On Fri, Jun 07, 2013 at 09:34:19PM +, Nicholas A. Bellinger wrote:
> From: Nicholas Bellinger
>
> This patch makes target_execute_cmd() set CMD_T_BUSY|CMD_T_SENT while
> holding se_cmd->t_state_lock, in order to avoid the extra aquire/release
> in __target_execute_cmd().
Can you also do an a
On Sat, Jun 08, 2013 at 12:00:36AM -0400, Jörn Engel wrote:
> > -static int transport_cmd_check_stop(struct se_cmd *cmd, bool
> > remove_from_lists)
> > +static int transport_cmd_check_stop(struct se_cmd *cmd, bool
> > remove_from_lists,
> > + bool write_pending)
> .
On Fri, 7 June 2013 21:34:16 +, Nicholas A. Bellinger wrote:
>
> -static int transport_cmd_check_stop(struct se_cmd *cmd, bool
> remove_from_lists)
> +static int transport_cmd_check_stop(struct se_cmd *cmd, bool
> remove_from_lists,
> + bool write_pending)
..
From: Nicholas Bellinger
Hi folks.
Here is an initial series for adding target descriptor pre-allocation
support based upon Kent's excellent per-cpu struct tag_pool library code.
It allows fabric drivers to avoid fast path memory allocation/free
overhead up to the number of registered struct ta
From: Nicholas Bellinger
This patch changes vhost/scsi to use transport_init_session_tagpool()
pre-allocation logic for per-cpu session tag pooling with internal
tag_alloc() + tag_free() calls based upon the saved se_cmd->map_tag id.
FIXME: Make transport_init_session_tagpool() number of tags
co
From: Nicholas Bellinger
This patch adds lib/tag.c based transport_init_session_tagpool() logic
that allows fabric drivers to setup a per-cpu se_sess->sess_tag_pool
and associated se_sess->sess_cmd_map for basic tagged pre-allocation
of fabric descriptor sized memory.
Cc: Kent Overstreet
Cc: Ro
From: Kent Overstreet
Allocates integers out of a predefined range - for use by e.g. a driver
to allocate tags for communicating with the device.
Signed-off-by: Kent Overstreet
Cc: Tejun Heo
Cc: Oleg Nesterov
Cc: Christoph Lameter
Cc: Ingo Molnar
Cc: Michael S. Tsirkin
Cc: Asias He
Signed
From: Nicholas Bellinger
This patch makes target_execute_cmd() set CMD_T_BUSY|CMD_T_SENT while
holding se_cmd->t_state_lock, in order to avoid the extra aquire/release
in __target_execute_cmd().
It also clears these bits in case of a target_handle_task_attr()
failure.
Cc: Christoph Hellwig
Cc:
From: Nicholas Bellinger
This patch drops an unnecessary acquire/release of se_cmd->t_state_lock within
transport_lun_remove_cmd() when checking CMD_T_DEV_ACTIVE for invoking
target_remove_from_state_list().
For all fast path completion cases, transport_lun_remove_cmd() is always
called ahead of
From: Nicholas Bellinger
This patch adds a new transport_cmd_check_stop() parameter for signaling
when TRANSPORT_WRITE_PENDING needs to be set.
This allows transport_generic_new_cmd() to avoid the extra lock acquire/release
of ->t_state_lock in the fast path for DMA_TO_DEVICE operations ahead of
From: Nicholas Bellinger
This patch changes vhost_scsi_free_cmd() to call transport_generic_free_cmd()
with wait_for_tasks=false in order to avoid the extra se_cmd->t_state_lock
access for the wait_for_tasks=true case.
This is unnecessary because vhost_scsi_free_cmd() is only ever called by
vhos
From: Nicholas Bellinger
In modern iscsi-target code, the setup and I/O submission is done within a
single process context, so there is no need to acquire se_cmd->t_state_lock
while
checking SCF_SUPPORTED_SAM_OPCODE for determining when unsolicited data-out
should be dumped.
Cc: Christoph Hellw
From: Nicholas Bellinger
This patch removes legacy se_cmd->t_fe_count usage in order to avoid
se_cmd->t_state_lock access within transport_put_cmd() during normal
fast path se_cmd descriptor release.
Also drop the left-over parameter usage within core_tmr_handle_tas_abort()
Cc: Christoph Hellwi
From: Nicholas Bellinger
This patch coverts vhost/scsi to se_cmd->cmd_kref TARGET_SCF_ACK_KREF
usage, instead of assuming that vhost_scsi_free_cmd() is always called
before TCM processing is completed in the response fast path.
This includes adding vhost_scsi_check_stop_free() -> target_put_sess
From: Nicholas Bellinger
This patch drops the se_cmd->t_state_lock access around SCF_SUPPORTED_SAM_OPCODE
assignment within target_setup_cmd_from_cdb().
Original v4.0 target code required this as fabrics would be checking for
this values in different process contexts for setup and I/O submission
From: Nicholas Bellinger
Now with iscsi-target using modern se_cmd->cmd_kref accounting in
v3.10 code, it's safe to go ahead and drop the legacy release
codepath + se_cmd->check_release bit in transport_release_cmd()
Cc: Christoph Hellwig
Cc: Roland Dreier
Cc: Kent Overstreet
Cc: Or Gerlitz
From: Nicholas Bellinger
Hi folks,
This series contains a number of target optimizations added during
recent prototype scsi-mq performance profiling in order to avoid
unnecessary se_cmd->t_state_lock acquire/release in fast path I/O
submission and completion code.
All of the target optimization
Hi Bart,
Thanks for the patches. Please share the warnings reported by Coverity, so
that its easy for us to review the patches.
Thanks,
~Saurav
>The current implementation of the qla2xxx Fibre Channel initiator driver
>triggers multiple warnings when analyzed with the Coverity static
>analyzer.
On Thu, 6 June 2013 11:43:54 +0200, Hannes Reinecke wrote:
>
> + spin_lock_irqsave(&sdev->list_lock, flags);
> + list_for_each_entry_safe(scmd, tmp, &sdev->eh_abort_list, eh_entry) {
> + list_del_init(&scmd->eh_entry);
> + spin_unlock_irqrestore(&sdev->list_lock, fl
On Fri, 7 June 2013 14:54:16 +0800, Ren Mingxin wrote:
> On 06/06/2013 05:43 PM, Hannes Reinecke wrote:
>
> This approach may not work for some LLDDs as you said, but I wonder
> whether SAS is applicable(whether there will be later patches for
> SAS).
We have tested a derived patch on SAS and it
With the driver now setup for default 32-bit DMA capabilities, remove the
broken make_local_pdev() implementation and usages.
Signed-off-by: Myron Stowe
---
drivers/scsi/megaraid.c | 98 ---
1 files changed, 16 insertions(+), 82 deletions(-)
diff -
With the driver now setup for default 32-bit DMA capabilities, remove the
64-bit DMA related dead code.
Signed-off-by: Myron Stowe
---
drivers/scsi/megaraid.c | 45 +
drivers/scsi/megaraid.h |1 -
2 files changed, 9 insertions(+), 37 deletions(-
If the megaraid device is 64-bit DMA capable then, once it is setup, any
subsequent DMA allocations for "internal commands" would not be properly
restricted due to megaraid_probe_one() having called pci_set_dma_mask() on
pdev with DMA_BIT_MASK(64). The driver attempts to solve this by using
make_l
While the megaraid device itself may be 64-bit DMA capable, 32-bit address
restricted DMA buffers are apparently required for "internal commands" as
is denoted by a couple of comments - "For all internal commands, the
buffer must be allocated in <4GB address range" - within the driver.
If the devi
On 06/07/2013 08:54 AM, Ren Mingxin wrote:
> Hi, Hannes:
>
> On 06/06/2013 05:43 PM, Hannes Reinecke wrote:
>> this is the first step towards a new non-blocking
>> error handler. This patch implements a new command
>> timeout handler which will be sending command aborts
>> inline without engaging
Hi, Hannes:
On 06/06/2013 05:43 PM, Hannes Reinecke wrote:
this is the first step towards a new non-blocking
error handler. This patch implements a new command
timeout handler which will be sending command aborts
inline without engaging SCSI EH.
In addition the commands will be returned directl
26 matches
Mail list logo
