kernel/linux.git/tools/testing/selftests/powerpc/alignment, branch v4.19.237

selftests/powerpc: Only test lwm/stmw on big endian

2021-02-07T13:48:38+00:00

[ Upstream commit dd3a44c06f7b4f14e90065bf05d62c255b20005f ] Newer binutils (>= 2.36) refuse to assemble lmw/stmw when building in little endian mode. That breaks compilation of our alignment handler test: /tmp/cco4l14N.s: Assembler messages: /tmp/cco4l14N.s:1440: Error: `lmw' invalid when little-endian /tmp/cco4l14N.s:1814: Error: `stmw' invalid when little-endian make[2]: *** [../../lib.mk:139: /output/kselftest/powerpc/alignment/alignment_handler] Error 1 These tests do pass on little endian machines, as the kernel will still emulate those instructions even when running little endian (which is arguably a kernel bug). But we don't really need to test that case, so ifdef those instructions out to get the alignment test building again. Reported-by: Libor Pechacek Signed-off-by: Michael Ellerman Tested-by: Libor Pechacek Link: https://lore.kernel.org/r/20210119041800.3093047-1-mpe@ellerman.id.au Signed-off-by: Sasha Levin

selftests/powerpc: Fix Makefiles for headers_install change

2018-09-28T05:07:45+00:00

Commit b2d35fa5fc80 ("selftests: add headers_install to lib.mk") introduced a requirement that Makefiles more than one level below the selftests directory need to define top_srcdir, but it didn't update any of the powerpc Makefiles. This broke building all the powerpc selftests with eg: make[1]: Entering directory '/src/linux/tools/testing/selftests/powerpc' BUILD_TARGET=/src/linux/tools/testing/selftests/powerpc/alignment; mkdir -p $BUILD_TARGET; make OUTPUT=$BUILD_TARGET -k -C alignment all make[2]: Entering directory '/src/linux/tools/testing/selftests/powerpc/alignment' ../../lib.mk:20: ../../../../scripts/subarch.include: No such file or directory make[2]: *** No rule to make target '../../../../scripts/subarch.include'. make[2]: Failed to remake makefile '../../../../scripts/subarch.include'. Makefile:38: recipe for target 'alignment' failed Fix it by setting top_srcdir in the affected Makefiles. Fixes: b2d35fa5fc80 ("selftests: add headers_install to lib.mk") Signed-off-by: Michael Ellerman

selftests/powerpc: Add more version checks to alignment_handler test

2018-08-07T14:32:28+00:00

The alignment_handler is documented to only work on Power8/Power9, but we can make it run on older CPUs by guarding more of the tests with feature checks. Signed-off-by: Michael Ellerman Reviewed-by: Andrew Donnellan

selftests/powerpc: Skip earlier in alignment_handler test

2018-08-07T14:32:28+00:00

Currently the alignment_handler test prints "Can't open /dev/fb0" about 80 times per run, which is a little annoying. Refactor it to check earlier if it can open /dev/fb0 and skip if not, this results in each test printing something like: test: test_alignment_handler_vsx_206 tags: git_version:v4.18-rc3-134-gfb21a48904aa [SKIP] Test skipped on line 291 skip: test_alignment_handler_vsx_206 Signed-off-by: Michael Ellerman Acked-by: Andrew Donnellan

selftests/powerpc: Consolidate copy/paste test logic

2018-07-20T02:50:52+00:00

This logic was shared between multiple tests, but now that we have removed all but one of them we can just move it into that test. Signed-off-by: Michael Ellerman

selftests/powerpc: Remove Power9 paste tests

2018-07-20T02:50:52+00:00

Paste on POWER9 only works to accelerators and not on real memory. So these tests just generate a SIGILL. So just delete them. Signed-off-by: Michael Ellerman Acked-by: Michael Neuling Signed-off-by: Michael Ellerman

selftests/powerpc: Remove Power9 copy_unaligned test

2018-07-20T02:50:51+00:00

This is a test of the ISA 3.0 "copy" instruction. That instruction has an L field, which if set to 1 specifies that "the instruction identifies the beginning of a move group" (pp 858). That's also referred to as "copy first" vs "copy". In ISA 3.0B the copy instruction does not have an L field, and the corresponding bit in the instruction must be set to 1. This test is generating a "copy" instruction, not a "copy first", and so on Power9 (which implements 3.0B), this results in an illegal instruction. So just drop the test entirely. We still have copy_first_unaligned to test the "copy first" behaviour. Signed-off-by: Michael Ellerman Acked-by: Michael Neuling Signed-off-by: Michael Ellerman

selftests/powerpc: Add missing .gitignores

2018-05-24T14:08:25+00:00

Signed-off-by: Michael Neuling Signed-off-by: Michael Ellerman

selftests/powerpc: Fix to use ucontext_t instead of struct ucontext

2018-02-13T11:44:18+00:00

With glibc 2.26 'struct ucontext' is removed to improve POSIX compliance, which breaks powerpc/alignment_handler selftest. Fix the test by using ucontext_t. Tested on ppc, works with older glibc versions as well. Fixes the following: alignment_handler.c: In function ‘sighandler’: alignment_handler.c:68:5: error: dereferencing pointer to incomplete type ‘struct ucontext’ ucp->uc_mcontext.gp_regs[PT_NIP] += 4; Signed-off-by: Harish Signed-off-by: Michael Ellerman

selftests/powerpc: Add alignment handler selftest

2018-01-21T18:48:34+00:00

Add a selftest to exercise the powerpc alignment fault handler. Signed-off-by: Michael Neuling Signed-off-by: Andrew Donnellan [mpe: Add 32-bit support to the signal handler] Signed-off-by: Michael Ellerman