[llvm-dev] [ARM] Should Use Load and Store with Register Offset (original) (raw)

Daniel Way via llvm-dev llvm-dev at lists.llvm.org
Tue Jul 21 20:33:58 PDT 2020

Thank you, Sjoerd.

Your high-level comments are very helpful and much appreciated. I ended up rebuilding the Newlib-nano source with -Oz instead of -Os and found an overall improvement in code size. The final size is still larger than the gcc-arm-none-eabi toolchain. Of course there are a few caveats to this:

For now, I've submitted BUG 46801 for the case when -Oz produces more instructions than -Os. I don't know if it needs to be a priority, but thought it should be recorded.

I may try benchmarking the memcpy implementations as well as a few other libc functions, but I haven't done this before. Of course, I'll share my results if I do end up testing.

Thank you for the help. Daniel Way

On Tue, Jul 21, 2020 at 6:05 PM Sjoerd Meijer <Sjoerd.Meijer at arm.com> wrote:

Hi Daniel,

Your observations seem valid to me. Some high-level comments from my side. As you said, the loops are quite similar. We have also observed that in general we generate more code around loops, in the function prologue and epilogue, where some data and arguments get moved and reshuffled etc. While this is very obvious in these micro-benchmarks, it hasn't bothered us enough yet for larger apps where this is less important (or where others things are more important). The outlier looks indeed to be Clang -Oz for memcpyalt2, that is perhaps a "code-size bug". As I haven't looked into it, it's too early for me to blame this on just the addressing modes as there could be several things going on. Since this is a micro-benchmark, and lowering memcpy is a bit of an art ;-), for which a specialised implementation is probably available, you might want to look at some other codes too that are important for you. Your remarks about execution times might be right too, and as you said, probably best confirmed with benchmark numbers. In our group, we have not really looked into performance for the Cortex-M0, probably because it's the only v6m core (although the Cortex-m23 and Armv8-M Baseline is very similar) and code-size would be more important for us, but there might be something to be gained here. Cheers, Sjoerd. ------------------------------ From: Daniel Way <p.waydan at gmail.com> Sent: 21 July 2020 08:12 To: Sjoerd Meijer <Sjoerd.Meijer at arm.com> Cc: llvm-dev at lists.llvm.org <llvm-dev at lists.llvm.org> Subject: Re: [llvm-dev] [ARM] Should Use Load and Store with Register Offset Hello Sjoerd, Thank you for your response! I was not aware that -Oz is a closer equivalent to GCC's -Os. I tried -Oz when compiling with clang and confirmed that the Clang's generated assembly is equivalent to GCC for the code snippet I posted above. clang --target=armv6m-none-eabi -Oz -fomit-frame-pointer memcpyalt1: push {r4, lr} movs r3, #0 .LBB01: cmp r2, r3 beq .LBB03 ldrb r4, [r1, r3] strb r4, [r0, r3] adds r3, r3, #1 b .LBB01 .LBB03: pop {r4, pc} On the other hand, -O2 in GCC still uses the register-offset load and store instructions while Clang -O2 generates the same assembly as -Os: immediate-offset (0 offset) load/store followed by incrementing the base register addresses. I have not tried to benchmark the Clang-generated code, it is possible that execution time is bounded by the load and store instructions and memory access latency. From an intuitive view, however, both GCC and Clang are generating code with 1 load and 1 store, so if Clang inserts two additional adds instructions, the binary size is larger, execution could be slower, and there's no improvement in register utilization over GCC. I wanted to try a couple other variants of memcpy-like functions. The https://godbolt.org/z/d7P6rG link includes memcpyalt2 which copies data from src to dst starting at the high address and memcpysilly which copies src to dst<0-4>. Here is the behavior I have noticed from GCC and Clang. memcpyalt2 - With -Os, GCC generates just 6 instructions. -O2 generates 7 but reduces branching to once per loop. - Clang with -Os or -O2 does a decent job of using a common register to offset the load and store bases. It adds some overhead, though, by pre-decrementing the base registers. 10 instructions generated. - Clang with -Oz is pathological, generating 13 instructions. It uses register-offset load/store instructions, but uses different registers for the offsets. memcpysilly - I created this case to see if clang would select load/store with a common offset register once enough load instructions were added. - Clang with -Os or -O2 does not seem to care about register-offset load/store and prefers to increment each base register address. - Clang with -Oz performs the optimization I want. It produces the same number of instructions as GCC, and avoids an issue where GCC has to re-read the same value from the stack each time through the loop.

I really think that, when limited to the Thumb1 ISA, register-offset load and store instructions should be used at -Oz, -Os, and -O2 optimization levels. Explicitly incrementing a register holding the base address seems unnecessary when the value seems wasteful and I cannot see how it will improve execution time in the examples I'm investigating. Id like to know if I'm wrong in assuming that LDR Rd, [Rn, Rm] and LDR Rd, [Rn, #] have the same execution time, but based on the Cortex-M0+ TRM they should both require 2 clock cycles. Best regards, Daniel Way On Mon, Jul 20, 2020 at 6:15 PM Sjoerd Meijer <Sjoerd.Meijer at arm.com> wrote: Hello Daniel, LLVM and GCC's optimisation levels are not really equivalent. In Clang, -Os makes a performance and code-size trade off. In GCC, -Os is minimising code-size, which is equivalent to -Oz with Clang. I have't looked into details yet, but changing -Os to -Oz in the godbolt link gives the codegen you're looking for? Cheers, Sjoerd. ------------------------------ From: llvm-dev <llvm-dev-bounces at lists.llvm.org> on behalf of Daniel Way via llvm-dev <llvm-dev at lists.llvm.org> Sent: 20 July 2020 06:54 To: llvm-dev at lists.llvm.org <llvm-dev at lists.llvm.org> Subject: [llvm-dev] [ARM] Should Use Load and Store with Register Offset Hello LLVM Community (specifically anyone working with ARM Cortex-M), While trying to compile the Newlib C library I found that Clang10 was generating slightly larger binaries than the libc from the prebuilt gcc-arm-none-eabi toolchain. I looked at a few specific functions (memcpy, strcpy, etc.) and noticed that LLVM does not tend to generate load/store instructions with a register offset (e.g. ldr Rd, [Rn, Rm] form) and instead prefers the immediate offset form. When copying a contiguous sequence of bytes, this results in additional instructions to modify the base address. https://godbolt.org/z/T1xhae void* memcpyalt1(void* dst, const void* src, sizet len) { char* save = (char*)dst; for (sizet i = 0; i < len; ++i)_ _*((char*)(dst + i)) = *((char*)(src + i));_ _return save;_ _}_ _clang --target=armv6m-none-eabi -Os -fomit-frame-pointer_ _memcpyalt1:_ _push {r4, lr}_ _cmp r2, #0_ _beq .LBB03_ _mov r3, r0_ _.LBB02:_ _ldrb r4, [r1]_ _strb r4, [r3]_ _adds r1, r1, #1_ _adds r3, r3, #1_ _subs r2, r2, #1_ _bne .LBB02_ _.LBB03:_ _pop {r4, pc}_ _arm-none-eabi-gcc -march=armv6-m -Os_ _memcpyalt1:_ _movs r3, #0_ _push {r4, lr}_ _.L2:_ _cmp r3, r2_ _bne .L3_ _pop {r4, pc}_ _.L3:_ _ldrb r4, [r1, r3]_ _strb r4, [r0, r3]_ _adds r3, r3, #1_ _b .L2_ _Because this code appears in a loop that could be copying hundreds of_ _bytes, I want to add an optimization that will prioritize load/store_ _instructions with register offsets when the offset is used multiple times._ _I have not worked on LLVM before, so I'd like advice about where to start._ _- The generated code is correct, just sub-optimal so is it appropriate_ _to submit a bug report?_ _- Is anyone already tackling this change or is there someone with more_ _experience interested in collaborating?_ _- Is this optimization better performed early during instruction_ _selection or late using c++ (i.e. ARMLoadStoreOptimizer.cpp)_ _- What is the potential to cause harm to other parts of the code gen,_ _specifically for other arm targets. I'm working with armv6m, but armv7m_ _offers base register updating in a single instruction. I don't want to_ _break other useful optimizations._ _So far, I am reading through the LLVM documentation to see where a change_ _could be applied. I have also:_ _- Compiled with -S -emit-llvm (see Godbolt link)_ _There is an identifiable pattern where a getelementptr function is_ _followed by a load or store. When multiple getelementptr functions appear_ _with the same virtual register offset, maybe this should match a tLDRr or_ _tSTRr._ _- Ran LLC with --print-machineinstrs_ _It appears that tLDRBi and tSTRBi are selected very early and never_ _replaced by the equivalent t<LDRB|STRB>r instructions. Thank you, Daniel Way -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.llvm.org/pipermail/llvm-dev/attachments/20200722/cedd419d/attachment-0001.html>

More information about the llvm-dev mailing list