From mboxrd@z Thu Jan 1 00:00:00 1970 From: Alex =?utf-8?Q?Benn=C3=A9e?= Subject: Re: [PATCH v9 6/6] tests/guest-debug: introduce basic gdbstub tests Date: Tue, 08 Dec 2015 12:02:11 +0000 Message-ID: <87poyhp2rg.fsf@linaro.org> References: <1447345251-22625-1-git-send-email-alex.bennee@linaro.org> <1447345251-22625-7-git-send-email-alex.bennee@linaro.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: In-reply-to: Sender: kvm-owner@vger.kernel.org To: Peter Maydell Cc: QEMU Developers , qemu-arm@nongnu.org, Christoffer Dall , Zhichao Huang , kvm-devel , arm-mail-list , "kvmarm@lists.cs.columbia.edu" , Marc Zyngier List-Id: kvmarm@lists.cs.columbia.edu Peter Maydell writes: > On 12 November 2015 at 16:20, Alex Benn=C3=A9e wrote: >> From: Alex Benn=C3=A9e >> >> The aim of these tests is to combine with an appropriate kernel >> image (with symbol-file vmlinux) and check it behaves as it should. >> Given a kernel it checks: >> >> - single step >> - software breakpoint >> - hardware breakpoint >> - access, read and write watchpoints >> >> On success it returns 0 to the calling process. >> >> I've not plumbed this into the "make check" logic though as we need = a >> solution for providing non-host binaries to the tests. However the t= est >> is structured to work with pretty much any Linux kernel image as it >> uses the basic kernel_init code which is common across architectures= =2E > > Do these tests pass if you run them on the TCG QEMU, just out > of interest? You'll be glad to know they do. > I'm not a great fan of tests that aren't in 'make check' > because IME they just bitrot, but as you say we have no > sensible approach for handling tests that need to run real > guest code :-( I was pondering if a git sub-project with large file support would work= =2E We could add pre-built binaries to the tree with appropriate meta-data (src tree, version, config) to rebuild if required. There would be some degree of trust implied in the original builder though. Maybe a signed commit? > > thanks > -- PMM -- Alex Benn=C3=A9e From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from zen.linaro.local ([81.128.185.34]) by smtp.gmail.com with ESMTPSA id z13sm2652219wjr.47.2015.12.08.04.02.12 (version=TLS1_2 cipher=ECDHE-RSA-AES128-GCM-SHA256 bits=128/128); Tue, 08 Dec 2015 04:02:12 -0800 (PST) Received: from zen.linaro.local (localhost [127.0.0.1]) by zen.linaro.local (Postfix) with ESMTPS id C18753E01DB; Tue, 8 Dec 2015 12:02:11 +0000 (GMT) References: <1447345251-22625-1-git-send-email-alex.bennee@linaro.org> <1447345251-22625-7-git-send-email-alex.bennee@linaro.org> User-agent: mu4e 0.9.15; emacs 24.5.50.4 From: Alex =?utf-8?Q?Benn=C3=A9e?= To: Peter Maydell Cc: QEMU Developers , qemu-arm@nongnu.org, Christoffer Dall , Zhichao Huang , kvm-devel , arm-mail-list , "kvmarm\@lists.cs.columbia.edu" , Marc Zyngier Subject: Re: [PATCH v9 6/6] tests/guest-debug: introduce basic gdbstub tests In-reply-to: Date: Tue, 08 Dec 2015 12:02:11 +0000 Message-ID: <87poyhp2rg.fsf@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-TUID: 0wwi/NtxhCQ7 Peter Maydell writes: > On 12 November 2015 at 16:20, Alex Benn=C3=A9e w= rote: >> From: Alex Benn=C3=A9e >> >> The aim of these tests is to combine with an appropriate kernel >> image (with symbol-file vmlinux) and check it behaves as it should. >> Given a kernel it checks: >> >> - single step >> - software breakpoint >> - hardware breakpoint >> - access, read and write watchpoints >> >> On success it returns 0 to the calling process. >> >> I've not plumbed this into the "make check" logic though as we need a >> solution for providing non-host binaries to the tests. However the test >> is structured to work with pretty much any Linux kernel image as it >> uses the basic kernel_init code which is common across architectures. > > Do these tests pass if you run them on the TCG QEMU, just out > of interest? You'll be glad to know they do. > I'm not a great fan of tests that aren't in 'make check' > because IME they just bitrot, but as you say we have no > sensible approach for handling tests that need to run real > guest code :-( I was pondering if a git sub-project with large file support would work. We could add pre-built binaries to the tree with appropriate meta-data (src tree, version, config) to rebuild if required. There would be some degree of trust implied in the original builder though. Maybe a signed commit? > > thanks > -- PMM -- Alex Benn=C3=A9e From mboxrd@z Thu Jan 1 00:00:00 1970 From: alex.bennee@linaro.org (Alex =?utf-8?Q?Benn=C3=A9e?=) Date: Tue, 08 Dec 2015 12:02:11 +0000 Subject: [PATCH v9 6/6] tests/guest-debug: introduce basic gdbstub tests In-Reply-To: References: <1447345251-22625-1-git-send-email-alex.bennee@linaro.org> <1447345251-22625-7-git-send-email-alex.bennee@linaro.org> Message-ID: <87poyhp2rg.fsf@linaro.org> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Peter Maydell writes: > On 12 November 2015 at 16:20, Alex Benn?e wrote: >> From: Alex Benn?e >> >> The aim of these tests is to combine with an appropriate kernel >> image (with symbol-file vmlinux) and check it behaves as it should. >> Given a kernel it checks: >> >> - single step >> - software breakpoint >> - hardware breakpoint >> - access, read and write watchpoints >> >> On success it returns 0 to the calling process. >> >> I've not plumbed this into the "make check" logic though as we need a >> solution for providing non-host binaries to the tests. However the test >> is structured to work with pretty much any Linux kernel image as it >> uses the basic kernel_init code which is common across architectures. > > Do these tests pass if you run them on the TCG QEMU, just out > of interest? You'll be glad to know they do. > I'm not a great fan of tests that aren't in 'make check' > because IME they just bitrot, but as you say we have no > sensible approach for handling tests that need to run real > guest code :-( I was pondering if a git sub-project with large file support would work. We could add pre-built binaries to the tree with appropriate meta-data (src tree, version, config) to rebuild if required. There would be some degree of trust implied in the original builder though. Maybe a signed commit? > > thanks > -- PMM -- Alex Benn?e From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:44568) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1a6GyH-0005J9-4f for qemu-devel@nongnu.org; Tue, 08 Dec 2015 07:02:18 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1a6GyD-0000Op-Vb for qemu-devel@nongnu.org; Tue, 08 Dec 2015 07:02:17 -0500 Received: from mail-wm0-x22e.google.com ([2a00:1450:400c:c09::22e]:35805) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1a6GyD-0000Ob-Nv for qemu-devel@nongnu.org; Tue, 08 Dec 2015 07:02:13 -0500 Received: by wmuu63 with SMTP id u63so178166122wmu.0 for ; Tue, 08 Dec 2015 04:02:13 -0800 (PST) References: <1447345251-22625-1-git-send-email-alex.bennee@linaro.org> <1447345251-22625-7-git-send-email-alex.bennee@linaro.org> From: Alex =?utf-8?Q?Benn=C3=A9e?= In-reply-to: Date: Tue, 08 Dec 2015 12:02:11 +0000 Message-ID: <87poyhp2rg.fsf@linaro.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v9 6/6] tests/guest-debug: introduce basic gdbstub tests List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell Cc: kvm-devel , Marc Zyngier , QEMU Developers , qemu-arm@nongnu.org, Christoffer Dall , Zhichao Huang , "kvmarm@lists.cs.columbia.edu" , arm-mail-list Peter Maydell writes: > On 12 November 2015 at 16:20, Alex Benn=C3=A9e w= rote: >> From: Alex Benn=C3=A9e >> >> The aim of these tests is to combine with an appropriate kernel >> image (with symbol-file vmlinux) and check it behaves as it should. >> Given a kernel it checks: >> >> - single step >> - software breakpoint >> - hardware breakpoint >> - access, read and write watchpoints >> >> On success it returns 0 to the calling process. >> >> I've not plumbed this into the "make check" logic though as we need a >> solution for providing non-host binaries to the tests. However the test >> is structured to work with pretty much any Linux kernel image as it >> uses the basic kernel_init code which is common across architectures. > > Do these tests pass if you run them on the TCG QEMU, just out > of interest? You'll be glad to know they do. > I'm not a great fan of tests that aren't in 'make check' > because IME they just bitrot, but as you say we have no > sensible approach for handling tests that need to run real > guest code :-( I was pondering if a git sub-project with large file support would work. We could add pre-built binaries to the tree with appropriate meta-data (src tree, version, config) to rebuild if required. There would be some degree of trust implied in the original builder though. Maybe a signed commit? > > thanks > -- PMM -- Alex Benn=C3=A9e