From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:39499) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1bSEsZ-0002jz-LA for qemu-devel@nongnu.org; Tue, 26 Jul 2016 22:47:30 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1bSEsV-0001n3-8b for qemu-devel@nongnu.org; Tue, 26 Jul 2016 22:47:26 -0400 Date: Wed, 27 Jul 2016 12:47:13 +1000 From: David Gibson Message-ID: <20160727024713.GA17429@voom.fritz.box> References: <1469571686-7284-1-git-send-email-benh@kernel.crashing.org> <1469571686-7284-32-git-send-email-benh@kernel.crashing.org> MIME-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="zYtRAvLPyLojxZ7Y" Content-Disposition: inline In-Reply-To: <1469571686-7284-32-git-send-email-benh@kernel.crashing.org> Subject: Re: [Qemu-devel] [PATCH 32/32] ppc: Speed up load/store multiple List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Benjamin Herrenschmidt Cc: qemu-ppc@nongnu.org, qemu-devel@nongnu.org --zYtRAvLPyLojxZ7Y Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Jul 27, 2016 at 08:21:26AM +1000, Benjamin Herrenschmidt wrote: > Use a single translate when not crossing a page boundary and avoid > going through layers of helpers. MacOS uses those instructions > a lot, so does OpenBIOS. >=20 > Signed-off-by: Benjamin Herrenschmidt > --- > target-ppc/mem_helper.c | 69 +++++++++++++++++++++++++++++++++++++++++++= ++++++ > 1 file changed, 69 insertions(+) >=20 > diff --git a/target-ppc/mem_helper.c b/target-ppc/mem_helper.c > index da3f973..511079b 100644 > --- a/target-ppc/mem_helper.c > +++ b/target-ppc/mem_helper.c > @@ -53,8 +53,48 @@ static inline target_ulong addr_add(CPUPPCState *env, = target_ulong addr, > } > } > =20 > +/* Reduce the length so that addr + len doesn't cross a page boundary. = */ > +static inline uint64_t adj_len_to_page(uint64_t len, uint64_t addr) > +{ > +#ifndef CONFIG_USER_ONLY > + if ((addr & ~TARGET_PAGE_MASK) + len - 1 >=3D TARGET_PAGE_SIZE) { > + return -addr & ~TARGET_PAGE_MASK; > + } > +#endif > + return len; > +} > + > void helper_lmw(CPUPPCState *env, target_ulong addr, uint32_t reg) > { > + uint32_t *src; > + uint64_t len, adjlen; > + > + if ((addr & 3)) { > + goto fallback; > + } > + len =3D (32 - reg) << 2; > + while (len) { > + src =3D tlb_vaddr_to_host(env, addr, MMU_DATA_LOAD, env->dmmu_id= x); > + if (!src) { > + goto fallback; > + } > + adjlen =3D adj_len_to_page(len, addr); > + len -=3D adjlen; > +#if defined(HOST_WORDS_BIGENDIAN) > + memcpy(&env->gpr[reg], src, adjlen); > + reg +=3D (adjlen >> 2); > + addr =3D addr_add(env, addr, adjlen); > +#else > + while(adjlen) { > + env->gpr[reg++] =3D bswap32(*(src++)); > + adjlen -=3D 4; > + addr =3D addr_add(env, addr, 4); > + } > +#endif Would it improve this any further to do the memcpy() unconditionally, then byteswap the GPRs in-place for the LE host case? > + } > + return; > + > + fallback: > for (; reg < 32; reg++) { > if (needs_byteswap(env)) { > env->gpr[reg] =3D bswap32(cpu_ldl_data_ra(env, addr, GETPC()= )); > @@ -67,6 +107,35 @@ void helper_lmw(CPUPPCState *env, target_ulong addr, = uint32_t reg) > =20 > void helper_stmw(CPUPPCState *env, target_ulong addr, uint32_t reg) > { > + uint32_t *dst; > + uint64_t len, adjlen; > + > + if ((addr & 3)) { > + goto fallback; > + } > + len =3D (32 - reg) << 2; > + while (len) { > + dst =3D tlb_vaddr_to_host(env, addr, MMU_DATA_STORE, env->dmmu_i= dx); > + if (!dst) { > + goto fallback; > + } > + adjlen =3D adj_len_to_page(len, addr); > + len -=3D adjlen; > +#if defined(HOST_WORDS_BIGENDIAN) > + memcpy(dst, &env->gpr[reg], adjlen); > + reg +=3D (adjlen >> 2); > + addr =3D addr_add(env, addr, adjlen); > +#else > + while(adjlen) { > + *(dst++) =3D bswap32(env->gpr[reg++]); > + adjlen -=3D 4; > + addr =3D addr_add(env, addr, 4); > + } > +#endif > + } > + return; > + > + fallback: > for (; reg < 32; reg++) { > if (needs_byteswap(env)) { > cpu_stl_data_ra(env, addr, bswap32((uint32_t)env->gpr[reg]), --=20 David Gibson | I'll have my music baroque, and my code david AT gibson.dropbear.id.au | minimalist, thank you. NOT _the_ _other_ | _way_ _around_! http://www.ozlabs.org/~dgibson --zYtRAvLPyLojxZ7Y Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- Version: GnuPG v1 iQIcBAEBAgAGBQJXmCCxAAoJEGw4ysog2bOSxbAP/3OBaf13WNyHZOuKKYofUwOz maEq4CWyKhrNmTUXEoCUwR6ww+9RuuMh1YAWouHm9Hu1WbTNE4TYcIV3PMDmY1hQ qZOfoH5uzVvOak5feL7zI4JHS0OPk52yfm4+dgrCuxeJfHNVlzTzdO05qlOQaMOf gR4ZAhuU9HodUFZ1XdToSPwHuRsaKEcQa5MQv0Iq3YtvRyHhQ3kGmrwIB9cp0l6R e2SXX8SPPkCQZLPPqBvbbba/sGaQpwVAjm22CywmYcj98GPd+98HpFBMVSgRR396 wm6Ebo9k2H0Ik0zFOVNNLPDDLbjBnxqoVqZB0fSmM7ByvFY4U+PkisM84WimaJSD 42xM5KbjgUKeNmBtjdsvRIKHOouySXNQjJlIG4nu8cQ5qsx1oevrun0b+7RVsP8+ 5mEf2r5gR+9zNx4ThgNHWbBmLYSmH//btQjbuBl1D4a0gpHzt0EUNb9e231o8sFK +4+opk33s5ZGBeotsCjhNYerfLMAAEBCdSClF2LYvp2jX8VbQ0932gGw0qjAZdXP CmlAHnEU0KtWc+gZMgNy4WULM99rKLj+xsmNkLt1dOpbz+nGDe1rQ8zUJVZm6J6M Blfpc8IPuZ70oAdoyDgSEFoY3vZuB3PBf5NrA7E9d9lndgIGDMVSk+sZlZ0X1ebH /c2py9yikKNGFqIzctLV =8dgr -----END PGP SIGNATURE----- --zYtRAvLPyLojxZ7Y--