Searched hist:q (Results 1 - 25 of 605) sorted by relevance
| /src/usr.bin/make/unit-tests/ | ||
| H A D | opt-query.exp | 1.4 Wed Aug 17 20:10:29 GMT 2022 rillig make: fix exit status for '-q' (since 1994) 1.3 Wed Aug 17 20:05:41 GMT 2022 rillig tests/make: demonstrate wrong exit status for '-q' (since 1994) Reported by Jeroen Ruigrok van der Werven via private mail. 1.2 Wed Aug 19 05:13:18 GMT 2020 rillig make(1): add test for the -q option |
| H A D | opt-query.mk | 1.7 Thu Aug 18 05:37:05 GMT 2022 rillig tests/make: fix test for option '-q' in ATF mode When running 'make test' in usr.bin/make, MAKE is set to '$PWD/make', and when that file is used as a dependency, everything works as expected. When running the tests via ATF, MAKE is set to simply 'make', based on argv[0]. Using 'make' as a dependency searches in the current directory but not in /usr/bin, so the file is not found, which makes the "up-to-date" target out of date. Switch the dependeny from ${MAKE} to ${MAKEFILE}, as that file does not involve any $PATH magic and is also guaranteed to be older than the 'up-to-date' file that is created while running the test. 1.6 Wed Aug 17 20:10:29 GMT 2022 rillig make: fix exit status for '-q' (since 1994) 1.5 Wed Aug 17 20:05:41 GMT 2022 rillig tests/make: demonstrate wrong exit status for '-q' (since 1994) Reported by Jeroen Ruigrok van der Werven via private mail. 1.3 Wed Aug 19 05:13:18 GMT 2020 rillig make(1): add test for the -q option |
| /src/bin/ksh/ | ||
| H A D | emacs-gen.sh | 1.4 Sat Oct 25 22:18:15 GMT 2008 apb branches: 1.4.62; In shell scripts run during the build, add a SED variable, defaulting to "sed". SED=${TOOL_SED:Q} should be passed in the environment to override this. 1.3 Sun Oct 19 22:10:04 GMT 2008 apb In shell scripts invoked during a build, and in crunchgen, use ${AWK} instead of plain "awk". The Makefiles that invoke these scripts or programs will pass AWK=${HOST_AWK:Q}. |
| H A D | mkman | 1.3 Sun Oct 19 22:10:04 GMT 2008 apb branches: 1.3.62; In shell scripts invoked during a build, and in crunchgen, use ${AWK} instead of plain "awk". The Makefiles that invoke these scripts or programs will pass AWK=${HOST_AWK:Q}. |
| /src/sys/dev/usb/ | ||
| H A D | auvitek_board.c | 1.2 Tue Dec 28 04:02:33 GMT 2010 jmcneill branches: 1.2.6; Hauppauge HVR-850 analog should be identical to HVR-950Q, so support it too 1.1 Mon Dec 27 15:42:11 GMT 2010 jmcneill add driver for the Auvitek AU0828 USB video controllers's analog video capture functions: auvitek0 at uhub6 port 2: AU0828 video0 at auvitek0: WinTV HVR-950Q uaudio0 at auvitek0 port 2 configuration 1 interface 1 uaudio0: vendor 0x2040 product 0x7200, rev 2.00/0.05, addr 2 uaudio0: audio rev 1.00 audio1 at uaudio0: full duplex, playback, capture, independent |
| H A D | auvitekreg.h | 1.1 Mon Dec 27 15:42:11 GMT 2010 jmcneill branches: 1.1.6; add driver for the Auvitek AU0828 USB video controllers's analog video capture functions: auvitek0 at uhub6 port 2: AU0828 video0 at auvitek0: WinTV HVR-950Q uaudio0 at auvitek0 port 2 configuration 1 interface 1 uaudio0: vendor 0x2040 product 0x7200, rev 2.00/0.05, addr 2 uaudio0: audio rev 1.00 audio1 at uaudio0: full duplex, playback, capture, independent |
| /src/sys/lib/libgnuefi/ | ||
| H A D | Makefile.inc | 1.5 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.5 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.5 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.5 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 |
| /src/sys/arch/acorn32/stand/lib/ | ||
| H A D | Makefile.inc | 1.4 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.4 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.4 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.4 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 |
| /src/sys/dev/pcmcia/ | ||
| H A D | Makefile.pcmciadevs | 1.2 Sun Oct 19 22:05:23 GMT 2008 apb Use ${TOOL_AWK} instead of ${AWK} or plain "awk" in make commands. Pass AWK=${TOOL_AWK:Q} to shell scripts that use awk. |
| /src/sys/dev/podulebus/ | ||
| H A D | Makefile.podules | 1.2 Sun Oct 19 22:05:23 GMT 2008 apb Use ${TOOL_AWK} instead of ${AWK} or plain "awk" in make commands. Pass AWK=${TOOL_AWK:Q} to shell scripts that use awk. |
| /src/sys/dev/videomode/ | ||
| H A D | Makefile.videomode | 1.2 Sun Oct 19 22:05:23 GMT 2008 apb Use ${TOOL_AWK} instead of ${AWK} or plain "awk" in make commands. Pass AWK=${TOOL_AWK:Q} to shell scripts that use awk. |
| /src/sys/arch/hppa/spmath/ | ||
| H A D | Makefile.inc | 1.10 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.10 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.10 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.10 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.8 Mon Jan 25 18:55:25 GMT 2016 christos use :Q to quote variables properly. |
| /src/sys/crypto/aes/arch/x86/ | ||
| H A D | files.aessse2 | 1.1 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.1 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.1 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.1 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.1 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.1 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. |
| /src/share/man/man8/man8.x68k/ | ||
| H A D | loadbsd.8 | 1.7 Sat Dec 23 20:15:19 GMT 2023 andvar Update documentation with -q and -N options. 1.6 Sat Dec 23 19:13:55 GMT 2023 andvar Remove obsolete -d flag from documentation. P.S. -q and -N flags need to be added. |
| /src/sys/dev/i2c/ | ||
| H A D | au8522reg.h | 1.1 Mon Dec 27 15:42:11 GMT 2010 jmcneill branches: 1.1.6; add driver for the Auvitek AU0828 USB video controllers's analog video capture functions: auvitek0 at uhub6 port 2: AU0828 video0 at auvitek0: WinTV HVR-950Q uaudio0 at auvitek0 port 2 configuration 1 interface 1 uaudio0: vendor 0x2040 product 0x7200, rev 2.00/0.05, addr 2 uaudio0: audio rev 1.00 audio1 at uaudio0: full duplex, playback, capture, independent |
| H A D | xc5kreg.h | 1.1 Mon Dec 27 15:42:11 GMT 2010 jmcneill branches: 1.1.6; add driver for the Auvitek AU0828 USB video controllers's analog video capture functions: auvitek0 at uhub6 port 2: AU0828 video0 at auvitek0: WinTV HVR-950Q uaudio0 at auvitek0 port 2 configuration 1 interface 1 uaudio0: vendor 0x2040 product 0x7200, rev 2.00/0.05, addr 2 uaudio0: audio rev 1.00 audio1 at uaudio0: full duplex, playback, capture, independent |
| H A D | xc5kvar.h | 1.1 Mon Dec 27 15:42:11 GMT 2010 jmcneill branches: 1.1.6; add driver for the Auvitek AU0828 USB video controllers's analog video capture functions: auvitek0 at uhub6 port 2: AU0828 video0 at auvitek0: WinTV HVR-950Q uaudio0 at auvitek0 port 2 configuration 1 interface 1 uaudio0: vendor 0x2040 product 0x7200, rev 2.00/0.05, addr 2 uaudio0: audio rev 1.00 audio1 at uaudio0: full duplex, playback, capture, independent |
| H A D | au8522var.h | 1.1 Mon Dec 27 15:42:11 GMT 2010 jmcneill branches: 1.1.2; 1.1.6; add driver for the Auvitek AU0828 USB video controllers's analog video capture functions: auvitek0 at uhub6 port 2: AU0828 video0 at auvitek0: WinTV HVR-950Q uaudio0 at auvitek0 port 2 configuration 1 interface 1 uaudio0: vendor 0x2040 product 0x7200, rev 2.00/0.05, addr 2 uaudio0: audio rev 1.00 audio1 at uaudio0: full duplex, playback, capture, independent |
| /src/sys/dev/isapnp/ | ||
| H A D | Makefile.isapnpdevs | 1.2 Sun Oct 19 22:05:22 GMT 2008 apb Use ${TOOL_AWK} instead of ${AWK} or plain "awk" in make commands. Pass AWK=${TOOL_AWK:Q} to shell scripts that use awk. |
| /src/sys/arch/ia64/stand/efi/libefi/ | ||
| H A D | Makefile.inc | 1.5 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.5 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.5 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 1.5 Sun May 27 01:14:50 GMT 2018 christos - Introduce :q modifier for make variables and make it double escape $'s so that passing variables to recursive makes with :q works as expected. - Revert :Q to work as before. - Adjust makefiles that use recursive make to use :q Discussed on tech-toolchain@ XXX: pullup 8 |
| /src/share/man/man3/ | ||
| H A D | __alignof__.3 | 1.5 Thu Apr 14 06:56:28 GMT 2011 wiz No need to use \*[q], use plain double quotes instead. |
| H A D | offsetof.3 | 1.5 Thu Apr 14 06:56:28 GMT 2011 wiz No need to use \*[q], use plain double quotes instead. |
| /src/sys/dev/mca/ | ||
| H A D | Makefile.mcadevs | 1.2 Sun Oct 19 22:05:22 GMT 2008 apb Use ${TOOL_AWK} instead of ${AWK} or plain "awk" in make commands. Pass AWK=${TOOL_AWK:Q} to shell scripts that use awk. |
| /src/sys/crypto/aes/ | ||
| H A D | aes.h | 1.2 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.2 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.2 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.2 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.2 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. 1.2 Mon Jun 29 23:47:54 GMT 2020 riastradh New SSE2-based bitsliced AES implementation. This should work on essentially all x86 CPUs of the last two decades, and may improve throughput over the portable C aes_ct implementation from BearSSL by (a) reducing the number of vector operations in sequence, and (b) batching four rather than two blocks in parallel. Derived from BearSSL'S aes_ct64 implementation adjusted so that where aes_ct64 uses 64-bit q[0],...,q[7], aes_sse2 uses (q[0], q[4]), ..., (q[3], q[7]), each tuple representing a pair of 64-bit quantities stacked in a single 128-bit register. This translation was done very naively, and mostly reduces the cost of ShiftRows and data movement without doing anything to address the S-box or (Inv)MixColumns, which spread all 64-bit quantities across separate registers and ignore the upper halves. Unfortunately, SSE2 -- which is all that is guaranteed on all amd64 CPUs -- doesn't have PSHUFB, which would help out a lot more. For example, vpaes relies on that. Perhaps there are enough CPUs out there with PSHUFB but not AES-NI to make it worthwhile to import or adapt vpaes too. Note: This includes local definitions of various Intel compiler intrinsics for gcc and clang in terms of their __builtin_* &c., because the necessary header files are not available during the kernel build. This is a kludge -- we should fix it properly; the present approach is expedient but not ideal. |
| /src/sys/arch/atari/stand/tostools/aptck/ | ||
| H A D | Makefile | 1.4 Sun Oct 19 22:05:21 GMT 2008 apb Use ${TOOL_AWK} instead of ${AWK} or plain "awk" in make commands. Pass AWK=${TOOL_AWK:Q} to shell scripts that use awk. |
Completed in 63 milliseconds