"Alex Bennée" <alex.ben...@linaro.org> wrote on 01/12/2021 06:26:50 AM:
> From: "Alex Bennée" <alex.ben...@linaro.org> > To: alar...@ddci.com > Cc: "Berto Furth" <bertofu...@sent.com>, qemu-discuss@nongnu.org > Date: 01/12/2021 06:29 AM > Subject: Re: Debugging 32 bit ARM guests running under QEMU KVM on > 64 bit ARM hosts with gdb > > > alar...@ddci.com writes: > > >> From: "Berto Furth" <bertofu...@sent.com> > >> To: qemu-discuss@nongnu.org > >> Date: 01/08/2021 05:48 AM > >> Subject: Debugging 32 bit ARM guests running under QEMU KVM on 64 > >> bit ARM hosts with gdb > > > >> I have noticed that when debugging a 32 bit arm (AA32) kvm guest > >> running in a 64 bit ARM (aarch64 / AA64) host with "qemu-system- > >> aarch64" and the "aarch64=off" flag set that a remote gdb debugger > >> still thinks that it's debugging ARM64. ... > > > > Just FYI, my group has been maintaining a similar patch so we can debug > > 32-bit guests. I don't have qemu commit privileges (and I shouldn't), > > but I would be willing to support this activity in terms of testing > > patches, which I can do on arm, ppc, and x86 guests. We haven't updated > > to recent QEMU versions recently, but for this I would. > > Re-reading the old thread I think this stalled because it wasn't > entirely clear moving the gdbstub initialisation in the second patch was > the cleanest way to deal with things. > > Does you patch take a similar approach? Sadly our patch is pretty brutal. It hard wires the GDB stub to 32-bit. We ran into several problems with the Eclipse version (at the time) stepping between 32 and 64-bit code. Since most of our code was 32-bit we just forced 32-bit access always. More recently we have more 64-bit code, hence the interest in looking for something that handles both more gracefully. I'm hoping the issues with Eclipse have been resolved in the mean time.
