This is an automated cleanup. This bug report has been moved
to QEMU's new bug tracker on gitlab.com and thus gets marked
as 'expired' now. Please continue with the discussion here:
https://gitlab.com/qemu-project/qemu/-/issues/61
** Changed in: qemu
Status: Confirmed => Expired
** Chan
I still want to look at this.
** Changed in: qemu
Status: Incomplete => Confirmed
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1918302
Title:
qemu-system-arm segfaults while servicing SYS
The state of this bug is "Fix committed", but the last comments look
like there is still work to do ... should this get reset to "New" or can
we close this bug now?
** Changed in: qemu
Status: Fix Committed => Incomplete
--
You received this bug notification because you are a member of qe
Yes, fair enough. I suppose what I meant was that that particular part
of the startup code was *regarding* SP as being uninitialised: it didn't
read it, or use it, or set it on purpose to any kind of interim temp
value before calling SYS_HEAPINFO.
It's true, of course, that this particular image d
Since this is an M-profile board, "SP is literally uninitialised at this
point" isn't correct unless the loaded image failed to provide a valid
vector table, because M-profile loads the initial SP from there.
Generally for bare metal programs, the linker has a much better idea of
what would be a go
Firstly, I agree with Peter's comment – this test image is exactly an
example of what he describes, in that it carefully doesn't make any use
of the value of SP it started up with (doesn't push or pop anything,
doesn't make sp-relative offsets). Very near the start, it invokes
SYS_HEAPINFO to decid
Oops – naturally, I realised just *after* hitting send that it would
have been a good idea to say exactly which git commit I was testing on.
It was 9e2e9fe3df9f539f8b6941ceb96d25355fdae47e .
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to Q
I think there's still work to do here -- we don't properly tell
semihosting where the memory is on M-profile or in all A-profile cases.
I don't think that "look at the stack pointer" is a very good heuristic.
--
You received this bug notification because you are a member of qemu-
devel-ml, which
I think this is fixed now - it would be useful if the OP could confirm
with the current state of master.
** Changed in: qemu
Status: In Progress => Fix Committed
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.lau
Looking at the current stack seems an odd approach, because often
semihosting guest programs use HEAPINFO to find out what they should set
SP to in the first place...
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launch
Ends doing:
(gdb)
0x0ce4 in ?? ()
=> 0xce4: b.n 0xce4
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1918302
Title:
qemu-system-arm segfaults while servicing SYS_HEAPINFO
Status in Q
Testing with:
Subject: [PATCH v5 0/5] semihosting/next (SYS_HEAPINFO)
Date: Fri, 12 Mar 2021 10:20:24 +
Message-Id: <20210312102029.17017-1-alex.ben...@linaro.org>
it doesn't seem to segfault QEMU anymore although the guest itself hangs
which probably means it's not happy with the numbers it
I think this was fixed earlier by:
commit 095f8c029319b79cce487e3b566cd826b93da3e6
Author: Keith Packard
Date: Fri Jan 8 22:42:51 2021 +
semihosting: Support SYS_HEAPINFO when env->boot_info is not set
env->boot_info is only set in some ARM startup paths, so we cannot
rely on
** Tags added: semihosting
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1918302
Title:
qemu-system-arm segfaults while servicing SYS_HEAPINFO
Status in QEMU:
New
Bug description:
I compiled
...oh, and also for A-profile where we do a boot of firmware and not a
Linux-kernel style boot.
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1918302
Title:
qemu-system-arm segfaults while servicin
Note that this only happens for M-profile in system emulation mode.
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1918302
Title:
qemu-system-arm segfaults while servicing SYS_HEAPINFO
Status in QE
** Tags added: arm
--
You received this bug notification because you are a member of qemu-
devel-ml, which is subscribed to QEMU.
https://bugs.launchpad.net/bugs/1918302
Title:
qemu-system-arm segfaults while servicing SYS_HEAPINFO
Status in QEMU:
New
Bug description:
I compiled QEMU ver
17 matches
Mail list logo
