From owner-freebsd-arch@FreeBSD.ORG Sun Mar 22 19:42:51 2015 Return-Path: Delivered-To: freebsd-arch@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:1900:2254:206a::19:1]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by hub.freebsd.org (Postfix) with ESMTPS id 2D68E287; Sun, 22 Mar 2015 19:42:51 +0000 (UTC) Received: from zxy.spb.ru (zxy.spb.ru [195.70.199.98]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id D2CBDC11; Sun, 22 Mar 2015 19:42:50 +0000 (UTC) Received: from slw by zxy.spb.ru with local (Exim 4.84 (FreeBSD)) (envelope-from ) id 1YZllh-000Nss-Jx; Sun, 22 Mar 2015 22:42:41 +0300 Date: Sun, 22 Mar 2015 22:42:41 +0300 From: Slawa Olhovchenkov To: Nathan Whitehorn Subject: Re: RFC: Simplfying hyperthreading distinctions Message-ID: <20150322194241.GA90188@zxy.spb.ru> References: <54FA1180.3080605@astrodoggroup.com> <1526311.uylCbgv5VB@ralph.baldwin.cx> <20150320123823.GA49621@zxy.spb.ru> <550DC564.5020802@freebsd.org> <20150321214336.334eaea5@nonamehost.local> <20150322100744.5b390591@ernst.home> <550EDF87.2070101@astrodoggroup.com> <20150322184349.GA86429@zxy.spb.ru> <550F0E66.6080908@freebsd.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <550F0E66.6080908@freebsd.org> User-Agent: Mutt/1.5.23 (2014-03-12) X-SA-Exim-Connect-IP: X-SA-Exim-Mail-From: slw@zxy.spb.ru X-SA-Exim-Scanned: No (on zxy.spb.ru); SAEximRunCond expanded to false Cc: freebsd-arch@freebsd.org X-BeenThere: freebsd-arch@freebsd.org X-Mailman-Version: 2.1.18-1 Precedence: list List-Id: Discussion related to FreeBSD architecture List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Sun, 22 Mar 2015 19:42:51 -0000 On Sun, Mar 22, 2015 at 11:48:06AM -0700, Nathan Whitehorn wrote: > > > On 03/22/15 11:43, Slawa Olhovchenkov wrote: > > On Sun, Mar 22, 2015 at 10:44:51AM -0700, Adrian Chadd wrote: > > > >> ... or maybe the scheduler shouldn't be the primary source of the CPU > >> topology information, but instead should be a consumer. :) > > And may be good think unified API (userland also) for get full > > topology (with PCIe, USB bus, USB HUB and other devices and IRQ > > handlers). > > > > Currently for detect IRQ handlers of NIC I am need to parse > > hw.intrnames with hacks for maping 't5nex0:0.1' to cxl0 and > > 'ix1:que 2' to ix1. > > > > devinfo doesn't work for you? devinfo also need hacks for map: pci130 t5nex0 Interrupt request lines: 0x120 0x121 0x122 0x123 0x124 0x125 0x126 0x127 0x128 0x129 0x12a 0x12b 0x12c 0x12d 0x12e 0x12f 0x130 0x131 0x132 0x133 0x134 0x135 pcib11 memory window: 0xf9000000-0xf9ffffff 0xfa300000-0xfa37ffff 0xfac04000-0xfac05fff cxl0 cxl1 devinfo don't show detailes about IRQ handelrs: # vmstat -i irq289: t5nex0:evt 17 0 irq290: t5nex0:0.0 235629333426 39932 irq291: t5nex0:0.1 235835938497 39967 irq292: t5nex0:0.2 236414122597 40065 irq293: t5nex0:0.3 234357043791 39716 irq294: t5nex0:0.4 228105465543 38657 irq295: t5nex0:0.5 229347649630 38867 irq296: t5nex0:0.6 230949738393 39139 irq297: t5nex0:0.7 230484724411 39060 irq298: t5nex0:0,0 78781 0 irq299: t5nex0:0,1 90592 0 and not devinfo not vmstat don't provide information about relations between CPU sockets and PCIe slots (cxl0 connect to PCIe handled second socket). Also, on other box: # acpidump -dt /* RSD PTR: OEM=ALASKA, ACPI_Rev=2.0x (2) XSDT=0x7f1e6080, length=36, cksum=95 */ /* XSDT: Length=132, Revision=1, Checksum=106, OEMID=ALASKA, OEM Table ID=A M I, OEM Revision=0x1072009, Creator ID=AMI, Creator Revision=0x10013 Assertion failed: (addr != 0), function acpi_print_rsdt, file /usr/src/usr.sbin/acpi/acpidump/acpi.c, line 1097. Entries={ 0x7f1efe90, 0x7f1eff88, 0x7f1f0098, 0x7f1f00d8, 0x7f1f0448, 0x7f1f0478, 0x7f1f04b0, 0x7f1f0570, 0x7f1f05b0, 0x7f2606b8, 0x7f260708, Abort (core dumped) And talk 'userland' I talk about API, from program, library call. Utility for shell need too.