| Message ID | 20220617043935.428083-1-rashmica@linux.ibm.com (mailing list archive) |
|---|---|
| State | Accepted |
| Headers | show |
| Series | selftests/powerpc: Fix matrix multiply assist test | expand |
| Context | Check | Description |
|---|---|---|
| snowpatch_ozlabs/github-powerpc_ppctests | success | Successfully ran 10 jobs. |
| snowpatch_ozlabs/github-powerpc_selftests | success | Successfully ran 10 jobs. |
| snowpatch_ozlabs/github-powerpc_clang | warning | kernel (ppc44x, ubuntu-21.10, ppc64) found 2 issues. |
| snowpatch_ozlabs/github-powerpc_sparse | success | Successfully ran 4 jobs. |
| snowpatch_ozlabs/github-powerpc_kernel_qemu | success | Successfully ran 23 jobs. |
On Fri, 17 Jun 2022 14:39:35 +1000, Rashmica Gupta wrote: > The ISA states: "when ACC[i] contains defined data, the contents of VSRs > 4×i to 4×i+3 are undefined until either a VSX Move From ACC instruction > is used to copy the contents of ACC[i] to VSRs 4×i to 4×i+3 or some other > instruction directly writes to one of these VSRs." We aren't doing this. > > This test only works on Power10 because the hardware implementation > happens to map ACC0 to VSRs 0-3, but will fail on any other implementation > that doesn't do this. So add xxmfacc between writing to the accumulator > and accessing the VSRs. > > [...] Applied to powerpc/next. [1/1] selftests/powerpc: Fix matrix multiply assist test https://git.kernel.org/powerpc/c/cd1e64935f79e31d666172c52c951ca97152b783 cheers
diff --git a/tools/testing/selftests/powerpc/math/mma.S b/tools/testing/selftests/powerpc/math/mma.S index 8528c9849565..61cc88b1b26b 100644 --- a/tools/testing/selftests/powerpc/math/mma.S +++ b/tools/testing/selftests/powerpc/math/mma.S @@ -20,6 +20,9 @@ test_mma: /* xvi16ger2s */ .long 0xec042958 + /* Deprime the accumulator - xxmfacc 0 */ + .long 0x7c000162 + /* Store result in image passed in r5 */ stxvw4x 0,0,5 addi 5,5,16
The ISA states: "when ACC[i] contains defined data, the contents of VSRs 4×i to 4×i+3 are undefined until either a VSX Move From ACC instruction is used to copy the contents of ACC[i] to VSRs 4×i to 4×i+3 or some other instruction directly writes to one of these VSRs." We aren't doing this. This test only works on Power10 because the hardware implementation happens to map ACC0 to VSRs 0-3, but will fail on any other implementation that doesn't do this. So add xxmfacc between writing to the accumulator and accessing the VSRs. Fixes commit 3527e1ab9a79 ("selftests/powerpc: Add matrix multiply assist (MMA) test") Signed-off-by: Rashmica Gupta <rashmica@linux.ibm.com> --- tools/testing/selftests/powerpc/math/mma.S | 3 +++ 1 file changed, 3 insertions(+)