From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:56001) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1dagsl-0008LK-Nz for qemu-devel@nongnu.org; Thu, 27 Jul 2017 07:23:08 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1dagsi-0002Zo-D0 for qemu-devel@nongnu.org; Thu, 27 Jul 2017 07:23:07 -0400 Received: from mx1.redhat.com ([209.132.183.28]:56628) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1dagsi-0002ZC-3t for qemu-devel@nongnu.org; Thu, 27 Jul 2017 07:23:04 -0400 Date: Thu, 27 Jul 2017 13:22:59 +0200 From: Cornelia Huck Message-ID: <20170727132259.7102a715@gondolin> In-Reply-To: <20170727015418.85407-3-bjsdjshi@linux.vnet.ibm.com> References: <20170727015418.85407-1-bjsdjshi@linux.vnet.ibm.com> <20170727015418.85407-3-bjsdjshi@linux.vnet.ibm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PATCH 2/3] s390x/css: generate solicited crw for rchp completion signaling List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Dong Jia Shi Cc: qemu-devel@nongnu.org, borntraeger@de.ibm.com, agraf@suse.de, rth@twiddle.net, pasic@linux.vnet.ibm.com, pmorel@linux.vnet.ibm.com On Thu, 27 Jul 2017 03:54:17 +0200 Dong Jia Shi wrote: > A successful completion of rchp should signal a solicited channel path > initialized CRW (channel report word), while the current implementation > always generates an un-solicited one. Let's fix this. Sounds legit. > > Reported-by: Halil Pasic > Signed-off-by: Dong Jia Shi > --- > hw/s390x/css.c | 15 +++++++++------ > include/hw/s390x/css.h | 2 +- > 2 files changed, 10 insertions(+), 7 deletions(-) > > diff --git a/hw/s390x/css.c b/hw/s390x/css.c > index 5321ca016b..60e1592d5c 100644 > --- a/hw/s390x/css.c > +++ b/hw/s390x/css.c > @@ -1745,10 +1745,10 @@ int css_do_rchp(uint8_t cssid, uint8_t chpid) > } > > /* We don't really use a channel path, so we're done here. */ > - css_queue_crw(CRW_RSC_CHP, CRW_ERC_INIT, > + css_queue_crw(CRW_RSC_CHP, CRW_ERC_INIT, 1, > channel_subsys.max_cssid > 0 ? 1 : 0, chpid); > if (channel_subsys.max_cssid > 0) { > - css_queue_crw(CRW_RSC_CHP, CRW_ERC_INIT, 0, real_cssid << 8); > + css_queue_crw(CRW_RSC_CHP, CRW_ERC_INIT, 1, 0, real_cssid << 8); > } > return 0; > } > @@ -2028,7 +2028,7 @@ void css_subch_assign(uint8_t cssid, uint8_t ssid, uint16_t schid, > } > } > > -void css_queue_crw(uint8_t rsc, uint8_t erc, int chain, uint16_t rsid) > +void css_queue_crw(uint8_t rsc, uint8_t erc, int s, int chain, uint16_t rsid) 's' is not a very speaking name... > { > CrwContainer *crw_cont; > > @@ -2040,6 +2040,9 @@ void css_queue_crw(uint8_t rsc, uint8_t erc, int chain, uint16_t rsid) > return; > } > crw_cont->crw.flags = (rsc << 8) | erc; > + if (s) { > + crw_cont->crw.flags |= CRW_FLAGS_MASK_S; ...as it obviously causes the S flag to be set ;) Let's call it 'solicited'? > + } > if (chain) { > crw_cont->crw.flags |= CRW_FLAGS_MASK_C; > } > @@ -2086,9 +2089,9 @@ void css_generate_sch_crws(uint8_t cssid, uint8_t ssid, uint16_t schid, > } > chain_crw = (channel_subsys.max_ssid > 0) || > (channel_subsys.max_cssid > 0); > - css_queue_crw(CRW_RSC_SUBCH, CRW_ERC_IPI, chain_crw ? 1 : 0, schid); > + css_queue_crw(CRW_RSC_SUBCH, CRW_ERC_IPI, 0, chain_crw ? 1 : 0, schid); > if (chain_crw) { > - css_queue_crw(CRW_RSC_SUBCH, CRW_ERC_IPI, 0, > + css_queue_crw(CRW_RSC_SUBCH, CRW_ERC_IPI, 0, 0, > (guest_cssid << 8) | (ssid << 4)); > } > /* RW_ERC_IPI --> clear pending interrupts */ > @@ -2103,7 +2106,7 @@ void css_generate_chp_crws(uint8_t cssid, uint8_t chpid) > void css_generate_css_crws(uint8_t cssid) > { > if (!channel_subsys.sei_pending) { > - css_queue_crw(CRW_RSC_CSS, CRW_ERC_EVENT, 0, cssid); > + css_queue_crw(CRW_RSC_CSS, CRW_ERC_EVENT, 0, 0, cssid); Should we want to support OS-triggered channel path vary (via SCLP or otherwise) in the future, we'll probably need a version that generates a solicited crw. > } > channel_subsys.sei_pending = true; > } > diff --git a/include/hw/s390x/css.h b/include/hw/s390x/css.h > index 5c5fe6b202..d03b4ffeac 100644 > --- a/include/hw/s390x/css.h > +++ b/include/hw/s390x/css.h > @@ -150,7 +150,7 @@ void copy_scsw_to_guest(SCSW *dest, const SCSW *src); > void css_inject_io_interrupt(SubchDev *sch); > void css_reset(void); > void css_reset_sch(SubchDev *sch); > -void css_queue_crw(uint8_t rsc, uint8_t erc, int chain, uint16_t rsid); > +void css_queue_crw(uint8_t rsc, uint8_t erc, int s, int chain, uint16_t rsid); > void css_generate_sch_crws(uint8_t cssid, uint8_t ssid, uint16_t schid, > int hotplugged, int add); > void css_generate_chp_crws(uint8_t cssid, uint8_t chpid); Otherwise, patch looks good.