[Qemu-devel] [PULL 00/10] SCSI patches for 2012-08-03

[Qemu-devel] [PULL 00/10] SCSI patches for 2012-08-03

[Paolo Bonzini]

Anthony,

The following changes since commit 5e3bc7144edd6e4fa2824944e5eb16c28197dd5a:

  Merge remote-tracking branch 'mst/tags/for_anthony' into staging (2012-07-30 10:00:48 -0500)

are available in the git repository at:


  git://github.com/bonzini/qemu.git scsi-next

for you to fetch changes up to b71760ef6180662cc2dff1f6cf673a53508c59f1:

  scsi: add support for ATA_PASSTHROUGH_xx scsi command (2012-08-03 10:04:37 +0200)

The main change here is re-enabling megasas, but there's also Herve's
retro HBA and a couple of random fixes.

----------------------------------------------------------------
Cong Meng (1):
      scsi: add support for ATA_PASSTHROUGH_xx scsi command

Hannes Reinecke (1):
      megasas: static SAS addresses

Hervé Poussineau (2):
      esp: add missing const on TypeInfo structures
      esp: add Tekram DC-390 emulation (PC SCSI adapter)

Paolo Bonzini (3):
      scsi-disk: fix compilation with DEBUG_SCSI
      Revert "megasas: disable due to build breakage"
      esp: enable for all PCI machines

Ronnie Sahlberg (2):
      SCSI: Update the sense code for PREVENT REMOVAL errors
      SCSI: STARTSTOPUNIT only eject/load media if powercondition is 0

Stefan Weil (1):
      megasas: Update function megasys_scsi_uninit

 default-configs/i386-softmmu.mak |    1 -
 default-configs/pci.mak          |    2 +
 hw/esp.c                         |  130 +++++++++++++++++++++++++++++++++++++-
 hw/megasas.c                     |   68 ++++++++++++++------
 hw/mfi.h                         |    1 +
 hw/scsi-bus.c                    |   92 +++++++++++++++++++++++++--
 hw/scsi-defs.h                   |    4 +-
 hw/scsi-disk.c                   |   29 +++++----
 8 files changed, 286 insertions(+), 41 deletions(-)
-- 
1.7.10.4

[Qemu-devel] [PATCH 01/10] SCSI: Update the sense code for PREVENT REMOVAL errors

[Paolo Bonzini]

From: Ronnie Sahlberg <ronniesahlberg@gmail.com>

Change the sense codes for failures to eject a device that is locked
by PREVENT_ALLOW_MEDIUM_REMOVAL from
the generic MEDIA_LOAD_OR_EJECT_FAILED to the more specific
MEDIUM_REMOVAL_PREVENTED.

The second sense code is more accurate, and is also listed in MMC annex F
for the recommended sense codes for MMC devices while the first sense code is not.

Signed-off-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/scsi-bus.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/hw/scsi-bus.c b/hw/scsi-bus.c
index e4ec19e..68049f6 100644
--- a/hw/scsi-bus.c
+++ b/hw/scsi-bus.c
@@ -1103,7 +1103,7 @@ const struct SCSISense sense_code_NO_MEDIUM = {
 
 /* LUN not ready, medium removal prevented */
 const struct SCSISense sense_code_NOT_READY_REMOVAL_PREVENTED = {
-    .key = NOT_READY, .asc = 0x53, .ascq = 0x00
+    .key = NOT_READY, .asc = 0x53, .ascq = 0x02
 };
 
 /* Hardware error, internal target failure */
@@ -1153,7 +1153,7 @@ const struct SCSISense sense_code_INCOMPATIBLE_FORMAT = {
 
 /* Illegal request, medium removal prevented */
 const struct SCSISense sense_code_ILLEGAL_REQ_REMOVAL_PREVENTED = {
-    .key = ILLEGAL_REQUEST, .asc = 0x53, .ascq = 0x00
+    .key = ILLEGAL_REQUEST, .asc = 0x53, .ascq = 0x02
 };
 
 /* Command aborted, I/O process terminated */
-- 
1.7.10.4

[Qemu-devel] [PATCH 02/10] SCSI: STARTSTOPUNIT only eject/load media if powercondition is 0

[Paolo Bonzini]

From: Ronnie Sahlberg <ronniesahlberg@gmail.com>

The START STOP UNIT command will only eject/load media if
power condition is zero.

If power condition is !0 then LOEJ and START will be ignored.

>From MMC (sbc contains similar wordings too)
  The Power Conditions field requests the block device to be placed
  in the power condition defined in
  Table 558. If this field has a value other than 0h then the Start
  and LoEj bits shall be ignored.

Signed-off-by: Ronnie Sahlberg <ronniesahlberg@gmail.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/scsi-disk.c |    6 ++++++
 1 file changed, 6 insertions(+)

diff --git a/hw/scsi-disk.c b/hw/scsi-disk.c
index 84b63ff..e2ec177 100644
--- a/hw/scsi-disk.c
+++ b/hw/scsi-disk.c
@@ -1247,6 +1247,12 @@ static int scsi_disk_emulate_start_stop(SCSIDiskReq *r)
     SCSIDiskState *s = DO_UPCAST(SCSIDiskState, qdev, req->dev);
     bool start = req->cmd.buf[4] & 1;
     bool loej = req->cmd.buf[4] & 2; /* load on start, eject on !start */
+    int pwrcnd = req->cmd.buf[4] & 0xf0;
+
+    if (pwrcnd) {
+        /* eject/load only happens for power condition == 0 */
+        return 0;
+    }
 
     if ((s->features & (1 << SCSI_DISK_F_REMOVABLE)) && loej) {
         if (!start && !s->tray_open && s->tray_locked) {
-- 
1.7.10.4

[Qemu-devel] [PATCH 03/10] megasas: Update function megasys_scsi_uninit

[Paolo Bonzini]

From: Stefan Weil <sw@weilnetz.de>

Commit f90c2bcdbc69e41e575f868b984c3e2de8f51bac changed
PCIUnregisterFunc, therefore the function prototype
needs an update.

megasas.o is currently not linked, so this bug was not
detected by the buildbots.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/megasas.c |    3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

diff --git a/hw/megasas.c b/hw/megasas.c
index 9a0eab1..8a4960f 100644
--- a/hw/megasas.c
+++ b/hw/megasas.c
@@ -2040,7 +2040,7 @@ static const VMStateDescription vmstate_megasas = {
     }
 };
 
-static int megasas_scsi_uninit(PCIDevice *d)
+static void megasas_scsi_uninit(PCIDevice *d)
 {
     MegasasState *s = DO_UPCAST(MegasasState, dev, d);
 
@@ -2050,7 +2050,6 @@ static int megasas_scsi_uninit(PCIDevice *d)
     memory_region_destroy(&s->mmio_io);
     memory_region_destroy(&s->port_io);
     memory_region_destroy(&s->queue_io);
-    return 0;
 }
 
 static const struct SCSIBusInfo megasas_scsi_info = {
-- 
1.7.10.4

[Qemu-devel] [PATCH 04/10] scsi-disk: fix compilation with DEBUG_SCSI

[Paolo Bonzini]

Reported-by: Gerhard Wiesinger <lists@wiesinger.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/scsi-disk.c |   23 ++++++++++++-----------
 1 file changed, 12 insertions(+), 11 deletions(-)

diff --git a/hw/scsi-disk.c b/hw/scsi-disk.c
index e2ec177..a9c7279 100644
--- a/hw/scsi-disk.c
+++ b/hw/scsi-disk.c
@@ -447,7 +447,7 @@ static void scsi_write_complete(void * opaque, int ret)
         return;
     } else {
         scsi_init_iovec(r, SCSI_DMA_BUF_SIZE);
-        DPRINTF("Write complete tag=0x%x more=%d\n", r->req.tag, r->qiov.size);
+        DPRINTF("Write complete tag=0x%x more=%zd\n", r->req.tag, r->qiov.size);
         scsi_req_data(&r->req, r->qiov.size);
     }
 
@@ -1277,7 +1277,7 @@ static void scsi_disk_emulate_read_data(SCSIRequest *req)
     int buflen = r->iov.iov_len;
 
     if (buflen) {
-        DPRINTF("Read buf_len=%zd\n", buflen);
+        DPRINTF("Read buf_len=%d\n", buflen);
         r->iov.iov_len = 0;
         r->started = true;
         scsi_req_data(&r->req, buflen);
@@ -1455,7 +1455,7 @@ static void scsi_disk_emulate_write_data(SCSIRequest *req)
 
     if (r->iov.iov_len) {
         int buflen = r->iov.iov_len;
-        DPRINTF("Write buf_len=%zd\n", buflen);
+        DPRINTF("Write buf_len=%d\n", buflen);
         r->iov.iov_len = 0;
         scsi_req_data(&r->req, buflen);
         return;
@@ -2093,23 +2093,24 @@ static SCSIRequest *scsi_new_request(SCSIDevice *d, uint32_t tag, uint32_t lun,
     const SCSIReqOps *ops;
     uint8_t command;
 
+    command = buf[0];
+    ops = scsi_disk_reqops_dispatch[command];
+    if (!ops) {
+        ops = &scsi_disk_emulate_reqops;
+    }
+    req = scsi_req_alloc(ops, &s->qdev, tag, lun, hba_private);
+
 #ifdef DEBUG_SCSI
-    DPRINTF("Command: lun=%d tag=0x%x data=0x%02x", lun, buf[0]);
+    DPRINTF("Command: lun=%d tag=0x%x data=0x%02x", lun, tag, buf[0]);
     {
         int i;
-        for (i = 1; i < r->req.cmd.len; i++) {
+        for (i = 1; i < req->cmd.len; i++) {
             printf(" 0x%02x", buf[i]);
         }
         printf("\n");
     }
 #endif
 
-    command = buf[0];
-    ops = scsi_disk_reqops_dispatch[command];
-    if (!ops) {
-        ops = &scsi_disk_emulate_reqops;
-    }
-    req = scsi_req_alloc(ops, &s->qdev, tag, lun, hba_private);
     return req;
 }
 
-- 
1.7.10.4

[Qemu-devel] [PATCH 05/10] megasas: static SAS addresses

[Paolo Bonzini]

From: Hannes Reinecke <hare@suse.de>

This patch introduces a new property 'sas_address' which
allows the user to specify the SAS address for the HBA.
The default address is following the NAA locally assigned
identifier format with the locally assigned address
0x525400 as used eg for the MAC addresses.
The lower bytes are set to the pci address which
will ensure uniqueness for the local machine.

The port addresses are now calculated based on the magic
number 0x1221 (which is found in real hardware, too) plus
the device number.

Signed-off-by: Hannes Reinecke <hare@suse.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Andreas Faerber <afaerber@suse.de>
Cc: Anthony Liguori <anthony@codemonkey.ws>
Cc: Alexander Graf <agraf@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/megasas.c |   65 +++++++++++++++++++++++++++++++++++++++++++---------------
 hw/mfi.h     |    1 +
 2 files changed, 49 insertions(+), 17 deletions(-)

diff --git a/hw/megasas.c b/hw/megasas.c
index 8a4960f..c35a15d 100644
--- a/hw/megasas.c
+++ b/hw/megasas.c
@@ -38,6 +38,9 @@
 #define MEGASAS_MAX_SECTORS 0xFFFF      /* No real limit */
 #define MEGASAS_MAX_ARRAYS 128
 
+#define NAA_LOCALLY_ASSIGNED_ID 0x3ULL
+#define IEEE_COMPANY_LOCALLY_ASSIGNED 0x525400
+
 #define MEGASAS_FLAG_USE_JBOD      0
 #define MEGASAS_MASK_USE_JBOD      (1 << MEGASAS_FLAG_USE_JBOD)
 #define MEGASAS_FLAG_USE_MSIX      1
@@ -89,6 +92,8 @@ typedef struct MegasasState {
     int shutdown_event;
     int boot_event;
 
+    uint64_t sas_addr;
+
     uint64_t reply_queue_pa;
     void *reply_queue;
     int reply_queue_len;
@@ -372,14 +377,16 @@ static uint64_t megasas_fw_time(void)
     return bcd_time;
 }
 
-static uint64_t megasas_gen_sas_addr(uint64_t id)
+/*
+ * Default disk sata address
+ * 0x1221 is the magic number as
+ * present in real hardware,
+ * so use it here, too.
+ */
+static uint64_t megasas_get_sata_addr(uint16_t id)
 {
-    uint64_t addr;
-
-    addr = 0x5001a4aULL << 36;
-    addr |= id & 0xfffffffff;
-
-    return addr;
+    uint64_t addr = (0x1221ULL << 48);
+    return addr & (id << 24);
 }
 
 /*
@@ -652,10 +659,7 @@ static int megasas_ctrl_get_info(MegasasState *s, MegasasCmd *cmd)
     size_t dcmd_size = sizeof(info);
     BusChild *kid;
     int num_ld_disks = 0;
-
-    QTAILQ_FOREACH(kid, &s->bus.qbus.children, sibling) {
-        num_ld_disks++;
-    }
+    uint16_t sdev_id;
 
     memset(&info, 0x0, cmd->iov_size);
     if (cmd->iov_size < dcmd_size) {
@@ -669,10 +673,29 @@ static int megasas_ctrl_get_info(MegasasState *s, MegasasCmd *cmd)
     info.pci.subvendor = cpu_to_le16(PCI_VENDOR_ID_LSI_LOGIC);
     info.pci.subdevice = cpu_to_le16(0x1013);
 
-    info.host.type = MFI_INFO_HOST_PCIX;
+    /*
+     * For some reason the firmware supports
+     * only up to 8 device ports.
+     * Despite supporting a far larger number
+     * of devices for the physical devices.
+     * So just display the first 8 devices
+     * in the device port list, independent
+     * of how many logical devices are actually
+     * present.
+     */
+    info.host.type = MFI_INFO_HOST_PCIE;
     info.device.type = MFI_INFO_DEV_SAS3G;
-    info.device.port_count = 2;
-    info.device.port_addr[0] = cpu_to_le64(megasas_gen_sas_addr((uint64_t)s));
+    info.device.port_count = 8;
+    QTAILQ_FOREACH(kid, &s->bus.qbus.children, sibling) {
+        SCSIDevice *sdev = DO_UPCAST(SCSIDevice, qdev, kid->child);
+
+        if (num_ld_disks < 8) {
+            sdev_id = ((sdev->id & 0xFF) >> 8) | (sdev->lun & 0xFF);
+            info.device.port_addr[num_ld_disks] =
+                cpu_to_le64(megasas_get_sata_addr(sdev_id));
+        }
+        num_ld_disks++;
+    }
 
     memcpy(info.product_name, "MegaRAID SAS 8708EM2", 20);
     snprintf(info.serial_number, 32, "QEMU%08lx",
@@ -761,7 +784,7 @@ static int megasas_mfc_get_defaults(MegasasState *s, MegasasCmd *cmd)
         return MFI_STAT_INVALID_PARAMETER;
     }
 
-    info.sas_addr = cpu_to_le64(megasas_gen_sas_addr((uint64_t)s));
+    info.sas_addr = cpu_to_le64(s->sas_addr);
     info.stripe_size = 3;
     info.flush_time = 4;
     info.background_rate = 30;
@@ -891,7 +914,7 @@ static int megasas_dcmd_pd_get_list(MegasasState *s, MegasasCmd *cmd)
         info.addr[num_pd_disks].scsi_dev_type = sdev->type;
         info.addr[num_pd_disks].connect_port_bitmap = 0x1;
         info.addr[num_pd_disks].sas_addr[0] =
-            cpu_to_le64(megasas_gen_sas_addr((uint64_t)sdev));
+            cpu_to_le64(megasas_get_sata_addr(sdev_id));
         num_pd_disks++;
         offset += sizeof(struct mfi_pd_address);
     }
@@ -994,7 +1017,7 @@ static int megasas_pd_get_info_submit(SCSIDevice *sdev, int lun,
     info->slot_number = (sdev->id & 0xFF);
     info->path_info.count = 1;
     info->path_info.sas_addr[0] =
-        cpu_to_le64(megasas_gen_sas_addr((uint64_t)sdev));
+        cpu_to_le64(megasas_get_sata_addr(sdev_id));
     info->connected_port_bitmap = 0x1;
     info->device_speed = 1;
     info->link_speed = 1;
@@ -2102,6 +2125,13 @@ static int megasas_scsi_init(PCIDevice *dev)
         msix_vector_use(&s->dev, 0);
     }
 
+    if (!s->sas_addr) {
+        s->sas_addr = ((NAA_LOCALLY_ASSIGNED_ID << 24) |
+                       IEEE_COMPANY_LOCALLY_ASSIGNED) << 36;
+        s->sas_addr |= (pci_bus_num(dev->bus) << 16);
+        s->sas_addr |= (PCI_SLOT(dev->devfn) << 8);
+        s->sas_addr |= PCI_FUNC(dev->devfn);
+    }
     if (s->fw_sge >= MEGASAS_MAX_SGE - MFI_PASS_FRAME_SIZE) {
         s->fw_sge = MEGASAS_MAX_SGE - MFI_PASS_FRAME_SIZE;
     } else if (s->fw_sge >= 128 - MFI_PASS_FRAME_SIZE) {
@@ -2136,6 +2166,7 @@ static Property megasas_properties[] = {
                        MEGASAS_DEFAULT_SGE),
     DEFINE_PROP_UINT32("max_cmds", MegasasState, fw_cmds,
                        MEGASAS_DEFAULT_FRAMES),
+    DEFINE_PROP_HEX64("sas_address", MegasasState, sas_addr, 0),
 #ifdef USE_MSIX
     DEFINE_PROP_BIT("use_msix", MegasasState, flags,
                     MEGASAS_FLAG_USE_MSIX, false),
diff --git a/hw/mfi.h b/hw/mfi.h
index 3045d4e..436b690 100644
--- a/hw/mfi.h
+++ b/hw/mfi.h
@@ -656,6 +656,7 @@ struct mfi_info_device {
 #define MFI_INFO_DEV_SAS3G      0x02
 #define MFI_INFO_DEV_SATA1      0x04
 #define MFI_INFO_DEV_SATA3G     0x08
+#define MFI_INFO_DEV_PCIE       0x10
     uint8_t reserved[6];
     uint8_t port_count;
     uint64_t port_addr[8];
-- 
1.7.10.4

[Qemu-devel] [PATCH 06/10] Revert "megasas: disable due to build breakage"

[Paolo Bonzini]

This reverts commit 92336855975805d88c7979f53bc05c2d47abab04.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 default-configs/pci.mak |    1 +
 1 file changed, 1 insertion(+)

diff --git a/default-configs/pci.mak b/default-configs/pci.mak
index 9d3e1db..4b49c00 100644
--- a/default-configs/pci.mak
+++ b/default-configs/pci.mak
@@ -10,6 +10,7 @@ CONFIG_EEPRO100_PCI=y
 CONFIG_PCNET_PCI=y
 CONFIG_PCNET_COMMON=y
 CONFIG_LSI_SCSI_PCI=y
+CONFIG_MEGASAS_SCSI_PCI=y
 CONFIG_RTL8139_PCI=y
 CONFIG_E1000_PCI=y
 CONFIG_IDE_CORE=y
-- 
1.7.10.4

[Qemu-devel] [PATCH 07/10] esp: enable for all PCI machines

[Paolo Bonzini]

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 default-configs/i386-softmmu.mak |    1 -
 default-configs/pci.mak          |    1 +
 2 files changed, 1 insertion(+), 1 deletion(-)

diff --git a/default-configs/i386-softmmu.mak b/default-configs/i386-softmmu.mak
index fee8cde..2c78175 100644
--- a/default-configs/i386-softmmu.mak
+++ b/default-configs/i386-softmmu.mak
@@ -25,4 +25,3 @@ CONFIG_HPET=y
 CONFIG_APPLESMC=y
 CONFIG_I8259=y
 CONFIG_PFLASH_CFI01=y
-CONFIG_ESP=y
diff --git a/default-configs/pci.mak b/default-configs/pci.mak
index 4b49c00..9febb47 100644
--- a/default-configs/pci.mak
+++ b/default-configs/pci.mak
@@ -17,3 +17,4 @@ CONFIG_IDE_CORE=y
 CONFIG_IDE_QDEV=y
 CONFIG_IDE_PCI=y
 CONFIG_AHCI=y
+CONFIG_ESP=y
-- 
1.7.10.4

[Qemu-devel] [PATCH 08/10] esp: add missing const on TypeInfo structures

[Paolo Bonzini]

From: Hervé Poussineau <hpoussin@reactos.org>

Signed-off-by: Hervé Poussineau <hpoussin@reactos.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/esp.c |    4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/hw/esp.c b/hw/esp.c
index a011347..77f5707 100644
--- a/hw/esp.c
+++ b/hw/esp.c
@@ -816,7 +816,7 @@ static void sysbus_esp_class_init(ObjectClass *klass, void *data)
     dc->vmsd = &vmstate_sysbus_esp_scsi;
 }
 
-static TypeInfo sysbus_esp_info = {
+static const TypeInfo sysbus_esp_info = {
     .name          = "esp",
     .parent        = TYPE_SYS_BUS_DEVICE,
     .instance_size = sizeof(SysBusESPState),
@@ -1176,7 +1176,7 @@ static void esp_pci_class_init(ObjectClass *klass, void *data)
     dc->vmsd = &vmstate_esp_pci_scsi;
 }
 
-static TypeInfo esp_pci_info = {
+static const TypeInfo esp_pci_info = {
     .name = "am53c974",
     .parent = TYPE_PCI_DEVICE,
     .instance_size = sizeof(PCIESPState),
-- 
1.7.10.4

[Qemu-devel] [PATCH 09/10] esp: add Tekram DC-390 emulation (PC SCSI adapter)

[Paolo Bonzini]

From: Hervé Poussineau <hpoussin@reactos.org>

Difference with AMD PCscsi is that DC-390 contains a EEPROM.

This has been successfully tested on:
- MS DOS 6.22 (using DC390 ASPI driver)
- MS Windows 98 SE (using DC390 driver)
- MS Windows NT 3.1 (using DC390 driver)
- MS Windows NT 4.0 (using DC390 driver)
- hard disk and cdrom boot

Signed-off-by: Hervé Poussineau <hpoussin@reactos.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/esp.c |  126 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++-
 1 file changed, 125 insertions(+), 1 deletion(-)

diff --git a/hw/esp.c b/hw/esp.c
index 77f5707..0bdc8be 100644
--- a/hw/esp.c
+++ b/hw/esp.c
@@ -26,6 +26,7 @@
 #include "sysbus.h"
 #include "pci.h"
 #include "scsi.h"
+#include "eeprom93xx.h"
 #include "esp.h"
 #include "trace.h"
 #include "qemu-log.h"
@@ -823,6 +824,9 @@ static const TypeInfo sysbus_esp_info = {
     .class_init    = sysbus_esp_class_init,
 };
 
+
+#define TYPE_AM53C974_DEVICE "am53c974"
+
 #define DMA_CMD   0x0
 #define DMA_STC   0x1
 #define DMA_SPA   0x2
@@ -1177,16 +1181,136 @@ static void esp_pci_class_init(ObjectClass *klass, void *data)
 }
 
 static const TypeInfo esp_pci_info = {
-    .name = "am53c974",
+    .name = TYPE_AM53C974_DEVICE,
     .parent = TYPE_PCI_DEVICE,
     .instance_size = sizeof(PCIESPState),
     .class_init = esp_pci_class_init,
 };
 
+typedef struct {
+    PCIESPState pci;
+    eeprom_t *eeprom;
+} DC390State;
+
+#define TYPE_DC390_DEVICE "dc390"
+#define DC390(obj) \
+    OBJECT_CHECK(DC390State, obj, TYPE_DC390_DEVICE)
+
+#define EE_ADAPT_SCSI_ID 64
+#define EE_MODE2         65
+#define EE_DELAY         66
+#define EE_TAG_CMD_NUM   67
+#define EE_ADAPT_OPTIONS 68
+#define EE_BOOT_SCSI_ID  69
+#define EE_BOOT_SCSI_LUN 70
+#define EE_CHKSUM1       126
+#define EE_CHKSUM2       127
+
+#define EE_ADAPT_OPTION_F6_F8_AT_BOOT   0x01
+#define EE_ADAPT_OPTION_BOOT_FROM_CDROM 0x02
+#define EE_ADAPT_OPTION_INT13           0x04
+#define EE_ADAPT_OPTION_SCAM_SUPPORT    0x08
+
+
+static uint32_t dc390_read_config(PCIDevice *dev, uint32_t addr, int l)
+{
+    DC390State *pci = DC390(dev);
+    uint32_t val;
+
+    val = pci_default_read_config(dev, addr, l);
+
+    if (addr == 0x00 && l == 1) {
+        /* First byte of address space is AND-ed with EEPROM DO line */
+        if (!eeprom93xx_read(pci->eeprom)) {
+            val &= ~0xff;
+        }
+    }
+
+    return val;
+}
+
+static void dc390_write_config(PCIDevice *dev,
+                               uint32_t addr, uint32_t val, int l)
+{
+    DC390State *pci = DC390(dev);
+    if (addr == 0x80) {
+        /* EEPROM write */
+        int eesk = val & 0x80 ? 1 : 0;
+        int eedi = val & 0x40 ? 1 : 0;
+        eeprom93xx_write(pci->eeprom, 1, eesk, eedi);
+    } else if (addr == 0xc0) {
+        /* EEPROM CS low */
+        eeprom93xx_write(pci->eeprom, 0, 0, 0);
+    } else {
+        pci_default_write_config(dev, addr, val, l);
+    }
+}
+
+static int dc390_scsi_init(PCIDevice *dev)
+{
+    DC390State *pci = DC390(dev);
+    uint8_t *contents;
+    uint16_t chksum = 0;
+    int i, ret;
+
+    /* init base class */
+    ret = esp_pci_scsi_init(dev);
+    if (ret < 0) {
+        return ret;
+    }
+
+    /* EEPROM */
+    pci->eeprom = eeprom93xx_new(DEVICE(dev), 64);
+
+    /* set default eeprom values */
+    contents = (uint8_t *)eeprom93xx_data(pci->eeprom);
+
+    for (i = 0; i < 16; i++) {
+        contents[i * 2] = 0x57;
+        contents[i * 2 + 1] = 0x00;
+    }
+    contents[EE_ADAPT_SCSI_ID] = 7;
+    contents[EE_MODE2] = 0x0f;
+    contents[EE_TAG_CMD_NUM] = 0x04;
+    contents[EE_ADAPT_OPTIONS] = EE_ADAPT_OPTION_F6_F8_AT_BOOT
+                               | EE_ADAPT_OPTION_BOOT_FROM_CDROM
+                               | EE_ADAPT_OPTION_INT13;
+
+    /* update eeprom checksum */
+    for (i = 0; i < EE_CHKSUM1; i += 2) {
+        chksum += contents[i] + (((uint16_t)contents[i + 1]) << 8);
+    }
+    chksum = 0x1234 - chksum;
+    contents[EE_CHKSUM1] = chksum & 0xff;
+    contents[EE_CHKSUM2] = chksum >> 8;
+
+    return 0;
+}
+
+static void dc390_class_init(ObjectClass *klass, void *data)
+{
+    DeviceClass *dc = DEVICE_CLASS(klass);
+    PCIDeviceClass *k = PCI_DEVICE_CLASS(klass);
+
+    k->init = dc390_scsi_init;
+    k->config_read = dc390_read_config;
+    k->config_write = dc390_write_config;
+    dc->desc = "Tekram DC-390 SCSI adapter";
+}
+
+static const TypeInfo dc390_info = {
+    .name = "dc390",
+    .parent = TYPE_AM53C974_DEVICE,
+    .instance_size = sizeof(DC390State),
+    .class_init = dc390_class_init,
+};
+
+
 static void esp_register_types(void)
 {
     type_register_static(&sysbus_esp_info);
     type_register_static(&esp_pci_info);
+    type_register_static(&dc390_info);
 }
 
 type_init(esp_register_types)
-- 
1.7.10.4

[Qemu-devel] [PATCH 10/10] scsi: add support for ATA_PASSTHROUGH_xx scsi command

[Paolo Bonzini]

From: Cong Meng <mc@linux.vnet.ibm.com>

Correct the command names of opcode 0x85 and 0xa1, and calculate
their xfer size from CDB.

Signed-off-by: Cong Meng <mc@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
 hw/scsi-bus.c  |   88 ++++++++++++++++++++++++++++++++++++++++++++++++++++++--
 hw/scsi-defs.h |    4 +--
 2 files changed, 87 insertions(+), 5 deletions(-)

diff --git a/hw/scsi-bus.c b/hw/scsi-bus.c
index 68049f6..6120cc8 100644
--- a/hw/scsi-bus.c
+++ b/hw/scsi-bus.c
@@ -733,6 +733,72 @@ static int scsi_get_performance_length(int num_desc, int type, int data_type)
     }
 }
 
+static int ata_passthrough_xfer_unit(SCSIDevice *dev, uint8_t *buf)
+{
+    int byte_block = (buf[2] >> 2) & 0x1;
+    int type = (buf[2] >> 4) & 0x1;
+    int xfer_unit;
+
+    if (byte_block) {
+        if (type) {
+            xfer_unit = dev->blocksize;
+        } else {
+            xfer_unit = 512;
+        }
+    } else {
+        xfer_unit = 1;
+    }
+
+    return xfer_unit;
+}
+
+static int ata_passthrough_12_xfer_size(SCSIDevice *dev, uint8_t *buf)
+{
+    int length = buf[2] & 0x3;
+    int xfer;
+    int unit = ata_passthrough_xfer_unit(dev, buf);
+
+    switch (length) {
+    case 0:
+    case 3: /* USB-specific.  */
+        xfer = 0;
+        break;
+    case 1:
+        xfer = buf[3];
+        break;
+    case 2:
+        xfer = buf[4];
+        break;
+    }
+
+    return xfer * unit;
+}
+
+static int ata_passthrough_16_xfer_size(SCSIDevice *dev, uint8_t *buf)
+{
+    int extend = buf[1] & 0x1;
+    int length = buf[2] & 0x3;
+    int xfer;
+    int unit = ata_passthrough_xfer_unit(dev, buf);
+
+    switch (length) {
+    case 0:
+    case 3: /* USB-specific.  */
+        xfer = 0;
+        break;
+    case 1:
+        xfer = buf[4];
+        xfer |= (extend ? buf[3] << 8 : 0);
+        break;
+    case 2:
+        xfer = buf[6];
+        xfer |= (extend ? buf[5] << 8 : 0);
+        break;
+    }
+
+    return xfer * unit;
+}
+
 static int scsi_req_length(SCSICommand *cmd, SCSIDevice *dev, uint8_t *buf)
 {
     switch (buf[0] >> 5) {
@@ -867,6 +933,17 @@ static int scsi_req_length(SCSICommand *cmd, SCSIDevice *dev, uint8_t *buf)
             cmd->xfer = buf[9] | (buf[8] << 8);
         }
         break;
+    case ATA_PASSTHROUGH_12:
+        if (dev->type == TYPE_ROM) {
+            /* BLANK command of MMC */
+            cmd->xfer = 0;
+        } else {
+            cmd->xfer = ata_passthrough_12_xfer_size(dev, buf);
+        }
+        break;
+    case ATA_PASSTHROUGH_16:
+        cmd->xfer = ata_passthrough_16_xfer_size(dev, buf);
+        break;
     }
     return 0;
 }
@@ -996,9 +1073,14 @@ static void scsi_cmd_xfer_mode(SCSICommand *cmd)
     case SEND_DVD_STRUCTURE:
     case PERSISTENT_RESERVE_OUT:
     case MAINTENANCE_OUT:
-    case ATA_PASSTHROUGH:
         cmd->mode = SCSI_XFER_TO_DEV;
         break;
+    case ATA_PASSTHROUGH_12:
+    case ATA_PASSTHROUGH_16:
+        /* T_DIR */
+        cmd->mode = (cmd->buf[2] & 0x8) ?
+                   SCSI_XFER_FROM_DEV : SCSI_XFER_TO_DEV;
+        break;
     default:
         cmd->mode = SCSI_XFER_FROM_DEV;
         break;
@@ -1335,7 +1417,7 @@ static const char *scsi_command_name(uint8_t cmd)
         [ PERSISTENT_RESERVE_OUT   ] = "PERSISTENT_RESERVE_OUT",
         [ WRITE_FILEMARKS_16       ] = "WRITE_FILEMARKS_16",
         [ EXTENDED_COPY            ] = "EXTENDED_COPY",
-        [ ATA_PASSTHROUGH          ] = "ATA_PASSTHROUGH",
+        [ ATA_PASSTHROUGH_16       ] = "ATA_PASSTHROUGH_16",
         [ ACCESS_CONTROL_IN        ] = "ACCESS_CONTROL_IN",
         [ ACCESS_CONTROL_OUT       ] = "ACCESS_CONTROL_OUT",
         [ READ_16                  ] = "READ_16",
@@ -1352,7 +1434,7 @@ static const char *scsi_command_name(uint8_t cmd)
         [ SERVICE_ACTION_IN_16     ] = "SERVICE_ACTION_IN_16",
         [ WRITE_LONG_16            ] = "WRITE_LONG_16",
         [ REPORT_LUNS              ] = "REPORT_LUNS",
-        [ BLANK                    ] = "BLANK",
+        [ ATA_PASSTHROUGH_12       ] = "BLANK/ATA_PASSTHROUGH_12",
         [ MOVE_MEDIUM              ] = "MOVE_MEDIUM",
         [ EXCHANGE_MEDIUM          ] = "EXCHANGE MEDIUM",
         [ LOAD_UNLOAD              ] = "LOAD_UNLOAD",
diff --git a/hw/scsi-defs.h b/hw/scsi-defs.h
index 8a73f74..d7a4019 100644
--- a/hw/scsi-defs.h
+++ b/hw/scsi-defs.h
@@ -100,7 +100,7 @@
 #define READ_REVERSE_16       0x81
 #define ALLOW_OVERWRITE       0x82
 #define EXTENDED_COPY         0x83
-#define ATA_PASSTHROUGH       0x85
+#define ATA_PASSTHROUGH_16    0x85
 #define ACCESS_CONTROL_IN     0x86
 #define ACCESS_CONTROL_OUT    0x87
 #define READ_16               0x88
@@ -117,7 +117,7 @@
 #define SERVICE_ACTION_IN_16  0x9e
 #define WRITE_LONG_16         0x9f
 #define REPORT_LUNS           0xa0
-#define BLANK                 0xa1
+#define ATA_PASSTHROUGH_12    0xa1
 #define MAINTENANCE_IN        0xa3
 #define MAINTENANCE_OUT       0xa4
 #define MOVE_MEDIUM           0xa5
-- 
1.7.10.4

Re: [Qemu-devel] [PULL 00/10] SCSI patches for 2012-08-03

[Anthony Liguori]

On 08/03/2012 03:06 AM, Paolo Bonzini wrote:
> Anthony,
>
> The following changes since commit 5e3bc7144edd6e4fa2824944e5eb16c28197dd5a:
>
>    Merge remote-tracking branch 'mst/tags/for_anthony' into staging (2012-07-30 10:00:48 -0500)
>
> are available in the git repository at:
>
>
>    git://github.com/bonzini/qemu.git scsi-next
>
> for you to fetch changes up to b71760ef6180662cc2dff1f6cf673a53508c59f1:
>
>    scsi: add support for ATA_PASSTHROUGH_xx scsi command (2012-08-03 10:04:37 +0200)
>
> The main change here is re-enabling megasas, but there's also Herve's
> retro HBA and a couple of random fixes.

   CC    sparc-softmmu/target-sparc/mmu_helper.o
   CC    sparc-softmmu/target-sparc/ldst_helper.o
   CC    sparc-softmmu/target-sparc/int32_helper.o
   LINK  sparc-softmmu/qemu-system-sparc
../libhw64/hw/esp.o: In function `dc390_read_config':
/home/anthony/git/qemu/hw/esp.c:1220: undefined reference to 
`pci_default_read_config'
../libhw64/hw/esp.o: In function `dc390_write_config':
/home/anthony/git/qemu/hw/esp.c:1245: undefined reference to 
`pci_default_write_config'
collect2: ld returned 1 exit status
make[1]: *** [qemu-system-sparc] Error 1
make: *** [subdir-sparc-softmmu] Error 2

It looks like sparc does link in esp.c but doesn't support pci (at least pci.o). 
  I must admit, it's not obvious to me whether pci.o should be added to the 
sparc build or whether esp.c needs to be refactored.

Regards,

Anthony Liguori

>
> ----------------------------------------------------------------
> Cong Meng (1):
>        scsi: add support for ATA_PASSTHROUGH_xx scsi command
>
> Hannes Reinecke (1):
>        megasas: static SAS addresses
>
> Hervé Poussineau (2):
>        esp: add missing const on TypeInfo structures
>        esp: add Tekram DC-390 emulation (PC SCSI adapter)
>
> Paolo Bonzini (3):
>        scsi-disk: fix compilation with DEBUG_SCSI
>        Revert "megasas: disable due to build breakage"
>        esp: enable for all PCI machines
>
> Ronnie Sahlberg (2):
>        SCSI: Update the sense code for PREVENT REMOVAL errors
>        SCSI: STARTSTOPUNIT only eject/load media if powercondition is 0
>
> Stefan Weil (1):
>        megasas: Update function megasys_scsi_uninit
>
>   default-configs/i386-softmmu.mak |    1 -
>   default-configs/pci.mak          |    2 +
>   hw/esp.c                         |  130 +++++++++++++++++++++++++++++++++++++-
>   hw/megasas.c                     |   68 ++++++++++++++------
>   hw/mfi.h                         |    1 +
>   hw/scsi-bus.c                    |   92 +++++++++++++++++++++++++--
>   hw/scsi-defs.h                   |    4 +-
>   hw/scsi-disk.c                   |   29 +++++----
>   8 files changed, 286 insertions(+), 41 deletions(-)

Re: [Qemu-devel] [PULL 00/10] SCSI patches for 2012-08-03

[Paolo Bonzini]

Il 03/08/2012 21:27, Anthony Liguori ha scritto:
> On 08/03/2012 03:06 AM, Paolo Bonzini wrote:
>> Anthony,
>>
>> The following changes since commit
>> 5e3bc7144edd6e4fa2824944e5eb16c28197dd5a:
>>
>>    Merge remote-tracking branch 'mst/tags/for_anthony' into staging
>> (2012-07-30 10:00:48 -0500)
>>
>> are available in the git repository at:
>>
>>
>>    git://github.com/bonzini/qemu.git scsi-next
>>
>> for you to fetch changes up to b71760ef6180662cc2dff1f6cf673a53508c59f1:
>>
>>    scsi: add support for ATA_PASSTHROUGH_xx scsi command (2012-08-03
>> 10:04:37 +0200)
>>
>> The main change here is re-enabling megasas, but there's also Herve's
>> retro HBA and a couple of random fixes.
> 
>   CC    sparc-softmmu/target-sparc/mmu_helper.o
>   CC    sparc-softmmu/target-sparc/ldst_helper.o
>   CC    sparc-softmmu/target-sparc/int32_helper.o
>   LINK  sparc-softmmu/qemu-system-sparc
> ../libhw64/hw/esp.o: In function `dc390_read_config':
> /home/anthony/git/qemu/hw/esp.c:1220: undefined reference to
> `pci_default_read_config'
> ../libhw64/hw/esp.o: In function `dc390_write_config':
> /home/anthony/git/qemu/hw/esp.c:1245: undefined reference to
> `pci_default_write_config'
> collect2: ld returned 1 exit status
> make[1]: *** [qemu-system-sparc] Error 1
> make: *** [subdir-sparc-softmmu] Error 2
> 
> It looks like sparc does link in esp.c but doesn't support pci (at least
> pci.o).  I must admit, it's not obvious to me whether pci.o should be
> added to the sparc build or whether esp.c needs to be refactored.

I think the latter.

I removed this patch and pushed the rest (commit e4b6526) at the same
location.  Blue/Herve, can you look at esp.c?

Paolo

Re: [Qemu-devel] [PULL 00/10] SCSI patches for 2012-08-03

[Anthony Liguori]

Paolo Bonzini <pbonzini@redhat.com> writes:

> Il 03/08/2012 21:27, Anthony Liguori ha scritto:
>> On 08/03/2012 03:06 AM, Paolo Bonzini wrote:
>>> Anthony,
>>>
>>> The following changes since commit
>>> 5e3bc7144edd6e4fa2824944e5eb16c28197dd5a:
>>>
>>>    Merge remote-tracking branch 'mst/tags/for_anthony' into staging
>>> (2012-07-30 10:00:48 -0500)
>>>
>>> are available in the git repository at:
>>>
>>>
>>>    git://github.com/bonzini/qemu.git scsi-next
>>>
>>> for you to fetch changes up to b71760ef6180662cc2dff1f6cf673a53508c59f1:
>>>
>>>    scsi: add support for ATA_PASSTHROUGH_xx scsi command (2012-08-03
>>> 10:04:37 +0200)
>>>
>>> The main change here is re-enabling megasas, but there's also Herve's
>>> retro HBA and a couple of random fixes.
>> 
>>   CC    sparc-softmmu/target-sparc/mmu_helper.o
>>   CC    sparc-softmmu/target-sparc/ldst_helper.o
>>   CC    sparc-softmmu/target-sparc/int32_helper.o
>>   LINK  sparc-softmmu/qemu-system-sparc
>> ../libhw64/hw/esp.o: In function `dc390_read_config':
>> /home/anthony/git/qemu/hw/esp.c:1220: undefined reference to
>> `pci_default_read_config'
>> ../libhw64/hw/esp.o: In function `dc390_write_config':
>> /home/anthony/git/qemu/hw/esp.c:1245: undefined reference to
>> `pci_default_write_config'
>> collect2: ld returned 1 exit status
>> make[1]: *** [qemu-system-sparc] Error 1
>> make: *** [subdir-sparc-softmmu] Error 2
>> 
>> It looks like sparc does link in esp.c but doesn't support pci (at least
>> pci.o).  I must admit, it's not obvious to me whether pci.o should be
>> added to the sparc build or whether esp.c needs to be refactored.
>
> I think the latter.
>
> I removed this patch and pushed the rest (commit e4b6526) at the same
> location.

Thanks.

Regards,

Anthony Liguori

> Blue/Herve, can you look at esp.c?
>
> Paolo

Re: [Qemu-devel] [PULL 00/10] SCSI patches for 2012-08-03

[Hervé Poussineau]

Paolo Bonzini a écrit :
> Il 03/08/2012 21:27, Anthony Liguori ha scritto:
>> On 08/03/2012 03:06 AM, Paolo Bonzini wrote:
>>> Anthony,
>>>
>>> The following changes since commit
>>> 5e3bc7144edd6e4fa2824944e5eb16c28197dd5a:
>>>
>>>    Merge remote-tracking branch 'mst/tags/for_anthony' into staging
>>> (2012-07-30 10:00:48 -0500)
>>>
>>> are available in the git repository at:
>>>
>>>
>>>    git://github.com/bonzini/qemu.git scsi-next
>>>
>>> for you to fetch changes up to b71760ef6180662cc2dff1f6cf673a53508c59f1:
>>>
>>>    scsi: add support for ATA_PASSTHROUGH_xx scsi command (2012-08-03
>>> 10:04:37 +0200)
>>>
>>> The main change here is re-enabling megasas, but there's also Herve's
>>> retro HBA and a couple of random fixes.
>>   CC    sparc-softmmu/target-sparc/mmu_helper.o
>>   CC    sparc-softmmu/target-sparc/ldst_helper.o
>>   CC    sparc-softmmu/target-sparc/int32_helper.o
>>   LINK  sparc-softmmu/qemu-system-sparc
>> ../libhw64/hw/esp.o: In function `dc390_read_config':
>> /home/anthony/git/qemu/hw/esp.c:1220: undefined reference to
>> `pci_default_read_config'
>> ../libhw64/hw/esp.o: In function `dc390_write_config':
>> /home/anthony/git/qemu/hw/esp.c:1245: undefined reference to
>> `pci_default_write_config'
>> collect2: ld returned 1 exit status
>> make[1]: *** [qemu-system-sparc] Error 1
>> make: *** [subdir-sparc-softmmu] Error 2
>>
>> It looks like sparc does link in esp.c but doesn't support pci (at least
>> pci.o).  I must admit, it's not obvious to me whether pci.o should be
>> added to the sparc build or whether esp.c needs to be refactored.
> 
> I think the latter.
> 
> I removed this patch and pushed the rest (commit e4b6526) at the same
> location.  Blue/Herve, can you look at esp.c?

I just sent an email to ML to add stubs for pci_default_read_config and 
pci_default_write_config. I compile-tested sparc-softmmu target and it 
works now well.

Hervé

[1]

Re: [Qemu-devel] [PULL 00/10] SCSI patches for 2012-08-03

[Anthony Liguori]

Hervé Poussineau <hpoussin@reactos.org> writes:

> Paolo Bonzini a écrit :
>> Il 03/08/2012 21:27, Anthony Liguori ha scritto:
>>> On 08/03/2012 03:06 AM, Paolo Bonzini wrote:
>>>> Anthony,
>>>>
>>>> The following changes since commit
>>>> 5e3bc7144edd6e4fa2824944e5eb16c28197dd5a:
>>>>
>>>>    Merge remote-tracking branch 'mst/tags/for_anthony' into staging
>>>> (2012-07-30 10:00:48 -0500)
>>>>
>>>> are available in the git repository at:
>>>>
>>>>
>>>>    git://github.com/bonzini/qemu.git scsi-next
>>>>
>>>> for you to fetch changes up to b71760ef6180662cc2dff1f6cf673a53508c59f1:
>>>>
>>>>    scsi: add support for ATA_PASSTHROUGH_xx scsi command (2012-08-03
>>>> 10:04:37 +0200)
>>>>
>>>> The main change here is re-enabling megasas, but there's also Herve's
>>>> retro HBA and a couple of random fixes.
>>>   CC    sparc-softmmu/target-sparc/mmu_helper.o
>>>   CC    sparc-softmmu/target-sparc/ldst_helper.o
>>>   CC    sparc-softmmu/target-sparc/int32_helper.o
>>>   LINK  sparc-softmmu/qemu-system-sparc
>>> ../libhw64/hw/esp.o: In function `dc390_read_config':
>>> /home/anthony/git/qemu/hw/esp.c:1220: undefined reference to
>>> `pci_default_read_config'
>>> ../libhw64/hw/esp.o: In function `dc390_write_config':
>>> /home/anthony/git/qemu/hw/esp.c:1245: undefined reference to
>>> `pci_default_write_config'
>>> collect2: ld returned 1 exit status
>>> make[1]: *** [qemu-system-sparc] Error 1
>>> make: *** [subdir-sparc-softmmu] Error 2
>>>
>>> It looks like sparc does link in esp.c but doesn't support pci (at least
>>> pci.o).  I must admit, it's not obvious to me whether pci.o should be
>>> added to the sparc build or whether esp.c needs to be refactored.
>> 
>> I think the latter.
>> 
>> I removed this patch and pushed the rest (commit e4b6526) at the same
>> location.  Blue/Herve, can you look at esp.c?
>
> I just sent an email to ML to add stubs for pci_default_read_config and 
> pci_default_write_config. I compile-tested sparc-softmmu target and it 
> works now well.

Shouldn't Sparc consume pci.o?

If Sparc doesn't have PCI, then stubs don't seem like the right answer.
If it has PCI, then it ought to use the generic code, no?

Regards,

Anthony Liguori

>
> Hervé
>
> [1]

Re: [Qemu-devel] [PULL 00/10] SCSI patches for 2012-08-03

[Blue Swirl]

On Fri, Aug 3, 2012 at 8:37 PM, Anthony Liguori <aliguori@us.ibm.com> wrote:
> Hervé Poussineau <hpoussin@reactos.org> writes:
>
>> Paolo Bonzini a écrit :
>>> Il 03/08/2012 21:27, Anthony Liguori ha scritto:
>>>> On 08/03/2012 03:06 AM, Paolo Bonzini wrote:
>>>>> Anthony,
>>>>>
>>>>> The following changes since commit
>>>>> 5e3bc7144edd6e4fa2824944e5eb16c28197dd5a:
>>>>>
>>>>>    Merge remote-tracking branch 'mst/tags/for_anthony' into staging
>>>>> (2012-07-30 10:00:48 -0500)
>>>>>
>>>>> are available in the git repository at:
>>>>>
>>>>>
>>>>>    git://github.com/bonzini/qemu.git scsi-next
>>>>>
>>>>> for you to fetch changes up to b71760ef6180662cc2dff1f6cf673a53508c59f1:
>>>>>
>>>>>    scsi: add support for ATA_PASSTHROUGH_xx scsi command (2012-08-03
>>>>> 10:04:37 +0200)
>>>>>
>>>>> The main change here is re-enabling megasas, but there's also Herve's
>>>>> retro HBA and a couple of random fixes.
>>>>   CC    sparc-softmmu/target-sparc/mmu_helper.o
>>>>   CC    sparc-softmmu/target-sparc/ldst_helper.o
>>>>   CC    sparc-softmmu/target-sparc/int32_helper.o
>>>>   LINK  sparc-softmmu/qemu-system-sparc
>>>> ../libhw64/hw/esp.o: In function `dc390_read_config':
>>>> /home/anthony/git/qemu/hw/esp.c:1220: undefined reference to
>>>> `pci_default_read_config'
>>>> ../libhw64/hw/esp.o: In function `dc390_write_config':
>>>> /home/anthony/git/qemu/hw/esp.c:1245: undefined reference to
>>>> `pci_default_write_config'
>>>> collect2: ld returned 1 exit status
>>>> make[1]: *** [qemu-system-sparc] Error 1
>>>> make: *** [subdir-sparc-softmmu] Error 2
>>>>
>>>> It looks like sparc does link in esp.c but doesn't support pci (at least
>>>> pci.o).  I must admit, it's not obvious to me whether pci.o should be
>>>> added to the sparc build or whether esp.c needs to be refactored.
>>>
>>> I think the latter.
>>>
>>> I removed this patch and pushed the rest (commit e4b6526) at the same
>>> location.  Blue/Herve, can you look at esp.c?
>>
>> I just sent an email to ML to add stubs for pci_default_read_config and
>> pci_default_write_config. I compile-tested sparc-softmmu target and it
>> works now well.
>
> Shouldn't Sparc consume pci.o?
>
> If Sparc doesn't have PCI, then stubs don't seem like the right answer.
> If it has PCI, then it ought to use the generic code, no?

Only JavaStation had PCI, but we don't emulate that.

Probably the same approach as lance & pcnet could be used.

>
> Regards,
>
> Anthony Liguori
>
>>
>> Hervé
>>
>> [1]
>

Re: [Qemu-devel] [PULL 00/10] SCSI patches for 2012-08-03

[Anthony Liguori]

Paolo Bonzini <pbonzini@redhat.com> writes:

> Il 03/08/2012 21:27, Anthony Liguori ha scritto:
>> On 08/03/2012 03:06 AM, Paolo Bonzini wrote:
>>> Anthony,
>>>
>>> The following changes since commit
>>> 5e3bc7144edd6e4fa2824944e5eb16c28197dd5a:
>>>
>>>    Merge remote-tracking branch 'mst/tags/for_anthony' into staging
>>> (2012-07-30 10:00:48 -0500)
>>>
>>> are available in the git repository at:
>>>
>>>
>>>    git://github.com/bonzini/qemu.git scsi-next
>>>
>>> for you to fetch changes up to b71760ef6180662cc2dff1f6cf673a53508c59f1:
>>>
>>>    scsi: add support for ATA_PASSTHROUGH_xx scsi command (2012-08-03
>>> 10:04:37 +0200)
>>>
>>> The main change here is re-enabling megasas, but there's also Herve's
>>> retro HBA and a couple of random fixes.
>> 
>>   CC    sparc-softmmu/target-sparc/mmu_helper.o
>>   CC    sparc-softmmu/target-sparc/ldst_helper.o
>>   CC    sparc-softmmu/target-sparc/int32_helper.o
>>   LINK  sparc-softmmu/qemu-system-sparc
>> ../libhw64/hw/esp.o: In function `dc390_read_config':
>> /home/anthony/git/qemu/hw/esp.c:1220: undefined reference to
>> `pci_default_read_config'
>> ../libhw64/hw/esp.o: In function `dc390_write_config':
>> /home/anthony/git/qemu/hw/esp.c:1245: undefined reference to
>> `pci_default_write_config'
>> collect2: ld returned 1 exit status
>> make[1]: *** [qemu-system-sparc] Error 1
>> make: *** [subdir-sparc-softmmu] Error 2
>> 
>> It looks like sparc does link in esp.c but doesn't support pci (at least
>> pci.o).  I must admit, it's not obvious to me whether pci.o should be
>> added to the sparc build or whether esp.c needs to be refactored.
>
> I think the latter.
>
> I removed this patch and pushed the rest (commit e4b6526) at the same
> location.  Blue/Herve, can you look at esp.c?

Pulled. Thanks.

Regards,

Anthony Liguori

>
> Paolo