From: Stefano Stabellini <sstabellini@kernel.org> To: peter.maydell@linaro.org, stefanha@gmail.com Cc: sstabellini@kernel.org, stefanha@redhat.com, anthony.perard@citrix.com, xen-devel@lists.xenproject.org, qemu-devel@nongnu.org, Roger Pau Monne <roger.pau@citrix.com> Subject: [Qemu-devel] [PULL 2/2] xen/pt: allow QEMU to request MSI unmasking at bind time Date: Wed, 20 Sep 2017 19:22:24 -0700 [thread overview] Message-ID: <1505960544-12202-2-git-send-email-sstabellini@kernel.org> (raw) In-Reply-To: <1505960544-12202-1-git-send-email-sstabellini@kernel.org> From: Roger Pau Monne <roger.pau@citrix.com> When a MSI interrupt is bound to a guest using xc_domain_update_msi_irq (XEN_DOMCTL_bind_pt_irq) the interrupt is left masked by default. This causes problems with guests that first configure interrupts and clean the per-entry MSIX table mask bit and afterwards enable MSIX globally. In such scenario the Xen internal msixtbl handlers would not detect the unmasking of MSIX entries because vectors are not yet registered since MSIX is not enabled, and vectors would be left masked. Introduce a new flag in the gflags field to signal Xen whether a MSI interrupt should be unmasked after being bound. This also requires to track the mask register for MSI interrupts, so QEMU can also notify to Xen whether the MSI interrupt should be bound masked or unmasked Signed-off-by: Roger Pau Monné <roger.pau@citrix.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Reported-by: Andreas Kinzler <hfp@posteo.de> Reviewed-by: Stefano Stabellini <sstabellini@kernel.org> Signed-off-by: Stefano Stabellini <sstabellini@kernel.org> --- hw/xen/xen_pt.h | 1 + hw/xen/xen_pt_config_init.c | 20 ++++++++++++++++++-- hw/xen/xen_pt_msi.c | 13 ++++++++++--- 3 files changed, 29 insertions(+), 5 deletions(-) diff --git a/hw/xen/xen_pt.h b/hw/xen/xen_pt.h index 191d9ca..aa39a9a 100644 --- a/hw/xen/xen_pt.h +++ b/hw/xen/xen_pt.h @@ -180,6 +180,7 @@ typedef struct XenPTMSI { uint32_t addr_hi; /* guest message upper address */ uint16_t data; /* guest message data */ uint32_t ctrl_offset; /* saved control offset */ + uint32_t mask; /* guest mask bits */ int pirq; /* guest pirq corresponding */ bool initialized; /* when guest MSI is initialized */ bool mapped; /* when pirq is mapped */ diff --git a/hw/xen/xen_pt_config_init.c b/hw/xen/xen_pt_config_init.c index 1f04ec5..a3ce33e 100644 --- a/hw/xen/xen_pt_config_init.c +++ b/hw/xen/xen_pt_config_init.c @@ -1315,6 +1315,22 @@ static int xen_pt_msgdata_reg_write(XenPCIPassthroughState *s, return 0; } +static int xen_pt_mask_reg_write(XenPCIPassthroughState *s, XenPTReg *cfg_entry, + uint32_t *val, uint32_t dev_value, + uint32_t valid_mask) +{ + int rc; + + rc = xen_pt_long_reg_write(s, cfg_entry, val, dev_value, valid_mask); + if (rc) { + return rc; + } + + s->msi->mask = *val; + + return 0; +} + /* MSI Capability Structure reg static information table */ static XenPTRegInfo xen_pt_emu_reg_msi[] = { /* Next Pointer reg */ @@ -1393,7 +1409,7 @@ static XenPTRegInfo xen_pt_emu_reg_msi[] = { .emu_mask = 0xFFFFFFFF, .init = xen_pt_mask_reg_init, .u.dw.read = xen_pt_long_reg_read, - .u.dw.write = xen_pt_long_reg_write, + .u.dw.write = xen_pt_mask_reg_write, }, /* Mask reg (if PCI_MSI_FLAGS_MASKBIT set, for 64-bit devices) */ { @@ -1404,7 +1420,7 @@ static XenPTRegInfo xen_pt_emu_reg_msi[] = { .emu_mask = 0xFFFFFFFF, .init = xen_pt_mask_reg_init, .u.dw.read = xen_pt_long_reg_read, - .u.dw.write = xen_pt_long_reg_write, + .u.dw.write = xen_pt_mask_reg_write, }, /* Pending reg (if PCI_MSI_FLAGS_MASKBIT set, for 32-bit devices) */ { diff --git a/hw/xen/xen_pt_msi.c b/hw/xen/xen_pt_msi.c index ff9a79f..6d1e3bd 100644 --- a/hw/xen/xen_pt_msi.c +++ b/hw/xen/xen_pt_msi.c @@ -24,6 +24,7 @@ #define XEN_PT_GFLAGS_SHIFT_DM 9 #define XEN_PT_GFLAGSSHIFT_DELIV_MODE 12 #define XEN_PT_GFLAGSSHIFT_TRG_MODE 15 +#define XEN_PT_GFLAGSSHIFT_UNMASKED 16 #define latch(fld) latch[PCI_MSIX_ENTRY_##fld / sizeof(uint32_t)] @@ -155,7 +156,8 @@ static int msi_msix_update(XenPCIPassthroughState *s, int pirq, bool is_msix, int msix_entry, - int *old_pirq) + int *old_pirq, + bool masked) { PCIDevice *d = &s->dev; uint8_t gvec = msi_vector(data); @@ -171,6 +173,8 @@ static int msi_msix_update(XenPCIPassthroughState *s, table_addr = s->msix->mmio_base_addr; } + gflags |= masked ? 0 : (1u << XEN_PT_GFLAGSSHIFT_UNMASKED); + rc = xc_domain_update_msi_irq(xen_xc, xen_domid, gvec, pirq, gflags, table_addr); @@ -273,8 +277,10 @@ int xen_pt_msi_setup(XenPCIPassthroughState *s) int xen_pt_msi_update(XenPCIPassthroughState *s) { XenPTMSI *msi = s->msi; + + /* Current MSI emulation in QEMU only supports 1 vector */ return msi_msix_update(s, msi_addr64(msi), msi->data, msi->pirq, - false, 0, &msi->pirq); + false, 0, &msi->pirq, msi->mask & 1); } void xen_pt_msi_disable(XenPCIPassthroughState *s) @@ -355,7 +361,8 @@ static int xen_pt_msix_update_one(XenPCIPassthroughState *s, int entry_nr, } rc = msi_msix_update(s, entry->addr, entry->data, pirq, true, - entry_nr, &entry->pirq); + entry_nr, &entry->pirq, + vec_ctrl & PCI_MSIX_ENTRY_CTRL_MASKBIT); if (!rc) { entry->updated = false; -- 1.9.1
WARNING: multiple messages have this Message-ID (diff)
From: Stefano Stabellini <sstabellini@kernel.org> To: peter.maydell@linaro.org, stefanha@gmail.com Cc: sstabellini@kernel.org, qemu-devel@nongnu.org, stefanha@redhat.com, anthony.perard@citrix.com, xen-devel@lists.xenproject.org, Roger Pau Monne <roger.pau@citrix.com> Subject: [PULL 2/2] xen/pt: allow QEMU to request MSI unmasking at bind time Date: Wed, 20 Sep 2017 19:22:24 -0700 [thread overview] Message-ID: <1505960544-12202-2-git-send-email-sstabellini@kernel.org> (raw) In-Reply-To: <1505960544-12202-1-git-send-email-sstabellini@kernel.org> From: Roger Pau Monne <roger.pau@citrix.com> When a MSI interrupt is bound to a guest using xc_domain_update_msi_irq (XEN_DOMCTL_bind_pt_irq) the interrupt is left masked by default. This causes problems with guests that first configure interrupts and clean the per-entry MSIX table mask bit and afterwards enable MSIX globally. In such scenario the Xen internal msixtbl handlers would not detect the unmasking of MSIX entries because vectors are not yet registered since MSIX is not enabled, and vectors would be left masked. Introduce a new flag in the gflags field to signal Xen whether a MSI interrupt should be unmasked after being bound. This also requires to track the mask register for MSI interrupts, so QEMU can also notify to Xen whether the MSI interrupt should be bound masked or unmasked Signed-off-by: Roger Pau Monné <roger.pau@citrix.com> Reviewed-by: Jan Beulich <jbeulich@suse.com> Reported-by: Andreas Kinzler <hfp@posteo.de> Reviewed-by: Stefano Stabellini <sstabellini@kernel.org> Signed-off-by: Stefano Stabellini <sstabellini@kernel.org> --- hw/xen/xen_pt.h | 1 + hw/xen/xen_pt_config_init.c | 20 ++++++++++++++++++-- hw/xen/xen_pt_msi.c | 13 ++++++++++--- 3 files changed, 29 insertions(+), 5 deletions(-) diff --git a/hw/xen/xen_pt.h b/hw/xen/xen_pt.h index 191d9ca..aa39a9a 100644 --- a/hw/xen/xen_pt.h +++ b/hw/xen/xen_pt.h @@ -180,6 +180,7 @@ typedef struct XenPTMSI { uint32_t addr_hi; /* guest message upper address */ uint16_t data; /* guest message data */ uint32_t ctrl_offset; /* saved control offset */ + uint32_t mask; /* guest mask bits */ int pirq; /* guest pirq corresponding */ bool initialized; /* when guest MSI is initialized */ bool mapped; /* when pirq is mapped */ diff --git a/hw/xen/xen_pt_config_init.c b/hw/xen/xen_pt_config_init.c index 1f04ec5..a3ce33e 100644 --- a/hw/xen/xen_pt_config_init.c +++ b/hw/xen/xen_pt_config_init.c @@ -1315,6 +1315,22 @@ static int xen_pt_msgdata_reg_write(XenPCIPassthroughState *s, return 0; } +static int xen_pt_mask_reg_write(XenPCIPassthroughState *s, XenPTReg *cfg_entry, + uint32_t *val, uint32_t dev_value, + uint32_t valid_mask) +{ + int rc; + + rc = xen_pt_long_reg_write(s, cfg_entry, val, dev_value, valid_mask); + if (rc) { + return rc; + } + + s->msi->mask = *val; + + return 0; +} + /* MSI Capability Structure reg static information table */ static XenPTRegInfo xen_pt_emu_reg_msi[] = { /* Next Pointer reg */ @@ -1393,7 +1409,7 @@ static XenPTRegInfo xen_pt_emu_reg_msi[] = { .emu_mask = 0xFFFFFFFF, .init = xen_pt_mask_reg_init, .u.dw.read = xen_pt_long_reg_read, - .u.dw.write = xen_pt_long_reg_write, + .u.dw.write = xen_pt_mask_reg_write, }, /* Mask reg (if PCI_MSI_FLAGS_MASKBIT set, for 64-bit devices) */ { @@ -1404,7 +1420,7 @@ static XenPTRegInfo xen_pt_emu_reg_msi[] = { .emu_mask = 0xFFFFFFFF, .init = xen_pt_mask_reg_init, .u.dw.read = xen_pt_long_reg_read, - .u.dw.write = xen_pt_long_reg_write, + .u.dw.write = xen_pt_mask_reg_write, }, /* Pending reg (if PCI_MSI_FLAGS_MASKBIT set, for 32-bit devices) */ { diff --git a/hw/xen/xen_pt_msi.c b/hw/xen/xen_pt_msi.c index ff9a79f..6d1e3bd 100644 --- a/hw/xen/xen_pt_msi.c +++ b/hw/xen/xen_pt_msi.c @@ -24,6 +24,7 @@ #define XEN_PT_GFLAGS_SHIFT_DM 9 #define XEN_PT_GFLAGSSHIFT_DELIV_MODE 12 #define XEN_PT_GFLAGSSHIFT_TRG_MODE 15 +#define XEN_PT_GFLAGSSHIFT_UNMASKED 16 #define latch(fld) latch[PCI_MSIX_ENTRY_##fld / sizeof(uint32_t)] @@ -155,7 +156,8 @@ static int msi_msix_update(XenPCIPassthroughState *s, int pirq, bool is_msix, int msix_entry, - int *old_pirq) + int *old_pirq, + bool masked) { PCIDevice *d = &s->dev; uint8_t gvec = msi_vector(data); @@ -171,6 +173,8 @@ static int msi_msix_update(XenPCIPassthroughState *s, table_addr = s->msix->mmio_base_addr; } + gflags |= masked ? 0 : (1u << XEN_PT_GFLAGSSHIFT_UNMASKED); + rc = xc_domain_update_msi_irq(xen_xc, xen_domid, gvec, pirq, gflags, table_addr); @@ -273,8 +277,10 @@ int xen_pt_msi_setup(XenPCIPassthroughState *s) int xen_pt_msi_update(XenPCIPassthroughState *s) { XenPTMSI *msi = s->msi; + + /* Current MSI emulation in QEMU only supports 1 vector */ return msi_msix_update(s, msi_addr64(msi), msi->data, msi->pirq, - false, 0, &msi->pirq); + false, 0, &msi->pirq, msi->mask & 1); } void xen_pt_msi_disable(XenPCIPassthroughState *s) @@ -355,7 +361,8 @@ static int xen_pt_msix_update_one(XenPCIPassthroughState *s, int entry_nr, } rc = msi_msix_update(s, entry->addr, entry->data, pirq, true, - entry_nr, &entry->pirq); + entry_nr, &entry->pirq, + vec_ctrl & PCI_MSIX_ENTRY_CTRL_MASKBIT); if (!rc) { entry->updated = false; -- 1.9.1 _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org https://lists.xen.org/xen-devel
next prev parent reply other threads:[~2017-09-21 2:22 UTC|newest] Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top 2017-09-21 2:21 [Qemu-devel] [PULL 0/2] xen-20170920-tag Stefano Stabellini 2017-09-21 2:21 ` Stefano Stabellini 2017-09-21 2:22 ` [Qemu-devel] [PULL 1/2] xen-disk: use g_new0 to fix build Stefano Stabellini 2017-09-21 2:22 ` Stefano Stabellini 2017-09-21 2:22 ` Stefano Stabellini [this message] 2017-09-21 2:22 ` [PULL 2/2] xen/pt: allow QEMU to request MSI unmasking at bind time Stefano Stabellini 2017-09-25 12:11 ` [Qemu-devel] [PULL 1/2] xen-disk: use g_new0 to fix build Olaf Hering 2017-09-25 12:11 ` Olaf Hering 2017-09-21 9:55 ` [Qemu-devel] [PULL 0/2] xen-20170920-tag Peter Maydell 2017-09-21 9:55 ` Peter Maydell
Reply instructions: You may reply publicly to this message via plain-text email using any one of the following methods: * Save the following mbox file, import it into your mail client, and reply-to-all from there: mbox Avoid top-posting and favor interleaved quoting: https://en.wikipedia.org/wiki/Posting_style#Interleaved_style * Reply using the --to, --cc, and --in-reply-to switches of git-send-email(1): git send-email \ --in-reply-to=1505960544-12202-2-git-send-email-sstabellini@kernel.org \ --to=sstabellini@kernel.org \ --cc=anthony.perard@citrix.com \ --cc=peter.maydell@linaro.org \ --cc=qemu-devel@nongnu.org \ --cc=roger.pau@citrix.com \ --cc=stefanha@gmail.com \ --cc=stefanha@redhat.com \ --cc=xen-devel@lists.xenproject.org \ /path/to/YOUR_REPLY https://kernel.org/pub/software/scm/git/docs/git-send-email.html * If your mail client supports setting the In-Reply-To header via mailto: links, try the mailto: linkBe sure your reply has a Subject: header at the top and a blank line before the message body.
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.