Hi, this is v11 of the IAA crypto driver, incorporating feedback from
v10.
v11 changes:
- Changed the wq size default from 16 to wq->max_wq_Size in
'dmaengine: idxd: Add support for device/wq defaults and updated
the Documentation to match.
- Added new Reviewed-by and Acked-by tags.
v10 changes:
- Fixed the warnings pointed out in v9 when using 'make W=1 C=1'.
- Added a new patch, 'dmaengine: idxd: Add support for device/wq
defaults' which auto-configures a default set of devices and wqs
so users don't have to do any IAA configuration/setup if they
don't want to. Also updated the Documentation accordingly.
- Removed the 'dmaengine: idxd: add wq driver name support for
accel-config user tool' patch since it's now upstream.
- Added a section (in the Appendix) to
Documentation/driver-api/crypto/iaa/iaa-crypto.rst describing the
sysfs interface.
- While converting my testing scripts to accel-config, realized
accel-config won't allow disabling the wqs if client_count is > 0,
so to address it moved the idxd_wq_get() into iaa_wq_get().
- Fixed a bug in the pr_debug() in wq_table_next_wq(), noticed while
testing.
v9 changes:
- Renamed drv_enable/disable_wq() to idxd_drv_enable/disable_wq()
and exported it, changing all existing callers as well as the
iaa_crypto driver.
- While testing, ran into a use-after-free bug in the irq support
flagged by KASAN so fixed that up in iaa_compress() (added missing
disable_async check).
- Also, while fixing the use-after-free bug, rearranged the out:
part of iaa_desc_complete() to make it cleaner.
- Also for the verify cases, reversed the dma mapping by adding and
calling a new iaa_remap_for_verify() function, since verify
basically does a decompress after reversing the src and dst
buffers.
- Added new Acked-by and Reviewed-by tags.
v8 changes:
- Rebased to current cryptodev tree.
- Added generic-deflate software fallback for decompression
in cases where there's a hardware failure that would
otherwise prevent the data being recovered.
- Changed the driver_name code to use strim().
- Changed the null-destination cases to use sgl_alloc_order() rather
than sgl_alloc().
- Added more Reviewed-by tags.
v7 changes:
- Rebased to current cryptodev tree.
- Removed 'canned' compression mode (deflate-iaa-canned) and fixed
up dependencies in other patches and Documentation.
- Removed op_block checks.
- Removed a stray debugging #ifdef.
- Changed sysfs-driver-dma-idxd driver_name version to 6.6.0.
v6 changes:
- Rebased to current cryptodev tree.
- Changed code to create/register separate algorithms for each
compression mode - one for 'fixed' (deflate-iaa) and one for
'canned' (deflate-iaa-canned).
- Got rid of 'compression_mode' attribute and all the compression
mode code that deals with 'active' compression modes, since
there's no longer a single active compression mode.
- Use crypto_ctx to allow common compress/decompress code to
distinguish between the different compression modes. Also use it
to capture settings such as verify_compress, use_irq, etc. In
addition to being cleaner, this will allow for easier performance
comparisons using different modes/settings.
- Update Documentation and comments to reflect the changes.
- Fixed a bug found by Rex Zhang in decompress_header() which
unmapped src2 rather than src as it should have. Thanks, Rex!
v5 changes:
- Rebased to current cryptodev tree.
- Changed sysfs-driver-dma-idxd driver_name version to 6.5.0.
- Renamed wq private accessor functions to idxd_wq_set/get_private().
v4 changes:
- Added and used DRIVER_NAME_SIZE for wq driver_name.
- Changed all spaces to tabs in CRYPTO_DEV_IAA_CRYPTO_STATS config
menu.
- Removed the private_data void * from wq and replaced with
wq_confdev() instead, as suggested by Dave Jiang.
- Added more Reviewed-by tags.
v3 changes:
- Reworked the code to only allow the registered crypto alg to be
unregistered by removing the module. Also added an iaa_wq_get()
and iaa_wq_put() to take/give up a reference to the work queue
while there are compresses/decompresses in flight. This is
synchronized with the wq remove function, so that the
iaa_wq/iaa_devices can't go away beneath active operations. This
was tested by removing/disabling the iaa wqs/devices while
operations were in flight.
- Simplified the rebalance code and removed cpu_to_iaa() function
since it was overly complicated and wasn't actually working as
advertised.
- As a result of reworking the above code, fixed several bugs such
as possibly unregistering an unregistered crypto alg, a memory
leak where iaa_wqs weren't being freed, and making sure the
compression schemes were registered before registering the driver.
- Added set_/idxd_wq_private() accessors for wq private data.
- Added missing XPORT_SYMBOL_NS_GPL() to [PATCH 04/15] dmaengine:
idxd: Export descriptor management functions
- Added Dave's Reviewed-by: tags from v2.
- Updated Documentation and commit messages to reflect the changes
above.
- Rebased to to cryptodev tree, since that has the earlier changes
that moved the intel drivers to crypto/intel.
v2 changes:
- Removed legacy interface and all related code; merged async
interface into main deflate patch.
- Added support for the null destination case. Thanks to Giovanni
Cabiddu for making me aware of this as well as the selftests for
it.
- Had to do some rearrangement of the code in order to pass all the
selftests. Also added a testcase for 'canned'.
- Moved the iaa crypto driver to drivers/crypto/intel, and moved all
the other intel drivers there as well (which will be posted as a
separate series immediately following this one).
- Added an iaa crypto section to MAINTAINERS.
- Updated the documenation and commit messages to reflect the removal
of the legacy interface.
- Changed kernel version from 6.3.0 to 6.4.0 in patch 01/15 (wq
driver name support)
v1:
This series adds Linux crypto algorithm support for Intel® In-memory
Analytics Accelerator (Intel IAA) [1] hardware compression and
decompression, which is available on Sapphire Rapids systems.
The IAA crypto support is implemented as an IDXD sub-driver. The IDXD
driver already present in the kernel provides discovery and management
of the IAA devices on a system, as well as all the functionality
needed to manage, submit, and wait for completion of work executed on
them. The first 7 patches (patches starting with dmaengine:) add
small bits of underlying IDXD plumbing needed to allow external
sub-drivers to take advantage of this support and claim ownership of
specific IAA devices and workqueues.
The remaining patches add the main support for this feature via the
crypto API, making it transparently accessible to kernel features that
can make use of it such as zswap and zram (patches starting with
crypto – iaa:).
These include both sync/async support for the deflate algorithm
implemented by the IAA hardware, as well as an additional option for
driver statistics and Documentation.
Patch 8 ('[PATCH 08/15] crypto: iaa - Add IAA Compression Accelerator
Documentation') describes the IAA crypto driver in detail; the
following is just a high-level synopsis meant to aid the following
discussion.
The IAA hardware is fairly complex and generally requires a
knowledgeable administrator with sufficiently detailed understanding
of the hardware to set it up before it can be used. As mentioned in
the Documentation, this typically requires using a special tool called
accel-config to enumerate and configure IAA workqueues, engines, etc,
although this can also be done using only sysfs files.
The operation of the driver mirrors this requirement and only allows
the hardware to be accessed via the crypto layer once the hardware has
been configured and bound to the the IAA crypto driver. As an IDXD
sub-driver, the IAA crypto driver essentially takes ownership of the
hardware until it is given up explicitly by the administrator. This
occurs automatically when the administrator enables the first IAA
workqueue or disables the last one; the iaa_crypto (sync and async)
algorithms are registered when the first workqueue is enabled, and
deregistered when the last one is disabled.
The normal sequence of operations would normally be:
< configure the hardware using accel-config or sysfs >
< configure the iaa crypto driver (see below) >
< configure the subsystem e.g. zswap/zram to use the iaa_crypto algo >
< run the workload >
There are a small number of iaa_crypto driver attributes that the
administrator can configure, and which also need to be configured
before the algorithm is enabled:
compression_mode:
The IAA crypto driver supports an extensible interface supporting
any number of different compression modes that can be tailored to
specific types of workloads. These are implemented as tables and
given arbitrary names describing their intent.
There are currently only 2 compression modes, “canned” and “fixed”.
In order to set a compression mode, echo the mode’s name to the
compression_mode driver attribute:
echo "canned" > /sys/bus/dsa/drivers/crypto/compression_mode
There are a few other available iaa_crypto driver attributes (see
Documentation for details) but the main one we want to consider in
detail for now is the ‘sync_mode’ attribute.
The ‘sync_mode’ attribute has 3 possible settings: ‘sync’, ‘async’,
and ‘async_irq’.
The context for these different modes is that although the iaa_crypto
driver implements the asynchronous crypto interface, the async
interface is currently only used in a synchronous way by facilities
like zswap that make use of it.
This is fine for software compress/decompress algorithms, since
there’s no real benefit in being able to use a truly asynchronous
interface with them. This isn’t the case, though, for hardware
compress/decompress engines such as IAA, where truly asynchronous
behavior is beneficial if not completely necessary to make optimal use
of the hardware.
The IAA crypto driver ‘sync_mode’ support should allow facilities such
as zswap to ‘support async (de)compression in some way [2]’ once
they are modified to actually make use of it.
When the ‘async_irq’ sync_mode is specified, the driver sets the bits
in the IAA work descriptor to generate an irq when the work completes.
So for every compression or decompression, the IAA acomp_alg
implementations called by crypto_acomp_compress/decompress() simply
set up the descriptor, turn on the 'request irq' bit and return
immediately with -EINPROGRESS. When the work completes, the irq fires
and the IDXD driver’s irq thread for that irq invokes the callback the
iaa_crypto module registered with IDXD. When the irq thread gets
scheduled, it wakes up the caller, which could be for instance zswap,
waiting synchronously via crypto_wait_req().
Using the simple madvise test program in '[PATCH 08/15] crypto: iaa -
Add IAA Compression Accelerator Documentation' along with a set of
pages from the spec17 benchmark and tracepoint instrumentation
measuring the time taken between the start and end of each compress
and decompress, this case, async_irq, takes on average 6,847 ns for
compression and 5,840 ns for decompression. (See Table 1 below for a
summary of all the tests.)
When sync_mode is set to ‘sync’, the interrupt bit is not set and the
work descriptor is submitted in the same way it was for the previous
case. In this case the call doesn’t return but rather loops around
waiting in the iaa_crypto driver’s check_completion() function which
continually checks the descriptor’s completion bit until it finds it
set to ‘completed’. It then returns to the caller, again for example
zswap waiting in crypto_wait_req(). From the standpoint of zswap,
this case is exactly the same as the previous case, the difference
seen only in the crypto layer and the iaa_crypto driver internally;
from its standpoint they’re both synchronous calls. There is however
a large performance difference: an average of 3,177 ns for compress
and 2,235 ns for decompress.
The final sync_mode is ‘async’. In this case also the interrupt bit
is not set and the work descriptor is submitted, returning immediately
to the caller with -EINPROGRESS. Because there’s no interrupt set to
notify anyone when the work completes, the caller needs to somehow
check for work completion. Because core code like zswap can’t do this
directly by for example calling iaa_crypto’s check_completion(), there
would need to be some changes made to code like zswap and the crypto
layer in order to take advantage of this mode. As such, there are no
numbers to share for this mode.
Finally, just a quick discussion of the remaining numbers in Table 1,
those comparing the iaa_crypto sync and async irq cases to software
deflate. Software deflate took average of 108,978 ns for compress and
14,485 ns for decompress.
As can be seen from Table 1, the numbers using the iaa_crypto driver
for deflate as compared to software are so much better that merging it
would seem to make sense on its own merits. The 'async' sync_mode
described above, however, offers the possibility of even greater gains
to be had against higher-performing algorithms such as lzo, via
parallelization, once the calling facilities are modified to take
advantage of it. Follow-up patchsets to this one will demonstrate
concretely how that might be accomplished.
Thanks,
Tom
Table 1. Zswap latency and compression numbers (in ns):
Algorithm compress decompress
----------------------------------------------------------
iaa sync 3,177 2,235
iaa async irq 6,847 5,840
software deflate 108,978 14,485
[1] https://cdrdv2.intel.com/v1/dl/getContent/721858
[2] https://lore.kernel.org/lkml/[email protected]/
Dave Jiang (1):
dmaengine: idxd: add external module driver support for dsa_bus_type
Tom Zanussi (13):
dmaengine: idxd: Rename drv_enable/disable_wq to
idxd_drv_enable/disable_wq, and export
dmaengine: idxd: Export descriptor management functions
dmaengine: idxd: Export wq resource management functions
dmaengine: idxd: Add wq private data accessors
dmaengine: idxd: add callback support for iaa crypto
crypto: iaa - Add IAA Compression Accelerator Documentation
crypto: iaa - Add Intel IAA Compression Accelerator crypto driver core
crypto: iaa - Add per-cpu workqueue table with rebalancing
crypto: iaa - Add compression mode management along with fixed mode
crypto: iaa - Add support for deflate-iaa compression algorithm
crypto: iaa - Add irq support for the crypto async interface
crypto: iaa - Add IAA Compression Accelerator stats
dmaengine: idxd: Add support for device/wq defaults
.../driver-api/crypto/iaa/iaa-crypto.rst | 824 +++++++
Documentation/driver-api/crypto/iaa/index.rst | 20 +
Documentation/driver-api/crypto/index.rst | 20 +
Documentation/driver-api/index.rst | 1 +
MAINTAINERS | 7 +
crypto/testmgr.c | 10 +
drivers/crypto/intel/Kconfig | 1 +
drivers/crypto/intel/Makefile | 1 +
drivers/crypto/intel/iaa/Kconfig | 19 +
drivers/crypto/intel/iaa/Makefile | 12 +
drivers/crypto/intel/iaa/iaa_crypto.h | 173 ++
.../crypto/intel/iaa/iaa_crypto_comp_fixed.c | 92 +
drivers/crypto/intel/iaa/iaa_crypto_main.c | 2181 +++++++++++++++++
drivers/crypto/intel/iaa/iaa_crypto_stats.c | 313 +++
drivers/crypto/intel/iaa/iaa_crypto_stats.h | 53 +
drivers/dma/idxd/Makefile | 2 +-
drivers/dma/idxd/bus.c | 6 +
drivers/dma/idxd/cdev.c | 6 +-
drivers/dma/idxd/defaults.c | 53 +
drivers/dma/idxd/device.c | 13 +-
drivers/dma/idxd/dma.c | 9 +-
drivers/dma/idxd/idxd.h | 83 +-
drivers/dma/idxd/init.c | 7 +
drivers/dma/idxd/irq.c | 12 +-
drivers/dma/idxd/submit.c | 9 +-
25 files changed, 3896 insertions(+), 31 deletions(-)
create mode 100644 Documentation/driver-api/crypto/iaa/iaa-crypto.rst
create mode 100644 Documentation/driver-api/crypto/iaa/index.rst
create mode 100644 Documentation/driver-api/crypto/index.rst
create mode 100644 drivers/crypto/intel/iaa/Kconfig
create mode 100644 drivers/crypto/intel/iaa/Makefile
create mode 100644 drivers/crypto/intel/iaa/iaa_crypto.h
create mode 100644 drivers/crypto/intel/iaa/iaa_crypto_comp_fixed.c
create mode 100644 drivers/crypto/intel/iaa/iaa_crypto_main.c
create mode 100644 drivers/crypto/intel/iaa/iaa_crypto_stats.c
create mode 100644 drivers/crypto/intel/iaa/iaa_crypto_stats.h
create mode 100644 drivers/dma/idxd/defaults.c
--
2.34.1
Add support for optional debugfs statistics support for the IAA
Compression Accelerator. This is enabled by the kernel config item:
CRYPTO_DEV_IAA_CRYPTO_STATS
When enabled, the IAA crypto driver will generate statistics which can
be accessed at /sys/kernel/debug/iaa-crypto/.
See Documentation/driver-api/crypto/iax/iax-crypto.rst for details.
Signed-off-by: Tom Zanussi <[email protected]>
---
drivers/crypto/intel/iaa/Kconfig | 9 +
drivers/crypto/intel/iaa/Makefile | 2 +
drivers/crypto/intel/iaa/iaa_crypto.h | 13 +
drivers/crypto/intel/iaa/iaa_crypto_main.c | 39 ++-
drivers/crypto/intel/iaa/iaa_crypto_stats.c | 313 ++++++++++++++++++++
drivers/crypto/intel/iaa/iaa_crypto_stats.h | 53 ++++
6 files changed, 427 insertions(+), 2 deletions(-)
create mode 100644 drivers/crypto/intel/iaa/iaa_crypto_stats.c
create mode 100644 drivers/crypto/intel/iaa/iaa_crypto_stats.h
diff --git a/drivers/crypto/intel/iaa/Kconfig b/drivers/crypto/intel/iaa/Kconfig
index fcccb6ff7e29..d53f4b1d494f 100644
--- a/drivers/crypto/intel/iaa/Kconfig
+++ b/drivers/crypto/intel/iaa/Kconfig
@@ -8,3 +8,12 @@ config CRYPTO_DEV_IAA_CRYPTO
decompression with the Intel Analytics Accelerator (IAA)
hardware using the cryptographic API. If you choose 'M'
here, the module will be called iaa_crypto.
+
+config CRYPTO_DEV_IAA_CRYPTO_STATS
+ bool "Enable Intel(R) IAA Compression Accelerator Statistics"
+ depends on CRYPTO_DEV_IAA_CRYPTO
+ default n
+ help
+ Enable statistics for the IAA compression accelerator.
+ These include per-device and per-workqueue statistics in
+ addition to global driver statistics.
diff --git a/drivers/crypto/intel/iaa/Makefile b/drivers/crypto/intel/iaa/Makefile
index cc87feffd059..b64b208d2344 100644
--- a/drivers/crypto/intel/iaa/Makefile
+++ b/drivers/crypto/intel/iaa/Makefile
@@ -8,3 +8,5 @@ ccflags-y += -I $(srctree)/drivers/dma/idxd -DDEFAULT_SYMBOL_NAMESPACE=IDXD
obj-$(CONFIG_CRYPTO_DEV_IAA_CRYPTO) := iaa_crypto.o
iaa_crypto-y := iaa_crypto_main.o iaa_crypto_comp_fixed.o
+
+iaa_crypto-$(CONFIG_CRYPTO_DEV_IAA_CRYPTO_STATS) += iaa_crypto_stats.o
diff --git a/drivers/crypto/intel/iaa/iaa_crypto.h b/drivers/crypto/intel/iaa/iaa_crypto.h
index de014ac53adb..014420f7beb0 100644
--- a/drivers/crypto/intel/iaa/iaa_crypto.h
+++ b/drivers/crypto/intel/iaa/iaa_crypto.h
@@ -48,6 +48,11 @@ struct iaa_wq {
bool remove;
struct iaa_device *iaa_device;
+
+ u64 comp_calls;
+ u64 comp_bytes;
+ u64 decomp_calls;
+ u64 decomp_bytes;
};
struct iaa_device_compression_mode {
@@ -69,6 +74,11 @@ struct iaa_device {
int n_wq;
struct list_head wqs;
+
+ u64 comp_calls;
+ u64 comp_bytes;
+ u64 decomp_calls;
+ u64 decomp_bytes;
};
struct wq_table_entry {
@@ -157,4 +167,7 @@ struct iaa_compression_ctx {
bool use_irq;
};
+extern struct list_head iaa_devices;
+extern struct mutex iaa_devices_lock;
+
#endif
diff --git a/drivers/crypto/intel/iaa/iaa_crypto_main.c b/drivers/crypto/intel/iaa/iaa_crypto_main.c
index 9dd52e1ea5ca..dbab926a342f 100644
--- a/drivers/crypto/intel/iaa/iaa_crypto_main.c
+++ b/drivers/crypto/intel/iaa/iaa_crypto_main.c
@@ -14,6 +14,7 @@
#include "idxd.h"
#include "iaa_crypto.h"
+#include "iaa_crypto_stats.h"
#ifdef pr_fmt
#undef pr_fmt
@@ -85,8 +86,8 @@ static void wq_table_clear_entry(int cpu)
memset(entry->wqs, 0, entry->max_wqs * sizeof(struct idxd_wq *));
}
-static LIST_HEAD(iaa_devices);
-static DEFINE_MUTEX(iaa_devices_lock);
+LIST_HEAD(iaa_devices);
+DEFINE_MUTEX(iaa_devices_lock);
/* If enabled, IAA hw crypto algos are registered, unavailable otherwise */
static bool iaa_crypto_enabled;
@@ -1051,6 +1052,7 @@ static inline int check_completion(struct device *dev,
ret = -ETIMEDOUT;
dev_dbg(dev, "%s timed out, size=0x%x\n",
op_str, comp->output_size);
+ update_completion_timeout_errs();
goto out;
}
@@ -1060,6 +1062,7 @@ static inline int check_completion(struct device *dev,
dev_dbg(dev, "compressed > uncompressed size,"
" not compressing, size=0x%x\n",
comp->output_size);
+ update_completion_comp_buf_overflow_errs();
goto out;
}
@@ -1072,6 +1075,7 @@ static inline int check_completion(struct device *dev,
dev_dbg(dev, "iaa %s status=0x%x, error=0x%x, size=0x%x\n",
op_str, comp->status, comp->error_code, comp->output_size);
print_hex_dump(KERN_INFO, "cmp-rec: ", DUMP_PREFIX_OFFSET, 8, 1, comp, 64, 0);
+ update_completion_einval_errs();
goto out;
}
@@ -1093,6 +1097,8 @@ static int deflate_generic_decompress(struct acomp_req *req)
kunmap_local(src);
kunmap_local(dst);
+ update_total_sw_decomp_calls();
+
return ret;
}
@@ -1160,6 +1166,15 @@ static void iaa_desc_complete(struct idxd_desc *idxd_desc,
ctx->req->dlen = idxd_desc->iax_completion->output_size;
}
+ /* Update stats */
+ if (ctx->compress) {
+ update_total_comp_bytes_out(ctx->req->dlen);
+ update_wq_comp_bytes(iaa_wq->wq, ctx->req->dlen);
+ } else {
+ update_total_decomp_bytes_in(ctx->req->dlen);
+ update_wq_decomp_bytes(iaa_wq->wq, ctx->req->dlen);
+ }
+
if (ctx->compress && compression_ctx->verify_compress) {
dma_addr_t src_addr, dst_addr;
u32 compression_crc;
@@ -1279,6 +1294,10 @@ static int iaa_compress(struct crypto_tfm *tfm, struct acomp_req *req,
goto err;
}
+ /* Update stats */
+ update_total_comp_calls();
+ update_wq_comp_calls(wq);
+
if (ctx->async_mode && !disable_async) {
ret = -EINPROGRESS;
dev_dbg(dev, "%s: returning -EINPROGRESS\n", __func__);
@@ -1293,6 +1312,10 @@ static int iaa_compress(struct crypto_tfm *tfm, struct acomp_req *req,
*dlen = idxd_desc->iax_completion->output_size;
+ /* Update stats */
+ update_total_comp_bytes_out(*dlen);
+ update_wq_comp_bytes(wq, *dlen);
+
*compression_crc = idxd_desc->iax_completion->crc;
if (!ctx->async_mode)
@@ -1509,6 +1532,10 @@ static int iaa_decompress(struct crypto_tfm *tfm, struct acomp_req *req,
goto err;
}
+ /* Update stats */
+ update_total_decomp_calls();
+ update_wq_decomp_calls(wq);
+
if (ctx->async_mode && !disable_async) {
ret = -EINPROGRESS;
dev_dbg(dev, "%s: returning -EINPROGRESS\n", __func__);
@@ -1539,6 +1566,10 @@ static int iaa_decompress(struct crypto_tfm *tfm, struct acomp_req *req,
if (!ctx->async_mode)
idxd_free_desc(wq, idxd_desc);
+
+ /* Update stats */
+ update_total_decomp_bytes_in(slen);
+ update_wq_decomp_bytes(wq, slen);
out:
return ret;
err:
@@ -2103,6 +2134,9 @@ static int __init iaa_crypto_init_module(void)
goto err_sync_attr_create;
}
+ if (iaa_crypto_debugfs_init())
+ pr_warn("debugfs init failed, stats not available\n");
+
pr_debug("initialized\n");
out:
return ret;
@@ -2125,6 +2159,7 @@ static void __exit iaa_crypto_cleanup_module(void)
if (iaa_unregister_compression_device())
pr_debug("IAA compression device unregister failed\n");
+ iaa_crypto_debugfs_cleanup();
driver_remove_file(&iaa_crypto_driver.drv,
&driver_attr_sync_mode);
driver_remove_file(&iaa_crypto_driver.drv,
diff --git a/drivers/crypto/intel/iaa/iaa_crypto_stats.c b/drivers/crypto/intel/iaa/iaa_crypto_stats.c
new file mode 100644
index 000000000000..0279edc6194e
--- /dev/null
+++ b/drivers/crypto/intel/iaa/iaa_crypto_stats.c
@@ -0,0 +1,313 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright(c) 2021 Intel Corporation. All rights rsvd. */
+
+#include <linux/module.h>
+#include <linux/kernel.h>
+#include <linux/highmem.h>
+#include <linux/mm.h>
+#include <linux/slab.h>
+#include <linux/delay.h>
+#include <linux/smp.h>
+#include <uapi/linux/idxd.h>
+#include <linux/idxd.h>
+#include <linux/dmaengine.h>
+#include "../../dma/idxd/idxd.h"
+#include <linux/debugfs.h>
+#include <crypto/internal/acompress.h>
+#include "iaa_crypto.h"
+#include "iaa_crypto_stats.h"
+
+static u64 total_comp_calls;
+static u64 total_decomp_calls;
+static u64 total_sw_decomp_calls;
+static u64 max_comp_delay_ns;
+static u64 max_decomp_delay_ns;
+static u64 max_acomp_delay_ns;
+static u64 max_adecomp_delay_ns;
+static u64 total_comp_bytes_out;
+static u64 total_decomp_bytes_in;
+static u64 total_completion_einval_errors;
+static u64 total_completion_timeout_errors;
+static u64 total_completion_comp_buf_overflow_errors;
+
+static struct dentry *iaa_crypto_debugfs_root;
+
+void update_total_comp_calls(void)
+{
+ total_comp_calls++;
+}
+
+void update_total_comp_bytes_out(int n)
+{
+ total_comp_bytes_out += n;
+}
+
+void update_total_decomp_calls(void)
+{
+ total_decomp_calls++;
+}
+
+void update_total_sw_decomp_calls(void)
+{
+ total_sw_decomp_calls++;
+}
+
+void update_total_decomp_bytes_in(int n)
+{
+ total_decomp_bytes_in += n;
+}
+
+void update_completion_einval_errs(void)
+{
+ total_completion_einval_errors++;
+}
+
+void update_completion_timeout_errs(void)
+{
+ total_completion_timeout_errors++;
+}
+
+void update_completion_comp_buf_overflow_errs(void)
+{
+ total_completion_comp_buf_overflow_errors++;
+}
+
+void update_max_comp_delay_ns(u64 start_time_ns)
+{
+ u64 time_diff;
+
+ time_diff = ktime_get_ns() - start_time_ns;
+
+ if (time_diff > max_comp_delay_ns)
+ max_comp_delay_ns = time_diff;
+}
+
+void update_max_decomp_delay_ns(u64 start_time_ns)
+{
+ u64 time_diff;
+
+ time_diff = ktime_get_ns() - start_time_ns;
+
+ if (time_diff > max_decomp_delay_ns)
+ max_decomp_delay_ns = time_diff;
+}
+
+void update_max_acomp_delay_ns(u64 start_time_ns)
+{
+ u64 time_diff;
+
+ time_diff = ktime_get_ns() - start_time_ns;
+
+ if (time_diff > max_acomp_delay_ns)
+ max_acomp_delay_ns = time_diff;
+}
+
+void update_max_adecomp_delay_ns(u64 start_time_ns)
+{
+ u64 time_diff;
+
+ time_diff = ktime_get_ns() - start_time_ns;
+
+ if (time_diff > max_adecomp_delay_ns)
+
+ max_adecomp_delay_ns = time_diff;
+}
+
+void update_wq_comp_calls(struct idxd_wq *idxd_wq)
+{
+ struct iaa_wq *wq = idxd_wq_get_private(idxd_wq);
+
+ wq->comp_calls++;
+ wq->iaa_device->comp_calls++;
+}
+
+void update_wq_comp_bytes(struct idxd_wq *idxd_wq, int n)
+{
+ struct iaa_wq *wq = idxd_wq_get_private(idxd_wq);
+
+ wq->comp_bytes += n;
+ wq->iaa_device->comp_bytes += n;
+}
+
+void update_wq_decomp_calls(struct idxd_wq *idxd_wq)
+{
+ struct iaa_wq *wq = idxd_wq_get_private(idxd_wq);
+
+ wq->decomp_calls++;
+ wq->iaa_device->decomp_calls++;
+}
+
+void update_wq_decomp_bytes(struct idxd_wq *idxd_wq, int n)
+{
+ struct iaa_wq *wq = idxd_wq_get_private(idxd_wq);
+
+ wq->decomp_bytes += n;
+ wq->iaa_device->decomp_bytes += n;
+}
+
+static void reset_iaa_crypto_stats(void)
+{
+ total_comp_calls = 0;
+ total_decomp_calls = 0;
+ total_sw_decomp_calls = 0;
+ max_comp_delay_ns = 0;
+ max_decomp_delay_ns = 0;
+ max_acomp_delay_ns = 0;
+ max_adecomp_delay_ns = 0;
+ total_comp_bytes_out = 0;
+ total_decomp_bytes_in = 0;
+ total_completion_einval_errors = 0;
+ total_completion_timeout_errors = 0;
+ total_completion_comp_buf_overflow_errors = 0;
+}
+
+static void reset_wq_stats(struct iaa_wq *wq)
+{
+ wq->comp_calls = 0;
+ wq->comp_bytes = 0;
+ wq->decomp_calls = 0;
+ wq->decomp_bytes = 0;
+}
+
+static void reset_device_stats(struct iaa_device *iaa_device)
+{
+ struct iaa_wq *iaa_wq;
+
+ iaa_device->comp_calls = 0;
+ iaa_device->comp_bytes = 0;
+ iaa_device->decomp_calls = 0;
+ iaa_device->decomp_bytes = 0;
+
+ list_for_each_entry(iaa_wq, &iaa_device->wqs, list)
+ reset_wq_stats(iaa_wq);
+}
+
+static void wq_show(struct seq_file *m, struct iaa_wq *iaa_wq)
+{
+ seq_printf(m, " name: %s\n", iaa_wq->wq->name);
+ seq_printf(m, " comp_calls: %llu\n", iaa_wq->comp_calls);
+ seq_printf(m, " comp_bytes: %llu\n", iaa_wq->comp_bytes);
+ seq_printf(m, " decomp_calls: %llu\n", iaa_wq->decomp_calls);
+ seq_printf(m, " decomp_bytes: %llu\n\n", iaa_wq->decomp_bytes);
+}
+
+static void device_stats_show(struct seq_file *m, struct iaa_device *iaa_device)
+{
+ struct iaa_wq *iaa_wq;
+
+ seq_puts(m, "iaa device:\n");
+ seq_printf(m, " id: %d\n", iaa_device->idxd->id);
+ seq_printf(m, " n_wqs: %d\n", iaa_device->n_wq);
+ seq_printf(m, " comp_calls: %llu\n", iaa_device->comp_calls);
+ seq_printf(m, " comp_bytes: %llu\n", iaa_device->comp_bytes);
+ seq_printf(m, " decomp_calls: %llu\n", iaa_device->decomp_calls);
+ seq_printf(m, " decomp_bytes: %llu\n", iaa_device->decomp_bytes);
+ seq_puts(m, " wqs:\n");
+
+ list_for_each_entry(iaa_wq, &iaa_device->wqs, list)
+ wq_show(m, iaa_wq);
+}
+
+static void global_stats_show(struct seq_file *m)
+{
+ seq_puts(m, "global stats:\n");
+ seq_printf(m, " total_comp_calls: %llu\n", total_comp_calls);
+ seq_printf(m, " total_decomp_calls: %llu\n", total_decomp_calls);
+ seq_printf(m, " total_sw_decomp_calls: %llu\n", total_sw_decomp_calls);
+ seq_printf(m, " total_comp_bytes_out: %llu\n", total_comp_bytes_out);
+ seq_printf(m, " total_decomp_bytes_in: %llu\n", total_decomp_bytes_in);
+ seq_printf(m, " total_completion_einval_errors: %llu\n",
+ total_completion_einval_errors);
+ seq_printf(m, " total_completion_timeout_errors: %llu\n",
+ total_completion_timeout_errors);
+ seq_printf(m, " total_completion_comp_buf_overflow_errors: %llu\n\n",
+ total_completion_comp_buf_overflow_errors);
+}
+
+static int wq_stats_show(struct seq_file *m, void *v)
+{
+ struct iaa_device *iaa_device;
+
+ mutex_lock(&iaa_devices_lock);
+
+ global_stats_show(m);
+
+ list_for_each_entry(iaa_device, &iaa_devices, list)
+ device_stats_show(m, iaa_device);
+
+ mutex_unlock(&iaa_devices_lock);
+
+ return 0;
+}
+
+static int iaa_crypto_stats_reset(void *data, u64 value)
+{
+ struct iaa_device *iaa_device;
+
+ reset_iaa_crypto_stats();
+
+ mutex_lock(&iaa_devices_lock);
+
+ list_for_each_entry(iaa_device, &iaa_devices, list)
+ reset_device_stats(iaa_device);
+
+ mutex_unlock(&iaa_devices_lock);
+
+ return 0;
+}
+
+static int wq_stats_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, wq_stats_show, file);
+}
+
+static const struct file_operations wq_stats_fops = {
+ .open = wq_stats_open,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .release = single_release,
+};
+
+DEFINE_DEBUGFS_ATTRIBUTE(wq_stats_reset_fops, NULL, iaa_crypto_stats_reset, "%llu\n");
+
+int __init iaa_crypto_debugfs_init(void)
+{
+ if (!debugfs_initialized())
+ return -ENODEV;
+
+ iaa_crypto_debugfs_root = debugfs_create_dir("iaa_crypto", NULL);
+ if (!iaa_crypto_debugfs_root)
+ return -ENOMEM;
+
+ debugfs_create_u64("max_comp_delay_ns", 0644,
+ iaa_crypto_debugfs_root, &max_comp_delay_ns);
+ debugfs_create_u64("max_decomp_delay_ns", 0644,
+ iaa_crypto_debugfs_root, &max_decomp_delay_ns);
+ debugfs_create_u64("max_acomp_delay_ns", 0644,
+ iaa_crypto_debugfs_root, &max_comp_delay_ns);
+ debugfs_create_u64("max_adecomp_delay_ns", 0644,
+ iaa_crypto_debugfs_root, &max_decomp_delay_ns);
+ debugfs_create_u64("total_comp_calls", 0644,
+ iaa_crypto_debugfs_root, &total_comp_calls);
+ debugfs_create_u64("total_decomp_calls", 0644,
+ iaa_crypto_debugfs_root, &total_decomp_calls);
+ debugfs_create_u64("total_sw_decomp_calls", 0644,
+ iaa_crypto_debugfs_root, &total_sw_decomp_calls);
+ debugfs_create_u64("total_comp_bytes_out", 0644,
+ iaa_crypto_debugfs_root, &total_comp_bytes_out);
+ debugfs_create_u64("total_decomp_bytes_in", 0644,
+ iaa_crypto_debugfs_root, &total_decomp_bytes_in);
+ debugfs_create_file("wq_stats", 0644, iaa_crypto_debugfs_root, NULL,
+ &wq_stats_fops);
+ debugfs_create_file("stats_reset", 0644, iaa_crypto_debugfs_root, NULL,
+ &wq_stats_reset_fops);
+
+ return 0;
+}
+
+void __exit iaa_crypto_debugfs_cleanup(void)
+{
+ debugfs_remove_recursive(iaa_crypto_debugfs_root);
+}
+
+MODULE_LICENSE("GPL");
diff --git a/drivers/crypto/intel/iaa/iaa_crypto_stats.h b/drivers/crypto/intel/iaa/iaa_crypto_stats.h
new file mode 100644
index 000000000000..c10b87b86fa4
--- /dev/null
+++ b/drivers/crypto/intel/iaa/iaa_crypto_stats.h
@@ -0,0 +1,53 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+/* Copyright(c) 2021 Intel Corporation. All rights rsvd. */
+
+#ifndef __CRYPTO_DEV_IAA_CRYPTO_STATS_H__
+#define __CRYPTO_DEV_IAA_CRYPTO_STATS_H__
+
+#if defined(CONFIG_CRYPTO_DEV_IAA_CRYPTO_STATS)
+int iaa_crypto_debugfs_init(void);
+void iaa_crypto_debugfs_cleanup(void);
+
+void update_total_comp_calls(void);
+void update_total_comp_bytes_out(int n);
+void update_total_decomp_calls(void);
+void update_total_sw_decomp_calls(void);
+void update_total_decomp_bytes_in(int n);
+void update_max_comp_delay_ns(u64 start_time_ns);
+void update_max_decomp_delay_ns(u64 start_time_ns);
+void update_max_acomp_delay_ns(u64 start_time_ns);
+void update_max_adecomp_delay_ns(u64 start_time_ns);
+void update_completion_einval_errs(void);
+void update_completion_timeout_errs(void);
+void update_completion_comp_buf_overflow_errs(void);
+
+void update_wq_comp_calls(struct idxd_wq *idxd_wq);
+void update_wq_comp_bytes(struct idxd_wq *idxd_wq, int n);
+void update_wq_decomp_calls(struct idxd_wq *idxd_wq);
+void update_wq_decomp_bytes(struct idxd_wq *idxd_wq, int n);
+
+#else
+static inline int iaa_crypto_debugfs_init(void) { return 0; }
+static inline void iaa_crypto_debugfs_cleanup(void) {}
+
+static inline void update_total_comp_calls(void) {}
+static inline void update_total_comp_bytes_out(int n) {}
+static inline void update_total_decomp_calls(void) {}
+static inline void update_total_sw_decomp_calls(void) {}
+static inline void update_total_decomp_bytes_in(int n) {}
+static inline void update_max_comp_delay_ns(u64 start_time_ns) {}
+static inline void update_max_decomp_delay_ns(u64 start_time_ns) {}
+static inline void update_max_acomp_delay_ns(u64 start_time_ns) {}
+static inline void update_max_adecomp_delay_ns(u64 start_time_ns) {}
+static inline void update_completion_einval_errs(void) {}
+static inline void update_completion_timeout_errs(void) {}
+static inline void update_completion_comp_buf_overflow_errs(void) {}
+
+static inline void update_wq_comp_calls(struct idxd_wq *idxd_wq) {}
+static inline void update_wq_comp_bytes(struct idxd_wq *idxd_wq, int n) {}
+static inline void update_wq_decomp_calls(struct idxd_wq *idxd_wq) {}
+static inline void update_wq_decomp_bytes(struct idxd_wq *idxd_wq, int n) {}
+
+#endif // CONFIG_CRYPTO_DEV_IAA_CRYPTO_STATS
+
+#endif
--
2.34.1
This patch registers the deflate-iaa deflate compression algorithm and
hooks it up to the IAA hardware using the 'fixed' compression mode
introduced in the previous patch.
Because the IAA hardware has a 4k history-window limitation, only
buffers <= 4k, or that have been compressed using a <= 4k history
window, are technically compliant with the deflate spec, which allows
for a window of up to 32k. Because of this limitation, the IAA fixed
mode deflate algorithm is given its own algorithm name, 'deflate-iaa'.
With this change, the deflate-iaa crypto algorithm is registered and
operational, and compression and decompression operations are fully
enabled following the successful binding of the first IAA workqueue
to the iaa_crypto sub-driver.
when there are no IAA workqueues bound to the driver, the IAA crypto
algorithm can be unregistered by removing the module.
A new iaa_crypto 'verify_compress' driver attribute is also added,
allowing the user to toggle compression verification. If set, each
compress will be internally decompressed and the contents verified,
returning error codes if unsuccessful. This can be toggled with 0/1:
echo 0 > /sys/bus/dsa/drivers/crypto/verify_compress
The default setting is '1' - verify all compresses.
The verify_compress value setting at the time the algorithm is
registered is captured in the algorithm's crypto_ctx and used for all
compresses when using the algorithm.
[ Based on work originally by George Powley, Jing Lin and Kyung Min
Park ]
Signed-off-by: Tom Zanussi <[email protected]>
---
crypto/testmgr.c | 10 +
drivers/crypto/intel/iaa/iaa_crypto.h | 36 +
drivers/crypto/intel/iaa/iaa_crypto_main.c | 1050 +++++++++++++++++++-
3 files changed, 1078 insertions(+), 18 deletions(-)
diff --git a/crypto/testmgr.c b/crypto/testmgr.c
index 59f0540d442e..273f294806d7 100644
--- a/crypto/testmgr.c
+++ b/crypto/testmgr.c
@@ -4815,6 +4815,16 @@ static const struct alg_test_desc alg_test_descs[] = {
.decomp = __VECS(deflate_decomp_tv_template)
}
}
+ }, {
+ .alg = "deflate-iaa",
+ .test = alg_test_comp,
+ .fips_allowed = 1,
+ .suite = {
+ .comp = {
+ .comp = __VECS(deflate_comp_tv_template),
+ .decomp = __VECS(deflate_decomp_tv_template)
+ }
+ }
}, {
.alg = "dh",
.test = alg_test_kpp,
diff --git a/drivers/crypto/intel/iaa/iaa_crypto.h b/drivers/crypto/intel/iaa/iaa_crypto.h
index 33e68f9d3d02..4c6b0f5a6b50 100644
--- a/drivers/crypto/intel/iaa/iaa_crypto.h
+++ b/drivers/crypto/intel/iaa/iaa_crypto.h
@@ -10,15 +10,42 @@
#define IDXD_SUBDRIVER_NAME "crypto"
+#define IAA_DECOMP_ENABLE BIT(0)
+#define IAA_DECOMP_FLUSH_OUTPUT BIT(1)
+#define IAA_DECOMP_CHECK_FOR_EOB BIT(2)
+#define IAA_DECOMP_STOP_ON_EOB BIT(3)
+#define IAA_DECOMP_SUPPRESS_OUTPUT BIT(9)
+
+#define IAA_COMP_FLUSH_OUTPUT BIT(1)
+#define IAA_COMP_APPEND_EOB BIT(2)
+
+#define IAA_COMPLETION_TIMEOUT 1000000
+
+#define IAA_ANALYTICS_ERROR 0x0a
+#define IAA_ERROR_DECOMP_BUF_OVERFLOW 0x0b
+#define IAA_ERROR_COMP_BUF_OVERFLOW 0x19
+#define IAA_ERROR_WATCHDOG_EXPIRED 0x24
+
#define IAA_COMP_MODES_MAX 2
#define FIXED_HDR 0x2
#define FIXED_HDR_SIZE 3
+#define IAA_COMP_FLAGS (IAA_COMP_FLUSH_OUTPUT | \
+ IAA_COMP_APPEND_EOB)
+
+#define IAA_DECOMP_FLAGS (IAA_DECOMP_ENABLE | \
+ IAA_DECOMP_FLUSH_OUTPUT | \
+ IAA_DECOMP_CHECK_FOR_EOB | \
+ IAA_DECOMP_STOP_ON_EOB)
+
/* Representation of IAA workqueue */
struct iaa_wq {
struct list_head list;
+
struct idxd_wq *wq;
+ int ref;
+ bool remove;
struct iaa_device *iaa_device;
};
@@ -119,4 +146,13 @@ int add_iaa_compression_mode(const char *name,
void remove_iaa_compression_mode(const char *name);
+enum iaa_mode {
+ IAA_MODE_FIXED,
+};
+
+struct iaa_compression_ctx {
+ enum iaa_mode mode;
+ bool verify_compress;
+};
+
#endif
diff --git a/drivers/crypto/intel/iaa/iaa_crypto_main.c b/drivers/crypto/intel/iaa/iaa_crypto_main.c
index 4ec7a9269243..6e2a8826ecfd 100644
--- a/drivers/crypto/intel/iaa/iaa_crypto_main.c
+++ b/drivers/crypto/intel/iaa/iaa_crypto_main.c
@@ -10,6 +10,7 @@
#include <uapi/linux/idxd.h>
#include <linux/highmem.h>
#include <linux/sched/smt.h>
+#include <crypto/internal/acompress.h>
#include "idxd.h"
#include "iaa_crypto.h"
@@ -20,6 +21,8 @@
#define pr_fmt(fmt) "idxd: " IDXD_SUBDRIVER_NAME ": " fmt
+#define IAA_ALG_PRIORITY 300
+
/* number of iaa instances probed */
static unsigned int nr_iaa;
static unsigned int nr_cpus;
@@ -29,9 +32,28 @@ static unsigned int nr_cpus_per_node;
/* Number of physical cpus sharing each iaa instance */
static unsigned int cpus_per_iaa;
+static struct crypto_comp *deflate_generic_tfm;
+
/* Per-cpu lookup table for balanced wqs */
static struct wq_table_entry __percpu *wq_table;
+static struct idxd_wq *wq_table_next_wq(int cpu)
+{
+ struct wq_table_entry *entry = per_cpu_ptr(wq_table, cpu);
+
+ if (++entry->cur_wq >= entry->n_wqs)
+ entry->cur_wq = 0;
+
+ if (!entry->wqs[entry->cur_wq])
+ return NULL;
+
+ pr_debug("%s: returning wq at idx %d (iaa wq %d.%d) from cpu %d\n", __func__,
+ entry->cur_wq, entry->wqs[entry->cur_wq]->idxd->id,
+ entry->wqs[entry->cur_wq]->id, cpu);
+
+ return entry->wqs[entry->cur_wq];
+}
+
static void wq_table_add(int cpu, struct idxd_wq *wq)
{
struct wq_table_entry *entry = per_cpu_ptr(wq_table, cpu);
@@ -66,6 +88,40 @@ static void wq_table_clear_entry(int cpu)
static LIST_HEAD(iaa_devices);
static DEFINE_MUTEX(iaa_devices_lock);
+/* If enabled, IAA hw crypto algos are registered, unavailable otherwise */
+static bool iaa_crypto_enabled;
+static bool iaa_crypto_registered;
+
+/* Verify results of IAA compress or not */
+static bool iaa_verify_compress = true;
+
+static ssize_t verify_compress_show(struct device_driver *driver, char *buf)
+{
+ return sprintf(buf, "%d\n", iaa_verify_compress);
+}
+
+static ssize_t verify_compress_store(struct device_driver *driver,
+ const char *buf, size_t count)
+{
+ int ret = -EBUSY;
+
+ mutex_lock(&iaa_devices_lock);
+
+ if (iaa_crypto_enabled)
+ goto out;
+
+ ret = kstrtobool(buf, &iaa_verify_compress);
+ if (ret)
+ goto out;
+
+ ret = count;
+out:
+ mutex_unlock(&iaa_devices_lock);
+
+ return ret;
+}
+static DRIVER_ATTR_RW(verify_compress);
+
static struct iaa_compression_mode *iaa_compression_modes[IAA_COMP_MODES_MAX];
static int find_empty_iaa_compression_mode(void)
@@ -250,6 +306,12 @@ int add_iaa_compression_mode(const char *name,
}
EXPORT_SYMBOL_GPL(add_iaa_compression_mode);
+static struct iaa_device_compression_mode *
+get_iaa_device_compression_mode(struct iaa_device *iaa_device, int idx)
+{
+ return iaa_device->compression_modes[idx];
+}
+
static void free_device_compression_mode(struct iaa_device *iaa_device,
struct iaa_device_compression_mode *device_mode)
{
@@ -268,6 +330,86 @@ static void free_device_compression_mode(struct iaa_device *iaa_device,
kfree(device_mode);
}
+#define IDXD_OP_FLAG_AECS_RW_TGLS 0x400000
+#define IAX_AECS_DEFAULT_FLAG (IDXD_OP_FLAG_CRAV | IDXD_OP_FLAG_RCR | IDXD_OP_FLAG_CC)
+#define IAX_AECS_COMPRESS_FLAG (IAX_AECS_DEFAULT_FLAG | IDXD_OP_FLAG_RD_SRC2_AECS)
+#define IAX_AECS_DECOMPRESS_FLAG (IAX_AECS_DEFAULT_FLAG | IDXD_OP_FLAG_RD_SRC2_AECS)
+#define IAX_AECS_GEN_FLAG (IAX_AECS_DEFAULT_FLAG | \
+ IDXD_OP_FLAG_WR_SRC2_AECS_COMP | \
+ IDXD_OP_FLAG_AECS_RW_TGLS)
+
+static int check_completion(struct device *dev,
+ struct iax_completion_record *comp,
+ bool compress,
+ bool only_once);
+
+static int decompress_header(struct iaa_device_compression_mode *device_mode,
+ struct iaa_compression_mode *mode,
+ struct idxd_wq *wq)
+{
+ dma_addr_t src_addr, src2_addr;
+ struct idxd_desc *idxd_desc;
+ struct iax_hw_desc *desc;
+ struct device *dev;
+ int ret = 0;
+
+ idxd_desc = idxd_alloc_desc(wq, IDXD_OP_BLOCK);
+ if (IS_ERR(idxd_desc))
+ return PTR_ERR(idxd_desc);
+
+ desc = idxd_desc->iax_hw;
+
+ dev = &wq->idxd->pdev->dev;
+
+ src_addr = dma_map_single(dev, (void *)mode->header_table,
+ mode->header_table_size, DMA_TO_DEVICE);
+ dev_dbg(dev, "%s: mode->name %s, src_addr %llx, dev %p, src %p, slen %d\n",
+ __func__, mode->name, src_addr, dev,
+ mode->header_table, mode->header_table_size);
+ if (unlikely(dma_mapping_error(dev, src_addr))) {
+ dev_dbg(dev, "dma_map_single err, exiting\n");
+ ret = -ENOMEM;
+ return ret;
+ }
+
+ desc->flags = IAX_AECS_GEN_FLAG;
+ desc->opcode = IAX_OPCODE_DECOMPRESS;
+
+ desc->src1_addr = (u64)src_addr;
+ desc->src1_size = mode->header_table_size;
+
+ src2_addr = device_mode->aecs_decomp_table_dma_addr;
+ desc->src2_addr = (u64)src2_addr;
+ desc->src2_size = 1088;
+ dev_dbg(dev, "%s: mode->name %s, src2_addr %llx, dev %p, src2_size %d\n",
+ __func__, mode->name, desc->src2_addr, dev, desc->src2_size);
+ desc->max_dst_size = 0; // suppressed output
+
+ desc->decompr_flags = mode->gen_decomp_table_flags;
+
+ desc->priv = 1;
+
+ desc->completion_addr = idxd_desc->compl_dma;
+
+ ret = idxd_submit_desc(wq, idxd_desc);
+ if (ret) {
+ pr_err("%s: submit_desc failed ret=0x%x\n", __func__, ret);
+ goto out;
+ }
+
+ ret = check_completion(dev, idxd_desc->iax_completion, false, false);
+ if (ret)
+ dev_dbg(dev, "%s: mode->name %s check_completion failed ret=%d\n",
+ __func__, mode->name, ret);
+ else
+ dev_dbg(dev, "%s: mode->name %s succeeded\n", __func__,
+ mode->name);
+out:
+ dma_unmap_single(dev, src_addr, 1088, DMA_TO_DEVICE);
+
+ return ret;
+}
+
static int init_device_compression_mode(struct iaa_device *iaa_device,
struct iaa_compression_mode *mode,
int idx, struct idxd_wq *wq)
@@ -300,6 +442,14 @@ static int init_device_compression_mode(struct iaa_device *iaa_device,
memcpy(device_mode->aecs_comp_table->ll_sym, mode->ll_table, mode->ll_table_size);
memcpy(device_mode->aecs_comp_table->d_sym, mode->d_table, mode->d_table_size);
+ if (mode->header_table) {
+ ret = decompress_header(device_mode, mode, wq);
+ if (ret) {
+ pr_debug("iaa header decompression failed: ret=%d\n", ret);
+ goto free;
+ }
+ }
+
if (mode->init) {
ret = mode->init(device_mode);
if (ret)
@@ -372,18 +522,6 @@ static struct iaa_device *iaa_device_alloc(void)
return iaa_device;
}
-static void iaa_device_free(struct iaa_device *iaa_device)
-{
- struct iaa_wq *iaa_wq, *next;
-
- list_for_each_entry_safe(iaa_wq, next, &iaa_device->wqs, list) {
- list_del(&iaa_wq->list);
- kfree(iaa_wq);
- }
-
- kfree(iaa_device);
-}
-
static bool iaa_has_wq(struct iaa_device *iaa_device, struct idxd_wq *wq)
{
struct iaa_wq *iaa_wq;
@@ -426,12 +564,8 @@ static int init_iaa_device(struct iaa_device *iaa_device, struct iaa_wq *iaa_wq)
static void del_iaa_device(struct iaa_device *iaa_device)
{
- remove_device_compression_modes(iaa_device);
-
list_del(&iaa_device->list);
- iaa_device_free(iaa_device);
-
nr_iaa--;
}
@@ -497,6 +631,85 @@ static void clear_wq_table(void)
pr_debug("cleared wq table\n");
}
+static void free_iaa_device(struct iaa_device *iaa_device)
+{
+ if (!iaa_device)
+ return;
+
+ remove_device_compression_modes(iaa_device);
+ kfree(iaa_device);
+}
+
+static void __free_iaa_wq(struct iaa_wq *iaa_wq)
+{
+ struct iaa_device *iaa_device;
+
+ if (!iaa_wq)
+ return;
+
+ iaa_device = iaa_wq->iaa_device;
+ if (iaa_device->n_wq == 0)
+ free_iaa_device(iaa_wq->iaa_device);
+}
+
+static void free_iaa_wq(struct iaa_wq *iaa_wq)
+{
+ struct idxd_wq *wq;
+
+ __free_iaa_wq(iaa_wq);
+
+ wq = iaa_wq->wq;
+
+ kfree(iaa_wq);
+ idxd_wq_set_private(wq, NULL);
+}
+
+static int iaa_wq_get(struct idxd_wq *wq)
+{
+ struct idxd_device *idxd = wq->idxd;
+ struct iaa_wq *iaa_wq;
+ int ret = 0;
+
+ spin_lock(&idxd->dev_lock);
+ iaa_wq = idxd_wq_get_private(wq);
+ if (iaa_wq && !iaa_wq->remove) {
+ iaa_wq->ref++;
+ idxd_wq_get(wq);
+ } else {
+ ret = -ENODEV;
+ }
+ spin_unlock(&idxd->dev_lock);
+
+ return ret;
+}
+
+static int iaa_wq_put(struct idxd_wq *wq)
+{
+ struct idxd_device *idxd = wq->idxd;
+ struct iaa_wq *iaa_wq;
+ bool free = false;
+ int ret = 0;
+
+ spin_lock(&idxd->dev_lock);
+ iaa_wq = idxd_wq_get_private(wq);
+ if (iaa_wq) {
+ iaa_wq->ref--;
+ if (iaa_wq->ref == 0 && iaa_wq->remove) {
+ __free_iaa_wq(iaa_wq);
+ idxd_wq_set_private(wq, NULL);
+ free = true;
+ }
+ idxd_wq_put(wq);
+ } else {
+ ret = -ENODEV;
+ }
+ spin_unlock(&idxd->dev_lock);
+ if (free)
+ kfree(iaa_wq);
+
+ return ret;
+}
+
static void free_wq_table(void)
{
int cpu;
@@ -580,6 +793,7 @@ static int save_iaa_wq(struct idxd_wq *wq)
ret = add_iaa_wq(new_device, wq, &new_wq);
if (ret) {
del_iaa_device(new_device);
+ free_iaa_device(new_device);
goto out;
}
@@ -587,6 +801,7 @@ static int save_iaa_wq(struct idxd_wq *wq)
if (ret) {
del_iaa_wq(new_device, new_wq->wq);
del_iaa_device(new_device);
+ free_iaa_wq(new_wq);
goto out;
}
}
@@ -721,6 +936,729 @@ static void rebalance_wq_table(void)
}
}
+static inline int check_completion(struct device *dev,
+ struct iax_completion_record *comp,
+ bool compress,
+ bool only_once)
+{
+ char *op_str = compress ? "compress" : "decompress";
+ int ret = 0;
+
+ while (!comp->status) {
+ if (only_once)
+ return -EAGAIN;
+ cpu_relax();
+ }
+
+ if (comp->status != IAX_COMP_SUCCESS) {
+ if (comp->status == IAA_ERROR_WATCHDOG_EXPIRED) {
+ ret = -ETIMEDOUT;
+ dev_dbg(dev, "%s timed out, size=0x%x\n",
+ op_str, comp->output_size);
+ goto out;
+ }
+
+ if (comp->status == IAA_ANALYTICS_ERROR &&
+ comp->error_code == IAA_ERROR_COMP_BUF_OVERFLOW && compress) {
+ ret = -E2BIG;
+ dev_dbg(dev, "compressed > uncompressed size,"
+ " not compressing, size=0x%x\n",
+ comp->output_size);
+ goto out;
+ }
+
+ if (comp->status == IAA_ERROR_DECOMP_BUF_OVERFLOW) {
+ ret = -EOVERFLOW;
+ goto out;
+ }
+
+ ret = -EINVAL;
+ dev_dbg(dev, "iaa %s status=0x%x, error=0x%x, size=0x%x\n",
+ op_str, comp->status, comp->error_code, comp->output_size);
+ print_hex_dump(KERN_INFO, "cmp-rec: ", DUMP_PREFIX_OFFSET, 8, 1, comp, 64, 0);
+
+ goto out;
+ }
+out:
+ return ret;
+}
+
+static int deflate_generic_decompress(struct acomp_req *req)
+{
+ void *src, *dst;
+ int ret;
+
+ src = kmap_local_page(sg_page(req->src)) + req->src->offset;
+ dst = kmap_local_page(sg_page(req->dst)) + req->dst->offset;
+
+ ret = crypto_comp_decompress(deflate_generic_tfm,
+ src, req->slen, dst, &req->dlen);
+
+ kunmap_local(src);
+ kunmap_local(dst);
+
+ return ret;
+}
+
+static int iaa_compress(struct crypto_tfm *tfm, struct acomp_req *req,
+ struct idxd_wq *wq,
+ dma_addr_t src_addr, unsigned int slen,
+ dma_addr_t dst_addr, unsigned int *dlen,
+ u32 *compression_crc,
+ bool disable_async)
+{
+ struct iaa_device_compression_mode *active_compression_mode;
+ struct iaa_compression_ctx *ctx = crypto_tfm_ctx(tfm);
+ struct iaa_device *iaa_device;
+ struct idxd_desc *idxd_desc;
+ struct iax_hw_desc *desc;
+ struct idxd_device *idxd;
+ struct iaa_wq *iaa_wq;
+ struct pci_dev *pdev;
+ struct device *dev;
+ int ret = 0;
+
+ iaa_wq = idxd_wq_get_private(wq);
+ iaa_device = iaa_wq->iaa_device;
+ idxd = iaa_device->idxd;
+ pdev = idxd->pdev;
+ dev = &pdev->dev;
+
+ active_compression_mode = get_iaa_device_compression_mode(iaa_device, ctx->mode);
+
+ idxd_desc = idxd_alloc_desc(wq, IDXD_OP_BLOCK);
+ if (IS_ERR(idxd_desc)) {
+ dev_dbg(dev, "idxd descriptor allocation failed\n");
+ dev_dbg(dev, "iaa compress failed: ret=%ld\n", PTR_ERR(idxd_desc));
+ return PTR_ERR(idxd_desc);
+ }
+ desc = idxd_desc->iax_hw;
+
+ desc->flags = IDXD_OP_FLAG_CRAV | IDXD_OP_FLAG_RCR |
+ IDXD_OP_FLAG_RD_SRC2_AECS | IDXD_OP_FLAG_CC;
+ desc->opcode = IAX_OPCODE_COMPRESS;
+ desc->compr_flags = IAA_COMP_FLAGS;
+ desc->priv = 1;
+
+ desc->src1_addr = (u64)src_addr;
+ desc->src1_size = slen;
+ desc->dst_addr = (u64)dst_addr;
+ desc->max_dst_size = *dlen;
+ desc->src2_addr = active_compression_mode->aecs_comp_table_dma_addr;
+ desc->src2_size = sizeof(struct aecs_comp_table_record);
+ desc->completion_addr = idxd_desc->compl_dma;
+
+ dev_dbg(dev, "%s: compression mode %s,"
+ " desc->src1_addr %llx, desc->src1_size %d,"
+ " desc->dst_addr %llx, desc->max_dst_size %d,"
+ " desc->src2_addr %llx, desc->src2_size %d\n", __func__,
+ active_compression_mode->name,
+ desc->src1_addr, desc->src1_size, desc->dst_addr,
+ desc->max_dst_size, desc->src2_addr, desc->src2_size);
+
+ ret = idxd_submit_desc(wq, idxd_desc);
+ if (ret) {
+ dev_dbg(dev, "submit_desc failed ret=%d\n", ret);
+ goto err;
+ }
+
+ ret = check_completion(dev, idxd_desc->iax_completion, true, false);
+ if (ret) {
+ dev_dbg(dev, "check_completion failed ret=%d\n", ret);
+ goto err;
+ }
+
+ *dlen = idxd_desc->iax_completion->output_size;
+
+ *compression_crc = idxd_desc->iax_completion->crc;
+
+ idxd_free_desc(wq, idxd_desc);
+out:
+ return ret;
+err:
+ idxd_free_desc(wq, idxd_desc);
+ dev_dbg(dev, "iaa compress failed: ret=%d\n", ret);
+
+ goto out;
+}
+
+static int iaa_remap_for_verify(struct device *dev, struct iaa_wq *iaa_wq,
+ struct acomp_req *req,
+ dma_addr_t *src_addr, dma_addr_t *dst_addr)
+{
+ int ret = 0;
+ int nr_sgs;
+
+ dma_unmap_sg(dev, req->dst, sg_nents(req->dst), DMA_FROM_DEVICE);
+ dma_unmap_sg(dev, req->src, sg_nents(req->src), DMA_TO_DEVICE);
+
+ nr_sgs = dma_map_sg(dev, req->src, sg_nents(req->src), DMA_FROM_DEVICE);
+ if (nr_sgs <= 0 || nr_sgs > 1) {
+ dev_dbg(dev, "verify: couldn't map src sg for iaa device %d,"
+ " wq %d: ret=%d\n", iaa_wq->iaa_device->idxd->id,
+ iaa_wq->wq->id, ret);
+ ret = -EIO;
+ goto out;
+ }
+ *src_addr = sg_dma_address(req->src);
+ dev_dbg(dev, "verify: dma_map_sg, src_addr %llx, nr_sgs %d, req->src %p,"
+ " req->slen %d, sg_dma_len(sg) %d\n", *src_addr, nr_sgs,
+ req->src, req->slen, sg_dma_len(req->src));
+
+ nr_sgs = dma_map_sg(dev, req->dst, sg_nents(req->dst), DMA_TO_DEVICE);
+ if (nr_sgs <= 0 || nr_sgs > 1) {
+ dev_dbg(dev, "verify: couldn't map dst sg for iaa device %d,"
+ " wq %d: ret=%d\n", iaa_wq->iaa_device->idxd->id,
+ iaa_wq->wq->id, ret);
+ ret = -EIO;
+ dma_unmap_sg(dev, req->src, sg_nents(req->src), DMA_FROM_DEVICE);
+ goto out;
+ }
+ *dst_addr = sg_dma_address(req->dst);
+ dev_dbg(dev, "verify: dma_map_sg, dst_addr %llx, nr_sgs %d, req->dst %p,"
+ " req->dlen %d, sg_dma_len(sg) %d\n", *dst_addr, nr_sgs,
+ req->dst, req->dlen, sg_dma_len(req->dst));
+out:
+ return ret;
+}
+
+static int iaa_compress_verify(struct crypto_tfm *tfm, struct acomp_req *req,
+ struct idxd_wq *wq,
+ dma_addr_t src_addr, unsigned int slen,
+ dma_addr_t dst_addr, unsigned int *dlen,
+ u32 compression_crc)
+{
+ struct iaa_device_compression_mode *active_compression_mode;
+ struct iaa_compression_ctx *ctx = crypto_tfm_ctx(tfm);
+ struct iaa_device *iaa_device;
+ struct idxd_desc *idxd_desc;
+ struct iax_hw_desc *desc;
+ struct idxd_device *idxd;
+ struct iaa_wq *iaa_wq;
+ struct pci_dev *pdev;
+ struct device *dev;
+ int ret = 0;
+
+ iaa_wq = idxd_wq_get_private(wq);
+ iaa_device = iaa_wq->iaa_device;
+ idxd = iaa_device->idxd;
+ pdev = idxd->pdev;
+ dev = &pdev->dev;
+
+ active_compression_mode = get_iaa_device_compression_mode(iaa_device, ctx->mode);
+
+ idxd_desc = idxd_alloc_desc(wq, IDXD_OP_BLOCK);
+ if (IS_ERR(idxd_desc)) {
+ dev_dbg(dev, "idxd descriptor allocation failed\n");
+ dev_dbg(dev, "iaa compress failed: ret=%ld\n",
+ PTR_ERR(idxd_desc));
+ return PTR_ERR(idxd_desc);
+ }
+ desc = idxd_desc->iax_hw;
+
+ /* Verify (optional) - decompress and check crc, suppress dest write */
+
+ desc->flags = IDXD_OP_FLAG_CRAV | IDXD_OP_FLAG_RCR | IDXD_OP_FLAG_CC;
+ desc->opcode = IAX_OPCODE_DECOMPRESS;
+ desc->decompr_flags = IAA_DECOMP_FLAGS | IAA_DECOMP_SUPPRESS_OUTPUT;
+ desc->priv = 1;
+
+ desc->src1_addr = (u64)dst_addr;
+ desc->src1_size = *dlen;
+ desc->dst_addr = (u64)src_addr;
+ desc->max_dst_size = slen;
+ desc->completion_addr = idxd_desc->compl_dma;
+
+ dev_dbg(dev, "(verify) compression mode %s,"
+ " desc->src1_addr %llx, desc->src1_size %d,"
+ " desc->dst_addr %llx, desc->max_dst_size %d,"
+ " desc->src2_addr %llx, desc->src2_size %d\n",
+ active_compression_mode->name,
+ desc->src1_addr, desc->src1_size, desc->dst_addr,
+ desc->max_dst_size, desc->src2_addr, desc->src2_size);
+
+ ret = idxd_submit_desc(wq, idxd_desc);
+ if (ret) {
+ dev_dbg(dev, "submit_desc (verify) failed ret=%d\n", ret);
+ goto err;
+ }
+
+ ret = check_completion(dev, idxd_desc->iax_completion, false, false);
+ if (ret) {
+ dev_dbg(dev, "(verify) check_completion failed ret=%d\n", ret);
+ goto err;
+ }
+
+ if (compression_crc != idxd_desc->iax_completion->crc) {
+ ret = -EINVAL;
+ dev_dbg(dev, "(verify) iaa comp/decomp crc mismatch:"
+ " comp=0x%x, decomp=0x%x\n", compression_crc,
+ idxd_desc->iax_completion->crc);
+ print_hex_dump(KERN_INFO, "cmp-rec: ", DUMP_PREFIX_OFFSET,
+ 8, 1, idxd_desc->iax_completion, 64, 0);
+ goto err;
+ }
+
+ idxd_free_desc(wq, idxd_desc);
+out:
+ return ret;
+err:
+ idxd_free_desc(wq, idxd_desc);
+ dev_dbg(dev, "iaa compress failed: ret=%d\n", ret);
+
+ goto out;
+}
+
+static int iaa_decompress(struct crypto_tfm *tfm, struct acomp_req *req,
+ struct idxd_wq *wq,
+ dma_addr_t src_addr, unsigned int slen,
+ dma_addr_t dst_addr, unsigned int *dlen,
+ bool disable_async)
+{
+ struct iaa_device_compression_mode *active_compression_mode;
+ struct iaa_compression_ctx *ctx = crypto_tfm_ctx(tfm);
+ struct iaa_device *iaa_device;
+ struct idxd_desc *idxd_desc;
+ struct iax_hw_desc *desc;
+ struct idxd_device *idxd;
+ struct iaa_wq *iaa_wq;
+ struct pci_dev *pdev;
+ struct device *dev;
+ int ret = 0;
+
+ iaa_wq = idxd_wq_get_private(wq);
+ iaa_device = iaa_wq->iaa_device;
+ idxd = iaa_device->idxd;
+ pdev = idxd->pdev;
+ dev = &pdev->dev;
+
+ active_compression_mode = get_iaa_device_compression_mode(iaa_device, ctx->mode);
+
+ idxd_desc = idxd_alloc_desc(wq, IDXD_OP_BLOCK);
+ if (IS_ERR(idxd_desc)) {
+ dev_dbg(dev, "idxd descriptor allocation failed\n");
+ dev_dbg(dev, "iaa decompress failed: ret=%ld\n",
+ PTR_ERR(idxd_desc));
+ return PTR_ERR(idxd_desc);
+ }
+ desc = idxd_desc->iax_hw;
+
+ desc->flags = IDXD_OP_FLAG_CRAV | IDXD_OP_FLAG_RCR | IDXD_OP_FLAG_CC;
+ desc->opcode = IAX_OPCODE_DECOMPRESS;
+ desc->max_dst_size = PAGE_SIZE;
+ desc->decompr_flags = IAA_DECOMP_FLAGS;
+ desc->priv = 1;
+
+ desc->src1_addr = (u64)src_addr;
+ desc->dst_addr = (u64)dst_addr;
+ desc->max_dst_size = *dlen;
+ desc->src1_size = slen;
+ desc->completion_addr = idxd_desc->compl_dma;
+
+ dev_dbg(dev, "%s: decompression mode %s,"
+ " desc->src1_addr %llx, desc->src1_size %d,"
+ " desc->dst_addr %llx, desc->max_dst_size %d,"
+ " desc->src2_addr %llx, desc->src2_size %d\n", __func__,
+ active_compression_mode->name,
+ desc->src1_addr, desc->src1_size, desc->dst_addr,
+ desc->max_dst_size, desc->src2_addr, desc->src2_size);
+
+ ret = idxd_submit_desc(wq, idxd_desc);
+ if (ret) {
+ dev_dbg(dev, "submit_desc failed ret=%d\n", ret);
+ goto err;
+ }
+
+ ret = check_completion(dev, idxd_desc->iax_completion, false, false);
+ if (ret) {
+ dev_dbg(dev, "%s: check_completion failed ret=%d\n", __func__, ret);
+ if (idxd_desc->iax_completion->status == IAA_ANALYTICS_ERROR) {
+ pr_warn("%s: falling back to deflate-generic decompress, "
+ "analytics error code %x\n", __func__,
+ idxd_desc->iax_completion->error_code);
+ ret = deflate_generic_decompress(req);
+ if (ret) {
+ dev_dbg(dev, "%s: deflate-generic failed ret=%d\n",
+ __func__, ret);
+ goto err;
+ }
+ } else {
+ goto err;
+ }
+ } else {
+ req->dlen = idxd_desc->iax_completion->output_size;
+ }
+
+ *dlen = req->dlen;
+
+ idxd_free_desc(wq, idxd_desc);
+out:
+ return ret;
+err:
+ idxd_free_desc(wq, idxd_desc);
+ dev_dbg(dev, "iaa decompress failed: ret=%d\n", ret);
+
+ goto out;
+}
+
+static int iaa_comp_acompress(struct acomp_req *req)
+{
+ struct iaa_compression_ctx *compression_ctx;
+ struct crypto_tfm *tfm = req->base.tfm;
+ dma_addr_t src_addr, dst_addr;
+ bool disable_async = false;
+ int nr_sgs, cpu, ret = 0;
+ struct iaa_wq *iaa_wq;
+ u32 compression_crc;
+ struct idxd_wq *wq;
+ struct device *dev;
+ int order = -1;
+
+ compression_ctx = crypto_tfm_ctx(tfm);
+
+ if (!iaa_crypto_enabled) {
+ pr_debug("iaa_crypto disabled, not compressing\n");
+ return -ENODEV;
+ }
+
+ if (!req->src || !req->slen) {
+ pr_debug("invalid src, not compressing\n");
+ return -EINVAL;
+ }
+
+ cpu = get_cpu();
+ wq = wq_table_next_wq(cpu);
+ put_cpu();
+ if (!wq) {
+ pr_debug("no wq configured for cpu=%d\n", cpu);
+ return -ENODEV;
+ }
+
+ ret = iaa_wq_get(wq);
+ if (ret) {
+ pr_debug("no wq available for cpu=%d\n", cpu);
+ return -ENODEV;
+ }
+
+ iaa_wq = idxd_wq_get_private(wq);
+
+ if (!req->dst) {
+ gfp_t flags = req->flags & CRYPTO_TFM_REQ_MAY_SLEEP ? GFP_KERNEL : GFP_ATOMIC;
+
+ /* incompressible data will always be < 2 * slen */
+ req->dlen = 2 * req->slen;
+ order = order_base_2(round_up(req->dlen, PAGE_SIZE) / PAGE_SIZE);
+ req->dst = sgl_alloc_order(req->dlen, order, false, flags, NULL);
+ if (!req->dst) {
+ ret = -ENOMEM;
+ order = -1;
+ goto out;
+ }
+ disable_async = true;
+ }
+
+ dev = &wq->idxd->pdev->dev;
+
+ nr_sgs = dma_map_sg(dev, req->src, sg_nents(req->src), DMA_TO_DEVICE);
+ if (nr_sgs <= 0 || nr_sgs > 1) {
+ dev_dbg(dev, "couldn't map src sg for iaa device %d,"
+ " wq %d: ret=%d\n", iaa_wq->iaa_device->idxd->id,
+ iaa_wq->wq->id, ret);
+ ret = -EIO;
+ goto out;
+ }
+ src_addr = sg_dma_address(req->src);
+ dev_dbg(dev, "dma_map_sg, src_addr %llx, nr_sgs %d, req->src %p,"
+ " req->slen %d, sg_dma_len(sg) %d\n", src_addr, nr_sgs,
+ req->src, req->slen, sg_dma_len(req->src));
+
+ nr_sgs = dma_map_sg(dev, req->dst, sg_nents(req->dst), DMA_FROM_DEVICE);
+ if (nr_sgs <= 0 || nr_sgs > 1) {
+ dev_dbg(dev, "couldn't map dst sg for iaa device %d,"
+ " wq %d: ret=%d\n", iaa_wq->iaa_device->idxd->id,
+ iaa_wq->wq->id, ret);
+ ret = -EIO;
+ goto err_map_dst;
+ }
+ dst_addr = sg_dma_address(req->dst);
+ dev_dbg(dev, "dma_map_sg, dst_addr %llx, nr_sgs %d, req->dst %p,"
+ " req->dlen %d, sg_dma_len(sg) %d\n", dst_addr, nr_sgs,
+ req->dst, req->dlen, sg_dma_len(req->dst));
+
+ ret = iaa_compress(tfm, req, wq, src_addr, req->slen, dst_addr,
+ &req->dlen, &compression_crc, disable_async);
+ if (ret == -EINPROGRESS)
+ return ret;
+
+ if (!ret && compression_ctx->verify_compress) {
+ ret = iaa_remap_for_verify(dev, iaa_wq, req, &src_addr, &dst_addr);
+ if (ret) {
+ dev_dbg(dev, "%s: compress verify remap failed ret=%d\n", __func__, ret);
+ goto out;
+ }
+
+ ret = iaa_compress_verify(tfm, req, wq, src_addr, req->slen,
+ dst_addr, &req->dlen, compression_crc);
+ if (ret)
+ dev_dbg(dev, "asynchronous compress verification failed ret=%d\n", ret);
+
+ dma_unmap_sg(dev, req->dst, sg_nents(req->dst), DMA_TO_DEVICE);
+ dma_unmap_sg(dev, req->src, sg_nents(req->src), DMA_FROM_DEVICE);
+
+ goto out;
+ }
+
+ if (ret)
+ dev_dbg(dev, "asynchronous compress failed ret=%d\n", ret);
+
+ dma_unmap_sg(dev, req->dst, sg_nents(req->dst), DMA_FROM_DEVICE);
+err_map_dst:
+ dma_unmap_sg(dev, req->src, sg_nents(req->src), DMA_TO_DEVICE);
+out:
+ iaa_wq_put(wq);
+
+ if (order >= 0)
+ sgl_free_order(req->dst, order);
+
+ return ret;
+}
+
+static int iaa_comp_adecompress_alloc_dest(struct acomp_req *req)
+{
+ gfp_t flags = req->flags & CRYPTO_TFM_REQ_MAY_SLEEP ?
+ GFP_KERNEL : GFP_ATOMIC;
+ struct crypto_tfm *tfm = req->base.tfm;
+ dma_addr_t src_addr, dst_addr;
+ int nr_sgs, cpu, ret = 0;
+ struct iaa_wq *iaa_wq;
+ struct device *dev;
+ struct idxd_wq *wq;
+ int order = -1;
+
+ cpu = get_cpu();
+ wq = wq_table_next_wq(cpu);
+ put_cpu();
+ if (!wq) {
+ pr_debug("no wq configured for cpu=%d\n", cpu);
+ return -ENODEV;
+ }
+
+ ret = iaa_wq_get(wq);
+ if (ret) {
+ pr_debug("no wq available for cpu=%d\n", cpu);
+ return -ENODEV;
+ }
+
+ iaa_wq = idxd_wq_get_private(wq);
+
+ dev = &wq->idxd->pdev->dev;
+
+ nr_sgs = dma_map_sg(dev, req->src, sg_nents(req->src), DMA_TO_DEVICE);
+ if (nr_sgs <= 0 || nr_sgs > 1) {
+ dev_dbg(dev, "couldn't map src sg for iaa device %d,"
+ " wq %d: ret=%d\n", iaa_wq->iaa_device->idxd->id,
+ iaa_wq->wq->id, ret);
+ ret = -EIO;
+ goto out;
+ }
+ src_addr = sg_dma_address(req->src);
+ dev_dbg(dev, "dma_map_sg, src_addr %llx, nr_sgs %d, req->src %p,"
+ " req->slen %d, sg_dma_len(sg) %d\n", src_addr, nr_sgs,
+ req->src, req->slen, sg_dma_len(req->src));
+
+ req->dlen = 4 * req->slen; /* start with ~avg comp rato */
+alloc_dest:
+ order = order_base_2(round_up(req->dlen, PAGE_SIZE) / PAGE_SIZE);
+ req->dst = sgl_alloc_order(req->dlen, order, false, flags, NULL);
+ if (!req->dst) {
+ ret = -ENOMEM;
+ order = -1;
+ goto out;
+ }
+
+ nr_sgs = dma_map_sg(dev, req->dst, sg_nents(req->dst), DMA_FROM_DEVICE);
+ if (nr_sgs <= 0 || nr_sgs > 1) {
+ dev_dbg(dev, "couldn't map dst sg for iaa device %d,"
+ " wq %d: ret=%d\n", iaa_wq->iaa_device->idxd->id,
+ iaa_wq->wq->id, ret);
+ ret = -EIO;
+ goto err_map_dst;
+ }
+
+ dst_addr = sg_dma_address(req->dst);
+ dev_dbg(dev, "dma_map_sg, dst_addr %llx, nr_sgs %d, req->dst %p,"
+ " req->dlen %d, sg_dma_len(sg) %d\n", dst_addr, nr_sgs,
+ req->dst, req->dlen, sg_dma_len(req->dst));
+ ret = iaa_decompress(tfm, req, wq, src_addr, req->slen,
+ dst_addr, &req->dlen, true);
+ if (ret == -EOVERFLOW) {
+ dma_unmap_sg(dev, req->dst, sg_nents(req->dst), DMA_FROM_DEVICE);
+ req->dlen *= 2;
+ if (req->dlen > CRYPTO_ACOMP_DST_MAX)
+ goto err_map_dst;
+ goto alloc_dest;
+ }
+
+ if (ret != 0)
+ dev_dbg(dev, "asynchronous decompress failed ret=%d\n", ret);
+
+ dma_unmap_sg(dev, req->dst, sg_nents(req->dst), DMA_FROM_DEVICE);
+err_map_dst:
+ dma_unmap_sg(dev, req->src, sg_nents(req->src), DMA_TO_DEVICE);
+out:
+ iaa_wq_put(wq);
+
+ if (order >= 0)
+ sgl_free_order(req->dst, order);
+
+ return ret;
+}
+
+static int iaa_comp_adecompress(struct acomp_req *req)
+{
+ struct crypto_tfm *tfm = req->base.tfm;
+ dma_addr_t src_addr, dst_addr;
+ int nr_sgs, cpu, ret = 0;
+ struct iaa_wq *iaa_wq;
+ struct device *dev;
+ struct idxd_wq *wq;
+
+ if (!iaa_crypto_enabled) {
+ pr_debug("iaa_crypto disabled, not decompressing\n");
+ return -ENODEV;
+ }
+
+ if (!req->src || !req->slen) {
+ pr_debug("invalid src, not decompressing\n");
+ return -EINVAL;
+ }
+
+ if (!req->dst)
+ return iaa_comp_adecompress_alloc_dest(req);
+
+ cpu = get_cpu();
+ wq = wq_table_next_wq(cpu);
+ put_cpu();
+ if (!wq) {
+ pr_debug("no wq configured for cpu=%d\n", cpu);
+ return -ENODEV;
+ }
+
+ ret = iaa_wq_get(wq);
+ if (ret) {
+ pr_debug("no wq available for cpu=%d\n", cpu);
+ return -ENODEV;
+ }
+
+ iaa_wq = idxd_wq_get_private(wq);
+
+ dev = &wq->idxd->pdev->dev;
+
+ nr_sgs = dma_map_sg(dev, req->src, sg_nents(req->src), DMA_TO_DEVICE);
+ if (nr_sgs <= 0 || nr_sgs > 1) {
+ dev_dbg(dev, "couldn't map src sg for iaa device %d,"
+ " wq %d: ret=%d\n", iaa_wq->iaa_device->idxd->id,
+ iaa_wq->wq->id, ret);
+ ret = -EIO;
+ goto out;
+ }
+ src_addr = sg_dma_address(req->src);
+ dev_dbg(dev, "dma_map_sg, src_addr %llx, nr_sgs %d, req->src %p,"
+ " req->slen %d, sg_dma_len(sg) %d\n", src_addr, nr_sgs,
+ req->src, req->slen, sg_dma_len(req->src));
+
+ nr_sgs = dma_map_sg(dev, req->dst, sg_nents(req->dst), DMA_FROM_DEVICE);
+ if (nr_sgs <= 0 || nr_sgs > 1) {
+ dev_dbg(dev, "couldn't map dst sg for iaa device %d,"
+ " wq %d: ret=%d\n", iaa_wq->iaa_device->idxd->id,
+ iaa_wq->wq->id, ret);
+ ret = -EIO;
+ goto err_map_dst;
+ }
+ dst_addr = sg_dma_address(req->dst);
+ dev_dbg(dev, "dma_map_sg, dst_addr %llx, nr_sgs %d, req->dst %p,"
+ " req->dlen %d, sg_dma_len(sg) %d\n", dst_addr, nr_sgs,
+ req->dst, req->dlen, sg_dma_len(req->dst));
+
+ ret = iaa_decompress(tfm, req, wq, src_addr, req->slen,
+ dst_addr, &req->dlen, false);
+ if (ret == -EINPROGRESS)
+ return ret;
+
+ if (ret != 0)
+ dev_dbg(dev, "asynchronous decompress failed ret=%d\n", ret);
+
+ dma_unmap_sg(dev, req->dst, sg_nents(req->dst), DMA_FROM_DEVICE);
+err_map_dst:
+ dma_unmap_sg(dev, req->src, sg_nents(req->src), DMA_TO_DEVICE);
+out:
+ iaa_wq_put(wq);
+
+ return ret;
+}
+
+static void compression_ctx_init(struct iaa_compression_ctx *ctx)
+{
+ ctx->verify_compress = iaa_verify_compress;
+}
+
+static int iaa_comp_init_fixed(struct crypto_acomp *acomp_tfm)
+{
+ struct crypto_tfm *tfm = crypto_acomp_tfm(acomp_tfm);
+ struct iaa_compression_ctx *ctx = crypto_tfm_ctx(tfm);
+
+ compression_ctx_init(ctx);
+
+ ctx->mode = IAA_MODE_FIXED;
+
+ return 0;
+}
+
+static void dst_free(struct scatterlist *sgl)
+{
+ /*
+ * Called for req->dst = NULL cases but we free elsewhere
+ * using sgl_free_order().
+ */
+}
+
+static struct acomp_alg iaa_acomp_fixed_deflate = {
+ .init = iaa_comp_init_fixed,
+ .compress = iaa_comp_acompress,
+ .decompress = iaa_comp_adecompress,
+ .dst_free = dst_free,
+ .base = {
+ .cra_name = "deflate",
+ .cra_driver_name = "deflate-iaa",
+ .cra_ctxsize = sizeof(struct iaa_compression_ctx),
+ .cra_module = THIS_MODULE,
+ .cra_priority = IAA_ALG_PRIORITY,
+ }
+};
+
+static int iaa_register_compression_device(void)
+{
+ int ret;
+
+ ret = crypto_register_acomp(&iaa_acomp_fixed_deflate);
+ if (ret) {
+ pr_err("deflate algorithm acomp fixed registration failed (%d)\n", ret);
+ goto out;
+ }
+
+ iaa_crypto_registered = true;
+out:
+ return ret;
+}
+
+static int iaa_unregister_compression_device(void)
+{
+ if (iaa_crypto_registered)
+ crypto_unregister_acomp(&iaa_acomp_fixed_deflate);
+
+ return 0;
+}
+
static int iaa_crypto_probe(struct idxd_dev *idxd_dev)
{
struct idxd_wq *wq = idxd_dev_to_wq(idxd_dev);
@@ -738,6 +1676,11 @@ static int iaa_crypto_probe(struct idxd_dev *idxd_dev)
mutex_lock(&wq->wq_lock);
+ if (idxd_wq_get_private(wq)) {
+ mutex_unlock(&wq->wq_lock);
+ return -EBUSY;
+ }
+
if (!idxd_wq_driver_name_match(wq, dev)) {
dev_dbg(dev, "wq %d.%d driver_name match failed: wq driver_name %s, dev driver name %s\n",
idxd->id, wq->id, wq->driver_name, dev->driver->name);
@@ -771,12 +1714,28 @@ static int iaa_crypto_probe(struct idxd_dev *idxd_dev)
rebalance_wq_table();
+ if (first_wq) {
+ iaa_crypto_enabled = true;
+ ret = iaa_register_compression_device();
+ if (ret != 0) {
+ iaa_crypto_enabled = false;
+ dev_dbg(dev, "IAA compression device registration failed\n");
+ goto err_register;
+ }
+ try_module_get(THIS_MODULE);
+
+ pr_info("iaa_crypto now ENABLED\n");
+ }
+
mutex_unlock(&iaa_devices_lock);
out:
mutex_unlock(&wq->wq_lock);
return ret;
+err_register:
+ remove_iaa_wq(wq);
+ free_iaa_wq(idxd_wq_get_private(wq));
err_save:
if (first_wq)
free_wq_table();
@@ -792,6 +1751,9 @@ static int iaa_crypto_probe(struct idxd_dev *idxd_dev)
static void iaa_crypto_remove(struct idxd_dev *idxd_dev)
{
struct idxd_wq *wq = idxd_dev_to_wq(idxd_dev);
+ struct idxd_device *idxd = wq->idxd;
+ struct iaa_wq *iaa_wq;
+ bool free = false;
idxd_wq_quiesce(wq);
@@ -800,12 +1762,38 @@ static void iaa_crypto_remove(struct idxd_dev *idxd_dev)
remove_iaa_wq(wq);
+ spin_lock(&idxd->dev_lock);
+ iaa_wq = idxd_wq_get_private(wq);
+ if (!iaa_wq) {
+ spin_unlock(&idxd->dev_lock);
+ pr_err("%s: no iaa_wq available to remove\n", __func__);
+ goto out;
+ }
+
+ if (iaa_wq->ref) {
+ iaa_wq->remove = true;
+ } else {
+ wq = iaa_wq->wq;
+ __free_iaa_wq(iaa_wq);
+ idxd_wq_set_private(wq, NULL);
+ free = true;
+ }
+ spin_unlock(&idxd->dev_lock);
+
+ if (free)
+ kfree(iaa_wq);
+
idxd_drv_disable_wq(wq);
rebalance_wq_table();
- if (nr_iaa == 0)
+ if (nr_iaa == 0) {
+ iaa_crypto_enabled = false;
free_wq_table();
+ module_put(THIS_MODULE);
+ pr_info("iaa_crypto now DISABLED\n");
+ }
+out:
mutex_unlock(&iaa_devices_lock);
mutex_unlock(&wq->wq_lock);
}
@@ -830,10 +1818,19 @@ static int __init iaa_crypto_init_module(void)
nr_nodes = num_online_nodes();
nr_cpus_per_node = nr_cpus / nr_nodes;
+ if (crypto_has_comp("deflate-generic", 0, 0))
+ deflate_generic_tfm = crypto_alloc_comp("deflate-generic", 0, 0);
+
+ if (IS_ERR_OR_NULL(deflate_generic_tfm)) {
+ pr_err("IAA could not alloc %s tfm: errcode = %ld\n",
+ "deflate-generic", PTR_ERR(deflate_generic_tfm));
+ return -ENOMEM;
+ }
+
ret = iaa_aecs_init_fixed();
if (ret < 0) {
pr_debug("IAA fixed compression mode init failed\n");
- goto out;
+ goto err_aecs_init;
}
ret = idxd_driver_register(&iaa_crypto_driver);
@@ -842,20 +1839,37 @@ static int __init iaa_crypto_init_module(void)
goto err_driver_reg;
}
+ ret = driver_create_file(&iaa_crypto_driver.drv,
+ &driver_attr_verify_compress);
+ if (ret) {
+ pr_debug("IAA verify_compress attr creation failed\n");
+ goto err_verify_attr_create;
+ }
+
pr_debug("initialized\n");
out:
return ret;
+err_verify_attr_create:
+ idxd_driver_unregister(&iaa_crypto_driver);
err_driver_reg:
iaa_aecs_cleanup_fixed();
+err_aecs_init:
+ crypto_free_comp(deflate_generic_tfm);
goto out;
}
static void __exit iaa_crypto_cleanup_module(void)
{
+ if (iaa_unregister_compression_device())
+ pr_debug("IAA compression device unregister failed\n");
+
+ driver_remove_file(&iaa_crypto_driver.drv,
+ &driver_attr_verify_compress);
idxd_driver_unregister(&iaa_crypto_driver);
iaa_aecs_cleanup_fixed();
+ crypto_free_comp(deflate_generic_tfm);
pr_debug("cleaned up\n");
}
--
2.34.1
Hi, Tom,
On 2023-12-01 at 14:10:32 -0600, Tom Zanussi wrote:
[snip]
> +static int iaa_wq_put(struct idxd_wq *wq)
> +{
> + struct idxd_device *idxd = wq->idxd;
> + struct iaa_wq *iaa_wq;
> + bool free = false;
> + int ret = 0;
> +
> + spin_lock(&idxd->dev_lock);
> + iaa_wq = idxd_wq_get_private(wq);
> + if (iaa_wq) {
> + iaa_wq->ref--;
> + if (iaa_wq->ref == 0 && iaa_wq->remove) {
> + __free_iaa_wq(iaa_wq);
> + idxd_wq_set_private(wq, NULL);
> + free = true;
> + }
> + idxd_wq_put(wq);
> + } else {
> + ret = -ENODEV;
> + }
> + spin_unlock(&idxd->dev_lock);
__free_iaa_wq() may cause schedule, whether it should be move out of the
context between spin_lock() and spin_unlock()?
> + if (free)
> + kfree(iaa_wq);
> +
> + return ret;
> +}
[snip]
> @@ -800,12 +1762,38 @@ static void iaa_crypto_remove(struct idxd_dev *idxd_dev)
>
> remove_iaa_wq(wq);
>
> + spin_lock(&idxd->dev_lock);
> + iaa_wq = idxd_wq_get_private(wq);
> + if (!iaa_wq) {
> + spin_unlock(&idxd->dev_lock);
> + pr_err("%s: no iaa_wq available to remove\n", __func__);
> + goto out;
> + }
> +
> + if (iaa_wq->ref) {
> + iaa_wq->remove = true;
> + } else {
> + wq = iaa_wq->wq;
> + __free_iaa_wq(iaa_wq);
> + idxd_wq_set_private(wq, NULL);
> + free = true;
> + }
> + spin_unlock(&idxd->dev_lock);
__free_iaa_wq() may cause schedule, whether it should be move out of the
context between spin_lock() and spin_unlock()?
> +
> + if (free)
> + kfree(iaa_wq);
> +
> idxd_drv_disable_wq(wq);
> rebalance_wq_table();
>
> - if (nr_iaa == 0)
> + if (nr_iaa == 0) {
> + iaa_crypto_enabled = false;
Is it necessary to add iaa_unregister_compression_device() here?
All iaa devices are disabled cause the variable first_wq will be true,
if enable wq, iaa_register_compression_device() will fail due to the
algorithm is existed.
> free_wq_table();
> + module_put(THIS_MODULE);
>
> + pr_info("iaa_crypto now DISABLED\n");
> + }
> +out:
> mutex_unlock(&iaa_devices_lock);
> mutex_unlock(&wq->wq_lock);
> }
[snip]
Thanks,
Rex Zhang
> --
> 2.34.1
>
Hi Rex,
On Mon, 2023-12-04 at 23:00 +0800, Rex Zhang wrote:
> Hi, Tom,
>
> On 2023-12-01 at 14:10:32 -0600, Tom Zanussi wrote:
>
> [snip]
>
> > +static int iaa_wq_put(struct idxd_wq *wq)
> > +{
> > + struct idxd_device *idxd = wq->idxd;
> > + struct iaa_wq *iaa_wq;
> > + bool free = false;
> > + int ret = 0;
> > +
> > + spin_lock(&idxd->dev_lock);
> > + iaa_wq = idxd_wq_get_private(wq);
> > + if (iaa_wq) {
> > + iaa_wq->ref--;
> > + if (iaa_wq->ref == 0 && iaa_wq->remove) {
> > + __free_iaa_wq(iaa_wq);
> > + idxd_wq_set_private(wq, NULL);
> > + free = true;
> > + }
> > + idxd_wq_put(wq);
> > + } else {
> > + ret = -ENODEV;
> > + }
> > + spin_unlock(&idxd->dev_lock);
> __free_iaa_wq() may cause schedule, whether it should be move out of
> the
> context between spin_lock() and spin_unlock()?
Yeah, I suppose it makes more sense to have it below anyway, will move
it there.
> > + if (free)
> > + kfree(iaa_wq);
> > +
> > + return ret;
> > +}
>
> [snip]
>
> > @@ -800,12 +1762,38 @@ static void iaa_crypto_remove(struct
> > idxd_dev *idxd_dev)
> >
> > remove_iaa_wq(wq);
> >
> > + spin_lock(&idxd->dev_lock);
> > + iaa_wq = idxd_wq_get_private(wq);
> > + if (!iaa_wq) {
> > + spin_unlock(&idxd->dev_lock);
> > + pr_err("%s: no iaa_wq available to remove\n",
> > __func__);
> > + goto out;
> > + }
> > +
> > + if (iaa_wq->ref) {
> > + iaa_wq->remove = true;
> > + } else {
> > + wq = iaa_wq->wq;
> > + __free_iaa_wq(iaa_wq);
> > + idxd_wq_set_private(wq, NULL);
> > + free = true;
> > + }
> > + spin_unlock(&idxd->dev_lock);
> __free_iaa_wq() may cause schedule, whether it should be move out of
> the
> context between spin_lock() and spin_unlock()?
Same.
> > +
> > + if (free)
> > + kfree(iaa_wq);
> > +
> > idxd_drv_disable_wq(wq);
> > rebalance_wq_table();
> >
> > - if (nr_iaa == 0)
> > + if (nr_iaa == 0) {
> > + iaa_crypto_enabled = false;
> Is it necessary to add iaa_unregister_compression_device() here?
> All iaa devices are disabled cause the variable first_wq will be
> true,
> if enable wq, iaa_register_compression_device() will fail due to the
> algorithm is existed.
No, this is required by review input from a previous version - the
compression device can only be unregistered on module exit.
Thanks,
Tom
> > free_wq_table();
> > + module_put(THIS_MODULE);
> >
> > + pr_info("iaa_crypto now DISABLED\n");
> > + }
> > +out:
> > mutex_unlock(&iaa_devices_lock);
> > mutex_unlock(&wq->wq_lock);
> > }
>
> [snip]
>
> Thanks,
> Rex Zhang
> > --
> > 2.34.1
> >
Hi Tom,
On 2023-12-04 at 15:41:46 -0600, Tom Zanussi wrote:
> Hi Rex,
>
> On Mon, 2023-12-04 at 23:00 +0800, Rex Zhang wrote:
> > Hi, Tom,
> >
> > On 2023-12-01 at 14:10:32 -0600, Tom Zanussi wrote:
> >
> > [snip]
> >
> > > +static int iaa_wq_put(struct idxd_wq *wq)
> > > +{
> > > + struct idxd_device *idxd = wq->idxd;
> > > + struct iaa_wq *iaa_wq;
> > > + bool free = false;
> > > + int ret = 0;
> > > +
> > > + spin_lock(&idxd->dev_lock);
> > > + iaa_wq = idxd_wq_get_private(wq);
> > > + if (iaa_wq) {
> > > + iaa_wq->ref--;
> > > + if (iaa_wq->ref == 0 && iaa_wq->remove) {
> > > + __free_iaa_wq(iaa_wq);
> > > + idxd_wq_set_private(wq, NULL);
> > > + free = true;
> > > + }
> > > + idxd_wq_put(wq);
> > > + } else {
> > > + ret = -ENODEV;
> > > + }
> > > + spin_unlock(&idxd->dev_lock);
> > __free_iaa_wq() may cause schedule, whether it should be move out of
> > the
> > context between spin_lock() and spin_unlock()?
>
> Yeah, I suppose it makes more sense to have it below anyway, will move
> it there.
>
> > > + if (free)
> > > + kfree(iaa_wq);
> > > +
> > > + return ret;
> > > +}
> >
> > [snip]
> >
> > > @@ -800,12 +1762,38 @@ static void iaa_crypto_remove(struct
> > > idxd_dev *idxd_dev)
> > >
> > > remove_iaa_wq(wq);
> > >
> > > + spin_lock(&idxd->dev_lock);
> > > + iaa_wq = idxd_wq_get_private(wq);
> > > + if (!iaa_wq) {
> > > + spin_unlock(&idxd->dev_lock);
> > > + pr_err("%s: no iaa_wq available to remove\n",
> > > __func__);
> > > + goto out;
> > > + }
> > > +
> > > + if (iaa_wq->ref) {
> > > + iaa_wq->remove = true;
> > > + } else {
> > > + wq = iaa_wq->wq;
> > > + __free_iaa_wq(iaa_wq);
> > > + idxd_wq_set_private(wq, NULL);
> > > + free = true;
> > > + }
> > > + spin_unlock(&idxd->dev_lock);
> > __free_iaa_wq() may cause schedule, whether it should be move out of
> > the
> > context between spin_lock() and spin_unlock()?
>
> Same.
>
> > > +
> > > + if (free)
> > > + kfree(iaa_wq);
> > > +
> > > idxd_drv_disable_wq(wq);
> > > rebalance_wq_table();
> > >
> > > - if (nr_iaa == 0)
> > > + if (nr_iaa == 0) {
> > > + iaa_crypto_enabled = false;
> > Is it necessary to add iaa_unregister_compression_device() here?
> > All iaa devices are disabled cause the variable first_wq will be
> > true,
> > if enable wq, iaa_register_compression_device() will fail due to the
> > algorithm is existed.
>
> No, this is required by review input from a previous version - the
> compression device can only be unregistered on module exit.
Do it mean disabling all WQs followed by enabling WQ is unacceptable?
User must do "rmmod iaa_crypto" before enabling WQ in this case.
Thanks.
>
> Thanks,
>
> Tom
>
> > > free_wq_table();
> > > + module_put(THIS_MODULE);
> > >
> > > + pr_info("iaa_crypto now DISABLED\n");
> > > + }
> > > +out:
> > > mutex_unlock(&iaa_devices_lock);
> > > mutex_unlock(&wq->wq_lock);
> > > }
> >
> > [snip]
> >
> > Thanks,
> > Rex Zhang
> > > --
> > > 2.34.1
> > >
>
Hi Rex,
On Tue, 2023-12-05 at 10:26 +0800, Rex Zhang wrote:
> Hi Tom,
>
> On 2023-12-04 at 15:41:46 -0600, Tom Zanussi wrote:
> > Hi Rex,
> >
> > On Mon, 2023-12-04 at 23:00 +0800, Rex Zhang wrote:
> > > Hi, Tom,
> > >
> > > On 2023-12-01 at 14:10:32 -0600, Tom Zanussi wrote:
> > >
> > > [snip]
> > >
> > > > +static int iaa_wq_put(struct idxd_wq *wq)
> > > > +{
> > > > + struct idxd_device *idxd = wq->idxd;
> > > > + struct iaa_wq *iaa_wq;
> > > > + bool free = false;
> > > > + int ret = 0;
> > > > +
> > > > + spin_lock(&idxd->dev_lock);
> > > > + iaa_wq = idxd_wq_get_private(wq);
> > > > + if (iaa_wq) {
> > > > + iaa_wq->ref--;
> > > > + if (iaa_wq->ref == 0 && iaa_wq->remove) {
> > > > + __free_iaa_wq(iaa_wq);
> > > > + idxd_wq_set_private(wq, NULL);
> > > > + free = true;
> > > > + }
> > > > + idxd_wq_put(wq);
> > > > + } else {
> > > > + ret = -ENODEV;
> > > > + }
> > > > + spin_unlock(&idxd->dev_lock);
> > > __free_iaa_wq() may cause schedule, whether it should be move out
> > > of
> > > the
> > > context between spin_lock() and spin_unlock()?
> >
> > Yeah, I suppose it makes more sense to have it below anyway, will
> > move
> > it there.
> >
> > > > + if (free)
> > > > + kfree(iaa_wq);
> > > > +
> > > > + return ret;
> > > > +}
> > >
> > > [snip]
> > >
> > > > @@ -800,12 +1762,38 @@ static void iaa_crypto_remove(struct
> > > > idxd_dev *idxd_dev)
> > > >
> > > > remove_iaa_wq(wq);
> > > >
> > > > + spin_lock(&idxd->dev_lock);
> > > > + iaa_wq = idxd_wq_get_private(wq);
> > > > + if (!iaa_wq) {
> > > > + spin_unlock(&idxd->dev_lock);
> > > > + pr_err("%s: no iaa_wq available to remove\n",
> > > > __func__);
> > > > + goto out;
> > > > + }
> > > > +
> > > > + if (iaa_wq->ref) {
> > > > + iaa_wq->remove = true;
> > > > + } else {
> > > > + wq = iaa_wq->wq;
> > > > + __free_iaa_wq(iaa_wq);
> > > > + idxd_wq_set_private(wq, NULL);
> > > > + free = true;
> > > > + }
> > > > + spin_unlock(&idxd->dev_lock);
> > > __free_iaa_wq() may cause schedule, whether it should be move out
> > > of
> > > the
> > > context between spin_lock() and spin_unlock()?
> >
> > Same.
> >
> > > > +
> > > > + if (free)
> > > > + kfree(iaa_wq);
> > > > +
> > > > idxd_drv_disable_wq(wq);
> > > > rebalance_wq_table();
> > > >
> > > > - if (nr_iaa == 0)
> > > > + if (nr_iaa == 0) {
> > > > + iaa_crypto_enabled = false;
> > > Is it necessary to add iaa_unregister_compression_device() here?
> > > All iaa devices are disabled cause the variable first_wq will be
> > > true,
> > > if enable wq, iaa_register_compression_device() will fail due to
> > > the
> > > algorithm is existed.
> >
> > No, this is required by review input from a previous version - the
> > compression device can only be unregistered on module exit.
> Do it mean disabling all WQs followed by enabling WQ is unacceptable?
> User must do "rmmod iaa_crypto" before enabling WQ in this case.
>
Right, as mentioned in the documentation:
+If a different configuration or set of driver attributes is required,
+the user must first disable the IAA devices and workqueues, reset the
+configuration, and then re-register the deflate-iaa algorithm with the
+crypto subsystem by removing and reinserting the iaa_crypto module.
Thanks,
Tom
> Thanks.
> >
> > Thanks,
> >
> > Tom
> >
> > > > free_wq_table();
> > > > + module_put(THIS_MODULE);
> > > >
> > > > + pr_info("iaa_crypto now DISABLED\n");
> > > > + }
> > > > +out:
> > > > mutex_unlock(&iaa_devices_lock);
> > > > mutex_unlock(&wq->wq_lock);
> > > > }
> > >
> > > [snip]
> > >
> > > Thanks,
> > > Rex Zhang
> > > > --
> > > > 2.34.1
> > > >
> >