co.uk> wrote:
Any updates on this? After hearing that NetSurf was crashing I
downloaded the 2.5 release and tried it myself on my beagleboard. The
first crash I got appeared to be in exactly the same part of Tinct as
shown in JMB's crash dump.
I don't believe we ever managed to agree on a licence for Tinct's
source code with Richard. My understanding is that we simply want to
remove the dependency on Tinct now.
B.
6:27 p.m.
In message <Pine.WNT.4.62.1005301649200.708@c203>
Jeffrey Lee <me(a)phlamethrower.co.uk> wrote:
Any updates on this? After hearing that NetSurf was crashing I
downloaded
the 2.5 release and tried it myself on my beagleboard. The first crash I
got appeared to be in exactly the same part of Tinct as shown in JMB's
crash dump.
"NetSurf configuration" -> "Image quality" : "Foreground = Use
OS"
and "Background = Use OS" might be a temporary workaround. Please let
us know.
John.
--
John Tytgat
joty(a)netsurf-browser.org
6:46 p.m.
On Sun, 30 May 2010, John Tytgat wrote:
In message <Pine.WNT.4.62.1005301649200.708@c203>
Jeffrey Lee <me(a)phlamethrower.co.uk> wrote:
> Any updates on this? After hearing that NetSurf was crashing I downloaded
> the 2.5 release and tried it myself on my beagleboard. The first crash I
> got appeared to be in exactly the same part of Tinct as shown in JMB's
> crash dump.
"NetSurf configuration" -> "Image quality" : "Foreground =
Use OS"
and "Background = Use OS" might be a temporary workaround. Please let
us know.
Yes, that seems to work. Also it's worth pointing out that it looks like
it's only 8bpp and 16bpp screen modes which result in a crash - 32bpp
seems to work fine, along with 4/2/1bpp (although I doubt many people
would accept those as a workaround). And any attempt to open the image
quality settings window in a 8bpp or 16bpp mode while NetSurf is set to
use Tinct will result in a crash, so you'd have to change to a working
mode to start using the workaround anyway.
In a week or two I'll hopefully have a working abort handler which will be
used to emulate the old rotated load behaviour for any code that attempts
an unaligned access. Relying on an abort handler obviously isn't an ideal
solution, but at least it will make NetSurf usable if the problems with
Tinct aren't resolved any time soon.
Cheers,
- Jeffrey
4576
days inactive
4905
days old
netsurf-dev@netsurf-browser.org
8 comments
5 participants
participants (5)
-
Jeffrey Lee -
John Tytgat -
John-Mark Bell -
Richard Wilson -
Rob Kendrick