Revision 9f28ffc03e93343ac04874fda9edb7affea45165 authored by David S. Miller on 19 December 2012, 23:19:11 UTC, committed by David S. Miller on 19 December 2012, 23:19:11 UTC
The basic scheme of the block mode assembler is that we start by
enabling the FPU, loading the key into the floating point registers,
then iterate calling the encrypt/decrypt routine for each block.
For the 256-bit key cases, we run short on registers in the unrolled
loops.
So the {ENCRYPT,DECRYPT}_256_2() macros reload the key registers that
get clobbered.
The unrolled macros, {ENCRYPT,DECRYPT}_256(), are not mindful of this.
So if we have a mix of multi-block and single-block calls, the
single-block unrolled 256-bit encrypt/decrypt can run with some
of the key registers clobbered.
Handle this by always explicitly loading those registers before using
the non-unrolled 256-bit macro.
This was discovered thanks to all of the new test cases added by
Jussi Kivilinna.
Signed-off-by: David S. Miller <davem@davemloft.net>
1 parent 4a9d194
| File | Mode | Size |
|---|---|---|
| mmp | ||
| mvebu | ||
| mxs | ||
| socfpga | ||
| spear | ||
| ux500 | ||
| versatile | ||
| Kconfig | -rw-r--r-- | 1.8 KB |
| Makefile | -rw-r--r-- | 1.0 KB |
| clk-bcm2835.c | -rw-r--r-- | 1.9 KB |
| clk-devres.c | -rw-r--r-- | 1.1 KB |
| clk-divider.c | -rw-r--r-- | 8.4 KB |
| clk-fixed-factor.c | -rw-r--r-- | 2.5 KB |
| clk-fixed-rate.c | -rw-r--r-- | 2.8 KB |
| clk-gate.c | -rw-r--r-- | 3.7 KB |
| clk-highbank.c | -rw-r--r-- | 8.1 KB |
| clk-ls1x.c | -rw-r--r-- | 2.8 KB |
| clk-max77686.c | -rw-r--r-- | 5.7 KB |
| clk-mux.c | -rw-r--r-- | 3.2 KB |
| clk-nomadik.c | -rw-r--r-- | 1.5 KB |
| clk-prima2.c | -rw-r--r-- | 25.8 KB |
| clk-sunxi.c | -rw-r--r-- | 906 bytes |
| clk-twl6040.c | -rw-r--r-- | 3.2 KB |
| clk-u300.c | -rw-r--r-- | 21.4 KB |
| clk-vt8500.c | -rw-r--r-- | 12.5 KB |
| clk-wm831x.c | -rw-r--r-- | 9.8 KB |
| clk-zynq.c | -rw-r--r-- | 9.3 KB |
| clk.c | -rw-r--r-- | 43.6 KB |
| clkdev.c | -rw-r--r-- | 7.4 KB |
Computing file changes ...
