vmnet: work around skb_frag_t build failure for SLE15-SP2 and openSUSE-15.2

The kernels used in (upcoming) SLE15 SP2 and openSUSE Leap 15.2 are based on upstream version 5.3 but have a backport of the skb_frag_t conversion to bio_vec which reached mainline in v5.4-rc1. As we cannot easily check availability compat_skb_frag_off() (which was also added in v5.4-rc1) and have to resort to version check, add an exception for SLE15-SP2 and openSUSE-15.2 kernels.
vmnet: handle switch of skb_frag_t to bio_vec
2019-11-25 08:24:56 +01:00 · 2019-09-21 01:22:38 +02:00 · 2019-09-21 01:18:29 +02:00 · 2019-09-21 01:18:29 +02:00 · 2019-09-21 01:18:29 +02:00 · 2019-09-21 01:18:29 +02:00
5 changed files with 20 additions and 25 deletions
@@ -1,7 +0,0 @@
-*-----------------------------------------------------------------------------
-* Branch workstation-15.5.1 has been retired on 2020-08-20 and has not been
-* receiving any updates since. Please switch to latest 15.5.x version
-* available.
-*
-* Latest kernel version this branch is intended to work with is 5.7.
-*-----------------------------------------------------------------------------
@@ -120,7 +120,6 @@ MSRQuery;
 #define MSR_ARCH_CAPABILITIES_RSBA                (1ULL << 2)
 #define MSR_ARCH_CAPABILITIES_NOL1F_VMENTRY       (1ULL << 3)
 #define MSR_ARCH_CAPABILITIES_SSB_NO              (1ULL << 4)
-#define MSR_ARCH_CAPABILITIES_MDS_NO              (1ULL << 5)

 #define MSR_FLUSH_CMD                        0x10b
 #define MSR_FLUSH_CMD_FLUSH_L1D                   (1ULL << 0)
@@ -131,7 +130,6 @@ MSRQuery;

 #define MSR_PRED_CMD_IBPB                         (1UL << 0)

-
 #ifndef MSR_MISC_FEATURES_ENABLES
 #define MSR_MISC_FEATURES_ENABLES            0x140
 #endif
@@ -2324,7 +2324,7 @@ SetVMAPICAddr(VMDriver *vm, // IN/OUT: driver state
   volatile void *hostapic;

   ASSERT_ON_COMPILE(APICR_SIZE <= PAGE_SIZE);
-   hostapic = (volatile void *) ioremap(ma, PAGE_SIZE);
+   hostapic = (volatile void *) ioremap_nocache(ma, PAGE_SIZE);
   if (hostapic) {
      if ((APIC_VERSIONREG(hostapic) & 0xF0) == 0x10) {
         vm->hostAPIC.base = (volatile uint32 (*)[4]) hostapic;
@@ -66,7 +66,19 @@
 #endif

 #if LOGLEVEL >= 4
-static u64 vnetTime;
+static struct timeval vnetTime;
+
+#if LINUX_VERSION_CODE >= KERNEL_VERSION(5, 0, 0)
+static void do_gettimeofday(struct timeval *tv)
+{
+	struct timespec64 now;
+
+	ktime_get_real_ts64(&now);
+	tv->tv_sec = now.tv_sec;
+	tv->tv_usec = now.tv_nsec / 1000;
+}
+#endif
+
 #endif

 typedef struct VNetBridge VNetBridge;
@@ -693,7 +705,7 @@ VNetBridgeReceiveFromVNet(VNetJack        *this, // IN: jack

 	 netif_rx_ni(clone);
 #	 if LOGLEVEL >= 4
-	 vnetTime = ktime_get_ns();
+	 do_gettimeofday(&vnetTime);
 #	 endif
      }
   }
@@ -1490,11 +1502,12 @@ VNetBridgeReceiveFromDev(struct sk_buff *skb,         // IN: packet to receive

 #  if LOGLEVEL >= 4
   {
-      u64 now;
-
-      now = ktime_get_ns();
+      struct timeval now;
+      do_gettimeofday(&now);
      LOG(3, (KERN_DEBUG "bridge-%s: time %d\n",
-	      bridge->name, (int)((now - vnetTime) / NSEC_PER_USEC)));
+	      bridge->name,
+	      (int)((now.tv_sec * 1000000 + now.tv_usec)
+                    - (vnetTime.tv_sec * 1000000 + vnetTime.tv_usec))));
   }
 #  endif

@@ -161,14 +161,6 @@ VNetProcOpen(struct inode *inode,   // IN:
 }

 /* Our procfs callbacks.  We only need to specialize open. */
-#ifdef DEFINE_PROC_SHOW_ATTRIBUTE
-static const struct proc_ops fops = {
-   .proc_open    = VNetProcOpen,
-   .proc_read    = seq_read,
-   .proc_lseek   = seq_lseek,
-   .proc_release = single_release,
-};
-#else
 static struct file_operations fops = {
   .open    = VNetProcOpen,
   .read    = seq_read,
@@ -176,7 +168,6 @@ static struct file_operations fops = {
   .release = single_release,
 };
 #endif
-#endif


 /*
Author	SHA1	Message	Date
Michal Kubecek	53696105a4	vmnet: work around skb_frag_t build failure for SLE15-SP2 and openSUSE-15.2 The kernels used in (upcoming) SLE15 SP2 and openSUSE Leap 15.2 are based on upstream version 5.3 but have a backport of the skb_frag_t conversion to bio_vec which reached mainline in v5.4-rc1. As we cannot easily check availability compat_skb_frag_off() (which was also added in v5.4-rc1) and have to resort to version check, add an exception for SLE15-SP2 and openSUSE-15.2 kernels.	2019-11-25 08:24:56 +01:00
Michal Kubecek	f404bc6855	vmnet: handle switch of skb_frag_t to bio_vec The switch from custom skb_frag_t implementation to bio_vec in v5.4-rc1 is mostly transparent for modules which use accessor for skb_frag_t members. Unfortunately many users access the members directly and function VNetCsumCopyDatagram() in vmnet is one of those. Use accessors everywhere so that vmnet code is compatible with kernel 5.4 and newer. Use "compat_" prefix to avoid clashes with backports adding the accessors to older codebase.	2019-09-21 01:22:38 +02:00
Michal Kubecek	413c125295	vmmon: use standard definition of MSR_TSX_FORCE_ABORT if available The MSR_TSX_FORCE_ABORT macro is defined in mainline since commit 52f64909409c ("x86: Add TSX Force Abort CPUID/MSR") in v5.1-rc1.	2019-09-21 01:18:29 +02:00
Michal Kubecek	d257b1293e	vmnet: bring back the do_gettimeofday() helper The do_gettimeofday() helper was removed by commit e4b92b108c6c ("timekeeping: remove obsolete time accessors") in v5.0-rc1. Bring it back for users in vmnet-only/bridge.c Note that Workstation/Player 15.1.0 did in fact resolve the absence of do_gettimeofday() in kernel 5.0 and newer but it did so only for calls in vmmon-only/linux/hostif.c. Calls in vmnet-only/bridge.c were left in place as those are only compiled when building with LOGLEVEL >= 4.	2019-09-21 01:18:29 +02:00
Michal Kubecek	8deac6bff4	vmmon: check presence of file_operations::poll() Commit 11c5ad0ec441 ("eventpoll: switch to ->poll_mask") in v4.18-rc1, made eventpoll switch from ->poll() to ->poll_mask(), resulting in a null pointer dereference as vmmon calls the callback directly without checking its availability. Rather than calling the callback directly (which would result in null pointer dereference), use vfs_poll() wrapper. As this wrapper is only available since 4.18-rc1 cycle as well, provide a copy to use when building against older kernels. Note: even if the commit revealing the problem was reverted in 4.18-rc3 (and is unlikely to go back in the same form) so that vmmon no longer crashes with 4.18 kernel, calling the callback without a check is a bad practice so let's keep the patch in place.	2019-09-21 01:18:29 +02:00
Michal Kubecek	7c7d92ff52	vmmon: fix indirect call with retpoline build Build against kernel with retpoline support issues warning objtool: Task_Switch()+0x425: indirect call found in RETPOLINE build This is because an indirect call in TaskSwitchToMonitor() is encoded using inline assembler so that it bypasses retpoline generation. For this purpose, macro CALL_NOSPEC exists since v4.15-rc8 (and has been backported into some distribution kernels with the rest of retpoline support). Use the macro if available and fallback to the original code if not.	2019-09-21 01:18:29 +02:00
Michal Kubecek	520a0d5e7b	vmmon: fix always_inline attribute usage Function declared with __attribute__((always_inline)) should also be declared as inline, otherwise gcc issues a warning "always_inline function might not be inlinable". It's just cosmetic but getting rid of known harmless warnings makes it easier to spot actual problems. Use the __always_inline macro for LinuxDriverSyncReadTSCs() as this is how always inline functions should be declared in kernel code.	2019-09-21 01:18:29 +02:00
Michal Kubecek	ac7f26b5d6	vmmon: use standard definition of MSR_K7_HWCR_SMMLOCK if available The MSR_K7_HWCR_SMMLOCK macro is defined in mainline since commit 18c71ce9c882 ("x86/CPU/AMD: Add the Secure Encrypted Virtualization CPU feature") in v4.16-rc1.	2019-09-21 01:18:29 +02:00
Theo Bertacchini	135c925e0f	modules: remove .cache.mk on make clean Remove .cache.mk inside vmmon-only and vmnet-only when executing "make clean". This file can cause issues when upgrading gcc as make will still look for includes inside older gcc includes directory. File .cache.mk with cache of generated variables was created by build since kernel v4.15-rc1, commit 3298b690b21c ("kbuild: Add a cache for generated variables") until the feature was removed in v4.18-rc1, commit e08d6de4e532 ("kbuild: remove kbuild cache").	2019-09-21 01:18:29 +02:00
Michal Kubecek	ab842681ee	vmmon: quick workaround for objtool warnings As discussed in https://bugzilla.suse.com/show_bug.cgi?id=1059674 the reason for multiple objtool warnings is the fact that vmmon module defines its own Panic() function which never returns. While it is marked as such which is used by the compiler for optimization, there is no way to find this from object file. While this seems innocuous, it might result in problems with unwinder later. The quickest way around is to replace vmmon's own Panic() with standard kernel panic() until a cleaner solution is found.	2019-09-21 01:18:29 +02:00
Michal Kubecek	a3f5b7bf8b	vmmon: use standard definition of CR3_PCID_MASK if available The CR3_PCID_MASK is defined in mainline asm/processor-flags.h since commit 6c690ee1039b ("x86/mm: Split read_cr3() into read_cr3_pa() and __read_cr3()") in v4.13-rc1.	2019-09-21 01:18:29 +02:00
Michal Kubecek	a3df62b5be	vmmon: use standard definition of MSR_MISC_FEATURES_ENABLES if available The MSR_MISC_FEATURES_ENABLES macro is defined in mainline since commit ab6d9468631a ("x86/msr: Rename MISC_FEATURE_ENABLES to MISC_FEATURES_ENABLES") in v4.12-rc1. The MSR_MISC_FEATURES_ENABLES_CPUID_FAULT macro is defined in mainline since commit e9ea1e7f53b8 ("x86/arch_prctl: Add ARCH_[GET\|SET]_CPUID") in v4.12-rc1.	2019-09-21 01:18:29 +02:00
Michal Kubecek	f1ff988206	vmnet: use standard definition of PCI_VENDOR_ID_VMWARE_VMXNET3 if available The PCI_DEVICE_ID_VMWARE_VMXNET3 macro is defined in mainline pci_ids.h since commit b1226c7db1d9 ("vmxnet3: Move PCI Id to pci_ids.h") in v4.10-rc1.	2019-09-21 01:18:29 +02:00
Michal Kubecek	e0db1933cc	vmnet: use standard definition of PCI_VENDOR_ID_VMWARE if available The PCI_VENDOR_ID_VMWARE macro is defined in mainline pci_ids.h since commit 94e57fea6202 ("PCI: Move PCI_VENDOR_ID_VMWARE to pci_ids.h") in v3.18-rc1.	2019-09-21 01:18:29 +02:00
Michal Kubecek	533f4e865b	modules: do not override CC command line options Workstation/Player 15.5.0 tries to guess the right retpoline (Spectre) related compiler command line options, resulting in objtool warnings like warning: objtool: .text.__x86_indirect_thunk_rbp+0x0: unsupported intra-function call warning: objtool: If this is a retpoline, please patch it in with alternatives and annotate e.g. in vmmon-only/common/memtrack.o (but also many others). This seems wrong; out-of-tree modules should be built with the same compiler options that were used to build the kernel they are built against. Distribution kernel devel packages usually make sure this is the way so let them do their job.	2019-09-21 01:18:29 +02:00
Michal Kubecek	277ec3a5e7	vmmon: fix wrong VMX86_DEVEL test If VMX86_DEVEL macro is not defined, preprocessor warns when it is tested with "#if" in vmmon-only/linux/driver.c. As VMX86_DEVEL is tested with "#ifdef" everywhere else, change this one test to match the rest.	2019-09-21 00:51:50 +02:00