- 10 Aug, 2017 10 commits
-
-
Steffen Maier authored
v4.9 commit aceeffbb ("zfcp: trace full payload of all SAN records (req,resp,iels)") fixed trace data loss of 2.6.38 commit 2c55b750 ("[SCSI] zfcp: Redesign of the debug tracing for SAN records.") necessary for problem determination, e.g. to see the currently active zone set during automatic port scan. While it already saves space by not dumping any empty residual entries of the large successful GPN_FT response (4 pages), there are seldom cases where the GPN_FT response is unsuccessful and likely does not have FC_NS_FID_LAST set in fp_flags so we did not cap the trace record. We typically see such case for an initiator WWPN, which is not in any zone. Cap unsuccessful responses to at least the actual basic CT_IU response plus whatever fits the SAN trace record built-in "payload" buffer just in case there's trailing information of which we would at least see the existence and its beginning. In order not to erroneously cap successful responses, we need to swap calling the trace function and setting the CT / ELS status to success (0). Example trace record pair formatted with zfcpdbf: Timestamp : ... Area : SAN Subarea : 00 Level : 1 Exception : - CPU ID : .. Caller : 0x... Record ID : 1 Tag : fssct_1 Request ID : 0x<request_id> Destination ID : 0x00fffffc SAN req short : 01000000 fc020000 01720ffc 00000000 00000008 SAN req length : 20 | Timestamp : ... Area : SAN Subarea : 00 Level : 1 Exception : - CPU ID : .. Caller : 0x... Record ID : 2 Tag : fsscth2 Request ID : 0x<request_id> Destination ID : 0x00fffffc SAN resp short : 01000000 fc020000 80010000 00090700 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] SAN resp length: 16384 San resp info : 01000000 fc020000 80010000 00090700 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] 00000000 00000000 00000000 00000000 [trailing info] The fix saves all but one of the previously associated 64 PAYload trace record chunks of size 256 bytes each. Signed-off-by:
Steffen Maier <maier@linux.vnet.ibm.com> Fixes: aceeffbb ("zfcp: trace full payload of all SAN records (req,resp,iels)") Fixes: 2c55b750 ("[SCSI] zfcp: Redesign of the debug tracing for SAN records.") Cc: <stable@vger.kernel.org> #2.6.38+ Reviewed-by:
Benjamin Block <bblock@linux.vnet.ibm.com> Signed-off-by:
Martin K. Petersen <martin.petersen@oracle.com>
-
Benjamin Block authored
Up until now zfcp would just ignore the FCP_RESID_OVER flag in the FCP response IU. When this flag is set, it is possible, in regards to the FCP standard, that the storage-server processes the command normally, up to the point where data is missing and simply ignores those. In this case no CHECK CONDITION would be set, and because we ignored the FCP_RESID_OVER flag we resulted in at least a data loss or even -corruption as a follow-up error, depending on how the applications/layers on top behave. To prevent this, we now set the host-byte of the corresponding scsi_cmnd to DID_ERROR. Other storage-behaviors, where the same condition results in a CHECK CONDITION set in the answer, don't need to be changed as they are handled in the mid-layer already. Following is an example trace record decoded with zfcpdbf from the s390-tools package. We forcefully injected a fc_dl which is one byte too small: Timestamp : ... Area : SCSI Subarea : 00 Level : 3 Exception : - CPU ID : .. Caller : 0x... Record ID : 1 Tag : rsl_err Request ID : 0x... SCSI ID : 0x... SCSI LUN : 0x... SCSI result : 0x00070000 ^^DID_ERROR SCSI retries : 0x.. SCSI allowed : 0x.. SCSI scribble : 0x... SCSI opcode : 2a000000 00000000 08000000 00000000 FCP rsp inf cod: 0x00 FCP rsp IU : 00000000 00000000 00000400 00000001 ^^fr_flags==FCP_RESID_OVER ^^fr_status==SAM_STAT_GOOD ^^^^^^^^fr_resid 00000000 00000000 As of now, we don't actively handle to possibility that a response IU has both flags - FCP_RESID_OVER and FCP_RESID_UNDER - set at once. Reported-by:Luke M. Hopkins <lmhopkin@us.ibm.com> Reviewed-by:
-
Steffen Maier authored
Since commit db007fc5 ("[SCSI] Command protection operation"), scsi_eh_prep_cmnd() saves scmd->prot_op and temporarily resets it to SCSI_PROT_NORMAL. Other FCP LLDDs such as qla2xxx and lpfc shield their queuecommand() to only access any of scsi_prot_sg...() if (scsi_get_prot_op(cmd) != SCSI_PROT_NORMAL). Do the same thing for zfcp, which introduced DIX support with commit ef3eb71d ("[SCSI] zfcp: Introduce experimental support for DIF/DIX"). Otherwise, TUR SCSI commands as part of scsi_eh likely fail in zfcp, because the regular SCSI command with DIX protection data, that scsi_eh re-uses in scsi_send_eh_cmnd(), of course still has (scsi_prot_sg_count() != 0) and so zfcp sends down bogus requests to the FCP channel hardware. This causes scsi_eh_test_devices() to have (finish_cmds == 0) [not SCSI device is online or not scsi_eh_tur() failed] so regular SCSI commands, that caused / were affected by scsi_eh, are moved to work_q and scsi_eh_test_devices() itself returns false. In turn, it unnecessarily escalates in our case in scsi_eh_ready_devs() beyond host reset to finally scsi_eh_offline_sdevs() which sets affected SCSI devices offline with the following kernel message: "kernel: sd H:0:T:L: Device offlined - not ready after error recovery" Signed-off-by:
-
Benjamin Block authored
Better form and cleans remaining warnings. Found with scripts/coccinelle/misc/boolinit.cocci. Signed-off-by:
-
Corentin Labbe authored
drivers/s390/scsi/zfcp_aux.c does not contain any miscdevice so the inclusion of linux/miscdevice.h is unnecessary. [maier@linux.vnet.ibm.com: just for the records, this is in fact a minor missing code cleanup of the following older "feature" which also dropped the only former use of a misc device in zfcp: commit 663e0890 ("[SCSI] zfcp: remove access control tables interface") commit b5dc3c48 ("[SCSI] zfcp: remove access control tables interface (keep sysfs files)") commit 1b33ef23 ("zfcp: remove access control tables interface (port leftovers)")] Signed-off-by:
Corentin Labbe <clabbe.montjoie@gmail.com> Signed-off-by:
-
Lukáš Korenčik authored
Use initialization with setup_timer function instead of using init_timer function and data fields. It improves readability. Signed-off-by:
Lukáš Korenčik <xkorenc1@fi.muni.cz> Signed-off-by:
Jiri Slaby <jslaby@suse.cz> Signed-off-by:
-
LABBE Corentin authored
The zfcp_qdio_sbale_count function do the same work than sg_nents(). So replace it by sg_nents() for removing duplicate code. Signed-off-by:
-
Varun Prakash authored
If cxgbi_ep_connect() is called with valid shost then find associated ndev and use ndev->ifindex to find route. Signed-off-by:
Varun Prakash <varun@chelsio.com> Signed-off-by:
-
Christophe JAILLET authored
In the lines above this test, 8 'kzalloc' are performed, but only 7 results are tested. Add the missing one (i.e. '!ioc->port_enable_cmds.reply'). Signed-off-by:
Christophe JAILLET <christophe.jaillet@wanadoo.fr> Signed-off-by:
-
Michał Mirosław authored
Regenerate firmware files to make cleaner base for following fix. This removes some unused definitions and reorders some #defines, but the code remains the same. Signed-off-by:
Michał Mirosław <mirq-linux@rere.qmqm.pl> Reviewed-by:
Hannes Reinecke <hare@suse.com> Signed-off-by:
-
- 08 Aug, 2017 2 commits
-
-
Michał Mirosław authored
We need to override Kbuild rules for copying shipped files, otherwise aic7xxx_reg.h and aic7xxx_reg_print.c will be ovewritten by old versions. Fixes: 516b7db5Signed-off-by:
-
Michał Mirosław authored
ahc_platform_dump_card_state() does nothing. Remove it. Signed-off-by:
-
- 07 Aug, 2017 28 commits
-
-
Benjamin Block authored
I have been working with Steffen on zFCP for quite a while now and we decided adding me as a co-maintainer might be a good thing. Acked-by:
-
Steffen Maier authored
Scsi_cmnd is an unsuitable argument for eh_device_reset_handler(), eh_target_reset_handler(), and eh_host_reset_handler() which do not have the scope of one single SCSI command. These callbacks tend to use fc_block_scsi_eh() requiring scsi_cmnd. In order to start decoupling above eh callbacks from scsi_cmnd, introduce a new variant of the function called fc_block_rport() taking an fc_rport as argument. Refactor the old fc_block_scsi_eh() to simply delegate to fc_block_rport(). Signed-off-by:
-
himanshu.madhani@cavium.com authored
Signed-off-by:
Duane Grigsby <duane.grigsby@cavium.com> Signed-off-by:
Himanshu Madhani <himanshu.madhani@cavium.com> Reviewed-by:
Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by:
-
himanshu.madhani@cavium.com authored
Simplified waiting for unregister local/remote FC-NVMe ports to complete cleanup. Signed-off-by:
Darren Trapp <darren.trapp@cavium.com> Signed-off-by:
Anil Gurumurthy <anil.gurumurthy@cavium.com> Signed-off-by:
-
Duane Grigsby authored
Add support to the driver to set the exchange threshold value for the number of outstanding AENs. Signed-off-by:
-
himanshu.madhani@cavium.com authored
Cc: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by:
-
himanshu.madhani@cavium.com authored
This patch does not change any functionality. Following cleanups have been done as requested by reviewer - Changed waitQ --> waitq - Collapsed multiple debug statements into single - Remove extra parentheses in if-else as per operator precedence - Remove unnecessary casting Cc: Johannes Thumshirn <jthumshirn@suse.de> Signed-off-by:
-
Greg Kroah-Hartman authored
It's better to use the DRIVER_ATTR_RW() and DRIVER_ATTR_RO() macros to explicitly show that this is a read/write or read/only sysfs file. So convert the remaining SCSI drivers that use the old style to use the newer macros. Bonus is that this removes some checkpatch.pl warnings :) This is part of a series to drop DRIVER_ATTR() from the tree entirely. Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: Kashyap Desai <kashyap.desai@broadcom.com> Cc: Sumit Saxena <sumit.saxena@broadcom.com> Cc: Shivasharan S <shivasharan.srikanteshwara@broadcom.com> Cc: Willem Riede <osst@riede.org> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org> Reviewed-by:
Bart Van Assche <bart.vanassche@wdc.com> Signed-off-by:
-
Rob Herring authored
Now that we have a custom printf format specifier, convert users of full_name to use %pOF instead. This is preparation to remove storing of the full path string for each node. Signed-off-by:
Rob Herring <robh@kernel.org> Cc: "James E.J. Bottomley" <jejb@linux.vnet.ibm.com> Cc: "Martin K. Petersen" <martin.petersen@oracle.com> Cc: linux-scsi@vger.kernel.org Signed-off-by:
-
Prabu Thangamuthu authored
As per internal decision, Joao Pinto will be maintainer for DWC UFS driver. Signed-off-by:
Prabu Thangamuthu <prabut@synopsys.com> Acked-by:
Joao Pinto <Joao.Pinto@synopsys.com> Signed-off-by:
-
Arnd Bergmann authored
gcc points out a theorerical string overflow: drivers/message/fusion/mptbase.c: In function 'mpt_detach': drivers/message/fusion/mptbase.c:2103:17: error: '%s' directive writing up to 31 bytes into a region of size 28 [-Werror=format-overflow=] sprintf(pname, MPT_PROCFS_MPTBASEDIR "/%s/summary", ioc->name); ^~~~~ drivers/message/fusion/mptbase.c:2103:2: note: 'sprintf' output between 13 and 44 bytes into a destination of size 32 We can simply double the size of the local buffer here to be on the safe side, and using snprintf() instead of sprintf() protects us if ioc->name was not terminated properly. Signed-off-by:Arnd Bergmann <arnd@arndb.de> Signed-off-by:
-
Arnd Bergmann authored
We print a 256 byte event string into a buffer that is only 161 bytes long, this is clearly wrong: drivers/scsi/gdth_proc.c: In function 'gdth_show_info': drivers/scsi/gdth.c:3660:41: error: '%s' directive writing up to 255 bytes into a region of size between 141 and 150 [-Werror=format-overflow=] sprintf(buffer,"Adapter %d: %s\n", ^~ /git/arm-soc/drivers/scsi/gdth.c:3660:13: note: 'sprintf' output between 13 and 277 bytes into a destination of size 161 sprintf(buffer,"Adapter %d: %s\n", ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ dvr->eu.async.ionode,dvr->event_string); ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ gcc calculates that the worst case buffer size would be 277 bytes, so we can use that. Signed-off-by: -
Arnd Bergmann authored
The MSI interrupt name can require 11 bytes in addition to the device name, for a total of 23 bytes: drivers/scsi/fnic/fnic_isr.c: In function 'fnic_request_intr': drivers/scsi/fnic/fnic_isr.c:192:4: error: '-fcs-rq' directive writing 7 bytes into a region of size between 5 and 16 [-Werror=format-overflow=] "%.11s-fcs-rq", fnic->name); drivers/scsi/fnic/fnic_isr.c:206:3: note: 'sprintf' output between 12 and 23 bytes into a destination of size 16 sprintf(fnic->msix[FNIC_MSIX_ERR_NOTIFY].devname, ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ "%.11s-err-notify", fnic->name); This extends the buffer to fit any possible value. Signed-off-by: -
Arnd Bergmann authored
gcc notices that we would overflow the buffer for the inquiry of the product name if we have too many adapters: drivers/scsi/gdth.c: In function 'gdth_next': drivers/scsi/gdth.c:2357:29: warning: 'sprintf' may write a terminating nul past the end of the destination [-Wformat-overflow=] sprintf(inq.product,"Host Drive #%02d",t); ^~~~~~~~~~~~~~~~~~~ drivers/scsi/gdth.c:2357:9: note: 'sprintf' output between 16 and 17 bytes into a destination of size 16 sprintf(inq.product,"Host Drive #%02d",t); This won't happen in practice, so just use snprintf to truncate the string. Signed-off-by: -
Arnd Bergmann authored
We print the driver name into one string and then add and ID and copy it into a second string of the same length, at which point gcc complains about a possible overflow: drivers/scsi/mpt3sas/mpt3sas_scsih.c: In function '_scsih_probe': drivers/scsi/mpt3sas/mpt3sas_scsih.c:8884:21: error: '_cm' directive writing 3 bytes into a region of size between 1 and 32 [-Werror=format-overflow=] printf(ioc->name, "%s_cm%d", ioc->driver_name, ioc->id); ^~~~~~~~~ drivers/scsi/mpt3sas/mpt3sas_scsih.c:8884:21: note: directive argument in the range [0, 255] drivers/scsi/mpt3sas/mpt3sas_scsih.c:8884:2: note: 'sprintf' output between 5 and 38 bytes into a destination of size 32 sprintf(ioc->name, "%s_cm%d", ioc->driver_name, ioc->id); ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Making the first string shorter is sufficient to avoid the warning here, as we know it can only contain either "mpt2sas" or "mpt3sas". Signed-off-by: -
Arnd Bergmann authored
gcc-7 complains that the firmware version strings might overflow for some values: drivers/scsi/megaraid.c: In function 'megaraid_probe_one': drivers/scsi/megaraid.c:314:33: error: '%d' directive writing between 1 and 2 bytes into a region of size between 1 and 2 [-Werror=format-overflow=] drivers/scsi/megaraid.c:314:33: note: directive argument in the range [0, 15] drivers/scsi/megaraid.c:314:3: note: 'sprintf' output between 7 and 9 bytes into a destination of size 7 drivers/scsi/megaraid.c:320:35: error: '%d' directive writing between 1 and 2 bytes into a region of size between 1 and 2 [-Werror=format-overflow=] drivers/scsi/megaraid.c:320:35: note: directive argument in the range [0, 15] drivers/scsi/megaraid.c:320:3: note: 'sprintf' output between 7 and 9 bytes into a destination of size 7 This makes the code use a truncating snprintf() instead, which shuts up that warning. Signed-off-by:
-
Romain Perier authored
The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by:
Romain Perier <romain.perier@collabora.com> Acked-by:
Peter Senna Tschudin <peter.senna@collabora.com> Tested-by:
-
Romain Perier authored
The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by:
-
Romain Perier authored
The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by:
-
Romain Perier authored
The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by:
Sumit Saxena <sumit.saxena@broadcom.com> Signed-off-by:
-
Romain Perier authored
The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. It also updates some comments, accordingly. Signed-off-by:
-
Romain Perier authored
The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. It also updates the name of some variables and the content of comments, accordingly. Signed-off-by:
-
Romain Perier authored
The PCI pool API is deprecated. This commit replaces the PCI pool old API by the appropriate function with the DMA pool API. Signed-off-by:
-
Ondrej Zary authored
Limit PDMA send to 512 B to avoid data corruption on DTC3181E. The corruption is always the same: one byte missing at the beginning of a 128 B block. It happens only with slow Quantum LPS 240 drive, not with faster IBM DORS-32160. It's not clear what causes this. Documentation for the DTC436 chip has not been made available. Signed-off-by:
Finn Thain <fthain@telegraphics.com.au> Tested-by:
Ondrej Zary <linux@rainbow-software.org> Signed-off-by:
-
Ondrej Zary authored
The polling loops in pread() and pwrite() can easily become infinite loops and hang the machine. Merge the IRQ check into host buffer wait loop and add polling limit. Also place a limit on polling for 53C80 registers accessibility. [Use NCR5380_poll_politely2() for register polling. Rely on polling for gated IRQ rather than polling for phase error, like the algorithm in the 53c400 datasheet. Move DTC436 workarounds into a separate patch. Factor-out common code as wait_for_53c80_access(). Rework the residual calculations. -- F.T.] Signed-off-by:
-
Finn Thain authored
The word "read" may be used to mean "DMA read operation" or "SCSI READ command", though a READ command implies writing to memory. Signed-off-by:
-
Finn Thain authored
Signed-off-by:
-
Ondrej Zary authored
When an IRQ arrives during PDMA transfer, pread() and pwrite() return without waiting for the 53C80 registers to be ready and this ends up messing up the chip state. This was observed with SONY CDU-55S which is slow enough to disconnect during 4096-byte reads. IRQ during PDMA is not an error so don't return -1. Instead, store the remaining byte count for use by NCR5380_dma_residual(). [Poll for the BASR_END_DMA_TRANSFER condition rather than remove the error message -- F.T.] Signed-off-by:
-
