[PATCH 0/1 v2] [RISC-V/RVV] optimize the memory probing for vector fault-only-first loads.

Paolo Savini posted 1 patch 1 day, 5 hours ago
target/riscv/vector_helper.c | 91 ++++++++++++++++++++++--------------
1 file changed, 56 insertions(+), 35 deletions(-)
[PATCH 0/1 v2] [RISC-V/RVV] optimize the memory probing for vector fault-only-first loads.
Posted by Paolo Savini 1 day, 5 hours ago
This version of the patch addresses the comments from the following review:

https://lore.kernel.org/all/2df9ae98-afb8-4647-be80-12540a1c4612@ventanamicro.com/

Previous version:

- v1: https://lore.kernel.org/all/20250129144435.82451-1-paolo.savini@embecosm.com/

The new version:

- fixes the "braod" typo in the comment.
- removes the "probe_size" variable in favour of "elems" that is already used
  for the same purpose.
- removes the duplication of the code calculating the page split and adds a
  separate "addr_i" variable for probing the memory without polluting "addr"
  used to perform the actual load.
- multiplies "elems" by "msize" when calling for the memory probing function so
  that we pass the number of bytes rather then the number of vector elements as
  that is what the probe function expects.

We also change the heading from RFC to PATCH.

I also take the opportunity to thanks Daniel Barboza for the review.

Cc: Richard Handerson <richard.henderson@linaro.org>
Cc: Palmer Dabbelt <palmer@dabbelt.com>
Cc: Alistair Francis <alistair.francis@wdc.com>
Cc: Bin Meng <bmeng.cn@gmail.com>
Cc: Weiwei Li <liwei1518@gmail.com>
Cc: Daniel Henrique Barboza <dbarboza@ventanamicro.com>
Cc: Liu Zhiwei <zhiwei_liu@linux.alibaba.com>
Cc: Helene Chelin <helene.chelin@embecosm.com>
Cc: Nathan Egge <negge@google.com>
Cc: Max Chou <max.chou@sifive.com>
Cc: Jeremy Bennett <jeremy.bennett@embecosm.com>
Cc: Craig Blackmore <craig.blackmore@embecosm.com>

Paolo Savini (1):
  target/riscv: optimize the memory probing for vector fault-only-first
    loads.

 target/riscv/vector_helper.c | 91 ++++++++++++++++++++++--------------
 1 file changed, 56 insertions(+), 35 deletions(-)

-- 
2.34.1