Re: [Qemu-devel] [PATCH v4 0/8] AHCI test suite framework

["Michael S. Tsirkin" <mst@redhat.com>]

On Tue, Sep 09, 2014 at 01:16:41PM -0400, John Snow wrote:
> 
> 
> On 08/21/2014 01:44 PM, John Snow wrote:
> >This submission does not re-send earlier patches in the series which
> >have already been merged into QEMU, which were primarily staging
> >fixes and small tweaks to support this smaller set of patches.
> >
> >This patch series introduces a number of small fixes and tweaks to
> >help support an AHCI test suite that in the future I hope to expand
> >to a fuller regression suite to help guide the development of the
> >AHCI device support under, in particular, the Q35 machine type in QEMU.
> >
> >Paolo Bonzini has contributed a number of cleanup and refactoring patches
> >that support changes to the PIO setup FIS packet construction code, which
> >is necessary for testing ths specification adherence of the IDENTIFY command,
> >which issues its data exclusively via PIO mechanisms.
> >
> >The ahci-test code being checked in represents a minimum of functionality
> >needed in order to issue and receive commands from the AHCI HBA under the
> >libqos / qtest environment.
> >
> >In V3+, there is one assertion for incorrect behavior that is expected to
> >be fixed in a future patch set, where the Descriptor Processed interrupt
> >is not set or posted within the identify test.
> >
> >V4:
> >     - Endianness fixes, primarily in ahci_test_identify, tested on ppc64
> >     - Style issues (define and structure names)
> >     - Reshuffled barsize changes into later, more relevant patches
> >     - Some additional comments for clarification.
> >
> >V3:
> >     - All tests complete successfully.
> >     - Fixed the PCI capabilities ordering for AHCI.
> >     - Corrected the boot-time values of the PxTFD and PxSIG registers
> >     - Corrected PxTFD register being updated prior to PxCMD.FRE being set
> >
> >V2:
> >"ide-test: add test for werror=stop"
> >     - changed filename variable name
> >     - altered the QMP event polling to avoid sleep
> >     - debug_path file cleanup on exit.
> >"ide: stop PIO transfer on errors"
> >     - Modified logic to be unconditional.
> >"ahci: construct PIO Setup FIS for PIO commands"
> >     - Added in dma_memory_map success checks and error pathways.
> >"libqtest: Correct small memory leak."
> >     - Corrected raw usage of free()
> >"ahci: Adding basic functionality qtest."
> >     - Removed HBA type.
> >     - Corrected cleanup order.
> >     - Corrected raw usage of free()
> >     - Removed needless conditional around g_free()
> >     - Altered ahci_boot to return by value instead of parameter.
> >"ahci: Add test_pci_spec to ahci-test."
> >     - Removed all ifdef logic in favor of runtime tuning. (--pedantic)
> >     - Removed all warnings, all infractions are now hard assertions.
> >     - Replaced g_assert with g_assert_cmphex in many cases
> >"ahci: add test_pci_enable to ahci-test."
> >     - Removed MSI codepaths, as it was incomplete and unused.
> >"ahci: Add test_hba_spec to ahci-test."
> >     - Removed unneeded macros
> >     - Added in an optional bar_size return parameter to qpci_iomap
> >"ahci: Add test_identify case to ahci-test."
> >     - Corrected raw usage of free()
> >
> >For convenience; https://github.com/jnsnow/qemu/tree/ahci-test-v4
> >
> >John Snow (8):
> >   ahci: Adding basic functionality qtest.
> >   ahci: MSI capability should be at 0x80, not 0x50.
> >   ahci: Add test_pci_spec to ahci-test.
> >   ahci: add test_pci_enable to ahci-test.
> >   ahci: properly shadow the TFD register
> >   ahci: Add test_hba_spec to ahci-test.
> >   ahci: Add test_hba_enable to ahci-test.
> >   ahci: Add test_identify case to ahci-test.
> >
> >  hw/ide/ahci.c      |   42 +-
> >  hw/ide/ich.c       |    7 +-
> >  tests/Makefile     |    2 +
> >  tests/ahci-test.c  | 1561 ++++++++++++++++++++++++++++++++++++++++++++++++++++
> >  tests/libqos/pci.c |    6 +
> >  5 files changed, 1603 insertions(+), 15 deletions(-)
> >  create mode 100644 tests/ahci-test.c
> >
> 
> 
> Ping?

Stefan?