- linaro-toolchain - lists.linaro.org

by Renato Golin

== Progress == * Support (2/10) - Having a go at PR25722, too hacky for a feature that can be easily worked around. - Reviewing some kernel issues with Arnd * Planning (6/10) - Drafting 2 year plans for LLVM * Background (2/10) - Code review, meetings, discussions, general support, etc. - Catching up from long holidays

9 years, 11 months

1
0
0 0

[ACTIVITY] 4-8 Jan 2016

by Yao Qi

# Progress # * Handle input interrupt in GDB. TCWG-424. Done. [3/10] Need to update GDB manual to clarify the expected behaviour of GDB. * Estimate the effort of GDB kernel-awareness work. Done. [1/10] * TCWG-491. Ongoing. [2/10]. Understand symbol handling in GDB. * Various patch review upstream. [2/10]. * Clean up code on arm software single step after some changes from Ericsson. Ongoing. [2/10]. # Plan # * TCWG-491. * Follow up of TCWG-424 to update GDB manual. * Clean up code on arm software single step. * Assess the ARM and AArch64 GDB test result, as 7.11 release is coming soon. -- Yao

9 years, 11 months

1
0
0 0

Re: gcc-linaro-5.1 vs gcc-linaro-4.8

by Rob Savoye

Is it in the 2015.11-1 release ? - rob - -------- Original message -------- From: Jim Wilson <jim.wilson(a)linaro.org> Date: 01/05/2016 19:45 (GMT-07:00) To: Xiaofeng Ren <xiaofeng.ren(a)nxp.com> Cc: linaro-toolchain(a)lists.linaro.org, Zhenhua Luo <zhenhua.luo(a)nxp.com> Subject: Re: gcc-linaro-5.1 vs gcc-linaro-4.8 On Tue, Jan 5, 2016 at 4:19 PM, Xiaofeng Ren <xiaofeng.ren(a)nxp.com> wrote: > Hello Jim, > Appreciate for your comments. > I will try to manually apply that patch on my side and try it. > BTW, may I know which released Linaro gcc version include that patch? Maybe I can download it and try it quickly. > https://gcc.gnu.org/ml/gcc-patches/2015-10/msg00025.html It was backported to our gcc-5 branch on Nov 24 by Yvan. This is after the latest release 2015-11 was made. The patch is in the December snapshot, but I think that is a source only release. http://snapshots.linaro.org/components/toolchain/gcc-linaro/5.3-2015.12/ You would have to build your own toolchain from that, perhaps by using abe. Jim _______________________________________________ linaro-toolchain mailing list linaro-toolchain(a)lists.linaro.org https://lists.linaro.org/mailman/listinfo/linaro-toolchain

9 years, 11 months

2
1
0 0

debug information in libc.a

by Fred Peterson

Linaro TCWG, In newer toolchains that are built with ABE, libc.a contains a lot of debugging information, including the paths to the source files on the build machine. I think that's because ABE builds the libraries with -g and never strips out the debug information. I verified this with both the 4.9-2015.05 and 5.2-2015.11 binary releases with the command: arm-linux-gnueabih-objdump -g libc.a | grep '\.c' In older toolchains that were built with crosstool-ng, libc.a did not contain the paths to the source files. I guess crosstool-ng either didn't build the libraries with -g, or it stripped out the debug information later. I verified this with the 4.9-2014.09 binary release. I'm not sure whether this change was intentional, or just an oversight during the switchover to ABE. Regardless, it makes the libraries a lot bigger, and it potentially affects the end user during debugging. The source files of libc, etc. are not typically included with the binary releases. So, when a user of an ABE-built binary release chooses to step into an extern function of libc, gdb will search for the source file. It likely won't be able to access the source file along the same path that worked for the build machine, so it will search its list of source directories. Ultimately, unless the user has downloaded the source files, gdb will likely display a message like "printf.c: No such file or directory". In contrast, when a user of a crosstool-ng-built toolchain tries to step into an extern function of libc, gdb will be unaware of the name of the source file. As a result, the user will not get a message about a missing file. So, should the toolchains' libraries really contain debug information? I think it could be useful for a theoretical multilib folder that covers a -g option. On the other hand, for the usual release builds, isn't the debug information a waste of space and a source of confusion? Thanks, Fred Peterson Engineer - Developer Tools NXP Semiconductors

9 years, 11 months

2
1
0 0

gcc-linaro-5.1 vs gcc-linaro-4.8

by Xiaofeng Ren

Hello All, I found one difference between gcc-linaro-5.1 vs gcc-linaro-4.8 while I'm doing lmbench benchmark test for our LS1043 (cortex-A53). While using gcc-linaro-4.8, gcc will generate advanced SIMD instructions (like as ld1, etc), however, gcc-linaro-5.1 will not generate advance SIMD instructions. This will cause big performance gap between gcc-4.8 and gcc-5.1 for lmbench memory bandwidth "fcp" test (bw_mem program). My compiler flags is "-O3 -mcpu=cortex-a53". I also tried several different compiler flags ("-O3 -mcpu=cortex-a53+fp+simd", "-O2 -ftree-vectorize -mcpu=cortex-a53", "-O3 -ftree-vectorize -mcpu=cortex-a53"), all of them doesn't work. Gcc-5.1 toolchain was downloaded from following link: https://snapshots.linaro.org/openembedded/sources/gcc-linaro-5.1-snapshot-2… Can I have your comments on this? Thanks Ron

9 years, 11 months

4
8
0 0

mixed LTO support for 'ld -r'

by Nicolas Pitre

Hello toolchain gurus, In the course of Linaro's kernel tinification project, the ability to compile the Linux kernel using LTO is a frequent requirement. However the kernel makes heavy usage of 'ld -r' with .o files resulting from LTO build of .c files as well as .o files resulting from pure assembly code. This mix of LTO and non-LTO object files is not supported by upstream binutils unless a patch from H.J. Lu is applied. That patch has been available since 2013 and was last refreshed in his 2.25.51.0.4 branch last September. It is accessible here: https://git.linaro.org/toolchain/binutils-gdb.git/commit/6da5456971 I've attached a very simple test case demonstrating the problem. With the binutils-lto-mixed.patch applied, this test case compiles to a working executable. Otherwise compilation fails at the 'ld -r' step. One question and one request: - What, if anything, has prevented this patch from being merged in the master branch upstream? - In the mean time, could we include this patch in the Linaro binutils package and releases? Having this available in our toolchain releases would greatly simplify the LTO related work on the kernel. It was included in all binutils releases from H.J. Lu since 2013 and therefore has obtained significant exposure already. Thanks. Nicolas

9 years, 11 months

4
10
0 0

[ACTIVITY] 21-23 Dec 2015

by Prathamesh Kulkarni

== This Week == * LTO 447.dealII segfault (2/10) - Breaks only with balanced partitioning, works with other types. - Worked with gcc-5.3, having issues in git bisect * Target Hook conversion (2/10) - Converted SET_ASM_OP to hook * TCWG-319 benchmarking (1/10) - fp benchmark results: a53: 7.16% r1-a12: ~3% a57: 0.35% a15: 0.64% * ipa (1/10) == Holidays upto 4th Jan ==

9 years, 11 months

1
0
0 0

[ACTIVITY] 14 - 18 Dec 2015

by Prathamesh Kulkarni

== This Week == * TCWG-72 (2/10) - Trying to address segfault with DImode * LTO ICE with 483.xalancbmk (6/10) - fighting with benchmark scripts - segfaults with ICE on -flto-partitions=none on symbol _ZThn8_N10xalanc_1_819XercesParserLiaison11resetErrorsEv demangler.com says the symbol is non-virtual thunk to: non-virtual thunk to xalanc_1_8::XercesParserLiaison::resetErrors() ./-lm.res (resolution map file) says PREVAILING_DEF_IRONLY - Trace: http://pastebin.com/vxzmQFHg - Possible fix: http://pastebin.com/rwq8z1N1 Patch passes test and bootstrap, not sure if that's correct approach however. * Target hook conversion (2/10) - Unconditionalizing ASM_OUTPUT_DEF on SET_ASM_OP and converting both to hooks.

9 years, 11 months

1
0
0 0

[ACTIVITY] 14th - 21st December

by Bernie Ogden

NB Last _6_ working days. Centralised benchmark source - TCWG-354 [6/10] * Abe integration * Wrote up some notes on collaborate * Enabled clang build, which flushed out some bugs, now fixed * LAVA reporting script Port to microinstance - TCWG-432 [1/10] * Investigating issues with Juno boot ** One turns out to be user error ** The other is a failed heath check, passed to the admins Backport benchmarking - TCWG-352 [2/10] * Difficulties with passing information from matrix children to parent ** Came up with an ugly workaround, may have thought of a better one Misc [3/10] =Plan= Holidays until 4th January

9 years, 12 months

1
0
0 0

[ACTIVITY] 14th-18th December

by Yao Qi

# Progress # * TCWG-156, GDB Test-Suite Parity Between Aarch64 and x86_64. Done. [4/10] After two patches are committed, except some tests written for x86_64 unnecessarily, the test results between aarch64 and x86_64 looks no difference. * TCWG-424, timeout when interrupt the inferior in remote debugging. [3/10] The fail is caused by different two problems. Two patches are ready, and being regression tested. * TCWG-171, Enable gdb core file tests when testing remotely. [1/10] Write down my conclusion as it can't be fixed. * Upstream review, [2/10] ** Review patch about handling ada aarch64 HVA array in GDB. ** Discuss target description of GDB for cortex-m device with openocd. # Plan # * Post patches upstream for TCWG-424, * Patches review. * On holiday since Wed. -- Yao

10 years

1
0
0 0

Linaro LLVM engagement

by Virendra Kumar Pathak

Hi All, I am interested in understanding Linaro LLVM activity. I have already gone through https://wiki.linaro.org/WorkingGroups/ToolChain/LLVM. Could you please guide me on below questions. 1. On which LLVM & clang version, linaro is actively working now ? 2. Where can I find the latest "linaro-llvm" source code & binary? I could not find any official git repo for "linaro-llvm" at https://git.linaro.org/. 3. Could you please explain Linaro LLVM working model? How similar/different it is when compared with Linaro-GCC engagement. 4. Certain links (e.g Roadmap) at https://wiki.linaro.org/WorkingGroups/ToolChain/LLVM ask for login credentials. Any comment on how to obtain the permission? Thanks in advance for your time. -- with regards, Virendra Kumar Pathak

10 years

3
2
0 0

[ACTIVITY] 14-17 December 2015

by Christophe Lyon

* 1 day off (Friday) (2/10) == Progress == * Validation: (2/10) - improvements to the comparison scripts - updated ABE stable and staging branches * GCC: (3/10) - TCWG-484: sent a new version of the patch to address problems with the testsuite and target attributes. - TCWG-485/PR68620: WIP, trying to understand the regressions introduced by my patch * Misc (conf calls, meetings, emails, ...) (3/10) == Next == Holidays until next year

10 years

1
0
0 0

Linaro-binutils release git tags

by Virendra Kumar Pathak

Hi All, As per release notes, linaro_binutils-2_25-2015_01-2_release should be present at http://git.linaro.org/toolchain/binutils-gdb.git. But I am not able to see any "linaro_binutils" tags on https://git.linaro.org/toolchain/binutils-gdb.git/tags. Am I doing any mistake? or on git it is renamed to something else (same as FSF binutils tags?). Apology for asking such a trivial question. Thanks for your time. -- with regards, Virendra Kumar Pathak

10 years

2
1
0 0

[ANNOUNCE] Linaro GCC 5 2015.12 snapshot

by Yvan Roux

The Linaro Toolchain Working Group (TCWG) is pleased to announce the 2015.12 snapshot of the Linaro GCC 5 source package. This monthly snapshot[1] is based on FSF GCC 5.3+svn231642 and includes performance improvements and bug fixes backported from mainline GCC. This snapshot contents will be part of the 2015.11 stable [1] quarterly release. This snapshot tarball is available on: http://snapshots.linaro.org/components/toolchain/gcc-linaro/5.3-2015.12/ Interesting changes in this GCC source package snapshot include: * Updates to GCC 5.3+svn231642 * Backport of [Bugfix] [AArch32] 1/4 PR63870 Add qualifiers for NEON builtins * Backport of [Bugfix] [AArch32] 2/4 PR63870 Mark lane indices of vldN/vstN with appropriate qualifier * Backport of [Bugfix] [AArch32] 3/4 PR63870 Remove error for invalid lane numbers * Backport of [Bugfix] [AArch32] 4/4 PR63870 Remove xfails for ARM targets * Backport of [Bugfix] [AArch32] PR67305, tighten neon_vector_mem_operand on eliminable registers * Backport of [Bugfix] [AArch32] remove unused variable * Backport of [Bugfix] [AArch64] PR 63304 Fix issue with global state * Backport of [Bugfix] [AArch64] PR63304 Handle literal pools for functions > 1 MiB in size * Backport of [Bugfix] [AArch64] PR 68088: Fix RTL checking ICE due to subregs inside accumulator forwarding check * Backport of [Bugfix] [AArch64] PR rtl-optimization/67218 * Backport of [Bugfix] PR 56036 fix typo in doc * Backport of [Bugfix] PR rtl-optimization/68236: Exit early from autoprefetcher lookahead if not in haifa sched * Backport of [Bugfix] PR tree-optimization/68234 Improve range info for loop Phi node * Backport of [AArch32] 1/4 Change GET_MODE_INNER to always return a non-void mode * Backport of [AArch32] 2/3 Make if_neg_move and if_move_neg into insn_and_split * Backport of [AArch32] 2/4 Control the FMA steering pass in tuning structures rather than as core property * Backport of [AArch32] 3/3 Expand mod by power of 2 * Backport of [AArch32] 3/4 Replace the pattern GET_MODE_BITSIZE (GET_MODE_INNER (m)) with GET_MODE_UNIT_BITSIZE (m) * Backport of [AArch32] 4/4 Fix - Introduce new inline functions for GET_MODE_UNIT_SIZE and GET_MODE_UNIT_PRECISION * Backport of [AArch32] 4/4 Introduce new inline functions for GET_MODE_UNIT_SIZE and GET_MODE_UNIT_PRECISION * Backport of [AArch32/AArch64] 2/2 Add a new Cortex-A53 scheduling model * Backport of [AArch32] Add missing v8a cpus to the t-aprofile file * Backport of [AArch32] Fix checking RTL error in cortex_a9_sched_adjust_cost * Backport of [AArch32] Fix for testcase after r228661 * Backport of [AArch32] Initialise cost to COSTS_N_INSNS (1) and increment in arm rtx costs * Backport of [AArch32] libgcc: include crtfastmath * Backport of [AArch32] Unified assembler in ARM state * Backport of [AArch64] 1/2 Give AArch64 ROR (Immediate) a new type attribute * Backport of [AArch64] 1/2 Rename SYMBOL_SMALL_GOT to SYMBOL_SMALL_GOT_4G * Backport of [AArch64] 1/2 Rename test source file for reuse * Backport of [AArch64] 1/3 Add the option -mtls-size * Backport of [AArch64] 1/3 Expand signed mod by power of 2 using CSNEG * Backport of [AArch64] 2/2 Implement -fpic for -mcmodel=small * Backport of [AArch64] 2/2 Implement TLS IE for tiny model * Backport of [AArch64] 2/3 Rename SYMBOL_TLSLE to SYMBOL_TLSLE24 * Backport of [AArch64] 3/3 Implement local executable mode for all memory model * Backport of [AArch64] Add initial qualcomm support * Backport of [AArch64] Add missing entries in iterator vwcore * Backport of [AArch64] Cleanup whitespace in aarch64.c * Backport of [AArch64] Cortex-A57 Choose some new branch costs * Backport of [AArch64] Define TARGET_UNSPEC_MAY_TRAP_P * Backport of [AArch64] Distinct costs for sign and zero extension * Backport of [AArch64] Do not ICE after apologising for -mcmodel=large -fPIC * Backport of [AArch64] Don't allow -mgeneral-regs-only to change the .arch assembler directives * Backport of [AArch64] Don't transform sign and zero extends inside mults * Backport of [AArch64] Fall back to -fPIC if no support of -fpic in binutils * Backport of [AArch64] Fix for branch offsets over 1 MiB * Backport of [AArch64] Fix ICE on (const_double:HF 0.0) * Backport of [AArch64] Fix insn types * Backport of [AArch64] Fix output assembly bug under TLSIE ILP32 * Backport of [AArch64] Handle vector float modes properly in aarch64_output_simd_mov_immediate * Backport of [AArch64] Mark GOT related MEM rtx as const to help RTL loop IV * Backport of [AArch64] Properly handle simple arith+extend ops in rtx costs * Backport of [AArch64] Rename SYMBOL_SMALL_GOTTPREL to SYMBOL_SMALL_TLSIE * Backport of [AArch64] Restrict pic-small.c by new test directive * Backport of [AArch64] Tighten direct call pattern for sibcall to repair -fno-plt * Backport of [AArch64] Tighten direct call pattern to repair -fno-plt * Backport of [Testsuite] [AArch32] Fix thumb2-slow-flash-data.c failures * Backport of [Testsuite] [AArch32] Switch ARM to unified asm * Backport of [Testsuite] [AArch64] Add more TLS local executable testcases * Backport of [Testsuite] [AArch64] Check branch types for noplt testcases * Backport of [Testsuite] [AArch64] Fix gcc.target/aarch64/vclz.c * Backport of [Testsuite] [AArch64] Fix some target attribute inlining tests for -fPIC * Backport of [Testsuite] [AArch64] Restrict got_mem_hoist_1.c with small memory model * Backport of [Testsuite] [AArch64] Skip tiny and large code model on gcc.target/aarch64/pic-small.c * Backport of [Testsuite] Add --param sra-max-scalarization-size-Ospeed to sra-12.c * Backport of [Testsuite] Fix target selector in gcc.target/i386/noplt-[1234].c testcases * Backport of [Misc] Allow sibcalls in no-PLT PIC * Backport of [Misc] Eliminate PLT stubs for specified external functions via -fno-plt * Backport of [Misc] Fix for ICE with -g on testcase with incomplete types * Backport of [Misc] Fix memory leak and wrong invariant dependence computation in IVOPT * Backport of [Misc] Improve rtl loop inv cost by checking if the inv can be propagated to address uses * Backport of [Cleanup] [AArch32/AArch64] fix ChangeLog * Backport of [cleanup] [AArch32] Remove uses of CONST_DOUBLE_HIGH/LOW * Backport of [Cleanup] [AArch64] Delete aarch64_symbol_context which is not used * Backport of [cleanup] [AArch64] Move iterators from atomics.md to iterators.md * Backport of [cleanup] [AArch64] Remove uses of CONST_DOUBLE_HIGH, CONST_DOUBLE_LOW * Backport of [Doc] [AArch64] Document several AArch64-specific test directives Feedback and Support Subscribe to the important Linaro mailing lists and join our IRC channels to stay on top of Linaro development. ** Linaro Toolchain Development "mailing list": http://lists.linaro.org/mailman/listinfo/linaro-toolchain ** Linaro Toolchain IRC channel on irc.freenode.net at @#linaro-tcwg@ * Bug reports should be filed in bugzilla against GCC product: http://bugs.linaro.org/enter_bug.cgi?product=GCC * Interested in commercial support? inquire at "Linaro support": mailto:support@linaro.org [1]. Stable source package releases are defined as releases where the full Linaro Toolchain validation plan is executed. [2]. Source package snapshots are defined when the compiler is only put through unit-testing and full validation is not performed.

10 years

1
0
0 0

[ACTIVITY] 7 Nov - 11 Nov 2015

by Kugan

== Progress == - LTO (8/10) * Looked at ipa-cp * Read LTO papers * Started with ipa-vrp experiment * Reproduced spec2006 issues - Misc (2/10) * gcc/bug list * LuaJIT == Plan == * LTO

10 years

1
0
0 0

[ACTIVITY] 7th-11th December

by Bernie Ogden

Centralised benchmark source - TCWG-354 [6/10] * Understood CoremarkPro run rules/behaviour better * Experimented with shortening runs while still giving meaningful results * Cleaned up build/run scaffolding Port to microinstance - TCWG-432 [1/10] * Got access to microinstance, learned a bit about SSL in the process Backport benchmarking - TCWG-352 [2/10] * Fixed 'build with sysroot' code for current deliverable shape * Fixed a problem with multiple manifests =Plan= Review, test, debug build-triggers-benchmark job Finish off Coremark Pro integration Review security with shared uinstance/main instance code Expose more data, benchmarks to bundles Debug/test Jenkins job in microinstance Create bootable image for at least 1 target, or know what the problems are Write up noise control report (if time) More support for SPEC-on-Android? =Absence= Holiday 22nd December - 1st January

10 years

1
1
0 0

[ACTIVITY] 7 Nov - 11 Nov 2015

by Prathamesh Kulkarni

== This Week == * TCWG-72 (2/10) - Another iteration. * LTO (4/10) - Building spec with different combinations of -flto -flto-partition -flto-compression-level - Builds cleanly on AArch64, ICE's on ARM for 447.dealII with -flto - Looked at segfault with 483.xalanpack * TCWG-319 benchmarking (1/10) - fp benchmark results on: r1-a12: -0.04 % cortex-a53: -0.09 % cortex-a57: -0.14% - int benchmarks completed running for reference revision on cortex-a15 - int benchmarks with-patch runnning on cortex-a15 * TCWG-310 benchmarking (1/10) - r1-a12: 0.11% - Benchmark jobs failed on a53, a57. * Misc (2/10) - Submitted patch to fix building 450.soplex - Looked at how to write ipa pass, ipa-pure-const and ipa-cp. - Meetings == Next Week == - tcwg-319, tcwg-310 benchmarking - tcwg-72 - LTO

10 years

1
0
0 0

[Weekly] 07-11 December 2015

by Michael Collison

== This week == * Bugzilla 68543 - [AArch64] Implement overflow arithmetic standard names (7/10) - Implemented signed and unsigned add, subtract, and multiply overflow standard patterns - Investigated testing of overflow patterns * Misc (3/10) - Conference calls - Misc ARM Housekeeping - Lost connectivity due to old ARM Unix VPN Client == Next week == * TCWG-317 - Exploit wide add operations when appropriate for Aarch32 - Minor code update to address upstream comment * Vacation until rest of year

10 years

1
0
0 0

[ACTIVITY] 7-11 December 2015

by Christophe Lyon

== Progress == * GCC (4/10) - pr68620 (fp16 transfers in big-endian mode) - checking regressions observed on trunk - had to revert my cleanup patch for target attributes tests, not sure how to handle all possible combinations of options/defaults * Validation (1/10) - small improvements in reporting * Misc (conf calls, meetings, emails, ...) (3/10) * Internal training (2/10) == Next == * Validation * GCC: bug fixes/cleanup

10 years

1
0
0 0

[ACTIVITY] Dec 7 - 11, 2015

by Yao Qi

* One day off on Wed. [2/10] # Progress # * Enable gdb core file tests when testing remotely, TCWG-171. I've almost had the conclusion that corefile remote testing can't be done due to limitations in dejagnu and nfs mount testing infrastructure. Need to write them down. [2/10] * Mutli-arch follow-up work, teach AArch64 GDBserver understand ARM breakpoint instructions. TCWG-460. Done. [3/10] * Review ARM GDBserver software single step patch V7. Almost OK, except some small things. [2/10] * Misc, meeting, email, [1/10] # Plan # * TCWG-424, Draft a fix for the fail in random-signal.exp. * TCWG-156, GDB test parity between AArch64 and X86_64. * One day off on Wed. or Fri. Planned absence: * Dec 24-Jan 3. -- Yao

10 years

1
0
0 0

[ACTIVITY]

by Renato Golin

== Progress == * Holiday (2/10) * Release (1/20) - 3.7.1 Final, all green * Maintenance (5/10) - Fixing PR25720 - Bug cleaning: PR18921, PR13259, PR13138, PR12926, PR17963, PR18187, PR22692 - Checked strided vectorizer on ToT, EEMBC 2x geomean! * Background (2/10) - Code review, meetings, discussions, general support, etc. - Power cut, restarting bots, testing TK1, plans, etc. == Holidays == I'll be back on the 5th Jan.

10 years

1
0
0 0

[ACTIVITY] Nov 30 - Dec 4, 2015

by Kugan

== Progress == - PR66726 (2/10) * Testing a patch - PR63586 (2/10) * Posted a patch * Revised the patch based on testing - LuaJIT (2/10) * Setup nginx * Still haven't figured out how to use mongodb with nginx (config required). - Misc (2/10) * gcc/bug list * LTO - sick (2/10) == Plan == * bug reports * LTO

10 years

1
0
0 0

incomplete .arch specifiers with latest Linaro gcc5 patches on armv8?

by Robert Schiele

Hi, when working with the Linaro patches I found that a particular commit breaks our aarch64 kernel build. The patch in question is that one: commit be09330da9d0777c4a58568d137e3f8a3dbe0a0b Author: Yvan Roux <yvan.roux(a)linaro.org> Date: Tue Oct 27 21:18:19 2015 +0100 One of the things it attempts to change apparently is moving the .arch specifiers in the assembler file from a global scope to individual functions. What also happens though is that they seem to lose some information after that transformation. I observed that when building arch/arm64/crypto/aes-ce-cipher.c from the Linux kernel. This code contains inline assembly like this: static void aes_cipher_decrypt(struct crypto_tfm *tfm, u8 dst[], u8 const src[]) { struct crypto_aes_ctx *ctx = crypto_tfm_ctx(tfm); struct aes_block *out = (struct aes_block *)dst; struct aes_block const *in = (struct aes_block *)src; void *dummy0; int dummy1; kernel_neon_begin_partial(4); __asm__(" ld1 {v0.16b}, %[in] ;" " ld1 {v1.2d}, [%[key]], #16 ;" " cmp %w[rounds], #10 ;" " bmi 0f ;" " bne 3f ;" " mov v3.16b, v1.16b ;" " b 2f ;" "0: mov v2.16b, v1.16b ;" " ld1 {v3.2d}, [%[key]], #16 ;" "1: aesd v0.16b, v2.16b ;" " aesimc v0.16b, v0.16b ;" "2: ld1 {v1.2d}, [%[key]], #16 ;" " aesd v0.16b, v3.16b ;" " aesimc v0.16b, v0.16b ;" "3: ld1 {v2.2d}, [%[key]], #16 ;" " subs %w[rounds], %w[rounds], #3 ;" " aesd v0.16b, v1.16b ;" " aesimc v0.16b, v0.16b ;" " ld1 {v3.2d}, [%[key]], #16 ;" " bpl 1b ;" " aesd v0.16b, v2.16b ;" " eor v0.16b, v0.16b, v3.16b ;" " st1 {v0.16b}, %[out] ;" : [out] "=Q"(*out), [key] "=r"(dummy0), [rounds] "=r"(dummy1) : [in] "Q"(*in), "1"(ctx->key_dec), "2"(num_rounds(ctx) - 2) : "cc"); kernel_neon_end(); } Now without this patch the compiler behaved like the following. It was invoked with: aarch64-linux-gnu-gcc -Wp,-MD,arch/arm64/crypto/.aes-ce-cipher.o.d -nostdinc -isystem /var/fpwork/rschiele/crossbuild/builds/aarch64-linux-gnu/linux-next/srcdir/bin/../lib/gcc/aarch64-linux-gnu/5.2.1/include -I/var/fpwork/rschiele/crossbuild/builds/aarch64-linux-gnu/linux-next/srcdir/src/linux/arch/arm64/include -Iarch/arm64/include/generated/uapi -Iarch/arm64/include/generated -I/var/fpwork/rschiele/crossbuild/builds/aarch64-linux-gnu/linux-next/srcdir/src/linux/include -Iinclude -I/var/fpwork/rschiele/crossbuild/builds/aarch64-linux-gnu/linux-next/srcdir/src/linux/arch/arm64/include/uapi -Iarch/arm64/include/generated/uapi -I/var/fpwork/rschiele/crossbuild/builds/aarch64-linux-gnu/linux-next/srcdir/src/linux/include/uapi -Iinclude/generated/uapi -include /var/fpwork/rschiele/crossbuild/builds/aarch64-linux-gnu/linux-next/srcdir/src/linux/include/linux/kconfig.h -I/var/fpwork/rschiele/crossbuild/builds/aarch64-linux-gnu/linux-next/srcdir/src/linux/arch/arm64/crypto -Iarch/arm64/crypto -D__KERNEL__ -mlittle-endian -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -std=gnu89 -mgeneral-regs-only -fno-delete-null-pointer-checks -O2 --param=allow-store-data-races=0 -Wframe-larger-than=2048 -fno-stack-protector -Wno-unused-but-set-variable -fno-omit-frame-pointer -fno-optimize-sibling-calls -fno-var-tracking-assignments -g -Wdeclaration-after-statement -Wno-pointer-sign -fno-strict-overflow -fconserve-stack -Werror=implicit-int -Werror=strict-prototypes -Werror=date-time -DCC_HAVE_ASM_GOTO -Werror -march=armv8-a+crypto -D"KBUILD_STR(s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(aes_ce_cipher)" -D"KBUILD_MODNAME=KBUILD_STR(aes_ce_cipher)" -c -o arch/arm64/crypto/aes-ce-cipher.o /var/fpwork/rschiele/crossbuild/builds/aarch64-linux-gnu/linux-next/srcdir/src/linux/arch/arm64/crypto/aes-ce-cipher.c As a result it created a file for the assembler with the global .arch armv8-a+fp+simd+crypto at the beginning of the file. After the patch it created individual .arch armv8-a at individual places. It is not clear to me, why the extensions (fp+simd+crypto) got lost. Is that intended, such that the code needs special adaption for inline assembly using those extensions or is that loss of extensions a bug of that patch? Greetings! Robert

10 years

2
5
0 0

[ACTIVITY] Week 48

by Yvan Roux

== Progress == o Linaro GCC (6/10) * More backports * Compared our various flavor of validation results * Looked at bug reports and mailing list questions on our last snapshot. o Upstream work (2/10) * Continue on sanitizing gfortran testsuite o Misc (2/10) * Various meetings * Discussed benchmarking infra with Bernie == Plan == o GCC 5.3 branch merge o Continue on-going tasks o Tuesday Off

10 years

1
0
0 0

[ACTIVITY] 30th November - 4th December

by Bernie Ogden

Holiday [2/10] Port to microinstance - TCWG-432 [1/10] * Set up reporting for CPU2006 * Learned how to generate metadata, but not how to use it Trigger benchmarks on backports - TCWG-352 [2/10] * Figured out the rough shape * Created, didn't test, rough implementation TCWG-354 [3/10] * Build/run scaffolding for CoreMark Pro * Working for manual runs Misc [2/10] * Input validation for dispatcher script * Meeting with Ade on LAVA benchmarking * Meetings/mail/etc =Plan= Review, test, debug build-triggers-benchmark job Check CoreMark Pro run configuration, enable for automatic runs Review security with shared uinstance/main instance code Expose more data, benchmarks to bundles Debug/test Jenkins job in microinstance Create bootable image for at least 1 target, or know what the problems are Write up noise control report (if time) More support for SPEC-on-Android?

10 years

1
0
0 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

linaro-toolchain