* [PATCH for-6.2 0/1] tcg/arm: Reduce vector alignment requirement for NEON
@ 2021-12-06 19:13 Richard Henderson
2021-12-06 19:13 ` [PATCH for-6.2 1/1] " Richard Henderson
0 siblings, 1 reply; 3+ messages in thread
From: Richard Henderson @ 2021-12-06 19:13 UTC (permalink / raw)
To: qemu-devel; +Cc: peter.maydell, elima, alex.bennee, rjones
Re-post of
https://patchew.org/QEMU/20210912174925.200132-1-richard.henderson@linaro.org/
fixing
https://bugzilla.redhat.com/show_bug.cgi?id=1999878
which seems to have gotten lost. In the qemu-devel thread,
I did successfully reproduce and fix the assertion failure,
but I never got Reviewed-by or Tested-by tags.
r~
Richard Henderson (1):
tcg/arm: Reduce vector alignment requirement for NEON
tcg/tcg.c | 8 +++++++-
tcg/arm/tcg-target.c.inc | 13 +++++++++----
2 files changed, 16 insertions(+), 5 deletions(-)
--
2.25.1
^ permalink raw reply [flat|nested] 3+ messages in thread
* [PATCH for-6.2 1/1] tcg/arm: Reduce vector alignment requirement for NEON
2021-12-06 19:13 [PATCH for-6.2 0/1] tcg/arm: Reduce vector alignment requirement for NEON Richard Henderson
@ 2021-12-06 19:13 ` Richard Henderson
2021-12-06 20:11 ` Philippe Mathieu-Daudé
0 siblings, 1 reply; 3+ messages in thread
From: Richard Henderson @ 2021-12-06 19:13 UTC (permalink / raw)
To: qemu-devel; +Cc: peter.maydell, elima, alex.bennee, rjones
With arm32, the ABI gives us 8-byte alignment for the stack.
While it's possible to realign the stack to provide 16-byte alignment,
it's far easier to simply not encode 16-byte alignment in the
VLD1 and VST1 instructions that we emit.
Remove the assertion in temp_allocate_frame, limit natural alignment
to the provided stack alignment, and add a comment.
Reported-by: Richard W.M. Jones <rjones@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20210912174925.200132-1-richard.henderson@linaro.org>
---
tcg/tcg.c | 8 +++++++-
tcg/arm/tcg-target.c.inc | 13 +++++++++----
2 files changed, 16 insertions(+), 5 deletions(-)
diff --git a/tcg/tcg.c b/tcg/tcg.c
index 57f17a4649..934aa8510b 100644
--- a/tcg/tcg.c
+++ b/tcg/tcg.c
@@ -3061,7 +3061,13 @@ static void temp_allocate_frame(TCGContext *s, TCGTemp *ts)
g_assert_not_reached();
}
- assert(align <= TCG_TARGET_STACK_ALIGN);
+ /*
+ * Assume the stack is sufficiently aligned.
+ * This affects e.g. ARM NEON, where we have 8 byte stack alignment
+ * and do not require 16 byte vector alignment. This seems slightly
+ * easier than fully parameterizing the above switch statement.
+ */
+ align = MIN(TCG_TARGET_STACK_ALIGN, align);
off = ROUND_UP(s->current_frame_offset, align);
/* If we've exhausted the stack frame, restart with a smaller TB. */
diff --git a/tcg/arm/tcg-target.c.inc b/tcg/arm/tcg-target.c.inc
index 633b8a37ba..9d322cdba6 100644
--- a/tcg/arm/tcg-target.c.inc
+++ b/tcg/arm/tcg-target.c.inc
@@ -2523,8 +2523,13 @@ static void tcg_out_ld(TCGContext *s, TCGType type, TCGReg arg,
tcg_out_vldst(s, INSN_VLD1 | 0x7d0, arg, arg1, arg2);
return;
case TCG_TYPE_V128:
- /* regs 2; size 8; align 16 */
- tcg_out_vldst(s, INSN_VLD1 | 0xae0, arg, arg1, arg2);
+ /*
+ * We have only 8-byte alignment for the stack per the ABI.
+ * Rather than dynamically re-align the stack, it's easier
+ * to simply not request alignment beyond that. So:
+ * regs 2; size 8; align 8
+ */
+ tcg_out_vldst(s, INSN_VLD1 | 0xad0, arg, arg1, arg2);
return;
default:
g_assert_not_reached();
@@ -2543,8 +2548,8 @@ static void tcg_out_st(TCGContext *s, TCGType type, TCGReg arg,
tcg_out_vldst(s, INSN_VST1 | 0x7d0, arg, arg1, arg2);
return;
case TCG_TYPE_V128:
- /* regs 2; size 8; align 16 */
- tcg_out_vldst(s, INSN_VST1 | 0xae0, arg, arg1, arg2);
+ /* See tcg_out_ld re alignment: regs 2; size 8; align 8 */
+ tcg_out_vldst(s, INSN_VST1 | 0xad0, arg, arg1, arg2);
return;
default:
g_assert_not_reached();
--
2.25.1
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH for-6.2 1/1] tcg/arm: Reduce vector alignment requirement for NEON
2021-12-06 19:13 ` [PATCH for-6.2 1/1] " Richard Henderson
@ 2021-12-06 20:11 ` Philippe Mathieu-Daudé
0 siblings, 0 replies; 3+ messages in thread
From: Philippe Mathieu-Daudé @ 2021-12-06 20:11 UTC (permalink / raw)
To: Richard Henderson, qemu-devel; +Cc: peter.maydell, elima, alex.bennee, rjones
On 12/6/21 20:13, Richard Henderson wrote:
> With arm32, the ABI gives us 8-byte alignment for the stack.
> While it's possible to realign the stack to provide 16-byte alignment,
> it's far easier to simply not encode 16-byte alignment in the
> VLD1 and VST1 instructions that we emit.
>
> Remove the assertion in temp_allocate_frame, limit natural alignment
> to the provided stack alignment, and add a comment.
>
> Reported-by: Richard W.M. Jones <rjones@redhat.com>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
> Message-Id: <20210912174925.200132-1-richard.henderson@linaro.org>
> ---
> tcg/tcg.c | 8 +++++++-
> tcg/arm/tcg-target.c.inc | 13 +++++++++----
> 2 files changed, 16 insertions(+), 5 deletions(-)
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2021-12-06 20:13 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-12-06 19:13 [PATCH for-6.2 0/1] tcg/arm: Reduce vector alignment requirement for NEON Richard Henderson
2021-12-06 19:13 ` [PATCH for-6.2 1/1] " Richard Henderson
2021-12-06 20:11 ` Philippe Mathieu-Daudé
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.