- 04 Dec, 2009 40 commits
-
-
Chris Leech authored
Allow FCP frames to bypass the FCoE receive processing threads and handle them directly in softirq context, if they are received on the correct CPU. This preserves the queuing to threads for scaling out receive processing to multiple CPUs, but allows FCoE-aware multi-queue network drivers that direct frames to the originating CPUs to handle FCP processing with less scheduling latency. Only FCP is handled directly, because libfc makes use of mutexes in ELS handling routines. The bulk of this change is just moving the FCoE receive processing out of the receive thread function, leaving behind just the thread and queue management. The interesting bits are in fcoe_rcv() Signed-off-by:
Chris Leech <christopher.leech@intel.com> Signed-off-by:
Robert Love <robert.w.love@intel.com> Signed-off-by:
James Bottomley <James.Bottomley@suse.de>
-
Yi Zou authored
Frame header room is already incluced, just pass the length of payload. Signed-off-by:
Yi Zou <yi.zou@intel.com> Signed-off-by:
-
James Smart authored
Signed-off-by:
James Smart <james.smart@emulex.com> Signed-off-by:
-
James Smart authored
Hardware related Fixes and Changes. - Added new Adapter IDs and update default Adapter names. - Added PCI read after EQarm doorbell PCI write to flush the write and avoid spurrious interrupts when in INTx mode. - Phase out use of ONLINE registers. - Fix for lost MSI interrupt Signed-off-by:
-
James Smart authored
Fix errors relating to crashes and hangs. - Fix crash due to list corruption while unloading driver. - Fix panic during pci-hot-plug testing. - Fix panic when unmapping luns. - Fixed total_scsi_bufs counting could cause exhausted memory. - Fixed locking issue causing hang. - Fixed the call from lpfc_new_scsi_buf_s3 to use lpfc_release_scsi_buf_s3. Signed-off-by:
-
James Smart authored
Fix AER issues. - Made AER sysfs entry point return "Operation not permitted" to OneConnect HBAs - Stop and abort all I/Os on HBA for AER uncorrectable non-fatal error handling Signed-off-by:
-
James Smart authored
FC protocol fixes. - Fix send sequence logic to handle multi SGL IOCBs. - Fix FDISC completion always setting VPORT state to failed. - Ported the fix on reporting of max_vpi to uppper layer. - Fix incorrect number of Vports allowed to be created. - Fixed Dead FCoE port after creating vports. - Added handling of ELS request for Reinstate Recovery Qualifier (RRQ) - Handle unsolicited CT exchange initiator receiving CT exchange ABTS - Migrate LUN queue depth ramp up code to scsi mid-layer. - Made ABTS WQE go to the same WQ as the WQE to be aborted. - Fix Vport does not rediscover after FCF goes away. - Fixed lpfc_unreg_vfi failure after devloss timeout. - Fixed RPI bit leak. - Fix hbq pointer corruption during target discovery. Signed-off-by:
-
James Smart authored
FCoE Protocol fixes. - Fixed FIP frame designation for ELS commands. - Fix CVL received on Port 1 not processed by driver. - Fix Zeroed frame on wire after FLOGI - Fix vport keep-alive does not contain the correct WWN. Signed-off-by:
-
Roel Kluin authored
value cannot logically be less than START and greater than BUFFERSIZE. #define EXTENDED_SENSE_START 18 // vi include/scsi/scsi_cmnd.h +105 #define SCSI_SENSE_BUFFERSIZE 96 [akpm@linux-foundation.org: fix warning] Signed-off-by:
Roel Kluin <roel.kluin@gmail.com> Signed-off-by:
Andrew Morton <akpm@linux-foundation.org> Acked-by:
Kai Makisara <kai.makisara@kolumbus.fi> Signed-off-by:
-
Randy Dunlap authored
Fix scsi_devinfo.c kernel-doc function names to match actual function names. Signed-off-by:
Randy Dunlap <randy.dunlap@oracle.com> Signed-off-by:
-
Martin Michlmayr authored
include/scsi/osd_protocol.h uses ALIGN() without an #include <linux/kernel.h>, leading to: | include/scsi/osd_protocol.h:362: error: implicit declaration of function 'ALIGN' Signed-off-by:
Martin Michlmayr <tbm@cyrius.com> Signed-off-by:
Boaz Harrosh <bharrosh@panasas.com> Cc: Stable Tree <stable@kernel.org> Signed-off-by:
-
Boaz Harrosh authored
Administer some love to the osd_req_decode_sense function * Fix a bad bug with osd_req_decode_sense(). If there was no scsi residual, .i.e the request never reached the target, then all the osd_sense_info members where garbage. * Add grossly missing in/out_resid to osd_sense_info and fill them in properly. * Define an osd_err_priority enum which divides the possible errors into 7 categories in ascending severity. Each category is also assigned a Linux return code translation. Analyze the different osd/scsi/block returned errors and set the proper osd_err_priority and Linux return code accordingly. * extra check a few situations so not to get stuck with inconsistent error view. Example an empty residual with an error code, and other places ... Lots of libosd's osd_req_decode_sense clients had this logic in some form or another. Consolidate all these into one place that should actually know about osd returns. Thous translating it to a more abstract error. Signed-off-by:
-
Boaz Harrosh authored
When an error was detected in an attribute list do to a target bug. We would print an error but spin endlessly regardless. Fix it. Signed-off-by:
-
Boaz Harrosh authored
The (never tested) osd_sense_attribute_identification case has never worked. The loop was never advanced on. Fix it to work as intended. On 10/30/2009 04:39 PM, Roel Kluin wrote: I found this by code analysis, searching for while loops that test a local variable, but do not modify the variable. Reported-by:
-
Boaz Harrosh authored
Define an osd_dev_info structure that Uniquely identifies an OSD device lun on the network. The identification is built from unique target attributes and is the same for all network/SAN machines. osduld_info_lookup() - NEW New API that will lookup an osd_dev by its osd_dev_info. This is used by pNFS-objects for cross network global device identification. And by exofs multy-device support, the device info is specified in the on-disk exofs device table. osduld_device_info() - NEW Given an osd_dev handle returns its associated osd_dev_info. The ULD fetches this information at startup and hangs it on each OSD device. (This is a fast operation that can be called at any condition) osduld_device_same() - NEW With a given osd_dev at one hand and an osd_dev_info at another, we would like to know if they are the same device. Two osd_dev handles can be checked by: osduld_device_same(od1, osduld_device_info(od2)); osd_auto_detect_ver() - REVISED Now returns an osd_dev_info structure. Is only called once by ULD as before. See added comments for how to use. Signed-off-by: -
Boaz Harrosh authored
The true logic of this patch will be clear in the next patch where we use the class_find_device() API. When doing so the use of an internal kref leaves us a narrow window where a find is started while the actual object can go away. Using the device's kobj reference solves this problem because now the same kref is used for both operations. (Remove and find) Core changes * Embed a struct device in uld_ structure and use device_register instead of devie_create. Set __remove to be the device release function. * __uld_get/put is just get_/put_device. Now every thing is accounted for on the device object. Internal kref is removed. * At __remove() we can safely de-allocate the uld_ structure. (The function has moved to avoid forward declaration) Some cleanups * Use class register/unregister is cleaner for this driver now. * cdev ref-counting games are no longer necessary I have incremented the device version string in case of new bugs. Note: Previous bugfix of taking the reference around fput() still applies. Signed-off-by: -
Boaz Harrosh authored
If scsi has released the device (logout), and exofs has last reference on the osduld_device it will be freed by osd_uld_release() within the call to fput(). But this will oops in cdev_release() which is called after the fops->release. (cdev is embedded within osduld_device). __uld_get/put pair makes sure we have a cdev for the duration of fput() Signed-off-by:
-
Boaz Harrosh authored
Add one more important cdb_field_offset that can be returned with scsi_invalid_field_in_cdb. It is the offset of the permissions_bit_mask field in the capabilities structure. Interestingly, the offset is the same for V1/V2 Signed-off-by:
-
Boaz Harrosh authored
define a new osd_dev_is_ver1 that operates on devices and the old osd_req_is_ver1 uses that new API. Signed-off-by:
-
James Smart authored
Signed-off-by:
-
James Smart authored
[SCSI] lpfc 8.3.5: fix fcp command polling, add FIP mode, performance optimisations and devloss timout fixes This patch includes the following changes: - Fixed Panic/Hang when using polling mode for fcp commands - Added support for Read_rev mbox bits indicating FIP mode of HBA - Optimize performance of slow-path handling of els responses - Add code to cleanup orphaned unsolicited receive sequences - Fixed Devloss timeout when multiple initiators are in same zone Signed-off-by:
-
James Smart authored
Signed-off-by:
-
James Smart authored
This patch include the following fixes and changes: - Fix crash when "error" is echoed to board_mode sysfs parameter - Fix FCoE Parameter parsing in regions 23 - Fix driver crash when creating vport with large number of targets on SLI4 - Fix bug with npiv message being logged when it is not supported by the adapter - Fix a potential dereferencing mailbox structure after free bug - Fix firmware crash after vport create with high target count - Error out requests to set board_mode to warm restart via sysfs on SLI4 HBAs - Fix Block guard logging - Fix a memory corruption issue during GID_FT IO prep - Fix crash while processing unsolicited FC frames - Fix failed to allocate XRI message is not a critical failure - Update and fix formatting in some log messages - Fix missing new line characters in log messages - Removed the use of the locally defined FC transport layer related macros - Check the rsplen in lpfc_handle_fcp_err function before using rsplen Signed-off-by:
-
James Smart authored
This patch includes the following fixes and new features: - Fix mask size for CT field in WQE - Fix VPI base not used when unregistering VPI on port 1. - Fix UNREG_VPI mailbox command to unreg the correct VPI - Fixed Check for aborted els command - Fix error when trying to load driver with wrong firmware on FCoE HBA. - Fix bug with probe_one routines not putting the Scsi_Host back upon error - Add support for Clear Virtual Link Async Events - Add support for unsolicited CT exchange sequence abort - Add 0x0714 OCeXXXXX PCI ID Signed-off-by:
-
James Smart authored
This patch includes the following fixes: - Fixed panic during HBA reset. - Fixed FCoE event tag passed in resume_rpi. - Fix out of order ELS commands - Fixed discovery issues found during VLAN testing. - Fix UNREG_VPI failure on extended link pull - Fixed crash while processing unsolicited FC frames. - Clear retry count in the delayed ELS handler - Fixed discovery failure during quick link bounce. Signed-off-by:
-
Mike Christie authored
It is rare to get a queue full with iscsi, because targets seem to just reduce the iscsi cmd window. However, there is at least one iscsi target that will throw a queue full when overloaded. This hooks the iscsi code in to the ramp up/down code, so we can handle it. Signed-off-by:
Mike Christie <michaelc@cs.wisc.edu> Signed-off-by:
-
Mike Christie authored
When iser enabled lu reset support it did not set the bit to allow userspace to get/set the timeout. This sets the tgt and lu reset timeout bits. Signed-off-by:
-
Mike Christie authored
This patch modifies the replacement/recovery_timeout so it works more like the fc fast io fail tmo. If userspace tries to set the replacement/recovery_timeout to less than zero, we will turn off the forced recovery cleanup. If userspace sets the value to 0 then we will force the recovery cleanup immediately. Signed-off-by:
-
Mike Christie authored
This implements warm target reset tmf support for the scsi-ml target reset callback. Previously we would just drop the session in that callback. This patch will now try a target reset and if that fails drop the session. Signed-off-by:
-
Mike Christie authored
Patch and mail from both MikeC and HannesR: Before we're trying to send a PDU we have to check whether a TMF is active. If so and if the PDU will be affected by the TMF we should allow only Data-out PDUs to be sent. If fast_abort is set, no Data-out PDUs will be sent while a LUN reset is being processed for a affected LUN. fast_abort is now ingored during a ABORT TASK tmf. We will not send any Data-outs for a task if the task is being aborted. Signed-off-by:
Hannes Reinecke <hare@suse.de> Signed-off-by:
-
Mike Christie authored
For some reason we used to check for the the immediate bit set and the opcocde in many places instead of just masking the opcode. In the passthrough code this is a problem because userspace may or may not have set the immediate bit and it does not have to. This fixes up the opcode checks in the passthrough code, so we mask off the opcode then check against the iscsi proto definition like is done in other places. Signed-off-by:
-
Mike Christie authored
This just has bnx2i use the iscsi_suspend_queue helper. The suspend works as follows: When ep_poll has succeeed iscsid will call conn_bind, the LLD will then call iscsi_conn_bind which will clear the suspend bit. When ep_disconnect is called (or if there is a conn error) we set the suspend bit. For the ep_disconnect case I added a helper in the previous kernel that will take the session lock to make sure iscsi_queuecommand/xmit_task is not running and it will set the suspend bit. Signed-off-by:
Anil Veerabhadrappa <anilgv@broadcom.com> Signed-off-by:
-
Takahiro Yasui authored
Four models, OPEN-/DF400/DF500/DISK-SUBSYSTEM, can handle REPORT_LUN, and the BLIST_REPORTLUN2 flag needs to be set. And DF600 doesn't require any flags because it returns ANSI 03h (SPC). Signed-off-by:
Takahiro Yasui <tyasui@redhat.com> Signed-off-by:
-
Mike Christie authored
The scsi ioctl code path was missing scsi target reset support. This patch just adds it. Signed-off-by:
-
Mike Christie authored
If the port state is blocked and the fast io fail tmo has fired then this patch will fail bsg requests immediately. This is needed if userspace is sending IOs to test the transport like with fcping, so it will not have to wait for the dev loss tmo. With this patch he bsg req fast io fail code behaves like the normal and sg io/passthrough fast io fail. Signed-off-by:
-
jack_wang authored
Allocate right size for bitmap tag,fix error goto and cleanup print message and undocable commemts. patch attached. Signed-off-by:
Lindar Liu <lindar_liu@usish.com> Signed-off-by:
Jack Wang <jack_wang@usish.com> Signed-off-by:
-
jack_wang authored
We set interupt cascading count of outbound queue to get better performance, correct some unnecessary return values and some noisy print messages. patch attached. Signed-off-by:
-
jack_wang authored
Signed-off-by:
-
James Bottomley authored
Some of our virtual SCSI hosts don't have a proper bus parent at the top, which can be a problem for doing DMA on them This patch makes the host device cache a pointer to the physical bus device and provides an extra API for setting it (the normal API picks it up from the parent). This patch also modifies the qla2xxx and lpfc vport logic to use the new DMA host setting API. Acked-By:
-
Chandra Seetharaman authored
Create the sysfs file, dh_state even if the new SCSI device is not in the any of the device handler's internal lists. Signed-Off-by:
Chandra Seetharaman <sekharan@us.ibm.com> Acked-by:
-
