From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-7.0 required=3.0 tests=DATE_IN_PAST_03_06, DKIM_INVALID,DKIM_SIGNED,HEADER_FROM_DIFFERENT_DOMAINS,INCLUDES_PATCH, MAILING_LIST_MULTI,SIGNED_OFF_BY,SPF_HELO_NONE,SPF_PASS,URIBL_BLOCKED, USER_AGENT_SANE_1 autolearn=ham autolearn_force=no version=3.4.0 Received: from mail.kernel.org (mail.kernel.org [198.145.29.99]) by smtp.lore.kernel.org (Postfix) with ESMTP id 4634CCA9EA0 for ; Fri, 18 Oct 2019 09:32:41 +0000 (UTC) Received: from lists.gnu.org (lists.gnu.org [209.51.188.17]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.kernel.org (Postfix) with ESMTPS id 136EF21852 for ; Fri, 18 Oct 2019 09:32:41 +0000 (UTC) Authentication-Results: mail.kernel.org; dkim=fail reason="signature verification failed" (1024-bit key) header.d=gibson.dropbear.id.au header.i=@gibson.dropbear.id.au header.b="W/P1gAm8" DMARC-Filter: OpenDMARC Filter v1.3.2 mail.kernel.org 136EF21852 Authentication-Results: mail.kernel.org; dmarc=none (p=none dis=none) header.from=gibson.dropbear.id.au Authentication-Results: mail.kernel.org; spf=pass smtp.mailfrom=qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Received: from localhost ([::1]:37052 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1iLOch-0002hM-On for qemu-devel@archiver.kernel.org; Fri, 18 Oct 2019 05:32:39 -0400 Received: from eggs.gnu.org ([2001:470:142:3::10]:46586) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1iLOax-0001CR-9a for qemu-devel@nongnu.org; Fri, 18 Oct 2019 05:30:54 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1iLOav-0006eV-0o for qemu-devel@nongnu.org; Fri, 18 Oct 2019 05:30:50 -0400 Received: from bilbo.ozlabs.org ([203.11.71.1]:59355 helo=ozlabs.org) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1iLOat-0006cT-EE; Fri, 18 Oct 2019 05:30:48 -0400 Received: by ozlabs.org (Postfix, from userid 1007) id 46vgkT6nc9z9sPJ; Fri, 18 Oct 2019 20:30:41 +1100 (AEDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=gibson.dropbear.id.au; s=201602; t=1571391041; bh=CrllFX9Yj4Jim06wHOVJCG0btafnhMARMN3EbZlNxgg=; h=Date:From:To:Cc:Subject:References:In-Reply-To:From; b=W/P1gAm8+Sv2h1Hubu6LITum6YOEh74LHJNWQMIrEo/5UQ7G81EgJOkpAnVVjde3i hP9xvfq2/MKw6umpPfSZlH6OgxKeShW4rTgRBWreJrrf+zLm/nhRZTudEhjyIRGIYp /m+HyoqDJqfxYYwSFmzldBoeMRtNApRVE7sl1Ab0= Date: Fri, 18 Oct 2019 14:55:57 +1100 From: David Gibson To: =?iso-8859-1?Q?C=E9dric?= Le Goater Subject: Re: [PATCH 2/2] spapr/xive: Set the OS CAM line at reset Message-ID: <20191018035557.GC2000@umbus.fritz.box> References: <20191017144241.12522-1-clg@kaod.org> <20191017144241.12522-3-clg@kaod.org> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="nmemrqcdn5VTmUEE" Content-Disposition: inline In-Reply-To: <20191017144241.12522-3-clg@kaod.org> User-Agent: Mutt/1.12.1 (2019-06-15) X-detected-operating-system: by eggs.gnu.org: GNU/Linux 2.2.x-3.x [generic] [fuzzy] X-Received-From: 203.11.71.1 X-BeenThere: qemu-devel@nongnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: qemu-ppc@nongnu.org, qemu-devel@nongnu.org, Greg Kurz Errors-To: qemu-devel-bounces+qemu-devel=archiver.kernel.org@nongnu.org Sender: "Qemu-devel" --nmemrqcdn5VTmUEE Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Thu, Oct 17, 2019 at 04:42:41PM +0200, C=E9dric Le Goater wrote: > When a Virtual Processor is scheduled to run on a HW thread, the > hypervisor pushes its identifier in the OS CAM line. When running in > TCG or kernel_irqchip=3Doff, QEMU needs to emulate the same behavior. >=20 > Introduce a 'os-cam' property which will be used to set the OS CAM > line at reset and remove the spapr_xive_set_tctx_os_cam() calls which > are done when the XIVE interrupt controller are activated. I'm not immediately seeing the advantage of doing this via a property, rather than poking it from the PAPR code which already knows the right values. Also, let me check my understanding: IIUC, on powernv the OS (running in HV mode) can alter the OS CAM lines for itself and/or its guests, but for pseries they're fixed in place. Is that right? > This change also has the benefit to remove the use of CPU_FOREACH() > which can be unsafe. >=20 > Signed-off-by: C=E9dric Le Goater > --- > include/hw/ppc/spapr_xive.h | 1 - > include/hw/ppc/xive.h | 4 +++- > hw/intc/spapr_xive.c | 31 +++++-------------------------- > hw/intc/xive.c | 22 +++++++++++++++++++++- > hw/ppc/pnv.c | 3 ++- > 5 files changed, 31 insertions(+), 30 deletions(-) >=20 > diff --git a/include/hw/ppc/spapr_xive.h b/include/hw/ppc/spapr_xive.h > index d84bd5c229f0..742b7e834f2a 100644 > --- a/include/hw/ppc/spapr_xive.h > +++ b/include/hw/ppc/spapr_xive.h > @@ -57,7 +57,6 @@ typedef struct SpaprXive { > void spapr_xive_pic_print_info(SpaprXive *xive, Monitor *mon); > =20 > void spapr_xive_hcall_init(SpaprMachineState *spapr); > -void spapr_xive_set_tctx_os_cam(XiveTCTX *tctx); > void spapr_xive_mmio_set_enabled(SpaprXive *xive, bool enable); > void spapr_xive_map_mmio(SpaprXive *xive); > =20 > diff --git a/include/hw/ppc/xive.h b/include/hw/ppc/xive.h > index 99381639f50c..e273069c25a9 100644 > --- a/include/hw/ppc/xive.h > +++ b/include/hw/ppc/xive.h > @@ -319,6 +319,7 @@ typedef struct XiveTCTX { > qemu_irq os_output; > =20 > uint8_t regs[XIVE_TM_RING_COUNT * XIVE_TM_RING_SIZE]; > + uint32_t os_cam; > } XiveTCTX; > =20 > /* > @@ -414,7 +415,8 @@ void xive_tctx_tm_write(XiveTCTX *tctx, hwaddr offset= , uint64_t value, > uint64_t xive_tctx_tm_read(XiveTCTX *tctx, hwaddr offset, unsigned size); > =20 > void xive_tctx_pic_print_info(XiveTCTX *tctx, Monitor *mon); > -Object *xive_tctx_create(Object *cpu, XiveRouter *xrtr, Error **errp); > +Object *xive_tctx_create(Object *cpu, XiveRouter *xrtr, uint32_t os_cam, > + Error **errp); > void xive_tctx_reset(XiveTCTX *tctx); > =20 > static inline uint32_t xive_nvt_cam_line(uint8_t nvt_blk, uint32_t nvt_i= dx) > diff --git a/hw/intc/spapr_xive.c b/hw/intc/spapr_xive.c > index 0c3acf1a4192..71f138512a1c 100644 > --- a/hw/intc/spapr_xive.c > +++ b/hw/intc/spapr_xive.c > @@ -205,21 +205,13 @@ void spapr_xive_mmio_set_enabled(SpaprXive *xive, b= ool enable) > memory_region_set_enabled(&xive->end_source.esb_mmio, false); > } > =20 > -/* > - * When a Virtual Processor is scheduled to run on a HW thread, the > - * hypervisor pushes its identifier in the OS CAM line. Emulate the > - * same behavior under QEMU. > - */ > -void spapr_xive_set_tctx_os_cam(XiveTCTX *tctx) > +static uint32_t spapr_xive_get_os_cam(PowerPCCPU *cpu) > { > uint8_t nvt_blk; > uint32_t nvt_idx; > - uint32_t nvt_cam; > - > - spapr_xive_cpu_to_nvt(POWERPC_CPU(tctx->cs), &nvt_blk, &nvt_idx); > =20 > - nvt_cam =3D cpu_to_be32(TM_QW1W2_VO | xive_nvt_cam_line(nvt_blk, nvt= _idx)); > - memcpy(&tctx->regs[TM_QW1_OS + TM_WORD2], &nvt_cam, 4); > + spapr_xive_cpu_to_nvt(cpu, &nvt_blk, &nvt_idx); > + return xive_nvt_cam_line(nvt_blk, nvt_idx); > } > =20 > static void spapr_xive_end_reset(XiveEND *end) > @@ -537,19 +529,14 @@ static int spapr_xive_cpu_intc_create(SpaprInterrup= tController *intc, > SpaprXive *xive =3D SPAPR_XIVE(intc); > Object *obj; > SpaprCpuState *spapr_cpu =3D spapr_cpu_state(cpu); > + uint32_t os_cam =3D spapr_xive_get_os_cam(cpu); > =20 > - obj =3D xive_tctx_create(OBJECT(cpu), XIVE_ROUTER(xive), errp); > + obj =3D xive_tctx_create(OBJECT(cpu), XIVE_ROUTER(xive), os_cam, err= p); > if (!obj) { > return -1; > } > =20 > spapr_cpu->tctx =3D XIVE_TCTX(obj); > - > - /* > - * (TCG) Early setting the OS CAM line for hotplugged CPUs as they > - * don't beneficiate from the reset of the XIVE IRQ backend > - */ > - spapr_xive_set_tctx_os_cam(spapr_cpu->tctx); > return 0; > } > =20 > @@ -650,14 +637,6 @@ static void spapr_xive_dt(SpaprInterruptController *= intc, uint32_t nr_servers, > static int spapr_xive_activate(SpaprInterruptController *intc, Error **e= rrp) > { > SpaprXive *xive =3D SPAPR_XIVE(intc); > - CPUState *cs; > - > - CPU_FOREACH(cs) { > - PowerPCCPU *cpu =3D POWERPC_CPU(cs); > - > - /* (TCG) Set the OS CAM line of the thread interrupt context. */ > - spapr_xive_set_tctx_os_cam(spapr_cpu_state(cpu)->tctx); > - } > =20 > if (kvm_enabled()) { > int rc =3D spapr_irq_init_kvm(kvmppc_xive_connect, intc, errp); > diff --git a/hw/intc/xive.c b/hw/intc/xive.c > index 0ae3f9b1efe4..be4f2c974178 100644 > --- a/hw/intc/xive.c > +++ b/hw/intc/xive.c > @@ -566,6 +566,18 @@ static void xive_tctx_reset_handler(void *dev) > ipb_to_pipr(tctx->regs[TM_QW1_OS + TM_IPB]); > tctx->regs[TM_QW3_HV_PHYS + TM_PIPR] =3D > ipb_to_pipr(tctx->regs[TM_QW3_HV_PHYS + TM_IPB]); > + > + /* > + * (TCG) Set the OS CAM line of the thread interrupt context. > + * > + * When a Virtual Processor is scheduled to run on a HW thread, > + * the hypervisor pushes its identifier in the OS CAM line. > + * Emulate the same behavior under QEMU. > + */ > + if (tctx->os_cam) { > + uint32_t qw1w2 =3D cpu_to_be32(TM_QW1W2_VO | tctx->os_cam); > + memcpy(&tctx->regs[TM_QW1_OS + TM_WORD2], &qw1w2, 4); > + } > } > =20 > void xive_tctx_reset(XiveTCTX *tctx) > @@ -667,11 +679,17 @@ static const VMStateDescription vmstate_xive_tctx = =3D { > }, > }; > =20 > +static Property xive_tctx_properties[] =3D { > + DEFINE_PROP_UINT32("os-cam", XiveTCTX, os_cam, 0), > + DEFINE_PROP_END_OF_LIST(), > +}; > + > static void xive_tctx_class_init(ObjectClass *klass, void *data) > { > DeviceClass *dc =3D DEVICE_CLASS(klass); > =20 > dc->desc =3D "XIVE Interrupt Thread Context"; > + dc->props =3D xive_tctx_properties; > dc->realize =3D xive_tctx_realize; > dc->unrealize =3D xive_tctx_unrealize; > dc->vmsd =3D &vmstate_xive_tctx; > @@ -689,7 +707,8 @@ static const TypeInfo xive_tctx_info =3D { > .class_init =3D xive_tctx_class_init, > }; > =20 > -Object *xive_tctx_create(Object *cpu, XiveRouter *xrtr, Error **errp) > +Object *xive_tctx_create(Object *cpu, XiveRouter *xrtr, uint32_t os_cam, > + Error **errp) > { > Error *local_err =3D NULL; > Object *obj; > @@ -698,6 +717,7 @@ Object *xive_tctx_create(Object *cpu, XiveRouter *xrt= r, Error **errp) > object_property_add_child(cpu, TYPE_XIVE_TCTX, obj, &error_abort); > object_unref(obj); > object_property_add_const_link(obj, "cpu", cpu, &error_abort); > + object_property_set_int(obj, os_cam, "os-cam", &local_err); > object_property_set_bool(obj, true, "realized", &local_err); > if (local_err) { > goto error; > diff --git a/hw/ppc/pnv.c b/hw/ppc/pnv.c > index 7cf64b6d2533..99c06842573e 100644 > --- a/hw/ppc/pnv.c > +++ b/hw/ppc/pnv.c > @@ -806,7 +806,8 @@ static void pnv_chip_power9_intc_create(PnvChip *chip= , PowerPCCPU *cpu, > * controller object is initialized afterwards. Hopefully, it's > * only used at runtime. > */ > - obj =3D xive_tctx_create(OBJECT(cpu), XIVE_ROUTER(&chip9->xive), &lo= cal_err); > + obj =3D xive_tctx_create(OBJECT(cpu), XIVE_ROUTER(&chip9->xive), 0, > + &local_err); > if (local_err) { > error_propagate(errp, local_err); > return; --=20 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 --nmemrqcdn5VTmUEE Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAEBCAAdFiEEdfRlhq5hpmzETofcbDjKyiDZs5IFAl2pN8oACgkQbDjKyiDZ s5ITohAAibSM8EYtmTKgC35Pyt9f6Mbfh2e14LdtWoJO5Gvu0VtluJVbPfSbJzCw XvTHTO3O7S6j0IXZzjiS/QgJttncpPpFm+5+/0uOdPXnaoTltqbP5ZG6wLP2M/Td 0bxe55Kvt0fwSKQiwL5Urfk9vLaefJEdJj0BXq/6i2I98dbal72zy0lE82Jn/+cQ WguTf0pDC/2gU5XTwyNlI7F/quUa6CMFHzzLCGtN4KPlfUt8pAuq1kFx9bXgfEN2 efbXWXjRUo1aBW6rtsJo6G6vtNhwZqla2JgZLya8wLxXeCvlkrh8iP+z1HoHF15N 1X45j4JARvss3ccXNh47dnX3HLv6nWABSQS5as0beO4quyZzUl6o0tKcwLpY2OR4 aE2ONmtWr0u14uPGvBhOoEDQ4UEtaTu0Ioy7Qqk+QjFL68B/8FjolTlumAs1Bu8l wCgnaIxdDzjCd2ME/E/dQG4JjuIGmleFfTfkS38Wc8yzI5CDAOyW+QURhsi/9/iw S40wLynPwZ4lZD6xs8ydnD4dxgHhAfuAE+XRfXqAJ3lhPqe47RrXELT+tjBcxAcD XEVyNIg++z5C1DP/6Fl2hZX2EkYPvWxPmY7gKKYvX07UlfXCwS0dekJu1lIO2iyt JcvTyqC2qQMd7uTyYezjsEf9fI5iTmyUWf7luqriDEmm6zdC+Ys= =ltw3 -----END PGP SIGNATURE----- --nmemrqcdn5VTmUEE--