From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([209.51.188.92]:56788) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1guQZp-00012q-58 for qemu-devel@nongnu.org; Thu, 14 Feb 2019 18:37:58 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1guQZi-0006NS-Df for qemu-devel@nongnu.org; Thu, 14 Feb 2019 18:37:53 -0500 Date: Thu, 14 Feb 2019 16:37:43 -0700 From: Alex Williamson Message-ID: <20190214163743.7162ad53@w520.home> In-Reply-To: <20190214052144.59541-1-aik@ozlabs.ru> References: <20190214052144.59541-1-aik@ozlabs.ru> MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH qemu v2 0/4] spapr_pci, vfio: NVIDIA V100 + POWER9 passthrough List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alexey Kardashevskiy Cc: qemu-devel@nongnu.org, qemu-ppc@nongnu.org, David Gibson , Reza Arbab , Piotr Jaroszynski , Jose Ricardo Ziviani , Daniel Henrique Barboza On Thu, 14 Feb 2019 16:21:40 +1100 Alexey Kardashevskiy wrote: > This is for passing through NVIDIA V100 GPUs on POWER9 systems. >=20 > This implements a subdriver for NVIDIA V100 GPU with coherent memory and > NPU/ATS support available in the POWER9 CPU. >=20 > 1/4 is a preparation for bigger DMA windows. > 2/4 is a small cleanup. >=20 > Here is the kernel driver: > https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit= /drivers/vfio/pci?h=3Dv5.0-rc6&id=3D7f92891778dff62303c070ac81de7b7d80de331a >=20 > SLOF changes already went in. >=20 > This depends on "pci: Move NVIDIA vendor id to the rest of ids" (posted s= eparately). TBH, I'm not sure it was the best idea to let it live or die on it's own when it now creates a build dependency for this series. > This is based on sha1 > 1ea6057 Mark Cave-Ayland "mac_newworld: change default NIC to sungem for = mac99 machine". Perhaps this is why it doesn't apply cleanly against qemu.git. Are there dependencies we need to wait for in the ppc tree as well? > Please comment. Thanks. Besides the build dependency on PCI_VENDOR_ID_NVIDIA, I also get this: .../qemu.git/hw/vfio/spapr.c: In function =E2=80=98vfio_spapr_create_window= =E2=80=99: .../qemu.git/hw/vfio/spapr.c:212:9: error: =E2=80=98ret=E2=80=99 may be use= d uninitialized in this function [-Werror=3Dmaybe-uninitialized] error_report("Failed to create a window, ret =3D %d (%m)", ret); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ cc1: all warnings being treated as errors Is this series expected to go in through the ppc branch given the bulk of the changes are there? Thanks, Alex