Re: [Qemu-devel] [PATCHv3 00/12] pseries: Consolidate guest device tree construction
On Tue, Oct 25, 2016 at 02:23:41PM +1100, David Gibson wrote: > On Tue, Oct 25, 2016 at 01:50:02PM +1100, David Gibson wrote: > > On Mon, Oct 24, 2016 at 04:04:31PM +1100, David Gibson wrote: > > > For historical reasons construction of the guest device tree in spapr > > > is divided between spapr_create_fdt_skel() which is called at init > > > time, and spapr_build_fdt() which runs at reset time. Over time, more > > > and more things have needed to be moved to reset time. > > > > > > This series consolidates all the device tree construction to reset > > > time, with some minor cleanups along the way. This will help to make > > > it more maintainable in future. > > > > > > Changes since v2: > > > * Removed a leftover reference to a variable called 'stdout', > > > shadowing the standard library stdout > > > Changes since v1: > > > * Fixed a memory leak introduced by 1/12 (spotted by Thomas Huth) > > > * Removed one patch that's already merged in ppc-for-2.8 > > > > Thanks for the acks, everyone. I've now merged this series into > > ppc-for-2.8, tentatively. > > > > Bharata Rao reported some boot failures caused by the series, but I > > haven't been able to reproduce them so far. Bharata, could you retest > > with the latest version (in ppc-for-2.8) and, if the problems are > > still present, send me some reproducer steps. > > Actually, I just realised the bug Bharata reported looks identical to > one someone else spotted. That was caused by one of Thomas' NVRAM > cleanups, rather than the DT cleanup. It's already been removed from > ppc-for-2.8 pending debug. David - After I reported the boot failure with this patchset to you, I figured out that the root cause was in fact the NVRAM patchset which I later reported separately. Regards, Bharata.
Re: [Qemu-devel] [PATCHv3 00/12] pseries: Consolidate guest device tree construction
On Tue, Oct 25, 2016 at 01:50:02PM +1100, David Gibson wrote: > On Mon, Oct 24, 2016 at 04:04:31PM +1100, David Gibson wrote: > > For historical reasons construction of the guest device tree in spapr > > is divided between spapr_create_fdt_skel() which is called at init > > time, and spapr_build_fdt() which runs at reset time. Over time, more > > and more things have needed to be moved to reset time. > > > > This series consolidates all the device tree construction to reset > > time, with some minor cleanups along the way. This will help to make > > it more maintainable in future. > > > > Changes since v2: > > * Removed a leftover reference to a variable called 'stdout', > > shadowing the standard library stdout > > Changes since v1: > > * Fixed a memory leak introduced by 1/12 (spotted by Thomas Huth) > > * Removed one patch that's already merged in ppc-for-2.8 > > Thanks for the acks, everyone. I've now merged this series into > ppc-for-2.8, tentatively. > > Bharata Rao reported some boot failures caused by the series, but I > haven't been able to reproduce them so far. Bharata, could you retest > with the latest version (in ppc-for-2.8) and, if the problems are > still present, send me some reproducer steps. Actually, I just realised the bug Bharata reported looks identical to one someone else spotted. That was caused by one of Thomas' NVRAM cleanups, rather than the DT cleanup. It's already been removed from ppc-for-2.8 pending debug. -- David Gibson| I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson signature.asc Description: PGP signature
Re: [Qemu-devel] [PATCHv3 00/12] pseries: Consolidate guest device tree construction
On Mon, Oct 24, 2016 at 04:04:31PM +1100, David Gibson wrote: > For historical reasons construction of the guest device tree in spapr > is divided between spapr_create_fdt_skel() which is called at init > time, and spapr_build_fdt() which runs at reset time. Over time, more > and more things have needed to be moved to reset time. > > This series consolidates all the device tree construction to reset > time, with some minor cleanups along the way. This will help to make > it more maintainable in future. > > Changes since v2: > * Removed a leftover reference to a variable called 'stdout', > shadowing the standard library stdout > Changes since v1: > * Fixed a memory leak introduced by 1/12 (spotted by Thomas Huth) > * Removed one patch that's already merged in ppc-for-2.8 Thanks for the acks, everyone. I've now merged this series into ppc-for-2.8, tentatively. Bharata Rao reported some boot failures caused by the series, but I haven't been able to reproduce them so far. Bharata, could you retest with the latest version (in ppc-for-2.8) and, if the problems are still present, send me some reproducer steps. -- David Gibson| I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson signature.asc Description: PGP signature
[Qemu-devel] [PATCHv3 00/12] pseries: Consolidate guest device tree construction
For historical reasons construction of the guest device tree in spapr is divided between spapr_create_fdt_skel() which is called at init time, and spapr_build_fdt() which runs at reset time. Over time, more and more things have needed to be moved to reset time. This series consolidates all the device tree construction to reset time, with some minor cleanups along the way. This will help to make it more maintainable in future. Changes since v2: * Removed a leftover reference to a variable called 'stdout', shadowing the standard library stdout Changes since v1: * Fixed a memory leak introduced by 1/12 (spotted by Thomas Huth) * Removed one patch that's already merged in ppc-for-2.8 David Gibson (12): pseries: Split device tree construction from device tree load pseries: Remove rtas_addr and fdt_addr fields from machinestate pseries: Make spapr_create_fdt_skel() get information from machine state pseries: Move adding of fdt reserve map entries pseries: Consolidate RTAS loading pseries: Move construction of /interrupt-controller fdt node pseries: Consolidate construction of /chosen device tree node pseries: Consolidate construction of /rtas device tree node pseries: Move /event-sources construction to spapr_build_fdt() pseries: Move /hypervisor node construction to fdt_build_fdt() pseries: Consolidate construction of /vdevice device tree node pseries: Remove spapr_create_fdt_skel() hw/intc/xics_spapr.c | 22 ++ hw/ppc/spapr.c | 560 + hw/ppc/spapr_events.c | 21 +- hw/ppc/spapr_rtas.c| 91 +++- hw/ppc/spapr_vio.c | 40 ++-- include/hw/ppc/spapr.h | 12 +- include/hw/ppc/spapr_vio.h | 4 +- include/hw/ppc/xics.h | 1 + 8 files changed, 361 insertions(+), 390 deletions(-) -- 2.7.4
